Geologia Depositos

Andrew D.
Miall
The Geology of Fluvial Deposits
Sedimentary Facies, Basin Analysis, and Petroleum Geology
4th corrected printing
Springer-Verlag Berlin Heidelberg GmbH
Andrew D. Miall
The Geology
of Fluvial Deposits
Sedimentary Facies, Basin Analysis,
and Petroleum Geology
4thcorrected printing
With 504 Figures and 30 Tables
Springer
Professor Or. Andrew O. Miall
University of Toronto
Oepartment of Geology
22 Russell Street
5S 3B1 Toronto
Canada
ISBN 978-3-642-08211-5 ISBN 978-3-662-03237-4 (eBook)

DOI 10.1007/978-3-662-03237-4
41h printing 2006
This work is subject to copyright. AII rights are reserved, whether the whole or part of the material is
concerned, specifically the nghts of translation, reprinting, reuse of illustrations, recitation, broadcasting,
reproduction on microfilm or in any other way, and storage in data banks. Duplication of this j)ublication or
parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its
current version, and permisslon for use must always be obtained from Springer-Verlag Berlin Heidelberg GmbH.
Violations are liable to prosecution under the German Copyright Law.
springer.com
© Springer-Verlag Berlin Heidelberg 2006
Originally published by Springer-Verlag Berlin Heidelberg New York in 2006
Softcover reprint of the hardcover 1st edition 2006
The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws
and regulations and therefore free for general use.
Cover design: KOnkel+Lopka, IIvesheim
Typesetting: Thomson Press (India), Ud., New Delhi
Produdion: Almas Schimmel
SPIN: 11751502 Printed on acid-free paper 30/3141/as 543

For Charlene,
Christopher,
and Sarah
Preface
Fluvial deposits represent the preserved record of one of the major nonmarine environ-
ments. They accumulate in large and small intermontane valleys, in the broad valleys of
trunk rivers, in the wedges of alluvial fans flanking areas of uplift, in the outwash plains
fronting melting glaciers, and in coastal plains. The nature of alluvial assemblages - their
lithofacies composition, vertical stratigraphic record, and architecture - reflect an inter-
play of many processes, from the wandering of individual channels across a floodplain, to
the long-term effects of uplift and subsidence. Fluvial deposits are a sensitive indicator of
tectonic processes, and also carry subtle signatures of the climate at the time of deposition.
They are the hosts for many petroleum and mineral deposits. This book is about all these
subjects.
The first part of the book, following a historical introduction, constructs the strati-
graphic framework of fluvial deposits, step by step, starting with lithofacies, combining
these into architectural elements and other facies associations, and then showing how
these, in turn, combine to represent distinctive fluvial styles. Next, the discussion turns to
problems of correlation and the building of large-scale stratigraphic frameworks. These
basin-scale constructions form the basis for a discussion of causes and processes, including
autogenic processes of channel shifting and cyclicity, and the larger questions of allogenic
(tectonic, eustatic, and climatic) sedimentary controls and the development of our ideas
about nonmarine sequence stratigraphy.
The final chapters address issues of concern to petroleum geologists. The geometry of
reservoirs is discussed, and primary reservoir heterogeneities are evaluated from the point
of view of facies architecture and porosity-permeability characteristics. Lastly, the strati-
graphic and tectonic distribution of petroleum reservoirs is analyzed, leading to a strati-
graphic-tectonic classification of oil and gas fields in fluvial deposits, and a description of
selected case examples.
The book is intended for advanced students, researchers, and professionals. Extensive
references have been made to the published literature, and hundreds of examples and
diagrams have been provided. It is to be hoped that the methods and classifications
proposed here will assist in future research work and resource production.
Acknowledgments
My earliest work on fluvial sedimentology was carried out as part of my doctoral thesis
research in the Canadian Arctic Islands, under the stimulating leadership of the late Brian
Rust. I was able to continue research in this areas during the 1970s, while undertaking
regional basin studies for the Geological Survey of Canada. I am grateful to the enlightened
management of the Survey at the time, for encouraging this "theoretical" interest, particu-
larly Don Stott, the Director of the Institute of Sedimentary and Petroleum Geology at
Calgary for most of my time there.
Since my move to the University of Toronto in 1979, my research has been supported by
the Natural Sciences and Engineering Research Council of Canada. I have also, at times,
received support from the Petroleum Research Fund of the American Chemical Society,
and from various petroleum companies, including British Petroleum (USA), Imperial Oil,
Union Pacific, Petro-Canada, Petrel-Robertson and Wascana. Some of the compilation of
case studies and mapping methods and, in particular, the sections dealing with petroleum
geology, formed part of a report prepared for the Japan National Oil Corporation in 1991,
and I am very grateful to the Corporation for permission to include this material in the
book.
Most of the chapters in the first half of the book were drafted at University of Oxford
while I was on saba tical leave there in 1991-1992. I am grateful to John Dewey, Chair of the
Department, and to Phil Allen, my host, for making my stay there a memorable one. Later
work on the book was completed at my home base, and I appreciate the flexibility I was
afforded by my Chair, John Westgate, in rearranging my teaching timetable to provide
blocks of time for work on the final chapters and for the production stage of the book.
Many of my students at the University of Toronto have become stimulating and
challenging colleagues, and I am grateful to them for the mutual learning experience. In
particular, Phil Fralick, David Eberth, Greg Nadon, Jun Cowan, Mike Bromley, Paul Godin,
Mark Stephens, Shuji Yoshida, Andrew Willis, and Octavian Catuneanu, have given me
much to think about. Most of them are quoted in this book.
I am grateful to various colleagues for reading parts of the book (including some
material that was original prepared for review articles and subsequently incorporated into
the book instead). Their input has been most valuable. They include Kevin Biddle, Lars
Clemmensen, Bob Dalrymple, Bill Galloway, Martin Gibling, John Horne, John Hubert,
Gary Kocurek, Mike Leeder, Jim Lowell, Peter McCabe, Judith Totman Parrish, Norm
Smith, Finn Surlyk, Jim Tucker, and Paul Wright. Mike Leeder drew my attention to the
early understanding of the power of rivers contained in works by Plato and Homer.
Photographs have been contributed by Kevin Crowley, Miles Hayes, Lloyd Homer (and
the Institute of Geological and Nuclear Sciences Ltd., Lower Hutt, New Zealand. My
thanks to Dale Leckie for the introduction), Tomasz Jerzykiewicz, Ray Levey, Grant
Mossop, Sarah Prosser, Derald Smith, and Norm Smith. The field photographs from the
Canadian Arctic were taken by myself, nearly all of them while I was employed by the
Geological Survey of Canada, Calgary, and the Survey is hereby gratefully acknowledged as
the source of these illustrations. Jim Dixon assisted with the preparation of some of these.
All other photographs are from my own research collection. Chris Fielding provided good
copies of some of his diagrams. Some original or redrawn line drawings were done by
x Acknowledgments
Subash Shanbhag. Finally, I must again thank my wife Charlene, and my children Christo-
pher and Sarah, for all their support, and for putting up with yet another book. This fluvial
enterprise owes much to Charlene's field assistance, companionship, professional advice,
support, and love over the years.
Contents
1 Introduction .................................................. . 1
1.1 Scope and Purpose of Book ...................................... . 1
1.2 Data Sources .................................................. . 2
2 Historical Background ......................................... . 5
2.1 Introduction .................................................. . 5

2.2 Early Developments in the Study of Fluvial Sediments .............. . 5
2.2.1 From the Ancient Greeks to Playfair .............................. . 5
2.2.2 From Lyell to Davis ............................................ . 8
2.3 Growth of Present-Day Concepts, up to 1977 ....................... . 13
2.3.1 Increasing Specialization of the Twentieth Century ................. . 13
2.3.2 Descriptive Fluvial Geomorphology .............................. . 13
2.3.3 Quantitative Fluvial Geomorphology ............................. . 16
2.3.4 Sediment Transport and Textural Studies ......................... . 17
2.3.5 Bedforms and Paleocurrents ..................................... . 19
2.3.6 Fluvial Facies Models ........................................... . 22
2.3.6.1 From Hobbs to Fisk ............................................ . 22
2.3.6.2 Meandering River Deposits:
Development of Modern Facies Model Concepts .................... 26
2.3.6.3 Braided Rivers ............ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3.6.4 Alluvial Fans ................................................... 31
2.3.6.5 Other Facies Models ............................................. 32
2.3.7 Fluvial Architecture ............................................ . 34
2.3.8 Paleohydraulics ................................................ . 35
2.4 Growth of Present-Day Concepts, 1978-1988 ....................... . 37
2.4.1 Bedforms and Sedimentary Structures ............................ . 37
2.4.2 The Decline and Fall of the Vertical Profile ........................ . 38
2.4.3 Fluvial Architecture ............................................ . 42
2.4.3.1 Architectural Scale and the Bounding-Surface Concept .............. . 42
2.4.3.2 Alluvial Basin Architecture ...................................... . 45
2.4.4 Fluvial Styles .................................................. . 47
2.4.4.1 High-Sinuosity Rivers .......................................... . 49
2.4.4.2 Low-Sinuosity Rivers ........................................... . 50
2.4.4.3 Anastomosed Rivers ............................................ . 52
2.4.4.4 Ephemeral Rivers .............................................. . 54
2.4.4.5 Large Rivers ................................................... . 54
2.4.4.6 Floodplain Environments ....................................... . 54
2.5 Conclusions ................................................... . 55
XII Contents
3 Concepts of Scale 57
3.1 Time Scales and Physical Scales in Sedimentation ................... 57
3.2 The Grouping of Architectural Units in Clastic Rocks
According to Depositional Time Scale ............................. 60
3.3 Definition of Sediment Groups by Bounding Surfaces ................ 69
3.4 Sedimentation Rate and Its Relation to Depositional Recurrence Interval 71
3.5 Application of Scale Concepts to Basin Analysis and Petroleum Geology 74
4 Methods of Architectural-Element Analysis ........................ 75

4.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.2 Construction of Outcrop Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.3 Classification of Lithofacies ...................................... 77
4.4 Principles of Paleocurrent Analysis ................................ 78
4.5 Classification of Bounding Surfaces ............................... 81
4.6 Classification of Architectural Elements ............... . . . . . . . . . . . . . 89
4.7 Classification of Channels and Larger Bodies ....................... 94
4.8 Annotation of Outcrop Profiles ................................... 95
4.9 Summary of Methods ............ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
5 Lithofacies ..................................................... 99
5.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.2 Gravel Facies ................................................... 99
5.2.1 Depositional Processes in Gravel-Bed Rivers. . . . . . . . . . . . . . . . . . . . . . . . 99
5.2.1.1 Introduction ................................................... 99
5.2.1.2 Traction Currents, Fluid Flows ................................... . 99
5.2.1.3 Sediment Gravity Flows ......................................... . 105
5.2.2 Gravel Lithofacies .............................................. . 106
5.3 Sand Facies ................................................... . 109
5.3.1 Sand Bedform Genesis and Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.3.2 Sand Lithofacies ................................................ 112
5.4 Fine-Grained Clastic Facies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.5 Nonelastic Facies ............................................... 127
5.6 Associated Facies ............................................... 130
6 Architectural Elements Formed Within Channels ................... 131

6.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
6.2 Channels (Element CH) .......................................... 131
6.3 Gravel Bars and Bedforms (Element GB) ........................... 139
6.4 Sediment-Gravity-Flow Deposits (Element SG) ...................... 145
6.5 Sandy Bedforms (Element SB) .................................... 146
6.6 Downstream-Accretion Macroforms (Element DA) .................. 151
6.7 Lateral-Accretion Deposits (Element LA) ....... . . . . . . . . . . . . . . . . . . . . 155
6.8 Laminated Sand Sheets (Element LS) .............................. 163
6.9 Hollows (Element HO) .................. .. . . . . . . .. . . . . . . . .. . . . . . . 163
7 Architectural Elements of the Overbank Environment ............... 169

7.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
7.2 Levee and Crevasse Deposits.................. ........... .... ..... 172
7.2.1 Levee Deposits (Element LV) ..................................... 172
7.2.2 Crevasse-Channel Deposits (Element CR) .......................... 173
Contents XIII
7.2.3 Crevasse-Splay Deposits (Element CS) ............................ . 174

7.3 Fine-Grained Clastic Deposits ................................... . 176
7.3.1 Floodplain Fines (Element FF) ................................... . 177
7.3.2 Abandoned Channel Fills (Element FF(CH)) ....................... . 177
7.4 Biochemical Sediments ......................................... . 178
7.4.1 Coal .......................................................... . 179
7.4.2 Paleosols ... , .................................................. . 183
7.4.3 Evaporites .................................................... . 190
8 Fluvial Styles and Facies Models ................................. . 191

8.1 Controls on Channel Style ....................................... . 191
8.2 Facies Models ................................................. . 198
8.2.1 Gravel-Bed Braided River with Sediment-Gravity-Flow Deposits ..... . 206
8.2.2 Shallow, Gravel-Bed Braided River ............................... . 208
8.2.3 Deep, Gravel-Bed Braided River .................................. . 209
8.2.4 Gravel-Bed, Wandering River .................................... . 211
8.2.5 Gravel-Bed, Meandering River ................................... . 212
8.2.6 Gravel-Sand Meandering River .................................. . 215
8.2.7 Sand-Bed Meandering River ..................................... . 217
8.2.8 Ephemeral, Sand-Bed Meandering River .......................... . 219
8.2.9 Fine-Grained Meandering River .................................. . 222
8.2.10 Anastomosed River ............................................ . 229
8.2.11 Low-Sinuosity River, with Alternate Bars .......................... . 233
8.2.12 Shallow, Perennial, Sand-Bed Braided River ....................... . 234
8.2.13 Deep, Perennial, Sand-Bed Braided River .......................... . 235
8.2.14 High-Energy, Sand-Bed Braided River ............................ . 238
8.2.15 Distal, Sheetflood, Sand-Bed River ............................... . 240
8.2.16 Flashy, Ephemeral, Sheetflood, Sand-Bed River .................... . 243
8.3 Alluvial Fans and Other Fluvial Distributary Systems ............... . 245
9 The Stratigraphic Architecture of Fluvial Depositional Systems ...... . 251

9.1 Introduction .................................................. . 251
9.2 Channel Belts .................................................. . 251
9.3 Depositional Systems .............•.............................. 261
9.4 Basin-fill Complexes .......................•..................... 265
9.5 Methods of Correlation and Mapping ............................. . 272
9.5.1 The Use of Marker Beds ......................................... . 272
9.5.2 Wireline Logs ................................................. . 273
9.5.3 Lithofacies Mapping ............................................ . 276
9.5.4 Seismic Methods ............................................... . 283
9.5.5 Ground-Penetrating Radar ...................................... . 292
9.5.6 Magnetostratigraphy ........................................... . 293
9.5.7 Paleocurrent Analysis .......................................... . 293
9.5.8 The Dipmeter ................................................. . 298
9.5.9 Surveillance Geology ........................................... . 302
9.6 Stratigraphic Nomenclature ..................................... . 306
10 Fluvial Depositional Systems and Autogenic Sedimentary Controls .. . 311

10.1 Introduction .................................................. . 311
10.2 The Evolution of Distributary Fluvial Systems ..................... . 311
10.3 Avulsion in Fluvial Systems and Its Effect on Alluvial Stratigraphy .... . 317
XIV Contents
10.3.1 The Development of Meander Belts ............................... . 318

10.3.2 Avulsion in Braided Fluvial Systems .............................. . 322
10.3.3 Avulsion in Anastomosed Fluvial Systems ......................... . 325
10.4 Quantitative Studies of Alluvial Architecture ...................... . 327
10.4.1 The Dimensions of Fluvial Sand Bodies ........................... . 328
10.4.2 Estimating Probabilities of Sand Body Penetration
and Interconnectedness in the Subsurface ......................... . 334
10.4.3 Alluvial Stratigraphy Models .................................... . 337
11 Tectonic Control of Fluvial Sedimentation ........................ . 343

11.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
11.2 Tectonic Control of Alluvial Stratigraphy. . . . . . . . . . . . . . . . . . . . . . . . . . . 343
11.2.1 The Effects of Syndepositional Fault and Fold Movements ............ 344
11.2.1.1 The Effects of Basin-Margin Faulting.............................. 344
11.2.1.2 The Effects of Faulting and Folding Within Basins ................... 348
11.2.2 Base-Level Changes ............................................. 352
11.3 Tectonic Control of Basin Style and Basin-Scale Fluvial Patterns. . . . . . . 362
11.3.1 Big Rivers ...................................................... 362
11.3.2 Axial and Transverse Drainage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
11.3.3 Regional Tectonic Control Revealed by Basin Analysis ............... 365
11.3.4 Tectonism and Sediment Supply .................................. 367
11.3.5 Intraplate Stress ................................................ 370
11.3.6 Quantitative Models of Sediment Supply, Transfer,
and Accumulation .............................................. 372
11.4 Plate-Tectonic Setting of Alluvial Basins ........................... 376
11.4.1 Basin Classification ............................................. 376
11.4.2 Extensional Basins .............................................. 377
11.4.2.1 Rift Basins ..................................................... 379
11.4.2.2 Continental-margin Basins...... .............................. ... 381
11.4.2.3 Failed Rifts and Aulacogens ...................................... 382
11.4.3 Convergent-margin Basins ....................................... 386
11.4.3.1 Forearc Basins .................................................. 386
11.4.3.2 Backarc Basins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
11.4.3.3 Retroarc (Foreland) Basins ....................................... 387
11.4.4 Basins Formed Along Strike-Slip Faults ............................ 393
11.4.4.1 Basins Associated with Intracontinental Transform Faults ............ 393
11.4.4.2 Basins Associated with Divergent Plate Boundaries .................. 394
11.4.5 Basins Related to Plate Collision .................................. 395
11.4.5.1 Peripheral Foreland Basins ....................................... 397
11.4.5.2 Hinterland Basins ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
11.4.6 Structural and Stratigraphic Patterns Common to Foreland Basins . . . . . 402
11.4.7 Sedimentary Basins and Allochthonous Terranes.................... 409
11.4.8 Cratonic Basins.................. ........... .................... 412
11.5 Basic Paleogeographic Models for Nonmarine Basins ................ 412
12 What Does Fluvial Lithofacies Reveal About Climate? .............. . 421

12.1 Introduction .................................................. . 421
12.2 Climatic Variables ............................................. . 422
12.3 Distinguishing Tectonic from Climatic Control .................... . 423
12.4 Review of Climatic Criteria ...................................... . 425
12.5 Conglomerates: The Significance of Texture and Petrology .......... . 425
Contents XV
12.5.1 Mass-flow Versus Traction-current Processes ....................... 425

12.5.1.1Arid Climates .................................................. 428
12.5.1.2Temperate-Humid Climates ...................................... 430
12.5.1.3Tropical-Humid Climates ........................................ 430
12.5.1.4Boreal-Paraglacial Climates ...................................... 431
12.5.1.5Summary and Conclusions ....................................... 431
12.5.2 The Influence of Climate on Texture and Composition of Gravels ..... 432
12.6 Sandstones: The Significance of Sand Body Architecture
and Sedimentary Structures ...................................... 433
12.6.1 Fluvial Style in Sand-bed Rivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
12.6.2 Sand Body Architecture ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434
12.6.3 Bedforms and Cycles ............................................ 436
12.7 Overbank Fines: The Significance of Bedding
and Minor Sedimentary Structures ................................ 437
12.8 The Significance of Color ........................................ 439
12.9 Associated Clastic, Chemical, and Biochemical Sediments ............ 439
12.9.1 Coal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
12.9.2 Paleosols ...................................................... . 440
12.9.3 Evaporites .................................................... . 441
12.9.4 Eolian Interbeds ............................................... . 441
12.9.5 Palustrine Limestones .......................................... . 442
12.10 Contrasting Climatic Indicators .................................. . 442
12.11 The Interrelationship Between Tectonics and Climate ............... . 443
12.12 Orbital Forcing ................................................ . 444
12.12.1 Sedimentary Evidence of Orbital Forcing .......................... . 444
12.12.2 Fluvial Response to the Late Cenozoic Glaciations .................. . 447
12.12.3 Conclusions ................................................... . 450
12.13 Discussion .................................................... . 450
13 Sequence Stratigraphy .......................................... . 453

13.1 Introduction .................................................. . 453
13.2 Accommodation Space ......................................... . 456
13.3 Main Components of the Fluvial Sequence Model .................. . 459
13.3.1 Sequence Boundary ............................................ . 459
13.3.2 Lowstand Systems Tract ........................................ . 466
13.3.3 Transgressive Systems Tract .........................•............ 471
13.3.4 Equivalent of Maximum Flooding Surface ......................... . 474
13.3.5 Highstand Systems Tract ........................................ . 474
13.3.6 Falling-stage Systems Tract ...................................... . 475
13.4 Time Scales of Nonmarine Sequences and Their Causes ............. . 475
13.4.1 First-order Cycles .............................................. . 476
13.4.2 Second-order Cycles ............................................ . 476
13.4.3 Third-order Cycles ............................................. . 477
13.4.4 Fourth- and Fifth-order Cycles ................................... . 477
13.5 Discussion .................................................... . 477
14 Stratigraphic and Tectonic Controls on the Distribution

and Architecture of Fluvial Oil and Gas Reservoirs ................. . 479
14.1 Introduction .................................................. . 479
14.2 The Geometry of Fluvial Reservoirs ............................... . 480
14.2.1 Geometry and Origin of Depositional Systems ..................... . 480
XVI Contents
14.2.1.1 Clastic Wedges ................................................. 480

14.2.1.2 Paleovalley Fills. . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482
14.2.2 Geometry of Reservoir Bodies .................................... 484
14.2.2.1 Sheet Sandstones ............................................... 484
14.2.2.2 Sandstone Ribbons and Lenses ................................... 484
14.2.2.3 Stratigraphic Variations in Reservoir Geometry ..................... 485
14.3 Tectonic Setting of Fluvial Reservoirs. . . . . . . . . . . . . • . . . . . . . . . . . . . . . . 486
14.3.1 Retroarc (Backarc) Foreland Basins ............................... 486
14.3.2 Backarc Basins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486
14.3.3 Forearc Basins ................................................. . 486
14.3.4 Collision-Related Basins ........................................ . 486
14.3.5 Basins in Continental-Transform Settings .............. ; .......... . 487
14.3.6 Rift Basins .................................................... . 487
14.3.7 Basins on Extensional Continental Margins ........................ . 487
14.3.8 Intracratonic Basins ............................................ . 487
14.4 Styles of Fluvial Reservoir ....................................... . 489
14.4.1 Paleovalley Bodies (PV Type) .................................... . 490
14.4.2 Sheet Bodies (SH Type) ......................................... . 491
14.4.3 Channel-and-Bar Bodies (CB Type) ............................... . 492
14.5 Conclusions ................................................... . 494
15 Case Studies of Oil and Gas Fields in Fluvial Reservoirs ............. . 495
15.1 Introduction .................................................. . 495
15.2 Paleovalley Fields (PV Type) ..................................... . 495
15.2.1 Little Bow Area, Alberta ......................................... . 495
15.2.2 Cut Bank Sandstone, Montana ................................... . 501
15.2.3 Zenith Field, Colorado .......................................... . 502
15.2.4 South Ceres Field, Oklahoma .................................... . 506
15.3 Sheet Reservoirs (SH Type) ...................................... . 506
15.3.1 Prudhoe Bay Field, Alaska ....................................... . 506
15.3.2 Messla Field, Libya ............................................. . 509
15.3.3 Statfjord Field; North Sea ....................................... . 511
15.4 Channel-and-Bar Reservoirs (CB Type) ........................... . 513
15.4.1 Little Creek Field, Mississippi .................................... . 513
15.4.2 Daqing Field, China ............................................ . 515
15.4.3 Red Wash Field, Utah ........................................... . 519
16 Future Research Trends ........................................ . 523
References ............................. ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525
Author Index 567
Subject Index 575

Chapter 1
Introduction
1.1 Scope and Purpose of Book tion of regional seismic-reflection data, geologists
are learning to appreciate the importance of regional
and global controls on basin architecture. The sedi-
Fluvial deposits represent the preserved record of mentological effects of orogenic uplift, of basin sub-
one of the major nonmarine environments. They sidence, and of base-level change brought about by
accumulate in large and small mountain-girt valleys, these processes, and by eustasy, are profound and
in the broad valleys of trunk rivers, in the wedges of pervasive. They influence the entire architecture of a
alluvial fans flanking areas of uplift, in the outwash fluvial succession, ranging from such small-scale
plains fronting melting glaciers, and in coastal characteristics as the vertical facies sequence in a
plains. The composition of fluvial detritus depends drill core, to the large-scale stratigraphy of a basin-
on the geology of the source areas from which the fIll complex. In this sense, fluvial sedimentology has
sediment was derived, and the climatic influences to now been incorporated into the larger field of se-
which the detritus was subjected before final burial. quence stratigraphy, which has revolutionized sedi~
The nature of the fluvial assemblage - its lithofacies mentology and basin analysis since the 1980s.
composition, vertical stratigraphic record and archi- At the same time, two other areas of research have
tecture - reflects an interplay of many processes, contributed significantly to the development of the
from the wandering of individual channels across a science. Firstly, observational and experimental geo-
floodplain, to the long-term effects of uplift and morphology have added much to our understanding
subsidence. Fluvial deposits are a sensitive indicator of sedimentological processes. In recent years, many
of tectonic processes, and also carry subtle signa- geomorphological studies have been stimulated by
tures of the climate at the time of deposition. They the needs of geologists to better understand the rock
are the hosts for many petroleum and mineral de- record. In particular, much geomorphic work has
posits. This book is about all these subjects. been carried out in order to explore the effects of
Why such a book now? The study of fluvial depo- geological processes, such as tectonism, acting over
sits has a lengthy history, as documented in some geologically significant lengths of time. Experimen-
detail in Chap. 2. Up until the early 1960s such study tal geomorphology, simulating natural processes in
constituted part of the broader study of "stratigra- laboratory flumes, has provided many Valuable in-
phy". During the 1960s and 1970s, the new subdis- sights, for example, into the effects of changing the
cipline called "sedimentology" came of age, one of sediment load, or the regional slope, or the base level
the major focuses of which was the development of of a fluvial system. At the time of writing this book,
our ideas about facies and intrabasinal, autogenic interaction between sequence stratigraphers and
depositional processes, leading to the concept of the geomorphologists was beginning to recognize the
process-response model and to generation of suites considerable complexity of the relationship between
of facies models. Since the early 1980s, sedimento- climate change, base-level change, and the fluvial
logy itself has undergone something of a revolution, response, and a renewed interest was developing in
and the term "basin analysis" is a more accurate the incised valleys and terrace deposits of rivers
description of what many sedimentary geologists formed during the Late Cenozoic cycle of glacial and
now do. During this most recent phase, a renewed interglacial phases.
interest in extrabasinal (allogenic) processes has led Secondly, numerical models of geological pro-
to the realization that fluvial basin architecture is cesses are providing a powerful means of quantita-
much more complicated than just a record of the tively testing complex, multivariate processes. The
meandering and switching of river channels. Stimu- effects of base-level change and tectonism on basin
lated primarily by developments in the interpreta- stratigraphy have now been modeled by many
2 Introduction
workers, and investigators are now exploring such regional journals and the publications of national
additional details as erosion rates, sediment transfer geological survey organizations. Reference is made
and sedimentary facies. in this book to many of these papers.
This complex interplay of processes and pro- One of the problems with the study of rivers and
ducts, and the modern concepts that are evolving their deposits is that they have been examined from
from our improved methods of observation and different standpoints by geologists, geographers,
analysis, require description, and it is the perceived and engineers, with consequent difficulties of com-
need for a synthesis of the welter of recent literature munication of results and ideas. One of the best ways
on the subject that has stimulated the writing of this to address this communication problem is to bring
book. the specialties together, and this was a principal
In carrying out the literature search for the sec- reason for convening the first international sympo-
tions in this book on petroleum geology, it became sium on fluvial sedimentology, which was held in
clear that there is still a vast amount of work to do to Calgary in 1977. Since then, there have been four
find and exploit oil and gas fields in the frontier areas other symposia, and a sixth is in the planning stage.
of the world. Despite a growing public belief that There have also been several other symposia at inter-
most of the world's petroleum has been found, thou- national meetings that have been devoted in part or
sands of geologists and geophysicists employed by entirely to fluvial deposits. The most important of
scores of companies and government organizations these events are listed below, with references to the
are still devoting their careers to carrying out facies sponsors and the resulting publications.
studies, stratigraphic correlation, seismic facies
analysis, sequence analysis, regional tectonic stu- 1. First international symposium on fluvial sedi-
dies, and production modeling in obscure corners of mentology. Canadian Society of Petroleum Geo-
the earth, in order to track down oil and gas pools logists, Calgary, October 1977 (MiallI978a).
and develop them for production (e.g., Ashton 1992; 2. Recent and ancient nonmarine depositional en-
Flint and Bryant 1993). This is as true for those pools vironments. Rocky Mountain Section, Society of
that are found to be trapped in fluvial reservoirs as Economic Paleontologists and Mineralogists,
for any other. This book is addressed, in part, to Casper, Wyoming, June 1979 (Ethridge and
these individuals. Flores 1981).
Most of the writer's experience has been in the 3. Modern and ancient fluvial systems: sedimen-
area of facies analysis, regional basin studies, and tology and processes (Second International
petroleum exploration and development, thus these Symposium). Sponsored by the International
topics form the main thrust of the book. The reader Association of Sedimentologists, Keele, UK,
will find very little here about diagenesis or September 1981 (Collinson and Lewin 1983).
geochemistry. It is to be hoped that the book will 4. Fluvial sedimentation and related tectonic
provde a useful source of data, analogs, references, framework in western North America. Cordil-
and ideas for those carrying out regional basin leran Section, Geological Society of America,
studies, as well as for those engaged in petroleum Anaheim, California, April 1982 (Nilsen 1984).
exploration and development, and research into 5. Rudites formed by unidirectional flow. Sympo-
groundwater and effluent disposal. sium, Eleventh Sedimentology Congress, Hamil-
ton, Canada, August 1982 (Koster and Steel
1984).
6. Modern and ancient fluvial sedimentation. Gen-
1.2 Data Sources eral Session, Eleventh International Congress on
Sedimentology, Hamilton, Canada, August 1982
Research in the area of fluvial sedimentology and (no proceedings publication).
basin analysis has appeared primarily in a half dozen 7. Third international fluvial sedimentology con-
geological journals, notably Sedimentology, Sedi- ference, Society of Economic Paleontologists
mentary Geology, Journal of Sedimentary Petrology and Mineralogists, Fort Collins, Colorado, Au-
(now the Journal of Sedimentary Research), Journal gust, 1985 (Ethridge et al. 1987).
of the Geological Society (London), Geological So- 8. The three-dimensional facies architecture of ter-
ciety of America Bulletin, and American Association rigenous clastic sediments, and its implications
of Petroleum Geologists Bulletin. Much important for hydrocarbon discovery and recovery. Society
data have, of course, appeared elsewhere, such as in of Economic Paleontologists and Mineralogists,
Data Sources 3
Research Symposium at Annual Meeting, San vidual journal articles have contained proposals for
Antonio, April 1989 (Miall and Tyler 1991). new facies models or variants on existing ones.
9. Fourth international conference on fluvial sedi- Bridge (1985, 1993b) developed a different approach,
mentology, Barcelona, International Associa- building theoretical models from flume experiments
tion of Sedimentologists, October 1989 (Marzo and dynamic studies of modern rivers.
and Puigdefabregas 1993). All the symposia have included large numbers of
10. Braided rivers: form, process and economic ap- case studies. Many of these, especially those pre-
plications; special two-day meeting, Geological sented at the second international meeting in Keele
Society of London, May 1992 (Best and Bristow (Collinson and Lewin 1983), focused on facies analy-
1993). sis. Contrasting with this body of work is the sympo-
11. Fifth International Conference on Fluvial Sedi- sium organized by Nilsen (1984), which set out to
mentology, Brisbane, July 1993 (Fielding 1993a). examine the regional effects of contemporary tec-
12. Alluvial Basin. European Research Conference, tonism on fluvial systems and the sedimentary
Lunteren, The Netherlands, September 1994 (no record. The 1993 Brisbane symposium saw the pre-
proceedings ). sentation of much work in the field of environmental
sedimentology.
These proceedings volumes represent a substan- The most recent developments in fluvial basin
tial body of literature and an essential basis for a analysis are as follows:
fluvial library, to which must be added various 1. Firstly, there is a focus on the detailed, formal
books, course notes, and review articles. Summaries documentation of three-dimensional facies ar-
of progress in research were provided in several of chitecture, stimulated by the pioneering work of
the edited volumes (Miall 1983a, 1987; Fielding Allen (1983a), Ramos and Sopena (1983), and
1993b; Bristow and Best 1993). Important review Ramos et al. (1986). A symposium focusing on
articles include those by Cant (1982), Rust and this type of work in fluvial and other clastic envi-
Koster (1984), Walker and Cant (1984), Bridge (1985, ronments was held at San Antonio in 1989, and
1993b), Collinson (1986), Miall (1992a), Zaitlin et al. has resulted in the production of a large-format
(1994), and Shanley and McCabe (1994). Books that atlas, in which several outcrop and subsurface
contain valuable syntheses of fluvial sedimentology case studies are documened by large-scale cross
include those by Turner (1980), Galloway and sections (Miall and Tyler 1991). The two most
Hobday (1983), Nilsen (1985), and Rachocki and recent fluvial symposia also devoted considerable
Church (1990), although none of these use the archi- attention to alluvial architecture (Marzo and
tectural approach that forms the basis of this book. Puigdefabregas 1993; Fielding 1993a), and publi-
Short courses by Miall (1981a), Flores (1983a), and cations dealing with reservoir heterogeneity are
Flores et al. (1985) provide useful summaries offlu- also giving much attention to this subject (Ashton
vial sedimentary processes, but are now largely out 1992; Flint and Bryant 1993).
of date. 2. Secondly, there is increasing interest in bringing
A noticeable characteristic of the First Interna- together our knowledge of the physics of sedi-
tional Symposium on Fluvial Sedimentology (Miall mentation and the physics of basin subsidence,
1978a) was the presentation of several detailed at- using the methods of quantitative modeling and
tempts to build suites of facies models (Friend 1978; computer simulation. The book by Cross (1990) is
Mial11978c; Rust 1978b; Cant 1978; Boothroyd and a landmark synthesis in this area, and contains
Nummedal1978; Heward 1978a), and several other much of importance to the fluvial sedimentolo-
papers providing critiques of the facies model con- gist. Other work of this type is referred to else-
cept (Jackson 1978; Collinson 1978). None of the where in this book, particularly papers by J.
subsequent symposia have yielded the same focused Bridge, C. Paola, and P. Heller.
examination of the facies model idea. The paper by 3. Observational and experimental geomorphology
Friend (1983) at the Keele meeting, providing a clas- has resulted in several important books and many
sification of channel architecture, is the only post- individual papers, particularly by S.A. Schumm
1977 symposium paper to attempt a generalized, (Schumm 1977; Schumm et al. 1987), and his stu-
theoretical examination of fluvial facies models. Sev- dents and coworkers, and these are also cited in
eral nonsymposium papers providing critiques of this book.
the ideas in this area include Miall (1980, 1985) and 4. Lastly, there are developments involving the
Bridge (1985). Many symposium papers and indi- larger sphere of sequence stratigraphy, requiring
4 Introduction
the researcher to examine the fluvial environment are discussed at length in this book. The geomor-
within the context of the nonmarine-marine tran- phological work referred to above has been par-
sition, and the effects of extrabasinal controls. ticularlyuseful in this regard (e.g., Schumm 1993;
Such books as those by Nummedal et al. (1987), Wescott 1993). A valuable review of the subject
Wilgus et al. (1988), and Van Wagoner et al. (1990, was provided by Shanley and McCabe (1994).
1991) provide a vital source of data and ideas, and
Chapter 2
Historical Background
2.1 Introduction Jopling (l975) discussed early research on fluvio-

glacial deposits. Middleton (1965, 1977) provided
brief histories of the work carried out on bedforms
It is difficult (and probably meaningless) to choose
and sedimentary structures. Pettijohn (l962) and
the moment at which modern studies of fluvial sedi-
Potter and Pettijohn (l963, 1977) reviewed the sub-
mentology could be said to have begun. Probably the
ject of paleocurrent analysis. Allen (1965a) described
monumental work of Fisk (l944, 1947) on the Missis-
the state of knowledge of alluvial sedimentology as it
sippi river and delta was the most important single
was in the mid-1960s, but this was not an historically
advance. Several developments of fundamental im-
oriented paper. The main historical developments
portance took place in the late 1950s and early 1960s.
up to the time of the first fluvial conference were
Flume and fieldwork on the origin of bedforms and
described by Miall (l978b), and later developments
sedimentary structures culminated in a major struc-
were detailed by Miall (1987), and it is from these two
ture classification by Allen (l963a) and statements
articles that this chapter is adapted. The intent of this
on the flow-regime concepts of bedform origin
chapter is to outline research progress up to the
(Simons and Richardson 1961; Simons et al. 1965;
1980s. Modern developments are dealt with in the
Harms and Fahnestock 1965). Facies models con-
detailed descriptive chapters later in this book.
cepts began to appear in the late 1950s, and had a
major impact on geologists following the develop-
ment of the fining-upward cyclothem model by
Allen (l963b, 1964) and Bernard et al. (l962). Mo- 2.2 Early Developments
dern research into stratigraphic traps for petroleum in the Study of Fluvial Sediments
essentially began at about the same time. The Shell
Development and Esso groups at Houston, USA, 2.2.1 From the Ancient Greeks to Playfair
in particular, devoted considerable energies to
researching the sedimentary products of modern Rivers and their deposits did not become subjects of
clastic and carbonate environments. They were in- specialized study until the nineteenth century. How-
terested in improving our knowledge of the geo- ever, many of the early naturalists and philosophers
metry of porous bodies in the subsurface, and made notes on this topic in the context of wide-
developing techniques for diagnosing sedimentary ranging discussions on the origins of the earth and
environments from such data as wireline logs. Nanz its physical features. The principles of uniformitari-
(l954), a member of the Shell group, was probably anism were not widely accepted until the time of
the first (at least, in the published record) to identify Lyell, but long before this it was realized that rivers
fining-upward fluvial deposits in the subsurface, but could transport and deposit large quantities of sedi-
they were not clearly understood as the products of ment, so that correct geological intepretations of
fluvial channel-fill and point-bar migration pro- Recent alluvium were quickly arrived at. Application
cesses for several years, until the work of Bernard et of the same ideas to consolidated and structurally
al. (1962), who were also members of the Shell group. deformed deposits took a little longer. The following
It is this history of the development of ideas that is is a brief review of some of the early writings on the
the subject of this chapter. Some of the areas of physical geology of rivers.
interest to fluvial sedimentologists have been the Plato, in a passage from Critias described the
subject of other historical reviews. Some aspects of destruction of fertile lands by fluvial erosion and
fluvial geomorphology which bear on fluvial sedi- transportation: " ... the soil which has kept breaking
mentology were summarized by Schumm (l972a). away from the high lands during these ages and these
6 Historical Background
disasters, forms no pile of sediment worth mention- Agricola, in his great treatise De re metallica,
ing, as in other regions, but keeps sliding away cease- published in 1556, made some observations on
lessly and disappearing in the deep." Homer, in The placer gold deposits. He recognized that "gold is not
Iliad, described the battle of Achilles with the river generated in the rivers and streams ... but is torn
god Scamander. The river is described as in full away from the veins and stringers and settled in the
flood, and the power of erosion and transportation is sands of torrents and water courses ... " He com-
graphically conveyed: mented on a prevailing theory of his time concerning
the action of the sun in drawing out the metallic
"He filled all his channels with foaming cataracts, and
roaring like a bull he flung up on dry land the innumerable material, which led to the idea that the orientation of
bodies of Achilles' victims that had choked him, protecting veins or gold-bearing streams was critical in their
the survivors by hiding them in the deep and ample pools economic exploitation:
that beautified his course.
The river god sweeps Achilles off his feet, an action "... a river, they say, or a stream, is most productive of fine
likened to a gardener digging an irrigation channel. and coarse grains of gold when it comes from the east and
Mattock in hand, he clears obstructions from the trench; flows to the west, and when it washes against the foot of
the water starts flowing; it sweeps the pebbles out of the mountains which are situated in the north, and when it has
way; and in a moment it runs singing down the slope and a level plain toward the south or west." (Hoover and
has outstripped its guide." Hoover 1950, p. 75).
Later, Scamander calls to Simols: Agricola devoted many pages to the extraction of
gold and tin from placers, and provided us with a
"Dear brother, let us unite to overpower this man ... fill picture of an early fluvial sedimentologist at work
your channels with water from the springs, replenish all with his gravels (Fig. 2.1).
your mountain streams, raise a great surge and send it
down, seething with logs and boulders .... [the armour of The German physician and geographer Bernhard
Achilles I shall lie deep in the slime beneath my flood; and Varenius (1622-1750) published the first text on
as for him, I'll roll him in the sand and pile up shingle high physical geography in 1734. He wrote "Of the
above him." (Rieu 1953) Changes on the Terraqueous Globe, viz. of Water
into Land or Land into Water", and listed a series of
The tendency for rivers to overflow and form "propositions" to explain such changes. Part of Pro-
thick blankets of floodplain deposits has been position V states, regarding rivers:
viewed with mixed feelings by those living in the
valley of the Nile since prehistoric times. One of the "If their water bring down a great deal of Earth, Sand and
first to reach correct conclusions about this process Gravel out of the high Places, and leave it upon the low, in
process of Time these will become as high as the other,
was the Greek philosopher Herodotus (born 484 from whence the Water flows: or when they leave this Filth
B.c.) who concluded that "Egypt is the gift of the in a certain Place on one side of the Channel, it hems in and
river". Aristotle (384-322 B.C.) was also aware of the raises part of the Channel which becomes dry Land."
depositional activity of the Nile and, in addition, he
Robert Hooke (1635-1703), the English physicist
noted the rapid silting up of navigable river channels
and mathematician, wrote about earthquakes
around the shores of the Black Sea and on the
changing the level of strata, and "that water counter-
Bosphorous. These observations were an improve-
acts these effects" by processes of erosion and sedi-
ment on that ofThales ofMiletus (636-546 B.C.) who
mentation:
thought the presence of new deltaic deposits project-
ing into the sea showed that water could change into "We have multitudes of instances of the wasting of the tops
earth. The Meander River in Turkey was mentioned of Hills, and of the filling or increasing the Plains or lower
by both Herodotus and Strabo, and its name became Grounds, of Rivers continually carrying along with them
great quantities of Sand, Mud, or other Substances from
used as a general term for highly sinuous rivers higher to lower places ... Egypt as lying very low and yearly
(Russell 1954). overflow'd, is inlarg'd by the sediments of the Nile, espe-
Leonardo da Vinci (1452-1519) observed the cially towards that part where the Nile falls into the Medi-
gravel deposits of the River Arno and wrote in his terranean. The Gulph of Venice is almost choked with the
notebooks "which deposits are still to be seen welded sand of the Po. The Mouth of the Thames is grown very
shallow by the continual supply of Sand brought down
together and forming one concrete mass of various with the Stream."
kinds of stones from different localities." Elsewhere
the river deposited mud containing shells, which da Nikolaus Steno (1638-1687) recognized the use of
Vinci used as evidence to dispute the origin of fossils certain fossils as environmental indicators, propos-
as products of the Flood. ing, for example, that the presence of a terrestrial
Early Developments in the Study of Fluvial Sediments 7
A-·RIVER. B-WEIR. C-GATE. D-AREA. E-MEADOW. F-FENCE. G-DITCH.
Fig. 2.1. Early interest in fluvial sedimentation: extraction of tin from placer gravels. From "De Re Metallica" by
Georgius Agricola, 1556. (Hoover and Hoover 1950)
fauna, rushes, grasses and tree stems indicated depo- Tozzetti, De Saussure and Guettard, had put forward
sition in fresh water. similar conclusions in earlier publications.
In 1770, the French naturalist Jeanne Etienne The theory of uniformitarianism is attributed to
Guettard (1715-1786) published a memoir "On the James Hutton (1726-1797), although many of the
degradation of mountains effected in our time by ideas that contributed to his theory had appeared
heavy rains, rivers and the sea" in which he noted the before his time. Hutton's first paper on this subject
petrographic variations in the detritus of different appeared as an extended abstract in 1785 (Bailey
river basins. He was aware of the way in which a river 1967), but his principal work was published 10 years
basin acts as a long-term, but nevertheless tempo- later as a four-part book (Hutton 1795). His main
rary, site of sediment storage in its journey to the sea, contribution to sedimentary geology was the prin-
and observed how the thickness of alluvium in- ciple that most of the features we see in ancient rocks
creases toward the river mouth. Guettard inter- can be explained by processes taking place at the
preted the "poudingues" (English: puddingstone, an present day, and that apparently imperceptible pro-
early term for conglomerate) of the basin as fluviatile cesses can have profound effects if they continue
in origin, comparing them to the modern gravels long enough. Hutton was also, of course, the chief
carried by the River Seine. protagonist of the Plutonist theories of earth his-
The origin of valleys by the erosive action of tory, in opposition to Abraham Gottlab Werner
streams was first clearly taught from actual examples (1749-1817), the principal Neptunist. According to
by Desmarest in a memoir published in 1774, even Werner, alluvial deposits were part of the youngest
though others before him, including Targioni- of his four layers of water-laid rocks, immediately
below the volcanics, and were always to be found in influence of the book were due in large part to the
the same stratigraphic position. wealth of field observations that Lyell was able to
Hutton's language is complex, and his uniformi- bring to bear on his thesis, based on his travels
tarianist theories were better explained by his friend throughout Britain, France, and Italy. Von Hoff
and colleague Playfair (1802), from whom the fol- (1822-1824) had published a treatise earlier on the
lowing observations on rivers and their deposits are theme "investigation of the changes that have taken
culled. He suggested that the thickness of alluvial place in the earth's surface conformation since his-
gravels may be used to estimate the depth of erosion toric times, and the application which can be made
in the source area, but concluded: "whether data of such knowledge in investigating revolutions be-
precise enough could be found to give weight to such yond the domain of history". This work provided
a computation must be left for future enquiry to copious documentation of recent geological activity
determine." Modern basin modeling techniques are around the shores of the Mediterranean Sea and the
only now coming to grips with this problem, nearly Black Sea, such as alluvial deposition and changes in
200 years later. sea level, but the work was based mainly on literature
Fluvial facies and textural studies could be said to research, Von Hoff not having the means for travel
have begun with Playfair's following statement: available to Lyell, and his treatise did not receive the
recognition it deserved (Zittel 1901, p. 188). Lyell
"It isa fact very generally observed, that where the valleys
among primitive mountains open into large plains, the acknowledged that much of the information that he
gravel of those plains consists of stones, evidently derived used in his Principles came from Von Hoff, "but he
from the mountains. The nearer that any spot is to the helped me not to my scientific view of causes" (letter
mountains, the larger are the gravel Stones, and the less quoted in Bailey 1962, p. 78; Wilson 1972, p. 276).
rounded is their figure; and, as the distance increases, this Among his more important pieces of fieldwork,
gravel, which often forms a stratum nearly level, is covered
with a thicker bed of earth or vegetable soil ... The reason of Lyell observed Tertiary fluvial deposits interbedded
this gradation is evident; the farther the stones have tra- with lacustrine sediments and lava flows in the
velled, and the more rubbing they have endured, the Auvergne, France, and Tertiary fluvial gravels con-
smaller they grow, the more regular is the figure they taining beds of clay with marine fossils near Nice
assume, and the greater quantity of that finer detritus (Wilson 1972, Chap. 7). These clearly demonstrated
which constitutes the soil. The washing of the rains and
rivers is here obvious." to Lyell the long and complex geological histories of
these regions, explainable in terms of processes act-
These observations state in a more succinct form ing at the present day over long periods of time, but
ideas expressed at length by Hutton (1795, Part II, quite incompatible with catastrophist hypotheses.
Chap. IV), although Hutton doubted that a high Five chapters in the first volume of the Principles
degree of clast roundness could be attained except in (1st edition) are devoted to rivers and deltas. Lyell
the sea. All these conclusions contrast strongly with described the way in which river meanders gradually
those of Finch (1823), Conybeare and Philips (1824), enlarge themselves and develop neck cutoffs by
Hayden (1821), and others, all of whom attributed wearing away at their banks (Fig. 2.2):
"diluvial gravels" to the Deluge (see Schumm 1972, "When the tortuous flexures of a river are extremely great,
p. 16), and ideas about the fluvial origin of valleys the aberration from the direct line of descent is often
and alluvial deposits continued to be resisted for restored by the river cutting through the isthmus which
many years, because they conflicted with these bibli- separates two neighboring curves."
cal teachings. Lyell thought that meanders were initiated as de-
flections caused by the walls of the valley or by
variations in the thickness or hardness of the alluvial
2.2.2 From Lyell to Davis valley deposits. He quoted earlier work of Flint on
the Mississippi River to the effect that opposite each
Charles Lyell's Principles of Geology, the first volume
of which was published in 1830, is undoubtedly
amongst the two or three most important books on
geology ever published. The intent of his book was to
establish uniformitarianism as the guiding principle
for geological interpretation, following Hutton's
teachings, and to discredit catastrophism and dilu" Fig. 2.2. A river meander, showing a potential neck cutoff
vialism. The immediate success and wide-ranging (a). (From Lyell 1830)
curve "there is always a sand-bar, answering, in the geology, in particular that of the Tertiary, for which
convexity of its form, the concavity of the bend", and Lyell proposed a detailed subdivision, including the
described the development of levees, which slope terms Eocene, Miocene, and Pliocene. Amongst the
back and become finer grained away from the chan- nonmarine deposits described are the alluvium, or
nel. "loess" of the Rhine, and the Swiss "molasse". He
Observations of many floods are quoted as testa- observed eolian sand dunes evolving near Calais and
ment to the power of rivers to erode and transport used his observations in interpreting ripple marks
large quantities of coarse debris (Playfair was aware and other cross-bedding structures (Fig. 2.4):
of this 30 years earlier), and historical evidence per-
"If a bank [ripple] has a steep side, it may grow by the
taining to the burial of Roman buildings and the successive apposition of thin strata thrown down upon its
isolation of ancient ports far inland is used to dem- slanting side, and the removal of matter from the top may
onstrate the fact that rivers can build up thick and proceed simultaneously with its lateral extension. The
extensive alluvial tracts in a few hundred years. The same curent may borrow from the top what it gives to the
velocity gradient of river channels (slowest near the sides ... Each ridge [ripple] had one side slightly inclined,
and the other steep, the lee side being always steep, as be,
bed) was already known, and some basic data on de [Fig. 2.4], the windward side a gentle slope, as ab, cd ...
sediment entrainment had been obtained: We think we shall not strain analogy too far if we suppose
A velocity of 3 inls is ascertained to be sufficient the same laws to govern the subaqueous and subaerial
to tear up fine clay; 6 inls, fine sand; 12 inls, fine phenomena ... We may refer to a drawing given in the first
gravel; and 3 ftls, stones of the size of an egg. volume [Fig. 2.3], to show the analogy of the arrangement
of the ... strata just considered, to that exhibited by depo-
Lyell provided what is probably the earliest illus- sits formed in the channel of rivers."
tration of fluvial cross-bedding, or "false" bedding
(Fig. 2.3), observed in a cutbank in the River RhOne Lyell was thus aware of the hydraulic significance
at the confluence with the Arve. He explained the of "false" stratification. For example, he illustrated
origin of the varying cross-bedding angle as follows: what we would call herringbone cross-bedding, and
attributed the opposite dips to "the set of tides and
"Those layers must have accumulated one on the other
by lateral apposition, probably when one of the rivers currents in opposite directions". However, the first
was very gradually increasing or diminishing in velocity, clear enunciation of the principles of paleocurrent
so that the point of greatest retardation caused by their analysis was left to Sorby.
conflicting currents shifted slowly, allowing the sediment In 1841 and again in 1845, Lyell visited North
to be thrown down in successive layers on a sloping bank. America; the second visit included some time spent
The same phenomenon is exhibited in older strata of all
ages." examining the Mississippi River and delta. Both
these trips provided him with much new informa-
Nowadays we can recognize a reactivation surface tion for revisions of his Principles, which subse-
in Lyell's illustration. quently ran to 11 editions (the last in 1872).
Volume 2 contains little of relevance to this book. Lyell's contributions to geology have been as-
However, Volume 3 of the Principles, published in sessed at length elsewhere (Bailey 1962; Wilson
1833 after Lyell's further travels to Spain, Germany, 1972), but it is useful here to repeat Middleton's
and Switzerland, deals primarily with stratigraphic (1978) comment that, although Lyell made many
Fig. 2.3. Cross-stratified sand (C) overlain by

laminated sand (B) and interbedded sand and
gravel (A). (From Lyell 1830)
Fig. 2.4. Development of an asymmetric

ripple. (Lyell 1830)
observations that we would regard as being sedi- He stated " ... when strata are deposited under the
mentological in nature, his primary concern was influence of a current, the character of the resulting
with establishing the working methods of historical structure must depend on the depth of the water, the
geology, and it would not be correct to regard him as velocity of the current, the nature of the deposits and
a founder of sedimentology. It is now suggested by the rate of deposition". He thought that it should be
historians of geology that Lyell overemphasized the possible to deduce the rate of deposition of ripple-
importance of slow, steady processes in his attempts drift cross-bedding from lamina thickness which
to discredit biblical ideas. Catastrophism, that is to "indicates the excess of material deposited on the
say, the rare occurrence of catastrophic events, in- sheltered side of ripples over that washed up again
cluding violent floods, may in fact be very important from the exposed side, during the time required for
processes in building the stratigraphic record (Ager each ripple to advance a distance equal to its own
1993). length", and he carried out hydraulic experiments in
The next important figure in the study of fluvial a stream passing through the grounds of his house
sediments is Sorby, whose earliest contribution on (see Sect. 2.3.5).
this topic constitutes the first paper on paleocurrent Sorby's contributions to the study of sedimentary
analysis (Sorby 1852; available, along with other pa- structures appear to have remained unappreciated
pers by this great innovator, in a collection edited by and largely unknown until exhumed by Pettijohn
Summerson 1976). The principles were stated as fol- (1962) and reinterpreted by Allen (1963b). They are
lows: barely mentioned in Higham's (1963) biography,
" ... By observing the direction of these ripple marks the
and Sorby has generally been much better known for
line in which the current moved may be known ... [when his work in establishing the use of the microscope in
ripple drift or false bedding] is observed in progress in petrographic studies and in metallurgy. However,
modern sand drifts ... it will be perceived that the line of the Sorby's accumulation of much data that never be-
dip of the talus [foreset] is not constantly in the true came published is, regrettably, a familiar tendency
direction of the current on account of the formation of
complicated deltoid deposits, in which the line of the dip of
amongst present -day geologists.
their sloping termination varies very considerably on each Several advances in the study of sedimentary
side of that of the current. Its mean direction, however, structures were made by other naturalists in this
coincides with it; and hence, if a number of properly placed period. For example, Jamieson (1860) first recog-
observations be made, their mean gives a result very nized clast imbrication, while studying fluvioglacial
closely agreeing with the true line of the current."
gravels in Scotland, and Reade (1884) described ex-
Sorby first studied paleocurrent patterns in the amples of ripple lamination in British fluvioglacial
Carboniferous coal measures near Edinburgh. He deposits. Gilbert (1884, 1899) described large-scale
classified the sandstones into four types: level bed- symmetrical ripples from some of the Paleozoic ma-
ded, ripple laminated (single ripple train), ripple rine sandstones of New York State and stated (1899)
drifted, and drift bedded (large-scale cross-bedding) "previous to 1882, ideas as to the origin of water-
and referred to them using the first known set of made sand-ripples were crude and unsatisfactory",
facies symbols. He recognized that structure size is but quoted several papers that had appeared since
partly dependent on flow velocity. that date that had improved theoretical knowledge.
In a paper published only 7 years later, Sorby He proceeded to supply an explanation of oscillation
(1859) was able to boast "I must now have, in my ripples and current ripples which, for the latter, in-
notebooks, not less than twenty thousand recorded cluded Darwin's (1883) discovery of the separation
observations" of paleocurrent directions, a record eddy which forms over ripple crests. "An eddy or
that few have equaled. Most were never published. vortex is created in the lee of the prominence, and
He went on to say "In various papers ... I have ex- the return current of this vortex checks travelling
plained many of my deductions, and I have shown particles, causing a growth of the prominence on its
that many peculiarities of physical geography at downstream side" (Fig. 2.5). Gilbert (1899) provided
former epochs may be learned from a knowledge of a classification of ripple-drift cross-bedding (Fig.
the directions of the currents in various localities". 2.6) which was, in effect, only replaced in the early
He lamented that others had not taken up his meth- 1960s.
ods of facies and paleocurrent analysis, and could The first investigations into the nature and quan-
not know that such methods would not be widely tity of sediment carried by rivers were those of
used again until the 1950s, over 90 years later. Everest (1832), who measured the suspended load of
Sorby, in his 1859 paper, also made the first tenta- the River Ganges under a variety of flow conditions.
tive attempts at understanding bedform hydraulics. Humphreys and Abbot (1861), working for the US
---------
Fig. 2.5. Flow lines over an asymmetric
--- ------
-------- ------- ----~-------- ripple, showing separation eddies. (Dar-
win 1883)
~~~~
"We should now consider the relationship of the fans to
the alluvium proper of the main valleys.
If the level of this latter alluvium is remaining, on the
whole, stationary (the main river neither deepening its bed
nor raising it), and if the fan is undergoing increase, then
the fan-stuff will just extend over the alluvium, gradually
encroach on it in area; and, stratigraphically, rest upon it.
But I think it is a more usual case that the river alluvium
has been increasing contemporaneously with the increase
of fan -stuff, that the river and the ravine-stream both were
raising their beds; then there will have been an inter-
stratification of the two deposits at the fan edges as they
were at successive epochs, a lapping for a short distance of
one set of alluvial beds over the other.
Fig. 2.6. Three types of ripple marks. (Gilbert 1899) I have sometimes observed in section an interstratifica-
tion such as to suggest the above origin - beds of well-
rounded materials and of sand among the less-worn
fan-stuff; for indeed the latter, being nearer its source, is
seldom thoroughly rounded."
Army Corps of Topographic Engineers, carried out Gilbert (1880) published what rapidly became a
detailed studies of the sediment load in the Missis- classic of geomorphological literature in his descrip-
sippi River, and their work was widely quoted by tion of the Henry Mountains, Utah. He dealt with
later writers (e.g., Dana, Geikie) as testament to the fluvial processes extensively, including the growth of
enormous quantities of detritus that can be moved alluvial fans and the development of meanders and
by fluvial transport. Here is a brief extract from their cutoffs, and provided a detailed discussion of the
report: balance between discharge, slope, and load in a river,
"... the sediment of the Mississippi is to the water, by showing how they tended towards an equilibrium, or
weight, nearly as 1 to 1500, and, by bulk, nearly as to 1 to graded, condition. However, from the point of view
2900; provided long periods of time be considered ... of fluvial sedimentology, as distinct from geomor-
812,500,000,000 pounds of sedimentary matter, constitut- phology, little of what he said was an advance on
ing one square mile of deposit 241 feet in depth [0.19 km3], Drew's (1873) work.
are yearly transported in a state of suspension into the gulf
.. , No exact measurement of the amount of the annual The importance of flash flood sedimentation was
contributions to the gulf from [bedload transport] can be recognized following work in the arid American
made, but from the yearly rate of progress of the bars into west. McGee (1897) described sheet floods4 km wide
the gulf, it appears to be about 750,000,000 cubic feet, and recorded flood deposits that extended for 16 km
which would cover a square mile about 27 feet deep [0.021 across strike, and 30 km down depositional dip.
k m 3]."
Summaries of the state of the art at this time are
Drew (1873) provided the first complete descrip- provided in two classic textbooks by Dana (1862)
tion of alluvial fans and their deposits, based on and Geikie (1882). Much is written in each about the
extensive field observations in the upper Indus Ba- erosive power of rivers, and details are provided
sin. A wealth of topographic details is given, and the regarding the slope, discharge, and sediment load of
process of fan sedimentation by distributaries wan- selected rivers around the world; the work of
dering laterally over their own deposits is well de- Humphreys and Abbot on the Mississippi is quoted
scribed: "the lateral changes of position of the extensively. More precise information than before
depositing stream, and the partial growth of each regarding sediment entrainment is given, and the
layer, are denoted by false bedding". Drew described meandering and avulsive habits of river channels are
fan-head trenching and discussed the concept ofthe described. However, even as late as the third editions
graded stream. From the point of view of this histori- of these books, published in 1880 and 1893, respec-
cal review, his most interesting contribution is as tively, little attention is paid to fluvial deposits them-
follows: selves, and no mention at all is made of Sorby's
paleocurrent and paleogeographic analyses. In deal- Dana also thought that floating ice was "con-
ing with ancient deposits, the authors' preoccupa- cerned in part of the deposition". A detailed report
tions are mainly with stratigraphic documentation. on these same rocks by the geographer and geomor-
However, some glimmerings of fluvial sedimento- phologist W.M. Davis (lS9Sa) included a detailed
logy do creep in. Dana wrote "in the alluvial deposits discussion of the various possible environments
from a flood the little layer is begun with a relatively in which they could have formed, including tidal,
rapid rate of deposition and finished with a slower, lacustrine, fluvial, glacial, or eolian setings. It is a
as the flood declines, and hence its upper and lower fine example of sedimentological reasoning, in
portions will differ as to coarseness and density". which modern analogs are clearly appealed to. For
Here, we have the first recognition of the cyclic na- example:
ture of flood deposits. Dana also provided a good
"The Pre-Triassic peneplain might have been warped so as
description of the shape and development of trans- to alter the action of the quiescent old rivers that had
verse sand bars, based on the work of General G.K. before flowed across it ... Such a change would set the
Warren, another of the US Army topographers, on streams to eroding in their steepened courses, and to de-
the Mississippi River. Geikie wrote: positing where their load increased above their ability of
transportation ... The heavy accumulation of river-borne
"The deposit of alluvium on river beds is characteristically waste on the broad plains of California, or of the Indo-
shown by the accumulation of sand or shingle at the con- Gangetic depression all agree in testifying that rivers may
cave side of each sharp bend of a river course. While the form extensive stratified deposits, and that the deposits
main upper current is making a more rapid sweep around may be fine as well as coarse. They are characteristically
the opposite bank, undercurrents pass across to the inner crossbedded and variable, and they may frequently con-
side of the curve and drop their freight of loose detritus tain rain-pitted or sun-cracked layers."
which, when laid bare in dry weather, forms the familiar
sand-bank or shingle beach." However, Davis was a cautious individual. He
concluded:
This is an eminently clear description of the heli-
cal flow patterns that occur on river meanders, and "Penck has suggested the name 'continental' for deposits
of the point-bar deposits that result. It remains only formed on land areas, whether in lakes, by rivers, by winds,
under the creeping action of waste slopes, or under all
to provide the idea of decreasing competency and these conditions combined. This term seems more appli-
sediment sorting up the point-bar slope to arrive at cable than any other to the Triassic deposits of Connecti-
the fining-upward model. However, this was not to cut. It withdraws them from necessary association with a
come for another SO years. marine origin, for which there is no sufficient evidence,
In spite of the general acceptance of the principle and at the same time it avoids what is today an impossible
task - that of assigning a particular origin to one or an-
of uniformitarianism, these excellent observations other member of the formation."
on modern fluvial environments were not, at this
time, being applied to interpretations of ancient The last sentence is revealing, for it shows that
rock successions. Copious descriptions of the litho- application of the growing body of knowledge about
logy, fauna, and flora of such nonmarine units as the modern environments to sedimentological prob-
Old Red Sandstone were given by Dana and Geikie, lems was still very much in its infancy. Davis was
but environmental and paleogeographic interpreta- somewhat more self-confident in his fluvial inter-
tions were limited to general statements concerning pretations of the Tertiary strata of the Rocky Moun-
the deposition of such rocks in lakes or "inland tains, pubished only 2 years later (Davis 1900).
seas". Their general nonmarine nature was recog- At about this time, Davis (lS9Sb, 1S99) also made
nized, but further interpretations did not proceed a major contribution to fluvial geomorphology with
beyond such statements as this, from Dana, concern- his classification of rivers according to their stage of
ing the Triassic of the Appalachians: development in the erosion cycle as youthful, ma-
ture, and old, and he recognized and described rivers
"The mud-cracks, raindrop-impressions and footprints -
these show, wherever they occur, that the layer was for the of braided and meandering character.
time a half-emerged mud-flat or sand-flat; and, as they A rather more advanced level of interpretation of
extend through much of the rock, there is evidence that the fluvial sediments was given earlier by Medlicott and
layers in general were not formed in deep water ... The Blanford (1S79), but their observations were not
occurrence ... of coarse conglomerate, some of the stones widely applied for some years. They described the
of which are very large and of a coarse .kind of oblique
laminatIon [cross-bedding?) in much of the rock, is evi- Siwalik Formation of the Himalayas, compared it
dence that some of the beds were deposited by a flood of with the "Molasse" of Switzerland and with modern
waters pouring violently down this valley." alluvium, as follows:
Growth of Present-Day Concepts, up to 1977 13
"In the upper Siwaliks conglomerates prevail largely; they lowing are the most important areas of special-
are often made up of coarsest shingle, precisely like that in ized study that were actively evolving into the late
the beds of the great Himalayan torrents. Brown days
occur often with the conglomerate, and sometimes almost
1970s.
entirely replace it. This day, even when tilted to the vertical 1. Descriptive fluvial geomorphology: the physical
is indistinguishable in hand specimens from that of the
characteristics of rivers, alluvial fans, floodplains,
recent plains deposit; and no doubt it was formed in a
similar manner, as alluvium. The sandstone, too, of this their evolution and their variations from source
zone is exactly like the sand forming the banks of the great to mouth.
rivers, but in a more or less consolidated condition." 2. Quantitative fluvial geomorphology: the inter-
This appeal to a specific modern analog is a good relationships between width, depth, slope, dis-
example of an early application of actualistic prin- charge, sinuosity, and sediment load (hydraulic
ciples to a sedimentological problem. geometry).
3. Sediment transport studies: investigations of
sediment entrainment, transport rates, and
downstream textural changes.
2.3 Growth of Present-Day Concepts, 4. Bedforms: their external form and internal struc-
upto 1977 ture, and their relationship to depth, velocity,
sediment grain size, and other parameters. Orien-
tation studies, in the form of paleocurrent analy-
The First International Symposium on Fluvial Sedi-
sis.
mentology was held in Calgary, in October 1977. This
5. Facies studies: recognition of the various discrete
event marks a watershed in the development of ideas
lithologic components of a fluvial deposit, with
about fluvial sedimentary structures and facies, be-
their associated sedimentary structures, and in-
cause it was the first time that specialists in a variety
vestigation of the vertical (cyclic) and lateral rela-
of related areas were brought together to deal with
tionships between them. Generalizations in the
this focused theme. The publication that resulted
form of facies models.
(Miall 1978a) is the only memoir produced by the
6. Paleohydraulics: attempts to construct quantita-
Canadian Society of Petroleum Geologists that has
tive descriptions of vanished rivers from evidence
ever been reprinted as a result of demand. This sec-
contained in their deposits. This area of study is
tion deals with developments up to that time, and is
the most recent of the six, and draws on all the
adapted from a review prepared for publication in
other facets of the science of fluviology listed
the proceedings volume (Miall1978b).
above.
In succeeding sections, an attempt will be made to

2.3.1 Increasing Specialization
trace the most significant developments in each of
of the Twentieth Century
these branches of study down to the present day.
Until the end of the nineteenth century, it was pos-
sible to be both a general geologist and a geomor-
phologist. Middleton (1978) regarded Walther 2.3.2 Descriptive Fluvial Geomorphology
(1893-1894) as the first "to draw together the many
scattered observations on modern sediments and to Rivers and drainage networks have been classified in
document the actualistic method as the basis for a four principal ways by geologists and geomorpholo-
specific science of sedimentary rocks". At about this gists:
time, several separate themes gradually began to
appear in the study of rivers, fluvial processes, 1. The cycle-of-erosion classification (youth, matu-
and their deposits. Each theme became a specialized rity, old age) of Davis (1898b, 1899).
subject, with its own body of workers, its own 2. The structural-control classification (conse-
purposes and applications and, eventually, its own quent, subsequent, obsequent, antecedent, super-
literature. To bridge these divergences has be- imposed, etc.) developed by Powell (1875) and
come an increasingly difficult objective for modern Davis (1898b, 1899; see also Johnson 1932).
research workers. Likewise it is a complex task 3. The morphological classification (braided, mean-
to trace the various threads that lead to our dering, anastomosed, straight) which emerged
present understanding of fluvial deposits. The fol- slowly, mainly during this century.
4. A process-response classification that relates Chamberlin and Salisbury (1909), in which detailed
channel morphology to discharge and sediment maps of the meanders, cutoffs, and bayous copied
type. from charts of the US Geological Survey and the
The first two classifications were the earliest, re- Mississippi River Commission are provided. Figures
flecting the preoccupation of nineteenth-century 2.7 and 2.8, taken from this book, show that the
geologists with the cycle of erosion, landscape evolu- surface processes of meander migration in a large,
tion, and the power of rivers to remove and transport highly sinuous stream, were well understood by this
large quantitites of debris (Dana 1850a,b; Davis time.
Davis (1898b) was probably the first to recognize
1899).
The morphological study and classification of ri- clearly that there is another distinctive type of chan-
nel pattern, which he called braided. He illustrated
vers do not appear to have started with the work of
any single individual. The terms "braided", "mean- this type using the Platte River. Chamberlin and
dering", and "anastomosing", though poorly de- Salisbury (1909) used the same river as an example of
fined, were all in use in the nineteenth century (e.g., what they termed an anastomosing stream. The term
see Davis 1898b, Chap. IX). Knowledge of the geo- had earlier been used by Jackson (1834) and Peale
(1879). The early recognition of the distinctiveness
morphology of alluvial fans and meandering rivers
was well advanced by the turn of the century, as of the Platte, and its more recent prominence in
sedimentological studies as a result of N.D. Smith's
quotes from Drew (1873), Geikie (1882), and others,
(1970, 1971) work, undoubtedly qualifies the Platte
earlier in this review, have indicated. The work of
as the type example of a braided river. Davis (1898b,
Gilbert (1880) on this subject has also been much
p. 243) stated:
quoted. The concept of the graded stream was begin-
ning to be understood, and Powell (1875) introduced
the concept of "base level" in downward erosion.
The Mississippi River has always loomed large RlmrBanlr., 18aJ
in the minds of alluvial sedimentologists and geo- Bar.s fic. 168J
-,
morphologists. Nowhere is this more obvious, in Rwer Ban.hs ,,,,jo
contemporary literature, than in the book by
Fig. 2.7. Development of meanders in a high-sinuosity Fig.2.B. "Meanders and cut-offs (ox-bow lakes) in the
river, showing location of cut-bank erosion (ticks on out- Mississippi Valley a little below Vicksburg. The figure also
side of curves) and point bars (stipple); based on know- shows the migration of meanders downstream and their
ledge of the Mississippi River. (Chamberlin and Salisbury tendency to increase themselves". (Chamberlin and
1909) Salisbury 1909)
"If a stream has a large load of coarse rock waste, its graded on the flood-plain surface during extreme stages".
flood plain must be relatively steep (a descent of from 5 to Single-crest streams are those which lack this second
20 ft or more a mile). In this case the stream does not turn
far aside from a direct course along the flood plain; but is
depositional stage. Melton distinguished between
constantly embarassed by the formation of bars and is- vertical-accretion and lateral-accretion deposits; he
lands of gravel and sand, splitting its current into many discussed the difference between braided and mean-
shifting channels." dering rivers and described some of the characteris-
Chamberlin and Salisbury (1909) stated: tics of intermediate morphological types; but all this
information was assigned a secondary importance in
"In general, anything which greatly increases the load of a his classification scheme.
stream near its head is likely to cause deposition, and so An important advance came with the extensive
the development of a flood plain, at some point farther
down the valley. Streams which are actively aggrading series of flume experiments carried out by Friedkin
their valleys are likely to anastomose. This results from the (1945), who investigated the effects of altering dis-
filling of the channels until they are too small to charge, bed materials, and slope on meander pat-
accomodate all the water. The latter then break out of the terns and the stability of bars. Friedkin also
channel at few or many points. The new channels thus produced braided channel patterns by increasing the
established suffer the same fate."
load and slope, and concluded that an important
Nowadays it is recommended that the term anas- contributary factor in the development of braided
tomosing be restricted to rivers characterized by a patterns was easily erodible banks. This work was
network of stable channels of low to high sinuosity, much quoted by Fisk (1947) in his description of the
and that the term braided should be used for the type Mississippi alluvial deposits.
of river described by Davis (1898b) and Chamberlin Leopold and Wolman (1957,1960) used both field
and Salisbury (1909), which show unstable, low- and flume experiments to study bar and channel
sinuosity channel patterns, a wide, shallow cross behavior. They demonstrated that channel patterns
section, and a large bedload (MiallI977). form a continuum resulting from variations in dis-
The fourth approach to river classification is the charge, load, and slope, and showed how channel
one of most interest to fluvial sedimentologists. Its patterns change downstream in response to varia-
development overlaps with that of the morphologi- tions in these controls.
cal classification. The link between channel cross Schumm (1963) divided rivers into three types:
section and sediment type was first suggested by bedload-, suspended-Ioad-, and mixed-Ioad-
Griffith (1927), who observed that natural streams streams, and showed how load characteristics are
with heavy bedload tend to flow in broad, shallow related to sinuosity and channel cross-section shape,
channels, and concluded that a river must tend to and to the tendencies toward vertical or lateral ero-
adopt the cross-section shape which gives it the sion and sedimentation within the channel. Several
maximum sediment-carrying capacity. Actual clas- workers, notably Galloway (1981), later used this
sification began with the work of Melton (1936). This classification as a basis for a basic subdivision of
was an important geomorphological contribution, fluvial styles (Sect. 2.4.4).
in which aerial photographs were first used exten- Galay et al. (1973) and Mollard (1973) carried
sively to illustrate fluvial landforms, but it was these studies further, showing, in particular, the
largely ignored by geologists (with the exception of wide range of variability that can exist in channel
Fisk). The reason may be that, although the paper plan and cross-section shapes and the sediment
contains numerous useful observations regarding characteristics associated with them. Research is
channels, bars, and floodplains, the classification needed to relate these features in more detail to the
proposed by Melton is complex and uses termino- sedimentary record, there being a particular need for
logy that is difficult to interpret. The word "plain" is quantitative studies. Rust (1978a) developed a new
used to refer to the bar-covered channel (as in braid- braiding index and proposed coupling this with a
plain, meander-plain) and, in a different context, its measure of sinuosity in an inclusive channel classifi-
use implies the entire width of the alluvial valley. cation.
Melton's primary subdivision is into "single-crest" The implications of much of this recent work for
and "double-crest" streams. Double-crest streams studies of ancient fluvial sediments have yet to be
are those which "possess two phases of geological explored in detail. (See Sect. 2.3.6 for additional
activity. For example, many streams build meanders discussion of this area, where its relevance for facies
during high water, and also deposit an alluvial cover analysis is explored.)
2.3.3 Quantitative Fluvial Geomorphology out the logical cause-and-effect nature of river be-
havior he provided an essential rationale for the
An understanding of the quantitative aspects of flu- detailed empirical work which was in progress at
vial geomorphology developed simultaneously with that time on rivers in the United States.
the work carried out on descriptive fluvial geomor- Mackin provided a warning to those engaged in
phology that was discussed in the previous section. the study of fluvial sediments to the effect:
The reason for separating these two aspects of the
work is that they have received very different degrees "(1) that deposits formed by or associated with aggrading
of attention by clastic sedimentologists. Those streams differ markedly from the loads carried by them;
(2) that distinguishing between channel and overbank de-
studying ancient fluvial sediments have generally posits is the first essential step in interpreting modern
made some attempts, with varying degrees of sophis- valley fills or ancient fluviatile sediments; and (3) that even
tication, to relate the features of their deposits to the after this distinction is made it is virtually impossible to
morphology and behavior of modern rivers, but work directly from the grade sizes represented in the
quantification of these relationships has been at- channel deposits to the characteristics of the depositing
streams because there is no simple relationship between
tempted only recently. the deposits of an aggrading stream and such partly inde-
In this section, the development of some of the pendent factors as slope, discharge, channel characteris-
principal ideas concerning quantitative aspects of tics, velocity, and load. We cannot proceed directly from
river behavior will be described briefly, emphasizing laboratory-determined laws relating to stream transporta-
those aspects most used by geologists. tion processes to interpretations of ancient stream depo-
sits. An alternative and promising route of attack on the
Early attempts to relate velocity and discharge to problem is via study of deposits now being formed by
channel morphology were carried out by engineers natural streams of many types to determine whether the
for the purpose of designing irrigation canals in sum total of all characteristics of given deposits is uniquely
India, Egypt, and arid parts of the United States. related to the particular modern streams by which they are
Kennedy (1895) published the first important work being formed, and to proceed thence to an understanding
of ancient streams by comparison of their deposits with
in this field, relating velocity to flow depth and width deposits of modern streams of known character."
in a series of empirical graphs derived from work in
India. Contributions by manv later workers were In this statement, Mackin succinctly outlined
summarized by Lane (935), who showed that em- the problems we are still tackling in the area of
pirical equations were not necessarily of universal paleohydraulics, and foreshadowed what was to be-
applicability, depending on a variety of factors that come the most powerful tool in the hands of sedi-
earlier workers, working within the confines of a mentologists: facies analysis of modern and ancient
limited geographic area, had not previously recog- sediments.
nized. Lane listed the following factors that may Lane's work was continued by Leopold and
enter into a determination of channel shapes: Ca) Maddock (1953), who drew on "the mass of data on
hydraulic factors (slope, roughness, hydraulic radius streamflow collected over a period of seventy years
or depth, mean velocity, velocity distribution, and representing rivers allover the United States". This
temperature); (b) channel shape (width, depth, and included concurrent measurements of mean velo-
side slopes); (c) nature of material transported (size, city, width, shape and area of cross section, dis-
shape, specific gravity, dispersion, quantity, and charge, and suspended-sediment concentrations.
bank and subgrade materials); and (d) miscella- They provided a series of graphs showing the rela-
neous (alignment, uniformity of flow, and aging). tionships between these parameters for specific ri-
"A graded stream responds to a change in conditions in vers, and arrived at some general relationships that
accordance with Le Chatelier's general law: - 'if a stress is theoretically could be applied to a variety of settings
brought to bear on a system in equilibrium, a reaction by adjusting the values of various exponents and
occurs, displacing the equilibrium in a direction that tends coefficients. This work was continued by the same
to absorb the effect of the stress'''. group of workers for some years (Leopold and
This is the primary conclusion of Mackin (1948) Wolman 1957; Leopold et al. 1964), and additional
in an important geomorphological paper on the con- data on the suspended-sediment load, the relation-
cept of the graded stream - the first to deal with this ship of meander wavelength to channel width and
subject in depth since Gilbert's (1880) pioneer work. radius of curvature, flood magnitude and recurrence
Mackin discussed the effects on a river of changes in interval, and the relationship between discharge and
discharge, load, slope, base level, and other para- drainage area were published. A later paper
meters. His discussion is qualitative, but in setting (Langbein and Leopold 1966) showed that:
" ... meanders are the result of erosion-deposition pro- sediment transported under various flow condi-
cesses tending toward the most stable form in which tions. Grabau (l913a) summarized earlier work on
the variability of certain essential properties is mini-
mized. This minimization involves the adjustment of
this topic, which included the use of hazelnuts, wal-
the planimetric geometry and the hydraulic factors of nuts, peas, pigeons' eggs, and hens' eggs as compara-
depth, velocity, and local slope. The planimetric geo- tive grain-size standards. However, this picturesque
metry of a meander is that of a random walk whose most approach to quantification was quickly lost by later
frequent form minimizes the sum of the squares of the workers. Grabau (1913a) also summarized early
changes in direction in each successive unit length. The
direction angles are then sine functions of channel dis-
work dealing with particle sorting and roundness, it
tance." being understood that the latter depended on the
size, specific gravity, and hardness of the grains,
This theoretical conclusion has been used in their distance of transport, and the agents by which
models of river meanders and in attempts to recon- they had been transported. Sorby (1908), in a classic
struct the sinuosity of paleostreams from the record paper on quantification in geological studies, re-
of preserved bedforms. ported on his investigations into particle settling
From the geological viewpoint, quantitative geo- velocities, and angle of repose. He was the first to
morphology has been brought to a relatively ad- attempt to use experimental laboratory studies in
vanced state by the work of Schumm, in a variety of investigations of geological problems (see also Sect.
papers culminating in a review article (Schumm 2.3.5), but he appears to have been been ignorant of
1972b). Schumm provided a family of empirical the many advances made by this time in fluid hy-
equations relating various hydraulic parameters draulics.
to features that could be studied and measured in An important, widely quoted paper by Gilbert
ancient fluvial deposits. Leeder (1973) provided a (1914) reported an extensive series of flume investi-
useful supplement to this, dealing with channel di- gations into the transport of sediment by streams,
mensions, and Ethridge and Schumm (1978) pro- the stimulus for which arose from "an investigation
vided a review of the methodology. There remains of problems occasioned by the overloading of cer-
the problem, mentioned earlier, regarding the appli- tain California rivers with waste from hydraulic
cability of these relationships to rivers flowing under mines". Gilbert was concerned mainly with investi-
different climatic regimes. Even in that classic river, gating stream capacity. He examined modes of
the Mississippi, the concept of the graded river and transport and concluded:
all its quantitative ramifications were not fully ap-
"Some particles of the bed load slide; many roll; the
preciated for many years, as shown by the history of multitide make short skips or leaps, the process being
the many man-made neck cutoffs made in the river called saltation. Saltation grades into suspension."
since 1931, the success of which appears to have
ranged from the dubious to the disastrous (Winkley Gilbert varied the grade and quantity of debris,
1977). width, and discharge, and measured the "slope to
which the stream automatically adjusted its bed so as
to enable the current to transport the load". He also
investigated the velocity required to move particles
2.3.4 Sediment Transport and Textural Studies of different sizes and concluded that competence
varied between the 3.2 and 4.0 power of mean velo-
The physics of particle movement is of importance to city, depending on which experimental conditions,
fluvial sedimentologists for a variety of reasons: (1) discharge, slope, and depth, were varied. Gilbert's
for an understanding of the gross variation in grain- (1914) paper was an important bridge between the
size and other textural parameters in fluvial depo- disciplines in that he brought many of the tech-
sits; (2) as a basis to the understanding of bedform niques of fluid hydraulics to bear on a subject of
generation and flow regime concepts; and (3) in considerable geological significance.
providing an explanation for density sorting of Several theoretical developments followed during
clastic grains, of importance in petrological studies, the 1920s, including the boundary layer theory, de-
particularly those dealing with economic placer developed mainly by aeronautical engineers, recogni-
posits. tion of the importance of bed roughness, and the
Early studies were concerned with empirical mea- suggestion by Jeffreys (1929) that particle movement
sures of the current velocities required to move ma- depends not on velocity or drag force, but upon lift
terial of different sizes and with the total volume of induced by the velocity gradient or the rate of shear
between adjacent fluid laminae. Turbulent flow in of sediment transport. A more complete discussion
open channels, pipes, and in air was investigated, is beyond the scope of this chapter, and the reader is
and the change from laminar to turbulent flow with referred to Bush and to Briggs and Middleton (in
increased velocity was documented with the use of Middleton 1965) for further historical details. Recent
the Reynolds number (Reynold's classic work in this reviews of the topic have been given by Church and
field dates back to 1883). In 1936, Shields provided Gilbert (1975), Middleton and Southard (1977), Shen
the first satisfactory theory regarding the initiation (1978), and Fielding (1993b).
of bedload movement on a flat bed, in which the Subaerial debris flows contribute much coarse,
bottom shear stress or shear velocity was shown to be poorly sorted sediment to alluvial fan surfaces, as
the critical parameter. Hjulstrom (1935) used aver- first documented in detail by Blackwelder (1928). He
age velocity to predict the critical conditions for the referred to these catastrophic events as mudflows,
beginning of sediment movement. This approach but data compiled by Bull (1964) and Lustig (1965)
has been used recently in paleohydraulic reconstruc- showed that their actual mud content may be less
tions, but is less accurate. Rubey (1938) reexamined than 10%. Observations on modern floods by
Gilbert's (1914) flume data in the light of these theo- Chawner (1935) and Sharp and Nobles (1953) de-
retical developments, although he does not seem to monstrated the transporting power of debris flows
have been aware of Shield's (1936) contributions. and the conditions under which such events occur,
Rubey was able to explain reasonably well the move- notably infrequent but torrential rainfall on unvege-
ment of material of medium sand to pebble grade, tated upland regions, where sufficient time has been
but stated that "a satisfactory theory of the force allowed for the accumulation of a mass of loose
required to start movement of fine sand and silt debris. More recent studies of the mechanics of de-
must await additional observations". bris flows have been published by Middleton and
Theoretical developments on the calculation of Hampton (1969), Statham (1976), and Rodine and
total load are incomplete. Rouse (1939), and later Johnson (1976). Other special processes operating
Vanoni (1946), used the concept of momentum dif- on alluvial fans, including the tendency of flows to
fusion to predict concentrations of suspended sedi- lose competence by inflltration, leaving "sieve de-
ment. The earliest successful work on bedload posits", were described by Hooke (1967).
tranportation was that of Du Boys (1879), who de- Textural studies, that is, investigations of grain-
duced that the rate of movement of bed load is size parameters (such as mean, sorting, skewness,
proportional to the product of the shear stress and and kurtosis) and other grain descriptions, includ-
the difference between shear stress and the critical ing roundness, shape, and sphericity, have long fas-
shear stress for the initiation of movement. This cinated clastic sedimentologists. Since the work of
physical model has been widely used. However, it Udden (1914), the hope has been that simple labora-
fails to take into account bed roughness (grain tory measurements on a suite of samples (better still,
roughness and form roughness - that induced by the a single sample) would reveal their depositional en-
presence of bedforms), a problem first tackled by vironment, and the literature on grain-size analysis
Einstein (1950). is now vast. Modern work on the subject began with
Developments in fluvial hydraulics up to the mid- the grain-size scale for pebble- to clay-sized material
1950s were summarized by Leliavsky (1955). Geolo- erected by Wentworth (1922; based on Udden 1914)
gists made little use of this field until Sundborg's and the investigation of the logarithmic phi scale
(1956) major paper on the River Klaralven was pub- for grain-size description by Krumbein (1934). At-
lished. In this publication, Sundborg applied avail- tempts to use sorting parameters as indicators of
able theory regarding flow dynamics, erosion, and mode of deposition began with the work of Inman
entrainment of sediment to a study of the bedload (1949, 1952), and Folk and Ward (1957) proposed
and suspended load of a specific river. The geomor- methods for describing the sorting of samples of
phology of the river, and the texture and structure of clastic rocks using percentile values of various grain-
its deposits were also described in great detail. The size intervals derived from the graphs of the cumula-
paper thus represented a major bridge between the tive grain-size distributions. These methods were
disciplines of hydraulics, geomorphology, and geo- reviewed by Folk (1966). Many attempts have been
logy, and it was widely quoted during the 1960s and made to define the sorting characteristics of sand
1970s. deposits formed in various environments using
This brief summary has attempted to cover most graphic (Passega 1957; Friedman 1961, 1967; Spencer
of the more fundamental developments in the theory 1963; Visher 1965a, 1969) and statistical (Sahu 1964;
Klovan 1966) methods. None of these methods ap- (1883) who demonstrated the existence of separation
pears to be completely satisfactory for discriminat- eddies in the lee of ripple crests (Fig. 2.5) using a
ing the various depositional environments, although rotating flume. Cornish (1899) observed the forma-
the method of factor analysis (Klovan 1966) and the tion of antidunes in natural systems and Hunt (1882)
system of graphic analysis proposed byVisher (1965, studied the growth of ripples on tidal flats. Hobbs
1969a) appear to be the most consistent. Most of this (1906) observed cross-bedding with a consistent dip
work has been strictly empirical in nature. orientation and used this as part of his evidence for
Attempts to relate grain-size distribution to sedi- interpreting the enclosing strata as "torrential" in
ment tranport mechanics were made by Middleton origin (see Sect. 2.3.6).
(1976) and Sagoe and Visher (1977). Middleton Sorby carried out experimental work on the "ef-
(1976) showed that the size break between the trac- fects of current on sand". He explained (in his major
tion and intermittent suspension sUbpopulations in 1908 paper, published shortly after his death):
a river bed depends on the dominant shear velocity
of the flow, a fact of some potential paleohydraulic "Fifty-nine years ago, when I was living at Woodbourne, a
significance. country-house on the east side of Sheffield, there was at the
bottom of the small park a brook entirely under my con-
Methods for the study of particle roundness, trol. In order to investigate a number of questions, I con-
sphericity, and shape (a more comprehensive de- structed a place for experiment with some self-registering
scription than sphericity) were provided by appliances. I could easily regulate and measure the depth
Krumbein (1941) and Sneed and Folk (1958). More and velocity of the current within certain limits."
recent developments were summarized by Pettijohn
et al. (1972). Roundness increases downstream due The work carried out here and elsewhere, plus
to abrasion; mean and maximum grain size decrease careful observations of sedimentary structures in
in the same direction. These parameters are there- ancient rocks, led Sorby to divide sandstone into
fore useful indicators of paleo slope and have been four types:
used in many studies of fluvial sediments too nu-
"1. Thinly or thickly bedded rock, without ripples or drift
merous to mention (see Pettijohn 1962, for an early bedding, and showing little or no graining of the sur-
review). However, their use demands much field face in the line of the current ... This could be explained
measurement and data processing, and, as paleo- by supposing that the water was at considerable depth,
slope indicators, they are less flexible tools than and the material mainly deposited from above, not
paleocurrent analysis. An additional problem is that drifted along the bottom where the velocity of the cur-
rent was much less than 6 inches per second.
of inherited textural characteristics - sand which has 2. Thinly bedded rock, with well marked graining in the
been recycled from earlier deposits may retain size surface in the line of the current, indicating a mean
and shape characteristics which obscure those im- velocity up to about 6 inches per second, but showing
printed by the later environment. For these reasons, few or no ripple marks.
grain size and shape studies largely went out of favor 3. More or less thick masses of rock almost entirely made
up of ripple drift. This must have been when the
during the 1980s, and further developments are not velocity of the current was something like a foot per
dealt with in detail in this book. second ...
4. What I have called drift bedding in numerous pub-
lished papers ... The velocity of the current is indicated
2.3.5 Bedforms and Paleocurrents by the nature of the sand; and probably further experi-
ments would enable us to learn the approximate depth,
which was probably small, since an increase of a very
It was realized early in sedimentological studies few feet made so great a difference in the strength of the
that the morphology of individual hydrodynamic current."
sedimentary structures, and their assemblage and
relative arrangement, might yield clues as to the As pointed out by Allen (1963b), sedimentary
depositional environment under which they were structures in types 2 to 4 would now be termed
formed, and that their scale was somehow related to primary current lineation, small-scale cross-lamina-
the energy level during deposition. Some of the con- tion, and large-scale cross-stratification, respec-
clusions of Lyell, Sorby, and others along these lines tively. The order in which Sorby described the
have been quoted above. Other early work included sandstone types was one of increasing flow velocity-
that of Spurr (1894a,b) who discussed the strength which was the beginning of the flow regime idea,
and constancy (directional consistency) of currents although, as Allen (1963b) stated, Sorby was mis-
required to produce ripple marks, and Darwin taken in that type 2 is now known to form at higher
flow velocities (or shallower depths) than any of the

other three.
The term dune was in use by Swiss workers before
the turn of the century, and had been proposed for
water-formed bedforms on the basis of their similar-
ity to eolian dunes. The term was introduced to
American workers by Gilbert (1914) in preference to
the term sand wave, also in current use (e.g., Hider
1882), and this started a nomenclature confusion
that is only now being resolved (Ashley 1990). The
term antidune was first coined by Gilbert (1914), and
the sequence dune, smooth bed, antidune, with in-
creased flow velocity, was documented by Owens
(1908).
Gilbert's detailed flume experiments confirmed
the results of Sorby's pioneer work, and also repro-
duced the bedform sequence described by Owens
(1908). The mechanism of grain and water move-
ment was observed in detail. It was recognized that
dune spacing depends on depth and velocity (Gilbert
quoted earlier workers who had made similar obser-
vations), and the significance of grain size was real-
ized (although not explored in detail). Similar
experimental work was carried out by Hahmann
(1912) but has not been widely quoted by English-
speaking workers. Fig. 2.9. Contour map representing the ideal form of
Several detailed descriptive papers on cross-bed- linguoid current-ripples. L Lowest point; dotted lines lines
ding structures appeared at about this time, most of of flow of water. (From Bucher 1919; copied from earlier
which attempted to develop criteria by which such work by Blasius)
structures could be used as environmental indica-
tors (Grabau 1907; King 1916; Kindle 1917; Bucher
1919). Epry (1913) stated of ripple marks that "no
one seems until now, to have definitely ascertained important results in fluviatile deposits, the true nature and
their cause", and "ripple marks are the work of the wide distribution of which among the sediments of the
tide and of that alone", but he obviously had not past we are just beginning to realize ... In the deposits of a
river obviously a great majority of all ripples should be
been keeping up with the literature. Bucher provided found facing approximately in the direction of flow of the
data concerning wavelength, amplitude, "horizontal river. Locally, of course, they can be found facing even an
form index" ("the degree of asymmetry may be ex- opposite direction, as along parts of meanders or under
pressed by the ratio of the horizontal length of the the influence oflocal eddies. Such cases can not, however,
lee-side to that of the stoss-side") and "vertical form seriously affect the average direction obtained from nu-
merous determinations."
index" (wavelength/amplitude). He proposed the
terms rhomboid and linguoid for the principal This was contrasted with the more varied ripple
three-dimensional ripple forms, and illustrated their patterns to be expected in marine environments, for
morphology and the flow lines passing over them which Bucher provided what are probaby the first
(Fig. 2.9). Oscillation and current ripples were published current rose diagrams. Bucher discussed
clearly distinguished on the basis of morphology, the initiation of ripples on a flat bed, a process that
origin, and environmental significance. puzzled experimenters for much of this century.
Bucher emphsized the usefulness of ripples as Darwin (1883) thought that discontinuous water
current -direction indicators: motion was necessary, whereas Gilbert (1914)
thought that a preexistent water rhythm was the
"While in marine sediments the trend of ripples seems of cause, and Bucher assumed the need for a small
little value to the paleogeographer, its study may yield obstacle to flow. The process is now understood to be
Growth of Present -Day Concepts, up to 1977 21
the result of bed defects produced by the turbulent the case of cross-bedding, was not fully resolved
burst-sweep cycle (see review by Leeder 1983). until the various structures were reproduced in ex-
A useful review of cross-bedding structures was haustive flume experiments, beginning in the 1950s.
provided by Twenhofel (1932; parts of this section of The work of Fisk (1944, 1947) stimulated a renewed
Twenhofel's book were contributed by Kindle and interest in alluvial sedimentation after World War II,
Bucher). A description of the development of sand- particularly amongst oil companies, who increas-
bars (probably what are now termed 2-D and 3-D ingly realized the value of studies of modern envi-
dunes) was provided: ronments, and the importance of statistics in
geological applications also gained general recogni-
"Sandbars move over bottoms of water as plateau-like
areas with steep slopes on the advancing sides. They may
tion. These developments all came together in the
range in height from less than an inch to 10 or more feet, early 1950s to start the surge in knowledge of fluvial
most of them being a foot or less. The sand is rolled over processes that has occurred since the first fluvial
the top, which usually undergoes some erosion. Reaching facies model was published by Bernard et al. in 1962
the edge of the plateau, the sands roll down the slope at the (see Sect. 2.3.6).
front. The direction of the front slope is usually not con-
stant and the inclinations have a similar range in direction
A contribution by McKee and Weir (1953) fo-
... Each bed thus formed has its cross-lamination in one cused attention on the internal arrangement and
direction, which mayor may not be the same as that of the contact relationships of hydrodynamic sedimentary
beds below and above." structures, and provided a structure classification
that became widely used. Their recognition of the
Apart from Sorby's work, the first systematic
importance of bounding surfaces between cross-bed
mapping of cross-bedding orientation in fluvial de-
sets was an important step forward that has been
posits appears to have been that of Rubey and Bass
built upon in modern architectural classifications of
(1925). According to Pettijohn (1962), "Brinkman
fluvial deposits (see Sect. 2.4.3.1, Chap. 4). Statistical
(1933) was, perhaps, the first of the 'modern' work-
techniques began to be applied to paleocurrent mea-
ers to have a clear concept of the objectives of
surements on a systematic basis, induding the use of
paleocurrent research and to develop methods to
perfected vector. methods of calculating mean and
fulfill these aims."
dispersion (Curray 1956), moving average methods
" ... through measured observations and statistical analysis for clarifying regional trends (Potter 1955; Pelletier
a simple facies characteristic - cross-bedding - might 1958), and the use of analysis-of-variance methods
allow a number of basic, reliable paleogeographic conclu-
to unravel the sources of dispersion in samples of
sions to be drawn."
different outcrop scale (Olson and Potter 1954). Tan-
Important advances in the statistical treatment of ner (1955) discussed, with a variety of examples
paleocurrent measurements were made by Reiche based on his own fieldwork, the use of paleocurrents
(1938), including the use of stereonets for plotting in basin analysis. A much expanded classification of
poles to cross-bed foresets, and vector methods for hydrodynamic sedimentary structures was pro-
data reduction. McKee (1938, 1939) described the posed by Allen (1963a), and this remains the most
sedimentary structures of the Colorado River in the comprehensive treatment of the subject, although
Grand Canyon, and proposed a classification of other classifications have been proposed since, espe-
ripple cross-lamination; Knight (1929) described cially by workers focusing on specific depositional
festoon cross-bedding. environments. The use of paleocurrent measure-
Except for the German work, led by Brinkman ments in basin analysis and regional paleogeo-
(1933), the pace of research in this area was slow, graphic reconstructions was demonstrated in an
especially compared to the explosion of paleocur- innovative case study by Pryor (1960) and reviewed
rent studies which took place in the 1950s. Before in two important state-of-the-art publications by
this time, the attention of most sedimentologists was Pettijohn (1962) and Potter and Pettijohn (1963; re-
held by petrographic studies, as revealed by the focus vised edition, 1977), and the latter authors produced
of the first edition of Pettijohn's (1949) textbook, an illustrated manual of sedimentary structures for
and by the title of the first professional sedimen- use in fieldwork, a year later (Pettijohn and Potter
tological journal, the Journal of Sedimentary Petro- 1964). Another useful book on this topic was pub-
logy, founded in 1931. As Pettijohn (1962) has lished by Conybeare and Crook (1968).
pointed out, there was much misunderstanding as to A renewed interest in flume work produced some
the origins of many sedimentary structures which, in results of fundamental importance in the early
1960s. Various workers experimented with the pro- (1975) showed the time-dependent nature of the ele-
duction of cross-bedding (McKee 1957; Brush 1958; ments in the bedform hierarchy, and Allen (1973,
Jopling 1963,1965), but the most important develop- 1974a) discussed the problems associated with
ments were those of Simons and Richardson (1961), bedform lag - their slowness to respond to changes
in which the flow-regime concept was established. in flow conditions and the implications for hydraulic
This work later appeared (with C.F. Nordin as an interpretation (the concept of lag had been intro-
additional co-author) in a collection of papers on the duced by Kennedy 1963). Southard (1971) and
hydrodynamics of primary sedimentary strucures Harms et al. (1975, 1982) demonstrated the impor-
edited by Middleton (1965). Much of the work of tance of a third parameter, depth, in controlling
earlier years on fluid mechanics and sediment trans- bedform morphology, and the recognition of this
port was brought to bear on sedimentological prob- important point has allowed a much greater degree
lems in this book. The statement of the flow-regime of precision in the prediction of bedform morpho-
concept and demonstrations in the field and the logy from flow conditions. Harms et al. (1975) also
flume of the relationship between bedforms and recorded the recognition of a new "bar" or "sand
sedimentary structures were the two principal fea- wave" bed phase, generally developed between the
tures of the book, and it had a major influence on ripple and dune phase. Such features had been first
sedimentologists over the next decade. The develop- recognized in experiments by French workers in
ment of ideas on the hydraulic interpretation of 1963, and were subsequently recorded in the natural
sedimentary structures up to the mid-1970s was de- state in intertidal environments (see Middleton
scribed in detail by Middleton (1977). 1977).
Parallel to the North American work, mentioned Pebbly rivers were studied by Johansson (1963),
above, was that of Allen, who reviewed much engi- Williams and Rust (1969), Rust (1975), N.D. Smith
neering work on bedforms and carried out his own (1974), and Gustavson (1974, 1978). Measurement
flume experiments on the flow patterns associated and interpretation of clast imbrication have been an
with the various ripple types. Most of this work is important part of this work. Hein and Walker (1977)
contained in his books (Allen 1968, 1970a, 1984). proposed a model for the evolution of bars in gravel
One of the most widely used parts of this work was rivers, and Shaw and Kellerhals (1977) and Koster
the diagram showing the control of flow velocity and (1978) discussed the interpretation of antidunes in
sediment grain size on the stability of the principal fluvial gravels.
bedforms.
Descriptive papers on bedforms and bars in mod-
ern sandy rivers published prior to the First Inter- 2.3.6 Fluvial Facies Models
national Symposium on Fluvial Sedimentology (held
in Calgary in 1977) include those of Carey and Keller 2.3.6.7 From Hobbs to Fisk
(1957), Harms et al. (1963), Harms and Fahnestock
(1965), Coleman (1969), Collinson (1970), N.D. In 1917, Grabau wrote:
Smith (1970, 1971), Williams (1971), and Jackson "From the very first, students of sediments have been in
(1976a). All provided much useful data regarding the main, marinists, though when the organic evidence
morphology and internal structures; orientation pointed unmistakably to the presence of fresh waters they
measurements were provided by some, and most of resorted to a lacustrine variant. Even today the interpreta-
these papers have been widely used in interpretation of sediments as of other than marine or lacustrine
origin meets with a good deal of skepticism in some circles,
tions of the ancient record. and the fluviatilists are not much in favour, especially
Paleocurrent variance in alluvial systems has when they intrude upon territory hitherto preemptied by
been discussed by a few workers. Potter and the marinists ... Fortunately however, American geologists
Pettijohn (1977, p. 108-111) summarized the early are the most tolerant of men, and even the most startling
research in this area. An important concept was idea gets a hearing if they are supported by fact and are the
result oflogical reasoning."
introduced by Allen (1966, 1967), who classified
sedimentary structures into a hierarchy of sizes and Some of the early attempts at fluvial interpreta-
attempted to relate this to the variability shown by tion have been quoted earlier in this section. That by
paleocurrent measurements. Miall (1974, 1976) de- Medlicott and Blanford (1879) was well ahead of its
veloped these ideas further (Fig. 2.10). Collinson time, probably because exposures of the Siwalik
(1971a) and Schwartz (1978) showed how paleo- sediments are located close to the great rivers of the
current dispersion varies with discharge. Jackson Indo-Gangetic plain - an inescapable modern ana-
~ ~
~
~~-t~'
____________~
RA 6 .~ L ____________~
~N~K~ ~N~K~5.~ L ____________~
RA RA
~ N~
K~4~.
Fig. 2.10. The bedform hierarchy, illustrated for a fluvial- tures are: rank 1 fluvial-deltaic system; 2 channels within a
deltaic system. Current rose diagrams illustrate direc- delta; 3 channel reaches; 4 lateral and midchannel bars; 5
tional variability for the structure of each rank, which must cross-bedding flanking a bar; 6 ripple crests on a bar
be added to that of the next larger structures to determine surface, oriented around a sheet run-off gulley. (Miall
the total variability of the structures of each rank. Small 1974)
squares show area enlarged for each lower rank. The struc-
log. A similar spatial proximity led to an interpreta- sources of supply of the material" (as in an alluvial
tion by Hobbs (1906) of various Tertiary sandstone fan) were good evidence of continental origin.
and conglomerate deposits of Spain and Italy as Virtually nobody was carrying out detailed work
"torrential" in origin, by comparison with modern in alluvial environments in the early part of the
flood deposits he observed in the process of forma- century. A major exception was Trowbridge (1911)
tion in the same area. Hobbs quoted Italian workers who made a careful study of the geomorphology and
who had reached the same conclusion in 1895, and deposits of some modern alluvial fans in California,
concluded that such deposits probably are very com- "to discuss the causes and processes involved in their
mon in the ancient record. Here are some of Hobb's deposition" with the objective of deducing "certain
sedimentological criteria: criteria whereby materials so deposited may be
"It should be characterized by included lenticular areas of distinguished from other deposits such as those of
coarser or finer material and by layers of fine material in lakes and seas, even after cementation has taken
sharply defined fIlms and plates which reveal its bedding .. . place". The criteria so deduced are similar to those of
Locally, at least, it may show a type of crossbedding .. . Hobbs (1906), to whom Trowbridge does not refer.
which, though often observed in ancient sandstone is not Fenneman (1906) used the terms lateral accretion
adequately explained by the changing currents along a
marine shore ... The dominance of ripple marks and the and vertical accretion, in reference to what we now
paucity of marine fossils just where marine life should term point-bar and overbank deposits, respectively,
have been most abundant are facts difficult to explain on in a discussion of the origin of floodplains in modern
the theory of marine origin for the great sandstone forma- streams. Melton (1936) made use of this sedimento-
tions." logical concept, but the first geologists to fully ex-
Kindle (1911) discussed cross-bedding and fossils ploit it were Happ et al. (1940) and Fisk (1944,1947),
as evidence of continental origin, and Grabau (1906) to whom reference is made later.
used gross stratigraphic relationships as indicators Barrell's (1912) classic paper on the "criteria for
of environment, suggesting that a "non-marine pro- the recognition of ancient delta deposits" provided a
gressive overlap" (what we would term prograda- detailed discussion of the differences between ma-
tion, with interfingering of marine and nonmarine rine, fluvial and eolian deposits and is, in fact, a
facies) or an unconformable overlap "away from the sophisticated approach to integrated facies studies
following Walther's (1893-1894) teachings, which Elsewhere Barrell (1913, p. 466) attributed the
belies Grabau's pessimistic statement published 5 alternation of fine and coarse units to "the lateral
years later (quoted above). The paper had a pro- shifting of distributaries, the red shales representing
found impact on the development of sedimentology. flood plain areas temporarily removed from the
As regards fluvial sedimentation Barrell stated (in presence of currents". Here, then, is the first inkling
part): of the fining-upward cycle and its origin, although
Barrell was not aware of the general significance of
«In rivers where sands are being deposited the channel
is subject to meandering. It cuts laterally into the banks this observed sequence. This seems to have first oc-
and scours down into older floodplain deposits. The sands curred to Dixon (1921, p. 32) in his studies of the
of the abandoned channels cut across the bedding of South Wales coalfield:
the floodplain deposits on the convex sides of the mean-
ders and on the concave side are interlaminated with «The relations of each sandstone-band (b) to the marls
them." (c and a) above and below are, typically, as follows:
This last statement is, of course, incorrect. Barrell (c) Red marls; passing down into
(b) Flaggy sandstones, the upper red, the lower green; at
continued:
the base of the whole band one or more conglomeratic
«The river works across the floodplain and buries channel cornstone; eroding
structures widely in the fluviatile deposits. The river bars (a) Marl, green immediately below (b) to a depth which
work regularly downstream, being continually cut out varies between a mere skin and several feet, but, in
above and deposited below ... Gravel tends to be concen- each case, is fairly uniform; often a purple band below;
trated along such channel bottoms. Lateral discontinuity lower still the marl is red and passes down into an-
of the sandstone lenses is also a feature, the ancient chan- other sandstone band, which repeats the features of
nel deposits forming a meshwork." (b).
This sequence is repeated interminably and in all parts of
Barrell was also aware of another style of fluvial the series."
sedimentation:
Other contributions made at this time include a
«Effects of sheet-flood deposition: Many aggrading study of the fine banding in Tertiary fluvial sedi-
streams overloaded with sand exhibit at low stage shallow
ments in Burma by Stamp (1925), who attributed it
braided channels within the main channel. At higher water
the main channel may likewise form a braided system, to the effects of seasonal floods, and a study of the
and at highest water the whole floodplain may be covered origins and transportation processes of gravel, in-
by a shallow moving water body ... It is such conditions cluding fluvial gravel, by Barrell (1925).
which seem to be required to produce the great depths A good summary of the state of knowledge of
of regularly bedded and widely extended sandstones
fluvial sedimentology is included in Twenhofel's
which mark certain continental deposits ... They succeed
each other without intet-Iamination of clays and com- (1932) great treatise. Alluvial-fan, valley-flat, and
monly show neither structure nor fossils. False bedding fluviodeltaic environments are distinguished, and
oblique to the even· regular bedding is occasionally the general facies descriptions are much as we would
observed, but the homogeneous material conceals its use today, except Twenhofel thought that "channel
frequent presence. Ripple and current marks, however, are deposits are entirely ephemeral" and that the distri-
rare or absent."
bution of the various lithologies is "extremely er-
A more recent description of this type of sheet- ratic".
flood deposition in Bijou Creek, Colorado, by McKee The next important contributions were those of
et al. (1967), formed the basis for one of the facies Happ et al. (1940) and Fisk (1944, 1947). Based on
models of (MiallI977). their work in modern rivers, Happ et al. divided
Detailed criteria such as those described by fluvial sediments into six types: channel fills, verti-
Barrell (1912) were used by Grabau (1913b) and cal-accretion deposits, floodplain splays, colluvial
Barrell (1913, 1914) in their lengthy descriptions of deposits (hill wash), lateral-accretion deposits, and
the Paleozoic deltas of the Appalachians. The alter- channel lag deposits. They thought that "the last two
nation of sandstone and shale units in these deposits types are of little importance under pre-modern
was noted, and Barrell (1913, p. 458) stated of the conditions". Good descriptions of the facies were
alternations in the Catskill Formation: given, and four facies associations were proposed:
normal floodplain or valley-flat, alluvial-fan, valley-
«A gray or olive sandstone member is commonly sharply
delimited at bottom, but at the top grades first into maroon
plug, and delta. Their description of the valley-flat
argillaceous sandstone, and this in turn into red sandy association includes the first block diagram of a
shale." fluvial facies model (Fig. 2.11) which clearly shows
Growth of Present -Day Concepts, up to 1977 2S
va . YlRTICAl ACCRETION DEPOSITS

1/1 • laTERAL ACCRETION DEPOSITS
.:!l • SPlAYS AlONe; 'IU.£O CHANNEL
P • CHAtlNEL 'ILL
~ • AI.l.VVIAL FANS
Fig. 2.11. A fluvial block-diagram model: "diagram illustrating typical relations of the various types of deposits in the
valley accumulation". (Happ et al. 1940)
the fining-upward nature of meandering stream de- Fisk's (1944, 1947) work, which undoubtedly consti-
posits, but this is nowhere explicitly stated. For ex- tutes the most widely quoted source in alluvial sedi-
ample: mentology. Some Russian work (e.g., Shantzer 1951;
Botvinkina et al. 1954) rivals that of Fisk in its scope
"The modern deposits of vertical accretion usually lie but, being in Russian, it had little influence on the
upon older deposits of similar origin, which in turn are
typically underlain by sandy channel deposits that were Western "school" of sedimentology (again, except
accumulated when the channel occupied other positions for Allen's work).
than at present. These older channel sands may represent Fisk's two classic reports (1944,1947) contributed
lateral accretion on the inside of gradually shifting channel a wealth of detail on the Mississippi fluvial and del-
bends, or they may represent fIlling of former channels taic plain. The postglacial history of the Mississippi
abandoned by avulsions. More recent deposits of lateral
accretion, in generally crescentic shapes along the inside of was documented in detail, paleochannels were
stream bends, may still be exposed at the surface but are mapped, the sediments were described, and the pro-
typically lower than the surface of the adjacent older flood- cesses of fluvial sedimentation, including meander
plain which has been covered by deposits of vertical accre- migration and cutoff, and the formation of clay
tion." plugs, levees, and splay deposits were all discussed in
Happ et al.'s (1940) work was descended from that detail. The reports were based on detailed field-
of Melton (1936; quoted earlier) and Mackin (1937), work, 16000 boreholes, and examination of aerial
who discussed lateral accretion in pebbly rivers and photos. They remain as one of the most detailed
illustrated diagrammatically the development of the studies of an alluvial system ever undertaken, and
now-familiar lateral-accretion surfaces (Fig. 2.12). this is the most valuable aspect of Fisk's contribu-
This paper was incorporated into the geological lit- tion. Most of the main fluvial sedimentary processes
erature on facies models by Allen (1964) but, apart and products were fairly well understood in a quali-
from this, it has rarely been referred to in the sedi- tative way by the 1940s, and Fisk filled out this know-
mentologicalliterature. This is in marked contrast to ledge with the necessary detailed documentation.
Fig. 2.12. Formation of the Shoshone valley floor. A

A Low water stage f, fine detritus, laid down in the
stream channel during periods of normal flow; c
coarse detritus, laid down in the channel at the end
of a period of high water. B High water stage. (From
Mackin 1937). This figure illustrates the principle
of lateral accretion, and shows the accumulation
of heterogeneous lithologic units in the point bar,
preserving large-scale, low-angle cross-bedding
surfaces - the structure later termed epsilon cross-
bedding by Allen (1963a)
- _- - . -
.. _ " ""
Fisk's main sedimentological contribution was to 2.3.6.2 Meandering River Deposits:

demonstrate the magnitude and complexity of the Development of Modern Facies Model Concepts
deposits of a major river, and to demonstrate the
response of that river to fluctuations in base level, in A complete description of the characteristics of fin-
the form of variations in channel morphology and ing-upward cycles was provided by Bersier (1948)
sedimentary facies. This work is still relevant to based on his work in the Alpine molasse (Fig. 2.13).
studies of channel avulsion (Chap. 10) and sequence The following is a free translation of his description
stratigraphy (Chap. 13). Fisk's contributions to del- of the typical cycle:
taic sedimentology were also of fundamental impor-
"The most marked characteristic is the base of the sand-
tance, in that they represented the first major stone beds, which overlie the shale with a sharp contact,
advance in the study of modern deltas since Gilbert's without transition. This is where the grain size is coarsest,
work on lacustrine deltas in the later part of the with minor gravel and reworked shale pebbles. The hard,
nineteenth century. In terms of the development of sandy bed sits on the surface of the shale with its impres-
scientific thought, Fisk provided an authoritative sions (sole structures?), borings and tracks. The basal sur-
face is not planar, but undulatory and irregular. The
demonstration of the value of combining detailed deposit is preceded by an erosion surface in the subjacent
sedimentological analysis with a close examination shale by which the beds are cut out.
of sedimentary processes in a modern environment. Higher in the sandstone bed the grain size decreases. At
This was essential groundwork for the development the same time fine argillaecous beds "ppear mixed with the
of actualistic facies models, which became a major sand. One passes progressively from the sandstone-
molasse to argillaceous, immature sandstone, then to fine-
trend of sedimentological progress in the 1960s, sandy shale, with intermediate gradations. The cycle is
following an innovative conference on the facies always terminated at the top by fine-grained units, colored
models concept in 1958 (Potter 1959). shale and calcareous mud or lithified fresh-water lime-
Fig. 2.13. A suite of typical fining-upward fluvial cyclo-

thems in the Alpine molasse. R Erosion surface. (Bersier
1948)
1 2 3 4 ,
-
NIVEAU DE ~'~~Ur--_ _---'
~----~
':-.:. -.:.., :-:~.:..-: - r-: ' -= :.-:' :,=-
.::"; '::' ::' :' ';':" . . . . .. .. . ...... .' ... .
, '
.
. ..: .:...
: . : :: . .. .., . . ., ...
'"
. .
,
..~ :~
.' ....
Fig. 2.14. Development of a molasse cyclothem. (Bersier 1948). See text for explanation
stone. A new erosion surface, with a recurrence of coarse of their origin. He ascribed them to vertical aggrada-
material marks the end of one cycle and the beginning of tion in channels which shift in position on a delta
the next."
plain, using the modern Mississippi delta as his ana-
Bersier thought that these cycles were tectonic in log. Allen (1962) proposed a similar mechanism for
origin, in keeping with prevailing theories of the some Old Red Sandstone (Devonian) cycles in En-
time that had been developed to explain, among gland, suggesting that the upward fining was due to a
other things, the Carboniferous coal-bearing cycles decrease in slope and current strength as the channel
of the United States. Figure 2.14 is his diagram show- became clogged with sediment. These proposals dis-
ing the development of the cycles, and the following pensed with the need for rhythmic subsidence as a
is a translation of his figure caption: cause, but there was still no general comprehension
by geologists of the sedimentary mechanism of
"Development of a molasse cycle by subsidence. The lateral accretion. This idea first appeared in the geo-
length of the horizontal 'arrows is proportional to the logical literature as a result of Wright's (1959) at-
strength of the currents. 1. Completion of underlying
cycle; 2. subsidence (vertical arrow) with simultaneous or
tempt to explain laterally-persistent cross-bedding
subsequent erosion by turbulent currents developed in the in fluvial deposits. Wright quoted Van Straaten
deeper channel; 3. sedimentation of coarse detritus com- (1954), who had "described deposition of material
mences; 4. deposition of fine sand, filling the channel, and on the inside curves of beach gullies", and Dunbar
slowing of currents by diminution of water depth; 5. depo- and Rodgers (1957) who "suggested that some wide
sition of fine mud at top. In beds of Chattian age fresh
water limestone and coal."
lenses of cross-stratified sediments may have been
deposited as point bars on the inside curves of a
Later Bersier (1958) provided further details on meandering stream". Wright does not seem to have
the molasse cyclothems, including their variability been aware of the work on alluvial sediments by
and thickness, and proposed a different explanation Fenneman (1906), Melton (1936), Mackin (1937), or
Fig.2.1S. Point-bar model, based on obser-

6 t t t
Overbank Flooding vations in tidal creeks, and applied to the
I I I Devonian Old Red Sandstone of Britain
(Allen 1963b). Reprinted by permission of
5 Kluwer Academic Publishers
Upper flow
regime
4
;z
o
(J)
o
D:
ILl
2.
o
IIII--Plane
- -
Beds - - - -
Channel Floor Erosion---- - ~-.,
Happ et al. (1940), and he confused point-bar accre- In a later paper, Allen (1964) documented the
tion surfaces with cross-bedding. Sundborg (1956) lateral-accretion model in detail. This paper was an
described growth of point bars in the River Klarlilven important bridge between the disciplines in that it
and illustrated what we now recognize to be lateral- drew on an extensive review of geological and geo-
accretion surfaces, but his description focuses on morphological literature, including European, Rus-
texture, and on smaller-scale bedforms. sian, and North American work. Elsewhere Allen
The lateral-accretion concept, as currently con- (1963a) described a type of large-scale, low-angle
ceived, was first proposed for modern estuarine cross-bedding, termed "epsilon cross-stratifica-
channel-fill sediments by Oomkens and Terwindt tion", and stated that:
(1960), and later was combined with the fluvial "Wright's (1959) interesting suggestion that point bar con-
fining-upward cyclothem into a complete process- struction generally may explain cross-stratified units
response facies model by Allen (1963b, 1964), who where the cross-strata are laid down parallel to the surface
was working on the Old Red Sandstone of southern of the bar, has not been borne out by a study of the
Britain (Fig. 2.15). More or less simultaneously, literature on point-bar deposits."
Bernard et al. (1962) and Bernard and Major (1963) Mackin (1937, Fig. 2) had, in fact, illustrated what
were studying Recent and Pleistocene fluvial depo- Allen termed epsilon cross-bedding based on his
sits (Fig. 2.16) as part of a major sedimentological work in the Big Horn Basin (Fig. 2.12), as had
research effort by petroleum geologists in the United Sundborg (1956, Fig. 47), but its significance for
States, particularly those of the Shell Development interpreting the ancient record was missed. But with
Group, based at Houston. They had identified up- the new fluvial facies model available, students of
ward-fining cyclic fluvial deposits in the subsurface ancient rocks rapidly began to find examples of
some years previously (Nanz 1954) but the origin of point-bar deposits containing large-scale, low-angle
such sequences was not discussed in detail at that epsilon cross-bedding, thereby providing abundant
time (at least, not in print in the public domain). evidence of lateral accretion as the primary deposi-
Allen's (1963b) depositional model is quite differ- tional mechanism (Allen 1965b; Moody-Stuart
ent from his earlier (1962) interpretation. He dis- 1966).
cussed the origins of the various facies present in a Fluvial and deltaic sedimentary processes and
fluvial cyclothem and showed that their vertical ar- their deposits were described in a detailed review
rangement is paper by Allen (1965a), which summed up the state
of knowledge as it existed in the mid-1960s. The
"... consistent with offlap deposition on the channel bar influence of this paper has been far-reaching, and
(?point bar) of a laterally migrating stream. As can be next to Fisk's work it became one of the most widely
observed on stream channel and point bars today ... both
the intensity and speed of flow tend to decline as the bar is quoted work yet published on fluvial sedimentology.
ascended from the deepest part of the river bed to the top In it, Allen provided, among other things, a series of
of the channel." block diagrams showing various theoretical archi-
B 'UP? as'N B'
I:J
.........
10'
-
••'0'. Low W.~ S toO. oc:posns 01' IIIOOEIitH 'OINT
'"
l
\. 1E111",., •• Mew ,.... ... ltI • .,., 10'
- - - - - -IN,"- ..,-... - -----------
... ,~
.,et.. , I. "OS 'II." 1:. DI[II'OII', 110' LOW 011 .d.llSINo : n II. - LtV (L ST,lGf; 0""_
!TIl mm mm mm'1nmrrnmrm
p ~ E 1 S
000'
10'
'"
, ! , ! !
-- -- K~' ~, -
c .17" _1771 .1"" .1"'0 c·
I:J I:J 1:.1 t:J
'''
. ..
I:J
u".,~.,
,
,I ",'
... I
AWl'.,. Lo. ".'" •• , •• I
-- otflOSITS " • UO;~SO'l'U
_\~!!!."'~!!.~~!!!!..!!r~.!-~ _________ _ tIf- ~ __ "'/
......... ;. I~to' ~ II ~ rt II r It "0'",,,,,
a[CINT ''',101
rTTI l m l ITTTl -n m
p \. DO, C E
... -- ...
~ ~ ~ T T
KAIo.! • "U
o 1000
'------....-----'
'[(1
Fig. 2.16. Log responses from borings in Recent point-bar deposits, Brazos River, Texas, (Bernard et al. 1962)
Piedmont formed of alluvial fans

Broided strean Low sinuosity stream Strongly meandering stream
lliJCocne
•..
~,....
Ion. cheme! bar. Of _ odeposilS
r bell • Many 'OIlstrc'um deposils
Fig. 2.17 A-D. The hypothetical architectural models of Allen (l965a), as redrawn by Friend (1983)
tectural models (Fig. 2.17) which have done much to deposits and showed that the point-bar cycle may be
stimulate the generation of detailed regional paleo- much more complex than commonly.envisaged. Ex-
geographic reconstructions by later workers. cellent examples of ancient point-bar deposits were
A useful paper by Beerbower (1964) divided described by Puigdefabregas (1973), Puigdefabregas
cycles into those of "autocyclic" origin, generated by and Van Vliet (1978), Nijman and Puigdefabregas
intrabasinal sedimentary processes, such as channel (1978), and Nami and Leeder (1978). Jackson (1978)
migration, and those of "allocyclic" or extrabasinal related depositional characteristics to load and dis-
origin, including such processes as tectonism, charge variations, and proposed several new facies
eustasy, and climate change. Nowadays the terms associations. These form the basis for the facies-
autogenic and allogenic are preferred, because the model classification of high-sinuosity rivers adopted
processes are not always strictly cyclic in nature. by Miall (1985) and elaborated in Chap. 8.
The importance of Allen's work on the fining-
upward cycle was emphasized by the work of Vis her
(1965a), and it was also seen that this cycle was one of 23.63 Braided Rivers
a range of possible stratigraphic variations that
could be developed in marine and continental sedi- The distinctiveness of the braided channel pattern
mentary environments, so that Visher's (1965b) was recognized in the nineteenth century, but not
other, more generalized paper (which actually ap- until 1962 was an attempt made to relate this pattern
peared first), firmly established vertical-profile to depositional behavior and to recognize a distinc-
models as one of the most useful tools available for tive sedimentary style, different from that of the
studying ancient clastic deposits. In subsurface meandering river. In that year Doeglas (1962) pub-
studies it remains a widely used technique, aided by lished his analysis of some braided rivers in southern
the fact that cyclic character can almost invariably be France. Ore (1964) discussed criteria for distinguish-
guessed at from geophysical logs alone, without de- ing the deposits of meandering and braided streams,
tailed core or sample studies (Pirson 1970). Prob- and Moody-Stuart (1966) suggested differences in
lems with the vertical-profile technique, and newer sedimentary facies that might be generated by low-,
methods of analysis, are discussed in Chap. 9. as distinct from high-, sinuosity rivers (not necessar-
The most definitive paper on the fining-upward ily the same distinction as that of Ore, as some of
cycle was that by Allen (1970b), in which statistical Moody-Stuart's criteria may apply to single-channel
studies of a suite of Devonian cycles in Britain and straight streams).
North America were described, and hydraulic inter- By the early 1970s the widespread distribution
pretations were provided. Little work on the study of of the braided environment was made apparent by
modern meandering river analogs was carried out the work of McKee et al. (1967), Coleman (1969),
until McGowen and Garner (1970) described some Williams and Rust (1969), and N.D. Smith (1970,
point bars in the southern United States, Bluck 1971), who described, respectively, flash-flood de-
(1971) studied the River Endrick in Scotland, and posits of a small ephemeral stream; the sedimento-
Jackson (1976b) analyzed a reach of the Wabash logy of the Brahmaputra - the first work of its kind
River (Indiana-Illinois). All of these workers docu- on one of the world's giant rivers; a gravel-domi-
mented the variability that can occur in point-bar nated mountain river; and a sandy river of the
American Great Plains. All these papers have now not all the depositional systems termed fans neces-
been extensively quoted in studies of ancient sarily contain this facies (MiaIl1978c; Rust 1978b).
braided stream deposits. Modern alluvial fans were studied by Tolman
By contrast with meandering-river deposits, (1909), who introduced the term "bajada" for the
those of braided rivers have generally been thought laterally amalgamated fans that occur at the foot of
of as somewhat random in depositional character, mountain slopes in many basins. The work of
and lacking recognizable cyclicity. Application of Trowbridge (1911) has been referred to earlier.
the statistical technique of Markov chain analysis to Lawson (1913) coined the word "fanglomerate", and
bedding sequences showed that this is not always the Blackwelder (1928) first appreciated the importance
case (Miall 1973; Cant and Walker 1976) and, in a of mudflows in alluvial-fan deposits. More recent,
review of the braided river environment, Miall detailed work on the geology of alluvial-fan deposits
(1977) developed four facies models for braided ri- by Blissenbach (1954) has been much quoted. Other
vers, one of which was strongly cyclic. Generation of descriptions were given by Bull (1972) and
these models was aided by the development of a Boothroyd and Ashley (1975). Ancient fan deposits
simple lithofacies coding system, which permitted were described by, amongst others, Nilsen (1968)
unified description of braided river deposits, includ- and Miall (1970a), both of whom used textural pa-
ing those already described in the literature. This rameters and clast composition to map fan outlines
method has been widely adopted, and forms the and transport directions. Heward (1978a) provided a
basis for lithofacies descriptions in this book (Chap. review of depositional processes on alluvial fans and
5). described some Carboniferous examples in Spain
Cyclic braided deposits were assigned by Miall (Heward 1978b). The first of these papers contains
(1977) to his Donjek model, but subsequent work an authoritative review of the autogenic and
showed that this name would be best confined to allogenic mechanisms of cycle development in allu-
gravel-dominated cyclic deposits, as in the original vial fans, and has become the standard work on the
description of the Donjek sediments (Williams and subject. Boothroyd and Nummedal (1978) docu-
Rust 1969; Rust 1972). Cant and Walker (1976) and mented the depositional facies occurring on humid
Cant (1978) described a sand-dominated cycle, fans.
which has been erected as a fifth model, and termed Difficulties with the generalized concept of an
the South Saskatchewan type (Miall 1978c). Rust alluvial-fan depositional model were already ap-
(1978b) discussed a slightly different grouping of parent by the late 1970s. The work quoted in the
braided facies associations using his own field ex- preceding paragraphs refers exclusively to gravel-
amples. Modern work on braided rivers and their dominated fans, but Friend (1978) described some
deposits is focusing on the details of the internal ancient cone-shaped sandstone units that he com-
architecture of their deposits (Sects. 2.4.3, 2.4.4.2, pared to the deposits of modern "terminal" fans in
Chap.8). arid regions ofIndia (Mukerji 1976). The giant fan of
the Kosi River was described by Gole and Chitale
(1966), and its deposits have been discussed by Singh
2.3.6.4 Alluvial Fans et al. (1993). This depositional system, so much
greater than the small fans normally conjured up by
Because of their distinctive morphology, alluvial the term alluvial fan, grades from boulder-domi-
fans have always been recognizable as discrete com- nated in proximal regions to a silt and fine sand
ponents of any alluvial system, and sedimentologists system at the toe, 60 km downslope. Later workers
have attempted to carry this distinction into their became uneasy with the inclusion of fluvioglacial
interpretations of the ancient record. Several au- outwash plains in the alluvial-fan category, even
thors have distinguished an alluvial-fan facies from a though many of them are distributary systems. Im-
braided-river facies, which is misleading, because portant "fan" papers, such as those by Boothroyd
much of the sedimentation that takes place on allu- and Ashley (1975) and Boothroyd and Nummedal
vial fans is brought about by the activity of braided (1978) would be excluded, according to this ap-
streams. The most distinctive components of alluvial proach. The current controversy in this area is dis-
fan environments are debris-flow deposits, although cussed in Sect. 8.3.
Fig.2.1B. The four basic fluvial chan-

nel styles. (Miall 1977)
~
~
straight
meandering
anastomosing
o bar surfaces covered

during flood stages
braided
Slims Firth Donjek Fig. 2.19. Three typical gravel-bed

braided rivers in Yukon Territory,
bid active areas 500m
Canada, drawn at the same scale. Note
1::::::::1vegetated, inactive areas L---.J I Gravelly Rivers 1 the major differences in fluvial style
2.3.6.5 Other Facies Models The Slims River is a typical braided river, multichan-
nel, and of low sinuosity. The Donjek, by contrast,
Geomorphological work in the 1970s (e.g., Schumm has only one or two active channels, and is of moder-
1972a,b; Galay et al. 1973; Mollard 1973) emphasized ate to high sinuosity. Point bars are present on some
the great variability in fluvial morphology, and this meander bends. In terms of lithofacies characteris-
revealed the inadequacy of the simplistic twofold tics (lithofacies types and vertical profiles), the three
subdivision of fluvial patterns into braided and rivers would be indistinguishable, yet their fluvial
meandering styles that had become commonplace architecture would be clearly quite different.
by that time. Miall (1977) used a fourfold channel The sedimentological ramifications of this geo-
classification (braided, meandering, anastomosed, morphological variability are still far from being
straight; Fig. 2.18), and Rust (1978a) showed how fully understood. Moody-Stuart (1966) discussed
this classification could be based on parameters that possible depositional products of straight channels,
are quantifiable, at least in modern rivers, namely and D.G. Smith (1973) recognized the distinctive-
sinuosity and a braiding index. However, this classi- ness of what has now become, almost entirely as a
fication has never been entirely satisfactory as a result of his research, the standard "anastomosed"
basis for facies modeling. The terms braided and fluvial model, although his work did not become
meandering refer to different parameters, and it is widely available until the end of the decade (D.G.
possible for a river to be defined by both terms. For Smith and N.D. Smith 1980). D.G. Smith (1973) pro-
example, note that the three rivers in Fig. 2.19 are all posed a quite different definition of "anastomosing"
braided, but differ considerably in their sinuosity. from that used by Miall (1977) and Rust (1978b),
whose ideas were based on those of Schumm (1963). ancient record. For example, it was shown that under
Schumm proposed the use of the term for a type of certain circumstances sequences of apparently
arid fluvial system in Australia, whereas those anas- braided origin could be formed in a single-channel,
tomosed rivers described by D.G. Smith are peren- high-sinuosity fluvial environment (Jackson 1978;
nial streams in a humid-temperate climate. The Nijman and Puigdefabregas 1978). Collinson (1978)
differences in sedimentological style between these pointed out that not all sandstone units in meander-
two types of rivers have yet to be clearly and defini- ing-stream successions are point-bar deposits, but
tively evaluated, although some common factors re- some are the product of overbank sheet floods.
lating to low slope and cohesive banks appear to be Jackson (1978) provided a devastating critique of
emerging (see Chap. 8). The Australian anastomosed the conventions being used at that time to identify
rivers are now regarded as something of a special fluvial style. Successions of "braided" and "mean-
case (Sect. 8.2.10). Baker (1978a) and Taylor and dering" origin may look very similar (Fig. 2.20). In-
Woodyer (1978) described modern fluvial systems deed, for some years it was thought that the two main
in, respectively, the Amazon basin and New South fluvial styles were those of the braided and the mean-
Wales (Australia) that obey very few of the the dering river, and that these styles could readily be
"norms" of depositional behavior, as they were un- distinguished in vertical profiles. For example, in the
derstood in the late 1970s. McKee et al. (1967) de- first two editions of the well-known Facies Models
scribed the deposits of a flash flood in a small book (Walker and Cant 1979, 1984) these two styles
ephemeral stream, Bijou Creek in Colorado, and this were contrasted, with suggestions given as to how
work was incorporated into the braided fluvial mod- they might be distinguished (Fig. 2.21). Walker
els of Miall (1977), establishing a category of fluvial and Cant (1979) suggested that the proportions
depositional styles that is now recognized to be of of "vertical accretion" deposits (bar-top and flood-
considerable importance in the ancient record (Sect. plain sediments) would be greater in the meander-
8.2.16). The study of ephemeral streams in Australia ing style, and the assemblage of sedimentary
by Williams (1971) also played an important part in structures different. If these criteria were correct,
the development of the ephemeral-stream model they would, indeed, be invaluable for subsurface
(Sect. 8.2.15). petroleum geology work, and there would have been
In the 1970s, difficulties were also appearing in little need for further research. However, already by
the application of geomorphological models to the the late 1970s, at the research level, it was realized
- - - Surface current
-- - ~Bottom current
j
E Fig. 2.20. Basic facies models for sand-bed
C\I
N braided and meandering rivers, as developed by
the late 1970s. That for the braided river (A) is
jbased on work in the South Saskatchewan River,
B
:~~~~~~~~i~~~~~~t~.~§~ reported by Cant river
the meandering and Walker
(B) is (1978) , and that
essentially
Bernard et al. (1962) and Allen (1963a)
that for
of
BATTERY POINT
SUMMARY SEQUENCE MEANDERING"
.,.'
FINING -UPWARD
SEQUENCE
V.A.
I
BAR
TOP'
V.A.
/' ...J I
W POINT BAR
Z
Z TOP
<l
"-
:r
u
~
I POINT
BAR
CHANNEL CHANNEL
FLOOR FLOOR
Fig. 2.21. Comparison of model vertical profiles for braided and meandering river deposits. (Walker 1979)
that the braided versus meandering separation was of a meandering stream system as a result of the
overly simplistic. interplay of various autogenic and allogenic mecha-
nisms. Again, as with several of Allen's earlier
papers, this work contained the seeds of what has
2.3.7 Fluvial Architecture evolved into a significant research topic in recent
years, that of sequence stratigraphy. As discussed in
The term "fluvial architecture" was used by J.R.L. Chap. 13, one of Allen's models contains many of the
Allen (at the First International Symposium on Flu- elements of a modern sequence model for non-
vial Sedimentology in his keynote address) to en- marine deposits.
compass the geometry and internal arrangement of In another early and tentative experiment to ex-
channel and overbank deposits in a fluvial sequence. plore controls on sand-body architecture and the
Most of the work up to the time of the Calgary likelihood of exploration holes encountering sand
symposium had concentrated on vertical profiles, bodies in the subsurface, Leeder (1978) provided a
and less attention had been paid to the way in which simulation model of subsurface channel distribu-
the various elements in a fluvial sequence stack on tion. This work was stimulated by the discovery
one another, although this is obviously of critical of petroleum in fluvial reservoirs in the North Sea
importance with regard to reservoir and orebody (M.R. Leeder, pers. comm., 1995), and formed
continuity, and the analysis of present-day and an- the beginning of attempts to develop numerical mo-
cient groundwater systems. Allen's (1965a) architec- dels of channel architecture, as described in Sect.
tural block diagrams, referred to earlier, were a lOA.
theoretical approach to this problem, following the Knowledge of the three-dimensional (or even
preliminary work of Happ et al. (1940). In a later two-dimensional) architecture of ancient fluvial de-
paper, Allen (1974b) emphasized the importance of posits was limited at this time. Campbell (1976) pro-
pedogenic carbonate units as indicators of stability vided a detailed cross section through a Jurassic
and non sedimentation in an alluvial plain, and used fluvial deposit of possible braided origin, and this
them in building a variety of hypothetical models became much quoted in later studies because it ap-
showing the variations to be expected in the deposits peared to demonstrate large-scale lateral shifting of
the channels in a fluvial distributary system, as in blages (Sect. lOA). Paleohydraulics is the study of the
giant modern fans. Analogies with the Kosi fan have quantitative relationships between the hydraulic pa-
commonly been drawn (e.g., Collinson 1986). Unfor- rameters of a river (depth, width, slope, discharge,
tunately, Campbell's reconstruction has now been sediment type, etc.) and its preserved deposits.
shown to be seriously flawed (Sect. 10.3.2). The study of paleohydraulics is fraught with
Other work included a study of the geometry of difficulties, and some fluviologists feel that our
sand units on a delta plain by Ferm and Cavaroc knowledge of the depositional and preservational
(1968). Horne and Ferm (1976) described the archi- processes of rivers is still so inadequate that numeri-
tecture of some Carboniferous coal-bearing rocks in cal estimates of past hydraulic parameters verge
the central Appalachians, based on large highway on the fictional. However, some progress has un-
and railroad exposures. Cant (1978) erected a hypo- doubtedly been made, and continued attempts at
thetical model for the two-dimensional geometry paleohydraulic reconstruction can only serve to en-
of a braided system of South Saskatchewan type. courage fluvial sedimentologists to structure their
Nami and Leeder (1978) described the geometry of a future research into modern rivers so as to ask the
well-exposed meandering river deposit. Friend right questions, in the hope that some of the difficul-
(1978) commented on gross architectural features of ties can gradually be overcome.
several ancient fluvial systems. There are essentially two approaches to paleo-
Vegetation is an architectural control of consi- hydraulics, the engineering approach, which uses
derable importance, affecting channel morphology theoretical relationships and empirical data regard-
and shifting behavior through increasing bank sta- ing sediment transport mechanics and bedform gen-
bility (D.G. Smith 1976) and, by its presence or ab- eration to reconstruct the depositional conditions
sence in headwater catchment areas, playing a large for individual beds or sedimentary structures, and
part in determining the nature of the river hydro- the geomorphological approach, which relates em-
graph and the sediment supply. The geomorphic pirical data about the morphology of modern
effects of vegetation were recognized in the nine- rivers to some of the gross features of their deposits.
teenth century (e.g., Surell1841, 1870), but Schumm Each method has its limitations - the engineering
(1968a) was the first to clearly point out its approach concerns itself primarily with instanta-
paleoclimatic implications. He suggested that rivers neous depositional conditions of specific bedding
of pre- and post-Devonian age were probably funda- units, the overall, long-term significance of which
mentally different because of the first appearance in may be hard to judge because of our lack of knowl-
the Devonian of an extensive cover of land vegeta- edge of facies preservability in fluvial environments.
tion. Long (1978) discussed the implications of this And, as Allen (1973, 1974a) has shown, bedforms are
difference for studies of the Proterozoic fluvial slow to react to changes in flow conditions and are
record, and Cotter (1978) examined the Paleozoic therefore inaccurate hydraulic indicators. The geo-
record of the Appalachians. morphological approach is potentially more useful,
On all scales ranging from the local to the conti- in that it deals with long-term, statistical averages,
nental, a knowledge of the geometry and dispersal but suffers, at present, from a gross lack of data from
patterns of alluvial deposits can lead to a better all but a limited range of modern climatic, discharge,
understanding of contemporary tectonics, as several and sediment-load conditions. Even at the time of
papers published in the 1970s were beginning to writing this book sedimentological concepts regard-
demonstrate (e.g., Steel and Aasheim 1978; McLean ing rivers are derived from very few, possibly atypi-
and Jerzykiewicz 1978; Friend 1978; Potter 1978; cal, modern examples, most of which are located in
MiaIl1978d). populated or otherwise accessible, relatively temper-
ate climates. The range of different conditions repre-
sented by ancient fluvial deposits of all ages has
2.3.8 Paleohydraulics barely been touched, and research which integrates
sedimentology with hydraulics is sparse. Weare
The study of fluvial facies in vertical profiles or in hampered, also, by the influence on modern rivers of
two or three dimensions is an essentially qualitative the rapid climatic and sea-level changes during and
approach to the reconstruction of past river systems, following the Quaternary glaciation, and an uncer-
even where statistical techniques are used to provide tainty as to how this distorts our perceptions of
quantitative information regarding facies assem- rivers at times of greater geological stability. Recent
work demonstrating the complexity of the effects of graphic reconstruction, whence the other hydraulic
climate and sea-level change is discussed in Chaps. parameters were estimated using empirical geomor-
12 and 13. phological relationships. The building of an edifice
Developments in sediment transport mechanics, of interpretation from one or two items of data by
bedform classification, flow regime concepts, and using one estimated parameter as input for the next
quantitative geomorphology, on which the study of estimate is one of the principal weaknesses of the
paleohydraulics rests, have been summarized earlier paleohydraulic method.
in this review. The remainder of this section contains Allen (1970b) used knowledge. of bedform hy-
a brief discussion of some of the published applica- draulics and flow patterns in meander bends to de-
tions of such knowledge to geological problems. duce a generalized paleohydraulic model for a suite
The fining-upward point-bar cycle has been the of Devonian fining-upward cycles. No attempt was
most widely used paleohydraulic indicator, almost made to proceed from this model to generalizations
since the genesis of the cycle as the product oflateral regarding discharge or drainage area.
accretion was first deduced in the early 1960s. The Cotter (1971) combined Allen's (1965b, 1970b)
discovery of "epsilon" cross-bedding and its inter- observations (in part quoted above) with Schumm's
pretation as superimposed point-bar accretion family of equations to produce a paleohydraulic in-
slopes helped to confirm this interpretation. Allen terpretation of a Cretaceous fluvial unit in Utah,
(1965b) was the first to realize the paleohydraulic including estimates of river length, drainage area,
significance of these cycles, in his study of the Old discharge, sinuosity, and slope. Cotter was aware of
Red Sandstone of Anglesey, North Wales: the possible sources of error in his estimates deriving
from differences in climate and vegetation in the
"It appears that the streams which deposited the alluvial
past, and emphasized that "the results obtained are
facies were already substantial rivers on reaching
Anglesey. Each epsilon cross-stratified unit has been inter- only reasonable estimates that must fit the nature of
preted as a point-bar deposit. The thickness of such units the environment interpreted in other ways."
therefore corresponds to the stream channel depth at the Similar exercises to those quoted above were used
bankfull state (Wolman and Leopold 1957, pp. 92, 95), and by Friend and Moody-Stuart (1972), Padgett and
the mean unit thickness is 6 ft 3 in. Leopold and Maddock Ehrlich (1976), Cant and Walker (1976), and Miall
(1953, Appendix A) gave data leading to relationships
between stream width, mean depth and drainage basin (1976) in paleohydraulic reconstructions performed
area, and Hack (1957, p. 63) later presented an equation on fluvial units in Spitzbergen, Morocco, Atlantic
connecting stream drainage area and stream length. When and Arctic Canada. The routes thro1lgh the various
the mean unit thickness is substituted for the mean chan- equations differ in each case; for example, Friend
nel depth in these relationships, it becomes statistically and Moody-Stuart (1972) and Cant and Walker
very unlikely that the streams were less than 40 miles long,
including meanders, and 70 ft wide. Statistically the (1976) based much of their work on flow velocity
streams were most likely to have been 400 miles long and and power deductions derived from sedimentary
300 ft wide. Needless to say, these figures do not represent structures and grain size. Miall (1976) introduced a
predictions, for very many factors combine to determine sinuosity estimation based on paleocurrent mea-
stream geometry, but are intended merely to convey the surements.
probable order of importance of the streams which depos-
ited the Old Red Sandstone of Anglesey." Baker (1973, 1978b) studied the catastrophic Lake
Missoula flood and provided a paleohydraulic re-
Schumm (1968b) studied the present and prior construction of some giant bedforms. Leeder (1973)
courses of the Murrumbidgee River, Australia, and compiled data concerning epsilon cross-bedding,
used the empirical geomorphological relationships and suggested some refinements of the methods for
he had been developing for American and Australian using this structure and cycle thickness as paleo-
rivers to determine their various paleohydraulic hydraulic indicators. Bridge (1975) used much of the
parameters. Sedimentological information, such as earlier work on bedform hydraulics and flow pat-
sedimentary structures and facies relationships, terns in a computer simulation of sedimentation in
were not used in this work. meandering streams. Allen (1977) and Shaw and
Eicher (1969) attempted to estimate river size, Kellerhals (1977) examined ripples and antidunes,
discharge, and slope for a Cretaceous fluvial system respectively, as paleohydraulic indicators.
in Colorado. His only input datum consisted of Several papers in Miall (1978a) reported field
stream length, derived from regional paleogeo- studies of paleohydraulics, and this section of the
Growth of Present-Day Concepts, 1978-1988 37
book also includes a major review of paleohydraulic meeting. Some post-1988 developments are referred
methodologies by Ethridge and Schumm (1978). to for the sake of completeness, but the details of
This collection of papers appears to have brought the most modern studies are given in the appropriate
subject of paleohydraulics to some kind of plateau. descriptive chapters later in this book.
Very little has been published in this area since the
late 1970s, perhaps because the review by Ethridge
2.4.1 Bedforms and Sedimentary Strudures
and Schumm (1978) revealed the weaknesses and
inadequacies of existing paleohydraulic methods. During this period, the definitive compilation of
Recent research has taken a quite different form, bedforms and sedimentary structures by J.R.L. Allen
that of computer modeling of fluvial systems, as was published. This first appeared as a two-volume
discussed in Sect. 10.4. work in 1982, and was republished as a single-vol-
ume edition 2 years later (Allen 1984). The book
contains an exhaustive literature review, and a
2.4 Growth of Present-Day Concepts, wealth of data dealing with the experimental pro-
1978-1988 duction of bedforms in flumes.
Advances in the understanding of the fluid dy-
The decade 1978-1988 included two fluvial sympo- namics of bedform generation were discussed by
sia, that at Keele in 1981 (Collinson and Lewin 1983), Jackson (1976c), Allen (1983b, 1984), and Leeder
and the third symposium, at Fort Collins, in 1985 (1983). Flow visualization studies revealed impor-
(Ethridge et al. 1987). This period marked the culmi- tant details about the turbulence of traction cur-
nation of efforts to define fluvial styles from vertical rents. It was demonstrated that a form of turbulence
profiles alone, and the growth of formal architec- termed the "burst-sweep cycle" is of critical impor-
tural methods for litho some description. Section 2.4 tance in generating ripples and dunes (Fig. 2.22). At
is adapted and expanded from Miall (1980) and the the base of a flow there is an inner layer consisting of
review prepared by Miall (1987) for the Fort Collins streaks of low- and high-speed fluid organized into
4-5 h
-
A
f low
1::::::::1 high speed region

_ low speed region
Fig. 2.22A,B. The two- and three-dimensional structure of turbulent boundary layers. (Leeder 1983)
small vortices and subject to periodic disruption and curred. The decreasing scale and erosive power of
uplift. This passes up into an outer layer containing separation eddies at higher flow speeds leads to
large breakup vortices that rise steeply to the surface, smaller scour pockets and the draping of sand over
where they produce boils. These large vortices de- rounded dune crests, rather than the production of
velop mainly over the lee of dunes, where the pres- angle-of-repose foreset lamination. The result is the
sure gradient encourages bursting. It has been found generation of "humpback" dunes with low-angle,
that the bursts occur, on average, every five water sigmoidal cross-bedding.
depths in the flow direction. As Leeder (1983, p. 7) Considerable problems of classification and
stated, "the feedback effect between flow and bed- nomenclature were apparent by the late 1970s, as
form is clearly of great importance." discussed by Miall (1977) and N.D. Smith (1978).
Leeder (1983) summarized progress in the appli- Leeder (1983) offered a tentative genetic classifica-
cation of these data on turbulence to the theory of tion of bedforms according to the dominant physical
bedform generation. Ripples are controlled by flow- process controlling their size limits and stability
separation dynamics, and are not influenced by the (Fig. 2.23). However, the problem of nomenclature
outer layer. They develop on hydraulically smooth was not satisfactorily resolved until an SEPM Re-
boundaries by the action of flow separation and search Symposium was held in 1987, the results of
reattachment. Their form is insensitive to water which were collated by Ashley (1990). This paper is
depth, and only loosely related to bed shear stress. discussed in a later section. Some other recent re-
With increased shear stress turbulent vortices be- search results have been reviewed by Fielding
come larger but more varied in scale and speed, (1993b).
leading to the generation of a wider range of heights,
wavelengths, and forward ripple speeds. Leeder
(1983) suggested that the larger ripples cannibalize 2.4.2 The Decline and Fall ofthe Vertical Profile
the smaller ones, leading to enhanced scour in the
ripple lee, and still further increases in ripple height. The first fluvial symposium in 1977 was held at a
Dunes, the characteristic bedform in the higher- time when a great deal of new information on fluvial
velocity ranges of sub critical flow, show strong cor- facies was being published. At this time the concept
relation of height and wavelength with water depth of the simplified facies model was much in vogue
(Allen 1968; Jackson 1976a), suggesting that their (Walker 1979). Such models were commonly pre-
form is controlled by the outer turbulent layer. It sented in simplified, abbreviated form, in terms of a
seems likely that bursts develop from large ripples as vertical stratigraphic profile, which sets out a typical
the ripples grow in size, and that these then domi- sequence of lithologies with their accompanying
nate the structure of the flow and become the princi- biofacies and sedimentary structures (e.g., Visher
pal mechanism in bedform production (Leeder 1965a,b). This practice developed from the applica-
1983). Sediment that is lifted into suspension by the tion of Walther's law, which states that only those
turbulent bursts travel downstream a maximum of facies that can be found forming side by side in
about five flow depths before deposition takes place. nature can occur in contact with one another in
Continued bursting, erosion and deposition will vertical succession, unless the succession contains
eventually cause the bedforms to adjust to a relation- internal erosion surfaces (Middleton 1973). The ver-
ship between bedform wavelength, height, and flow tical profiles are cyclic, in the sense that they are
depth. A stable dune bed will gradually develop in repeated several or many times in succession, gener-
which macroseparation as well as burst macrotur- ally with some internal variation. The model may
bulence play an important role in determining dune also contain information about lateral variability,
morphology and magnitude (Leeder 1983). particularly with reference to the position of a shore-
The transition to upper-stage plane beds may line or sediment source area and the direction of
take place because the high grain concentration that prevailing air or water currents. Most models were
develops at high flow speeds blankets and inhibits based on a combination of observations from mod-
turbulence development. Also, the frequency of tur- ern environments and the ancient rock record.
bulent sweeps may overcome the scouring effect of In subsurface analysis, application of Walther's
leeside separation eddies. Saunderson and Lockett law enables us to interpret lateral facies relationships
(1983) reported a series of flume experiments car- from their vertical sequence as expressed in a facies
ried out to examine this transition. They found that a model. Several textbooks appeared during this pe-
mixture of flat-bed and dune-like conditions oc- riod which provided catalogues of vertical profiles as
Fig. 2.23. Bedform phase diagram showing how

bedforms may be classified according to the physical
processes which control their size limits and stability.
(Leeder 1983)
N
I 10
E
2
.
<II
<II
~
<II
.
;;
J:
on 1-0
"0
:;
.
"0
C.
c.
<l
o - I !----~~-~-----~--~~JO-
0-1 0-3 0-5 0 -7 0-9 I- I 1-3
GrO In dlo meTer.d (mm)
a basis for subsurface environmental interpretations Amongst the problems that had been recognized
(e.g., Pirson 1977; Berg 1986). Commonly, these were with the Markov chain method was the fact that it
identified not on the basis oflithologic logs in cores, contained no information on the nature of facies
but from the profiles displayed by electric logs - the contacts. This is, of course, of critical importance,
classic "funnel-shaped" and "bell-shaped" gamma- following Walther's law. Cant and Walker (1976)
ray, spontaneous potential, and resistivity logs that attempted to address the problem by erecting a sepa-
had first been identified during the 1950s (e.g., Nanz rate facies state for scour surfaces that, in their field
1954). case, were typically accompanied by a distinctive
The value of a facies model in providing a synthe- poorly-sorted sandstone. Miall and Gibling (1978)
sis of a diverse array of facts is offset by the danger took a different approach, documenting the number
that uncritical use of the model may lead to a loss of of erosional and gradational contacts between each
information or to misinterpretation, because it is facies state and developing a "contact matrix" that
tempting to observe strata in terms of a precon- showed the preponderance of each type of contact in
ceived model. Objective field measurements com- tabular form. Although this was a systematic ap-
monly are difficult to make, particularly in areas of proach to the problem, the method has not been used
poor exposure. One of the trends in sedimentologi- by other workers because attention at this time be-
cal research in the 1970s was to quantify observa- gan to turn away from vertical profiles. As shown by
tions of lithological successions in order to enable Godin (1991), cyclic relationships can be developed
model sequences or cycles to be defined statistically. at several nested levels within fluvial systems, and
Markov chain analysis was commonly used for simple one-step Markov methods are not suitable for
this purpose (e.g., Allen 1970b; Miall 1973, 1977; analyzing these complexities (Chaps. 8, 10).
Cant and Walker 1976). Although this approach al- A somewhat different approach was taken by
lows more precision in defining the model, it can Friend et al. (1976), who measured many lengthy
cause valuable observational detail to be ignored. sections through the Devonian nonmarine sedi-
Collinson (1978, p. 579) made this point about the ments of East Greenland. The sections were divided
Battery Point Formation, a highly varied, low-sinu- into 10-m segments, and the abundance of a range of
osity-fluvial deposit from which Cant and Walker facies attributes (e.g., grain size, sedimentary struc-
(1976) extracted a single "summary" cyclic sequence tures) was determined for each segment. The data
(Fig. 2.24a). were then subjected to factor analysis to determine
Fig. 2.24. Comparison of four vertical profIles,

showing how similar fming-upward cycles may
FI be produced in several different ways. a Model
cyclic sequence; Battery Point Formation (De-
vonian), Quebec, probably formed by vertical
bar aggradation in a low-sinuosity (braided)
river (Cant and Walker 1976). b Sequence
s m9
c f c formed by lateral accretion on a point bar;
Lli..1.U
Castisent Sandstone (Eocene), Spain (Nijman
and Puigdefabregas 1978). c Sequence formed
by lateral accretion within a point bar; modern
Amite River, Louisiana (McGowen and Garner
1970). d Sequence formed by vertical aggrada-
tion and progressive channel abandonment on
an alluvial fan, under conditions of tectonic
quiescence (allogenic control); Upper Carbon-
iferous coal measures, northern Spain (Heward
1978a). Diagram from Miall (1980)
(a) (b) (c) (d)
COMPARISON OF
MEDIUM SCALE
sm\l
c f c COARSENING UP
.--_ _UJ...llJ.., CYCLES (3-llm)
Fig. 2.25. A Progradation and vertical

aggradation in distal part of an alluvial
fan, Upper Carboniferous coal measures,
northern Spain (Heward 1978). B Pro-
gradation and vertical aggradation in
proximal part of an alluvial fan, Devonian
of Hornelen Basin, Norway (Steel and
Aasheim 1978). C Vertical aggradation on
a sandur plain in front of an advancing
glacier. Sequence is capped by till (not
..... ...... shown); Pleistocene of Denmark (C .
Heinberg, pers. comm., 1979). D Cycle
.- .. ..""':;.&" .. .
""" .4;.
.~
formed by bar progradation or by fIll of a
..&
channel deepened by progressively melt-
..& ~ ing ice in a glacial retreat phase; Pleis-
tocene of Ontario (Costello and Walker
8 C
A D 1972). Diagram from Miall (1980)
associations ("reaction groups") between the vari- tions as all other techniques that rely on the vertical
ous facies attributes, and the segments were com- profile - it is not concerned with three-, or even two-
pared using cluster analysis of the reaction groups, dimensional facies architecture.
in order to define lithofacies associations. These The proliferation of facies studies led Dott and
could then be expressed graphically as lithologic logs Bourgeois (1983) to remark that by the early 1980s
of typical segments from each association. The re- fluvial facies models had "multipled like rabbits so
sults revealed a wide range of sedimentary styles, not that every real-world example now seems to require
necessarily cyclic in nature. Although quantitatively a new model. Such proliferation defeats the whole
rigorous, this method suffers from the same limit a- purpose of the conceptual model by encouraging
channel-avulsion events (an allogenic process) or

whether it can be accomplished by autogenic lateral
channel migration. Information on lateral variabi-
lity of the section (not available in this case) may
supply some clues.
s m g
c f c The cycles in Fig. 2.24b,c are unquestionably
UJ...l.J..J autogenic in origin. Figure 2.24b shows autogenicity
on two nested levels. The smaller cycles may be
seasonal in origin or represent infrequent floods,
whereas the overall cycle represents a migrating
point bar, as does Fig. 2.24c. Figure 2.24d, which is
similar to the others in thickness and fining-upward
character, represents the deposits of shallow allu-
vial-fan distributaries in an environment dominated
by tectonic controls. The basin in which it formed
was a tectonically active one, with differential move-
ment between basin and source area adequate to
create and maintain a set of alluvial fans at the basin
margin.
The question of scale must also be considered
when comparing the columns in Fig. 2.24. Diagrams
A, B, and C represent relatively large rivers, with
channel depths of several meters. Diagram D repre-
sents the deposits of fan distributaries, which prob-
ably were much shallower. Autogenic cycles, related
to such causes as channel migration and bankfull
A
floods, would be on a correspondingly smaller scale.
Fig.2.26. Stacking of three hypothetical point-bar succes- Several thin fining-upward cycles, of probable auto-
sions under conditions ofA high andB low subsidence rate genic origin, can be discerned in diagram D, nested
relative to aggradation rate. InA, complete successions are
within the overall fining-upward succession.
preserved intact, together with overlying floodplain units.
In B, upper point-bar and floodplain units have been The coarsening-upward cycles of Fig. 2.25A,B re-
removed by deep scour, and the resulting succession sulted from rapid progadation following pulses of
is very similar to those produced in some low-sinuosity relative uplift on the source areas of alluvial fans in
multiple-channel rivers, particularly those of South two different basins; they are therefore of tectonic
Saskatchewan type. (Miall1980)
origin. They are similar to the succession of Fig.
2.25C, in terms of an upward increase in grain size
excessive pigeon-holing, which obscures rather than and scale of sedimentary structures. Cycle C, how-
reveals whatever unity may exist among the vari- ever, represents a climatic effect. During periods of
ants". glacial advance, envirol,1ments in fluvioglacial rivers
There are two main kinds of problem with the are shifted progressively downstream, so that at a
vertical profile as a precise analytic tool: (1) in some given point the fluvial setting changes from distal to
cases, allogenic and autogenic processes can lead to proximal. The result is a coarsening-upward succes-
the same vertical succession; (2) similar vertical pro- sion. The example illustrated is from a Pleistocene
files can be generated in rivers of different channel succession. Other examples were quoted by Miall
pattern by different autogenic processes. These (1980).
points were explored at length by Miall (1980) from The question of preservation is also of impor-
whom Figs. 2.24, 2.25, and 2.26 were taken. tance in interpreting vertical profiles. Consider Fig.
The cycle shown in Fig. 2.24a may be of autogenic 2.26, which shows the effects of variation in subsid-
or allogenic origin, or a combination of both. It ence rates in a basin undergoing deposition by high-
probably represents the filling and progressive aban- sinuosity streams. The main sedimentary process is
donment of a relatively deep, braided channel sys- the formation of coarse channel-belt deposits by
tem, but it is unclear whether cyclic repetition of this lateral accretion. During periods of rapid subsi-
type of sequence requires tectonic triggering of dence, these will be rapidly buried and preserved,
largely intact (Fig. 2.26A), but during intervals of some data illustrating the validity of the concept
slow subsidence most of the floodplain deposits and (Fig. 2.10).
the upper, finer part of the point-bar succession, Brookfield (1977) discussed the concept of an
may be planed off, and the result would be multi- eolian bedform hierarchy and tabulated the charac-
story coarse units that preserve only the evidence of teristics of four orders of eolian bedform elements:
high-energy channel-floor sedimentation (Fig. draas, dunes, aerodynamic ripples, and impact
2.26B). This part of the succession may well be inter- ripples. These four orders occur simultaneously, su-
preted as the product of a braided stream system, perimposed on each other. Brookfield showed that
because of the relatively high proportion of sandy this superimposition resulted in the formation of
channel deposits, and the paucity of floodplain fines. three types of internal bounding surfaces. His first-
The fluctuation in fluvial style might be attributed to order surfaces are major, laterally extensive, flat-
climatic or tectonic causes. Allen (1978, p. 145) ar- lying, or convex-up bedding planes between draas.
gued that in a basin undergoing a slowly decreasing Second-order surfaces are low to moderately dip-
subsidence rate this effect could be the cause of ping surfaces bounding sets of cross-strata formed
large-scale coarsening-upward (or at least sandier- by the passage of dunes across draas. Third-order
upward) cycles (cycles similar to those in Fig. 2.26B surfaces are reactivation surfaces bounding bundles
overlying those more like 2.24a), whereas such cycles of laminae within cross-bed sets and are caused by
are commonly attributed to progradation caused by localized changes in wind direction or velocity.
increased differential movement, such as the coars- (mesoforms to microforms).
ening-upward successions of Steel and Aasheim Similar hierarchies of internal bounding surfaces
(1978), Steel et al. (1977), and Heward (1978a). The have been recognized in some subaqueous bed-
stacking and separation of channel bodies and pres- forms. A particularly useful approach was that of
ervation of intervening fines are discussed in Chap. Jackson (1975). Building on Russian work, he de-
9, and the interpretation of allogenic cyclicity is dis- fined three classes of bedforms: microforms (e.g.,
cussed further in Sect. 1l.3.6. ripples), mesoforms (e.g., dunes, sand waves, pro-
duced by "dynamic events", such as floods), and
macroforms (longer-term geomorphic products,
2.4.3 Fluvial Architecture such as point bars, sand flats, eolian draas). Reacti-
vation surfaces (Collinson 1970) may develop in
2.4.3.1 Architectural Scale and the Bounding-Surface Concept bedforms as a result of changes in stage or flow
direction or under conditions of constant stage,
As noted by Allen (l983a, p. 249): where bedforms randomly interact (McCabe and
Jones 1977). They may also separate individual
"The idea that sandstone bodies are divisible internally smaller-scale elements within a larger unit, such as
into 'packets' of genetically related strata by an hierarchi-
cally ordered set of bedding contacts has been exploited the component packages of mesoforms within a
sedimentologic ally for many years, although not always in macroform. They have been documented in small-
an explicit manner. For example, McKee and Weir (1953) scale (Collinson 1970) and large-scale fluvial cross-
distinguished the hierarchy of the stratum, the set of bed sets (Jones and McCabe 1980; Fig. 2.27) and in
strata, and the coset of sets of strata, bedding contacts tidal sand waves (Allen 1980). In the latter case, the
being used implicitly to separate these entities."
sand waves rest on horizontal surfaces analogous to
In order to trace the development of these ideas, it Brookfield's first-order surfaces and contain sec-
is necessary to reach back to the 1960s. Allen (1966) ond-order reactivation surfaces that relate to indi-
showed that flow fields in such environments as vidual tidal cycles. Third-order surfaces bound
rivers and deltas could be classified into a hierarchi- individual cross-bed sets.
cal order. His hierarchy was designed as an aid to the Brookfield's (1977) development of the relation-
interpretation of variance in paleocurrent data col- ship between the time duration of a depositional
lected over various areal scales, from the individual event, the physical scale of the depositional product,
bed to large outcrops or outcrop groups. The hierar- and the geometry of the resulting litho some was a
chy consists of five categories, small-scale ripples, major step forward that has been of considerable use
large-scale ripples, dunes, channels, and the inte- in the analysis of eolian deposits. Brookfield (1977),
grated system, meaning the sum of the variances Gradzinski et al. (1979), and Kocurek (1981) showed
over the four scales. Miall (1974) added the scale of how these ideas could be applied to the interpreta-
the entire river system to this idea, and compiled tion of ancient eolian deposits. Kocurek (1988) has
Growth of Present -Day Concepts, 1978-1988 43
Fig. 2.27. Giant cross-bedding in a Carboniferous fluvial-deltaic sandstone. Surfaces A, B, and C are reactivation
surfaces within cross-bed set 4. (Jones and McCabe 1980)
cross-bedding, and related them to changes in

N
bedform orientation and to stage changes in the
river. A similar type of analysis was performed by
Haszeldine (1983a,b) on the bounding surfaces
within a Carboniferous sand flat deposit. Bridge and
Diemer (1983) and Bridge and Gordon (1985) re-
ferred to major and minor bounding surfaces within
Paleozoic fluvial sequences (Fig. 2.29). The major
surfaces are typically horizontal and planar or
slightly concave up and enclose tabular sheets repre-
senting channel-fill successions.
Allen's (1983a) study of the Devonian Brown-
stones of the Welsh Borders represents the first ex-
plicit attempt to formalize the concept of a hierarchy
of bounding surfaces in fluvial deposits (Fig. 2.30)
and makes reference to Brookfield's work in eolian
151, 2nd, 3rd, = Ch,lnl\Cl (Orders)
strata as a point of comparison. Allen described
1st, 2nd, 3rd, = Bu (Orders)
three types of bounding surfaces. He reversed the
Fig. 2.28. The hierarchy of channels and bars in Donjek order of numbering from that used by Brookfield
River, Yukon. (Williams and Rust 1969) (1977), such that the surfaces with the highest
number are the most laterally extensive. No reason
was offered for this reversal, but the result is an
found that first-order surfaces include two types of open-ended numbering scheme that can readily
surfaces, the most laterally extensive of which he accomodate developments in our understanding of
termed super-surfaces. Characterization of eolian larger-scale depositional units, as discussed below.
deposits therefore now requires a fourfold hierarchy First-order contacts, in Allen's scheme, are set
of bounding surfaces. boundaries, in the sense of McKee and Weir (1953).
Several workers have attempted to develop a Second-order contacts "bound clusters of sedimen-
breakdown of the range of physical scales present in tation units of the kinds delineated by first-order
fluvial deposits. Williams and Rust (1969) proposed contacts." They are comparable to the coset bound-
an ordering of the scales of channels and bars in the aries of McKee and Weir (1953), except that more
modern Donjek River, Yukon (Fig. 2.28). Campbell than one type oflithofacies may comprise a cluster.
(1976), in an analysis of the Westwater Canyon Allen (1983a) stated that "these groupings, here
Member of the Morrison Formation in New Mexico, termed complexes, comprise sedimentation units
recognized several scales of fluvial sequence that that are genetically related by facies and/or paleo-
occurred in tabular channel-fill sandstone bodies of current direction." Many of the complexes in the
a range of dimensions (although his work has now Brownstones are macro forms, in the sense defined
been shown to be largely incorrect: Sect. 10.3.2). by Jackson (1975). Third-order surfaces are compa-
Jones and McCabe (1980) described three types of rable to the major surfaces of Bridge and Diemer
reactivation surfaces occurring within sets of giant (1983). No direct relationship is implied between
minor ;~ljl~~~;~~~~~~~~~~
erosion
surface
major
erosion
surface
minor :---+-1S;;;;::~~~~~~s;:;~~~~t1
erosion
surface
Fig. 2.29. Schematic geometry of a Devonian-Carboniferous channel sand body, Ireland, showing architectural subdivi-
sions proposed by Bridge and Diemer (1983)
~ Mudston,-clast Itringers (GI) ~ ·Troullh cross-bedded sandstone (S3)
r:~\ij Masslv. intraformational canlliomerate (G 2 )

~ Cross-bedded sandy conlliomerat. (G 3 ) ~ Plone-b.dded sandstone (S4' S~ ,
~ Cross-bedd.d pebbly .ands~an. (S,) . . Mudstone (M)
~ Cross-bedded landstone (S21 5- Position in depositional sequence

~,-(f)- Order-number in contact hierarchy
Fig. 2.30. Schematic summary of the lithofacies and bounding surfaces in the Devonian Brownstones of the Welsh
Borderlands. (Allen 1983a)
Allen's three orders of surfaces and those of together of some dissimilar types of surfaces, re-
Brookfield because of the different hydraulic behav- sulting in a loss of significant information on
ior and depositional patterns of eolian and aqueous macroform geometry. A more elaborate and com-
currents. prehensive classification was developed as a result
Miall used Allen's (1983a) classification of (Miall 1988a,b), consisting of a sixfold hierarchy of
bounding surfaces in his first synthesis of architec- lithosomes and bounding surfaces. This is the sys-
tural-element analysis (Miall 1985), but soon found, tem used in this book (with additions) and is des-
in actual field studies that it resulted in the grouping cribed in Chap. 4.
2.4.3.2 Alluvial Basin Architecture gested that channel migration patterns could be ana-
lyzed by studying paleosol maturity. The "pedo-
The next step beyond a reconstruction of the channel facies" of the paleosols matures with time, indicating
style of a river system is an attempt to reconstruct the the extent of exposure and distance from channel
basin architecture. As noted by Friend (1983), an sediment source (Fig. 2.31). These concepts are dis-
influential set of diagrams was that published by cussed further in Chaps. 7 and 10. Fourthly, coal
Allen (1965a) and reproduced here as Fig. 2.17. seams have a similar local utility for local litho-
These diagrams show a variety of styles of alluvial stratigraphic correlation.
stratigraphy that reflect the patterns of channel Very few studies of carefully documented re-
aggradation and migration of various river types. A gional alluvial architecture have been published.
great deal has been learned about fluvial styles since Those by Behrensmeyer and Tauxe (1982) and
these diagrams were published. Behrensmeyer (1987) have been referred to above,
In order to reconstruct alluvial stratigraphy it is and one set of reconstructions is illustrated in Fig.
necessary to carry out regional surface or subsurface 2.32. Campbell (1976) and Blakey and Gubitosa
correlation of outcrops or well sections. Correlation (1984) attempted detailed, large-scale architectural
is commonly a problem in fluvial deposits because of reconstructions. That by Blakey and Gubitosa (1984)
rapid lateral facies changes, a lack of regional marker described paleovalley-fill styles, and examined the
beds, and numerous internal erosion surfaces (chan- causes of architectural variation, such as regional
nels and interfluve surfaces). Four types of medium- changes in subsidence patterns. Campbell (1976)
to long-distance correlations can be carried out un- provided a reconstruction of what has been inter-
der certain conditions, however, by focusing on preted as a large-scale, braided, alluvial distributary
floodplain deposits (see Sect. 9.S for a more exten- system, in which he defined large-scale channel belts
sive discussion of these modern mapping methods). and several scales of channel filling. However,
First, magnetostratigraphy is proving to be extre- Cowan (1991) has shown that most of Campbell's
mely valuable for providing local chronostrati- (1976) outcrop units were based on weathering char-
graphic datum planes, particularly in the more acteristics that reflect diagenetic oxidation and
laterally extensive and possibly less episodically pre- groundwater leaching, not primary depositional ar-
served floodplain sequences. Behrensmeyer and chitecture (Sect. 1O.3.2).
Tauxe (1982) and Behrensmeyer (1987) provided A useful approach to channel classification was
excellent examples of architectural interpretation developed by Friend et al. (1979) and Friend (1983).
based on this type of control. They were able to Friend (1983) defined channels as "elongate depres-
distinguish and relate to each other interbedded sions in the alluvial surface, with more or less clearly
trunk-river and tributary deposits and to calculate defined margins or banks between which the river
local sedimentation rates. flow is restricted for most of the year". In many
Second, the same kind of refined lateral control rivers, there are channels and other types of "hollow"
may be possible using tuff beds. These are rarely of more than one size (tributaries, crevasse channels,
preserved in channel deposits because of the high chute channels, scour hollows), all of which will
transport energy, but may form widespread marker show variations in size and morphology. A careful
horizons on the floodplain. Allen and Williams analysis of these channels is an essential component
(1982) provided an example of a unit less than 30 m of a facies analysis, and Friend (1983) provided three
thick that can be subdivided into seven allostrati- diagrams to assist in this work (Figs. 2.33, 2.34, 2.35).
graphic units (these are "chronosomes", to use the Note that this classification is strictly descriptive.
term suggested by Schultz 1982), permitting a very No attempt need be made, until the final interpreta-
detailed reconstruction of architectural evolution. tion, to assign terms such as "braided" or "anasto-
Paleosols are the third type of floodplain marker mosed" to the deposits. Sheet flood deposits are,
horizon. These can be distinguished from each other strictly speaking, unchannelized by definition. They
on the basis of petrological and geochemical signa- are characteristic of many distal braid-plain depo-
tures, and have the added advantage of yielding a sits, as discussed in Chap. 8. Fixed channels are
considerable amount of evidence for local environ- typical of anastomosed rivers. They typically are de-
mental conditions, sedimentation rates, and local scribed as "ribbon" sandstones, and have width!
flooding and avulsion frequencies (Bown and Kraus depth ratios of less than 15. Some ribbons have
1981, 1987; Kraus 1987). They may even be detect- "wings" of sandstone extending laterally from the
able in seismic-reflection data. Kraus (1987) sug- top of the channel margin. These are probably levee
Fig. 2.31. Development of compound pedo-

facies sequence. Arabic numerals indicate de-
velopmental stage (maturity) of paleosol. Ro-
man numerals indicate depositional sequence
of channels. Note initial development of in-
creasingly mature pedofacies at left side of area
Avulsion as channels migrate laterally away to right (A,
B). Less mature paleosols cap this succession as
the channel migrates back to the left (C). (Kraus
3-IV
1987)
,
2- 111
2-11
1-1
Channel Sandston,
deposits. Mobile-channel belt deposits include Miall (1985) attempted to synthesize these new
braided and meandering channels, and the various data into an improved method of field facies analysis
types of channel deposits intermediate between and classification, termed "architectural-element
these well-known end members. The deposits are analysis". He showed how eight basic elements could
sheet-like in geometry, with width/depth ratios ex- be combined to form a variety of fluvial styles, but
ceeding 15, and commonly greater than 100. emphasized the limitation of the fixed-model ap-
In the 1980s, several important studies docu- proach. To some extent, every river and every an-
mented in considerable detail the internal architec- cient fluvial succession is unique. Miall (1985)
ture of ancient complex channel-fill and bar depo- suggested, therefore, that modeling might be simpler
sits. Okolo (1983) and Hopkins (1985) described at the level of the component elements, which did
some well-exposed channels filled by vertical aggra- seem to show some constant characteristics through
dation. Allen and Matter (1982) and Okolo (1983) rivers of all kinds.
described other, more complex, channel fills. Other Channel morphology changes downstream in re-
writers, including Allen (1983a), Haszeldine sponse to changes in valley slope, sediment load,
(1983a,b), Kirk (1983), and Ramos and Sopena bank materials, climate, or tectonic regime, and the
(1983) provided detailed two- and three-dimen- same controls may cause changes through time in
sional cross sections oflarge compound bars, reveal- the morphology of a particular river reach (Schumm
ing a range of complex structures enclosed by and 1969; Burnett and Schumm 1983; Peterson 1984;
containing major bounding surfaces. These studies Carson 1984a,b,c). It is therefore unwise to assume
provided the first data on the internal geometry and that fluvial style will remain constant throughout a
facies architecture of compound bars comparable to given stratigraphic unit.
the modern side bars of Collinson (1970) and the Modern techniques for basin correlation in
sand flats of Cant and Walker (1978). These are nonmarine strata are discussed in Chap. 9. Current
macro forms (Jackson 1975), comparable to, andgra- studies of alluvial architecture are at the center of a
dational with, point bars, in terms of their scale, debate in sequence stratigraphy regarding the inter-
complexity, and depositional longevity. The discov- pretations of base-level change and tectonism that
ery of the truly three-dimensional nature of these can be made from alluvial architecture. This is dis-
deposits was one of the most significant contribu- cussed at length in Chaps. 11 and 13.
tions to fluvial studies of this period.
o 2 km
~!----------------~--------------~I
... .... ..... .......... .. ..... .......................... .......................... ... ..... ........ .
' ,'
....
. .. ...... ....... .......
.. .. . ....
. .... ......... . .
' '
. . ' ... ...............

, '
, , , , ,
." , . .
Fig. 2.32. Five lateral architectural sections through the mit estimates of sedimentation rate (shown at left). Coarse
upper Miocene Siwalik Group of northern Pakistan. Later- stippling Major channels; fine stippling minor channels;
ally continuous pedogenic horizons and magnetic reversal black dots vertebrate localities. (Behrensmeyer 1987)
stratigraphy were used for correlation purposes, and per-
2.4.4 Fluvial Styles duced in Chaps. 5-8, where the modern framework
of fluvial facies analysis is formally documented.
Some of the developments in facies studies that oc- The gradual extension of fluvial research into less
curred during the 1970s and 1980s (and more re- well-studied regions of the earth, including the Ama-
cently) are summarized in this section. Many useful zon basin, southern Africa, and Australia, has led to
illustrations were published by the authors to whom a considerable broadening of our ideas regarding
reference is made below. Some of these are repro- fluvial style. Baker (l978a) showed that many of our
i) ChcmeI defi1itioo ii) 0mneI
variety
iiil Field de5CrJ>tion hollow UI
2d ~-
~".' • • •••• ~dcn
~t nmjn base/margi1
;,. ) Genetic ardysis fA hollows

2d .'. '': :.:: .:..:
~
? sub.Dqoe<lS scar (/ 3d
slrucIu!es 8\j fuIes C- Go
~. '. :-'.< . ' .;c::::;

. ~ .......... . F-
---~
·:::···· ·~~km . ~cuttxril
? ctuneI
? volley
rricro/meso forms /
Fig. 2.33. Channel definition and recognition. (Friend 1983)
I ) Wos flow chonnelised?

Don't know conceptions concerning fluvial geomorphology are
No - -_____ ISheetflOOdI
_ _ _ _ _ yes appropriate only under certain climatic conditions,
particularly temperate climates, and that rivers can
2 ) Describe cImnel perimeter
show quite different patterns of sinuosity, channel
Size -width_ surfoce
~
erosion
~Idepth mfOlmum
shape, and sediment type under tropical conditions,
a
~
Form width I depth rotio
width depth and where rivers flow from one climatic zone to
another. The distinctive climate of Australia has led
~ to the development of several unusual, or even
bO'lks,
mergin ~ ~ .,'>-. unique, fluvial styles within that continent. Some of
steep gentle mud sand these were reviewed by Fielding (1993b), in a volume
3)
t where much new work on these rivers was presented
Chav1e1 behaviour
(Fielding 1993a). A useful classification of fluvial
i ) channel stobie between
switches
ii) chO'lnel steody loterol migrotion floodplains, reflecting a geomorphological perspec-
tive, was provided by Nanson and Croke (1992), and
~ ~
is discussed in Chaps. 7 and 8.
!f .i ed chomeI! !mobile channel be" - meandering I There has been considerable debate regarding the
ribbon bodies sheet bodies
classification of alluvial fans, especially those that
prograde directly into standing bodies of water (fan
iii) channel steady migrotion . O'ld switching
deltas). It is now generally accepted that alluvial fans
/' /1
vary widely in size and may include a wide variety of
facies assemblages. Recent work on this subject is
!mobile channel bell - me<Jnderinc;I or braided I discussed in Sect. 8.3.
sheet bodies Schumm's (1963) classification of rivers into
Fig. 2.34. Key for classification of channel behavior. bed-load, mixed-load, and suspended-load types
(Friend 1983) has been very influential. A recent manifestation
of this classification is given as Fig. 8.6. It was
used as a basis for a classification of fluvial styles
by Galloway (1981) in his analysis of Cenozoic
fluvial systems of the Gulf Coast of the United States
(Fig. 2.36).
Growth of Present -Day Concepts, 1978-1988 49
SEDIMENT Fig. 2.35. Model of distinctive types of alluvial

(0) (b) architecture. (Friend 1983)
MAINLY COARSE COARSE AND FINE
(BED-TRANSPORT (BED-TRANSPORT At-.()
DEPOSITS) SUSPENSION DEPOSITS)
SHEET
FIXED CHANNEL
(not known)
MOBILE -CHANNEL
BELT
D Coorse _Fine sediment
2.4.4.7 High-Sinuosity Rivers derably during the decade up to 1988. Modern ex-
amples have been studied by Gustavson (1978) and
Jackson's (1978) critique offacies models was a ma- Forbes (1983). Point-bar structures are difficult to
jor contribution to fluvial studies. Preliminary work define in the deposits because of the complexities of
by him led to a tentative classification of high-sinu- small- to medium-scale bedforms and minor chan-
osity rivers (those with a sinuosity> 1.5) into five nels. Well-exposed ancient examples of gravel-bed
classes, based mainly on the grain size of the sedi- rivers were described by Ori (1979) and Ramos and
ment load. Several of these categories were poorly Sopena (1983). These examples contain gravellat-
known at the time of writing, but various studies of eral-accretion deposits, indicating that the rivers
modern rivers and ancient deposits have confirmed, contained sinuous channel reaches.
with some modifications, Jackson's original work. Rivers with medium to coarse sand or pebbly
These developments have been integrated into the sand beds have been studied by several workers.
architectural studies ofMiall (1985, 1988a,b) and are Such rivers are similar to the classic examples de-
discussed in Chap. 8. Only the main points are noted scribed by Bernard and Major (1963), McGowen and
below. Garner (1970), Puigdefabregas (1973), and Jackson
It must be emphasized that high-sinuosity (1976a,b), and more recent work has not introduced
reaches can occur in almost any fluvial setting. Such any major modifications to these earlier models.
reaches tend to migrate laterally, developing point Jackson (1976a) described the development of the
bars with the well-known lateral-accretion internal helical flow pattern in meander bends of the Wabash
structure (epsilon cross-bedding of Allen 1963a). River, Illinois. He showed that there is typically a
The occurrence of this structure is therefore not the transition zone in the upstream part of most mean-
simple diagnostic criterion of fluvial style that it is der bends, in which the helical flow is reversing itself
often assumed to be, but must be evaluated in con- from the previous bend. In this reach, the typical
text with the other associated facies. point-bar sorting process does not occur, and fining-
Gravel-bed, high-sinuosity rivers are not com- upward successions are not developed.
mon, but our knowledge of them improved consi-
CHANNEL COMPOSITION OF CHANNEL GEOMETRY INTERNAL STRUCTURE LATERAL
TYPE CHANNEL FILL CROSS SECTION MAP VIEW SANO ISOLI TH SECIMENTARY FABRIC VERTICAL SEOUENCE RELATIONS
lnumIHwlllmlWI
.J
. <:::______7
W
Z
Z
l: IoII1Q111W11fflr1/dtrptt1rorlo
u
o I.ow \10 ft'IIOdtf-ott ,.10,. 011
bcIwIlCOIoIf Wf toc.
~
w
CD
) f( cb i!;
Sf> l.tTk
.J
-
.
'"zz
t::::)9
~ €f27
!t~\
l:
u
--
0 Mutt"IIOI"f~ I I,t ...
~
0
W
... ...., ~, ~II, onG IIIud ~.ICI\tI / CltfJII'I
"'"
Hqlrf'l"'-OI'IboIGIKCIIII s' ......
6I)nlI OI'CI I)K OCCft 1101'l
tIo'Il'I
"""",1\',11
"ftwf~ 'II ,...,. .
"Oft' l .,olt~ . 1ICI

orol.1K wt41 ~
.......
9f\'If'f'OI1yMbot'cl~1O
""'f'CNfIid'l'Q~tIO,.
)( ColflClo'l.. , 'Jp.cOII~ " bHdecI"

i flo,h
.J
..is
W
Z
Z
o
~ 00trw0tIt, ~II ona I'PkIII ~ to .,...) ro. width I
diIOfI't'CIhO
-'..
MuIM.'Ot y cf'lcJnt'tt l tilts
o HtQh-rIfI;.t KOUI' .... tII tnCOM ~ ~~tIc
~ """"
Fig. 2.36. Classification of fluvial styles, following the fluvial-geomorphological classification of Schumm (1963), as
given in the left-hand column. (Galloway 1981)
Modern, fine-grained rivers with a fine-grained 2.4.4.2 Low-Sinuosity Rivers

sediment load have been described by Nanson
(1980) and Jackson (1981), and ancient examples by Few studies have been carried out on modern, mul-
Miall (l979a) and Stewart (1983). Point bars consist tiple-channel, low-sinuosity (braided) rivers during
of simple epsilon sets with banks or benches formed the decade 1978-1988. Blodgett and Stanley (1980)
as a result of flow separation eddies on tight bends and Crowley (1983) reexamined the Platte River,
(Nanson 1980; Nanson and Page 1983; Fig. 2.37). long regarded as the classic sandy-braided river,
Subordinate bedforms on the epsilon foreset are following the work of N.D. Smith (1970, 1971, 1972).
small or absent. Point-bar dip is steep (as much as Both of the papers focus on the linguoid "bars"
25°). D.G. Smith (1987) reviewed fine-grained point which characterize much of the lower Platte River.
bars, including those deposited in tidally influenced, Crowley (1983) demonstrated that in at least some
estuarine environments, and developed three facies cases these bedforms form part of much larger struc-
models. tures 200-400 m in length and 0.7-1.5 m high, which
Thomas et al. (1987) added a useful new acronym, he interpreted as macro forms. These ideas are reex-
IHS, which stand for inclined heterolithic stratifi- amined in Chap. 6. Other depositional processes
cation (Fig. 2.38). Observations on a variety of mod- were described by Bluck (1979, 1980). He demon-
ern meandering rivers in western Canada, and strated that coarse bar deposits commonly develop
several ancient (Cretaceous) deposits in the Alberta preferentially in the upstream part of bends and may
Basin, led to the recognition of a distinctive style of migrate downstream. This process produces coars-
lateral accretion, characterized by alternating fine- ening-upward sequences, the reverse of that usually
and coarse-grained beds. Tidal influence is sus- expected in fluvial deposits.
pected in many cases, as in estuarine distributary The other major sandy-braided fluvial model is
channels. that developed for the South Saskatchewan River by
Cant and Walker (1978), and applied by them to an
earlier study of the Battery Point Sandstone, Quebec,
A
\ by Cant and Walker (1976). This model is illustrated
in Fig. 2.20, and was used by Miall (1978c), together
with that of the Platte River, as the two major classes
of sandy-braided river for the purposes of defining
vertical-profile facies models. Cant and Walker
(1978) described the development of "sand flats" by
the process of nucleation of small sand bars, which
then grow by lateral and downstream capture and
turning of migrating transverse and linguoid bed-
forms. However, in their work, they did not consider
the internal architecture of the sand flats, and their
diagrams imply that the structure is tabular. Subse-
quent architectural work by Miall (1988a,b,c), dis-
CholPnel Mlgr.1tlOll Shear
cussed in detail in Chap. 6, shows that the sand
Zone flats contain significant internal three-dimensional
structure.
N.D. Smith and D.G. Smith (1984) described the
William River, a suspended-load stream showing
steady discharge that becomes characteristically
braided, with linguoid bedforms and sand flats simi-
lar to those of the Platte River, where abundant sand
Fig. 2.37. The formation of a flow-separation zone and silt bedload is added by eolian processes.
banks on a point bar. (Nanson 1980)
Bristow (1987) carried out a reconnaissance of the
Brahmaptura River, one of the world's giant braided
rivers. Earlier work on this river by Coleman (1969)
was of considerable significance in establishing
the style of bedform generation in deep, vigorous
•
KEY TO SYMaOLS
Mud
D', ' '.': Sand
Interlaminate d
.~ sand and mud
Ed Inclined strata ®
Intraformational
~ conglomerate
~ Trough cross-beds
g Erosional
Gradational
l Contact ' " . #
Fig. 2.38. Nomenclature for the description of inclined showing the three types of IHS geometry; E dip section,
heterolithic stratification (IHS). A Vertically stacked sets, showing three types of geometry of discontinuous lamina-
each individually may be 1-30 m thick; B individual IHS tion (1-3); F plan view of bar showing three possible types
units, comprising graded couplets 1 cm to several meters of along-strike continuity (1-3). (Thomas et al. 1987)
in thickness; C imbricate sets; D composite sets; IHS set
channels. Bristow's (1987) main contribution was to posed of fine-grained sediment (silt/clay) and veg-
document the growth of the large bar forms etation ... Separating the channels are floodplains
(macroforms), both bank-attached, lateral and point consisting of vegetated islands, natural levees, and
bars, and midchannel bars. Lateral and downstream wetlands. Occasionally, crevasse channels and splay
accretion was demonstrated to be of considerable deposits occur in wetland areas" (D.G. Smith and
importance. He also demonstrated that the giant N.D. Smith 1980). Channels have low width/depth
"sand waves" described by Coleman (l969) have ratios « 15) and, for the most part, aggrade verti-
downstream accretion faces that dip at only 4°. They cally. They may correspond to the ribbon channel
are, therefore, not the simple avalanche-face bed- bodies of Friend et al. (1979) and Friend (l983).
forms that many authors have assumed them to be, D.G. Smith (l983) focused on anastomosed rivers
but are probably complex macro forms, comparable in humid climates. Rust (1981), Rust and Legun
to the downstream-accretion deposits described in (l983), and Nanson and Rust (l985) showed that a
Chap. 6. similar fluvial style could occur in arid regions, but it
The first attempts at a systematic documentation has now been shown that these rivers are probably
of the architectural details of a sandstone succession anomalous (Sect. 8.2.10).
were those of Allen (l983a) and Ramos et al. (l986). Several misconceptions regarding anastomosed
Their classifications of architectural elements are rivers have been clarified by later work in this envi-
shown in Figs. 2.39 and 2.40. ronment. For example, because the channels do not
sweep across their floodplains, as do braided and
meandering systems, wetland areas were thought by
2.4.4.3 Anastomosed Rivers D.G. Smith and N.D. Smith (1980) to be prime sites
for long-term peat accumulation, and several coal
A facies model for anastomosed rivers was first deposits were initially interpreted as the product of
brought to the general attention of sedimentologists anastomosed fluvial systems (D.G. Smith and
by D.G. Smith and N.D. Smith (1980). The term Putnam 1980; Leblanc Smith and Eriksson 1979;
"anastomosed" is used "for an interconnected net- Cairncross 1980; Flores and Hanley 1984). However,
work of low-gradient, relatively deep and narrow, it is now realized that the process of channel
straight to sinous channels with stable banks com- crevassing and avulsion, which is the normal style of
(e) ;:::;
~
-
.- ---
~
(d)
Fig. 2.39. The main types of depositional features in the channels; e lateral-accretion units; f symmetrical com-
Brownstones of the Welsh Borderlands. a Tabular layers of plexes (sand shoals) of lateral-accretion units with gravel
trough cross-bedding; b assemblages of down-climbing cores. (Allen 1983a)
(forward accreting) bar units; c minor channels; d major
SEODI NG ANO SEDIMENTARY

FACIES
STRUCTURES
® -..
--D-O--Q--:
..<$>--()-.. -o---,·=-C>~"
- _ - -..
I1ASS I VE 0" o~ Q de ~Q c;:Io 00_ 0 a
II1BRICATED CLASTS .<Js>o ~G'l 0
:::
'" ...'"...
i~
... ~ CRUDE FLAT- BEDD I NG
a
... a . ... .. b
J:
II1BRICATED CLASTS
'",
0' ~a
0 ...
:0: $) e 9 Q d -0
........
",0
...... '"z '"

.,;
"" ...
,.;0
COItYEX UPWARD TOPS

<=> ...
I HBR I CATED CLASTS
UN ITS OF TABULAR
CROSS- STRAT I FI ED
CONGlOI1ERATES
TA8ULAR CROSS - STRAT I FI ED .
GO
o
LATERAL ACCRET I ON UN ITS

...,:0:
WI TM SANDSTONE DRAPES
IMBRICATED CLASTS
a 00
NW
....
""'"o ><
...'"N ,..'"
• • V)
LATERAL AND VERT I CAL b

;;;~~
........
.
'".,;
... ~>-
ACCRET I ONARY SURFACES "" .,0<>
<C>z
..... 0<
"':0:'"
@
I1ASS I VE a
..
~
......
~
COI1PLEX - FILL
~~
STRATIFIED ., 0 , • b
, ...'" , --
.
;:ll!i
Z"'"
z'" TRANSVERSE FI LL
......
~i!i CROSS-STRATI F ICAT I ON
o
I1UL TI-STOREY FILL

d
TROUGH CROSS-STRATI FICATION
Fig. 2.40. The major depositional facies of the Buntsandstein, central Spain. (Ramos et al. 1986)
channel evolution in anastomosed systems (N.D. More recent studies of possible anastomosed chan-
Smith et al. 1989), leads to the influx oflarge volumes nel systems in the ancient record (Eberth and Miall
of clastic detritus into the floodplains (e.g., D.G. 1991; Nadon 1991, 1994) indicate that this is a mis-
Smith and Locking 1989), and therefore prevents the conception. Present ideas about this fluvial style are
accumulation of ash-free peat deposits, which is the discussed in Sect. 8.2.lO.
prerequisite of coal formation (McCabe 1984). The Thirdly, the anastomosed model was applied with
coal deposits referenced above will therefore need great enthusiasm to the interpretation of some
reevaluation in the light of this improved under- channelized petroleum-bearing sandstone bodies in
standing. the subsurface, notably the Mannville Sandstones
Secondly, the original facies model of D.G. Smith (Cretaceous) of the Lloydminster area, Alberta
and N.D. Smith (1980) overemphasized the idea of (Putnam and Oliver 1980; D.G. Smith and Putnam
vertical-channel aggradation, to the extent of pro- 1980). This interpretation was controversial, in part
posing that "walls" of channel-fill sand or gravel because the modern analogs of the channel sand-
could accumulate beneath an aggrading channel. stones undergoing interpretation seemed quite in-
appropriate, being small rivers in Banff National both the ribbon and sheet geometries of Friend
Park, in narrow mountain valleys undergoing degra- (1983).
dation, quite different from the large channels Deposits of some recent floods have been de-
within the broad, aggrading alluvial plain of the scribed by Lucchitta and Suneson (1981), Sneh
Alberta foreland basin (as revealed by sandstone (1983), and Stear (1985), and include useful data on
body geometries). Putnam (1993) has noted one of sediment characteristics. Parkash et al. (1983) de-
the major difficulties in relating surface facies stud- scribed the sedimentology of a modern low-gradient
ies to subsurface interpretations, and that is that "terminal alluvial fan," on which flow decreases
"sand" bodies interpreted from wireline logs may downstream to zero at the distal edges, due to evapo-
include crevasse and levee deposits, which, com- ration and infiltration. Some ancient deposits that
bined with the width of the actual channel deposit, were possibly formed in this type of environment
would indicate much wider sand-body dimensions were described by Friend (1978).
than that for the channel alone, as observed in stud- Detailed facies and architectural work has now
ies of modern rivers. This controversy, and an alter- led to the recognition of several subtly different flu-
native interpretation, are discussed further in Sect. vial styles that develop under arid, ephemeral condi-
9.5.3. tions. These are described at length in Chap. 8.
Nanson and Croke (1992) indicated that cohesive
banks and low stream power are the key characteris-
tics that lead to the development of anastomosed 2.4.4.5 Large Rivers
channels. N.D. Smith et al. (1989) suggested that the
anastomosed fluvial pattern may be but a transi- Few thorough sedimentological studies of any
tional style that develops during a long-term process modern giant river have been undertaken since the
of channel avulsion within a meandering system. seminal contribution by Coleman (1969) on the
However, this has yet to be confirmed by later work. Brahmaputra. Bristow (1987, 1993a) reexamined this
The reversion to a single, stable channel observed river, as discussed in Sect. 2.4.4.2 (and in more detail
by these authors may be a reflection of the stable, in Sect. 6.6).
cratonic location of the river they studied. Work Recognition of the deposits of giant rivers (in-
by other researchers in steadily-subsiding basins cluding delta distributaries) requires identification
(e.g., D.G. Smith 1986; Shuster and Steidtmann 1987; of bar complexes or channel scour surfaces with
Kirschbaum and McCabe 1992; Nadon 1994) sug- appropriately large relief. A few recent studies have
gests that the anastomosed pattern may be dynami- achieved this. Jones and McCabe (1980) described
cally stable for significant periods of geological time. complex cross-bed sets 40 m thick (Fig. 2.27) which
they attributed to deposition on alternate bars in a
deep deltaic distributary channel. Several studies of
2.4.4.4 Ephemeral Rivers the Athabasca Oil Sands, Alberta (Mossop and Flach
1983; Flach and Mossop, 1985) documented point
Miall (1977) proposed using the flood deposits of bars 25 m thick, also interpreted as the deposits of
Bijou Creek, Colorado, as the type example of an distributary channels, probably in a tidally influ-
ephemeral stream, based on the work of McKee et al. enced environment. These deposits are discussed
(1967). Subsequently, it has been shown that some of further in Sect. 8.2.9. The deep fluvial channels in the
the details of this model need modification, notably Morrison Formation, which were reconstructed by
the presumed predominance of the upper flow-re- Campbell (1976), have now been shown to be a result
gime plane-bed lithofacies (flat-bedded sandstone) of misinterpretation of the outcrop record (Cowan
has been shown to be incorrect (F.G. Ethridge, pers. 1991; see Sects. 2.4.3.2, 10.3.2).
comm., 1985).
Important new details of facies associations and
bed geometries in ephemeral streams were pre- 2.4.4.6 Floodplain Environments
sented by Tunbridge (1981, 1984), Stear (1983), and
Wells (1983), based on well-exposed ancient ex- The predominantly fine-grained deposits of the
amples. Tunbridge (1984) and Stear (1983) showed floodplain have generally received much less de-
that point-bar deposits, with lateral-accretion cross- tailed study than the coarse "framework" gravel and
bedding, could develop into ephemeral streams. sand units of the channels. Several commmentators
Stear (1983) illustrated channel sandstones showing noted this at the first fluvial symposium. The major
Conclusions 55
exception was comprised of those geologists con- constrains channel migration and avulsion (Flores
cerned with the stratigraphy and sedimentology of 1984; McCabe 1984).
coal (e.g., Horne et al. 1978; Flores 1981, 1984; Field- The use of floodplain deposits in reconstructing
ing 1984; Rahmani and Flores 1984; McCabe 1984). fluvial architecture is discussed in Sect. 2.4.3.2. A
One of the most distinctive characteristics of the modern classification of floodplains by Nanson and
anastomosed fluvial environment is the presence of Croke (1993) is presented in Chap. 8 (Table 8.4) and
large, stable floodplain areas, typically wetlands with is discussed in Chaps. 7 and 8.
large crevasse splays. Some of the early references to
this model are noted in Sect. 2.4.4.3, although most
of these studies are concerned with the composition
2.S Conclusions
and geometry of the channel and crevasse-splay
units rather than the fine-grained deposits of the
floodplain. Similarly, Bridge (1984), in a paper on Although sedimentology and basin analysis have al-
facies sequences in overbank environments of high- ways been concerned with the geology of rocks in
sinuosity rivers, dealt exclusively with the types of three dimensions, during the 1970s the main focus
fining-upward and coarsening-upward sandstone- was on the elaboration of facies models, relying
dominated successions found in levee and crevasse- heavily on the use of the vertical profile. The weak-
splay deposits. nesses in this method were apparent by the end of
Several studies of coal-bearing successions sub- that decade, and several unrelated attempts to sys-
divided floodplain deposits into a series of distinct tematize architectural methods were made during
facies assemblages. For example, Ethridge et al. the 1980s. It is only since the late 1980s that we have
(1981) recognized crevasse-splay, levee, abandoned- had the tools to carry out systematic three-dimen-
channel, well-drained swamp, poorly drained sional investigations of fluvial strata, using the ar-
swamp, and lacustrine assemblages. A similar range chitectural techniques described in later chapters of
of facies was described by Gersib and McCabe this book.
(1984). Both of these studies focused on vertical Bersier, in 1948, attributed fluvial fining-upward
profiles, and on the building of generalized facies cycles to a tectonic cause. However, the 1960s and
models, but did not deal in detail with floodplain 1970s were a time when sedimentological studies
architecture. Wing (1984) subdivided plant-bearing were dominated by a fascination with autogenic pro-
floodplain fines into two main facies assemblages, cesses. Much of the fascination of the facies-model
and was able to construct a separate depositonal revolution was the discovery that appeals to such
model for each, based on lithofacies, plant assem- awkward mechanisms as pulsating tectonism were
blages, and unit geometry. Tabular units were inter- not necessary. In the case of fluvial sedimentology,
preted as the deposits of gradually aggrading this led to the development of the point-bar model
swamps, whereas lenticular bodies represent the fill and, later, an unraveling of the complexities of
of abandoned channels. This is one of the few studies braided and anastomosed rivers. To some exent,
to have exploited the floodplain deposits themselves however, we are coming full circle. The latest strati-
for their potential to yield architectural information. graphic revolution - that of sequence stratigraphy -
As discussed in Chap. 9, floodplain sequences may is returning us to some old ideas about allogenic
hold the key for unlocking a vast amount of architec- causes. The field of sequence stratigraphy, currently
tural detail, but few workers have made use of this one of the most exciting areas of research in strati-
potential. graphy, evolved from new ideas about basin devel-
Coal is commonly interpreted as an integral part opment that derived from developments in the
of the floodplain succession in fluvial and delta- analysis of seismic-reflection data during the 1970s.
plain environments. McCabe (1984) pointed out, These burst onto the geological scene in 1977 with
however, that in an active clastic environment char- the publication of AAPG Memoir 26 (Payton 1977),
acterized by crevassing and avulsion, floodplains but the implications of this research for fluvial sedi-
receive so much detrital input that high-quality, low- mentology only became apparent with the work of
ash coals are unlikely to form. Raised peat swamps Posamentier and Vail (1988) and Posamentier et al.
may develop where rainfall permits an elevated wa- (1988), at the end of the period dealt with in this
ter table to be maintained as rapid peat accumula- chapter. The ideas in these latter papers are still in
tion outpaces channelized clastic aggradation. The the category of current research, and are discussed in
relief of these swamps prevents detrital influx, and Chaps. 11 and 13. It is being shown that many clastic
successions consist of sequences that are basin-wide styles and facies models now used by sedimentolo-
or regional in extent, and are developed by allogenic gists (Chap. 8) are based largely on studies of the
mechanisms such as source-area tectonism, changes ancient record.
in rates of basin subsidence, basin tilting, or eustasy. Several areas of research that were active up to the
As discussed later in this book, many fluvial succes- end of the 1980s, and are discussed at length in this
sions contain fining- and coarsening-upward se- chapter, are not dealt with in detail elsewhere in this
quences of regional extent. Correlation and mapping book. Grain-size analysis, once thought to hold great
of these sequences have led to a renewed interest in promise as a basin-analysis tool (Sect. 2.3.4), is a far
vertical profiles, but have emphasized the need for more cumbersome technique for field research than
architectural methods to discriminate the various facies analysis, requiring considerable laboratory
cycles and sequences that have similar vertical pro- followup work, and the results have been shown to
fIles, but are of several different origins. be ambiguous or plain wrong. Paleohydraulic analy-
Miall (1978b) noted the paucity of studies of mod- sis reached a kind of plateau of development with the
ern rivers, and the consequently limited scope of the work of Ethridge and Schumm (1978), and has been
data base on which fluvial sedimentology rested. pursued by very few workers since that time, pre-
Little has changed in the intervening years. There is sumably because of the pitfalls and imprecisions in
still considerable room for detailed sedimentologi- the method that were revealed by that review. How-
cal studies of the facies composition and architec- ever, some of the methods of paleohydraulic analysis
ture of modern rivers, and their relationship to flow have been adapted for the investigation of sand-
patterns and discharge characteristics. As noted by body geometry in the subsurface, and are described
N.D. Smith (pers. comm., 1995) many of the fluvial in Chap. 10.
Chapter 3
Concepts of Scale
3.1 Time Scales and Physical Scales length of the measured stratigraphic record in-
in Sedimentation creases. Breaks in the record include such events as
the nondeposition or erosion that takes place in
front of an advancing bedform (a few seconds to
It has become a geological truism that many sedi- minutes), the nondeposition due to drying out at ebb
mentary units accumulate as a result of short inter- tide (a few hours), up to the major regional un-
vals of rapid sedimentation separated by long conformity generated by orogeny (millions of
intervals of time when little or no sediment is depo- years). The variation in sedimentation rate also re-
sited (Ager 1981, 1993). It is also now widely realized flects the variation in actual rates of continuous
that rates of sedimentation measured in modern accumulation, from the rapid sand flow or grain-fall
depositional environments or the ancient record accumulation of a cross-bed foreset lamina (time
vary in proportion to the time scale over which they measured in seconds), and the dumping of graded
are measured. Sadler (1981) documented this in de- beds from a turbidity current (time measured in
tail, and showed that measured sedimentation rates hours to days), to the slow pelagic fill of an oceanic
vary by 11 orders of magnitude, from 10-4 to 107 abyssal plain (undisturbed in places for hundreds or
m/ka. This wide variation reflects the increasing thousands of years, or more). There clearly exists a
number and length of intervals of nondeposition or wide variety of time scales of sedimentary processes
erosion factored into the measurements as the {Fig. 3.1).
FLUVIAL ARCHITECTURE:
SCALE, CONTROLS AND TIME
CONTROLS MACRO·ARCHITECTURE PROCESSES

6 e
IN BASIN TIME (10 - 10
TECTONICS I 3 •
10-10 SUBSIDENCE STYLE
AND RATE
laletal
/ill
TECTONICS SUBSIDENCE RATE

CLIMATE SLOPE CHANGES
BASE LEVEL MIGRATION/AVULSION
COMPACTION INCISION &
AGGRADATION
CHANNEL TYPE
BANKFULL SCOUR
BEDFORMS
INCISION/
HYDROGRAPH AGGRADATION
-. MIGRATION/AVULSION Fig. 3.1. Hierarchies of scale and time in flu-

L...:....;.:...:.:.....!:.......:.;..;....:....:;c......:;._;.....;..-,--_-, vial deposits. (Leeder 1993)
58 Concepts of Scale
There also exists a hierarchy of physical scales, the questions that we can ask." Schumm (1985a)
which the same two examples illustrate - the cross- showed that the significance of an event diminishes
bed foreset at one extreme to the basin-fill at the as the time-scale increases. Thus, an individual
other extreme (Fig. 3.1). At least 14 orders of magni- volcanic eruption, a spectacular geological event at
tude are represented, from the few square centime- the time of its occurrence (a "megaevent", to use
ters in area of the smallest scale of ripple foreset, to Schumm's term), diminishes in geological impor-
the tens of thousands of square kilometers of a major tance as the millenia go by and other eruptions take
sedimentary basin. This chapter, which is based on place, until eventually, after perhaps millions of
an earlier review by Miall (1991a), is a systematic years, all evidence of the eruption is lost (it becomes
exploration of these wide ranges in scales. a "nonevent") as a result of erosion or burial of
The ways by which earth scientists study sedi- the rocks and landforms formed by the eruption.
mentary processes and the resultant depositional Events that seem random in the short term (such
products vary according to the scale of interest. as turbidity-current events) may assume a regular
Bedforms in flumes are studied during experimental episodicity, or even cylicity, with definable recur-
runs of, at most, few days duration. Nonmarine and rence intervals, if studied over a long enough time
marginal-marine sediments and processes have scale. Many events occur only when some critical
been much analyzed in modern environments, using threshold has been passed, such as the buildup of
studies of surface processes, and by sampling the deposists on a depositional slope leading to gravita-
sediments themselves in trenches and shallow cores. tional instability and failure. In several essays,
The use of old maps and aerial photographs extends Schumm (1977, 1979, 1985a, 1988; Schumm and
the record as far back as about 100 years, and 14C Brakenridge 1987) has discussed the concept of
dates may enable stratigraphic records of the last few "geomorphic thresholds" and their impact on sedi-
tens of thousands of years to be calibrated. Many mentary processes. Such thresholds reflect both
sedimentological studies draw on geomorphological autogenic and allogenic processes, and are charac-
work on landforms and Recent sediments. However, terized by a wide range of time scales (Fig. 3.2) and
such work is hampered by the specific, and possibly scales of cyclicity (Fig. 3.3). Schumm (1985a) also
nongeneralizable nature of the Recent record, such provided a table which illustrates the magnitude of
as the Holocene deglaciation, climatic change, and various geological events as related to the time scale
rapid rise of global sea levels. Stratigraphic studies over which they are considered (Table 3.1).
typically deal with much longer time periods, as The incorporation of hierarchical scale concepts
represented by the deposits of basin fills, which may into fluvial studies requires an architectural ap-
have taken hundreds of thousands to millions of proach. Earlier approaches to the architectural study
years to accumulate. Intermediate scales, repre- of fluvial deposits are described in Chap. 2, notably
sented by such major depositional elements as large the work of ].R.L. Allen and A. Ramos and his col-
channels and bars, delta lobes, draas, coastal barriers leagues. The main classification used in this book is
and shelf sand ridges, which may represent thou- described in Chap. 4. The current explosion of inter-
sands to tens of thousands of years of accumulation, est in sequence stratigraphy represents an increasing
are particularly difficult to document in the ancient interest in large-scale stratigraphic architecture, and
record and to analyze in modern environments. The its dependence on such allogenic controls as tecton-
time scales of the relevant sedimentary processes are ics and sea-level change. This work is referred to
difficult to resolve, and the physical scale of the briefly here, but the main discussion is presented in
deposits falls between the normal size of large out- Chap. 13. The applications of the concepts to the
crops and the well spacing or the scale of geophysical study of reservoir heterogeneities are discussed in
resolution in the subsurface. Yet, it is this scale of Chap. 14.
deposit that is of particular interest to economic A review of the depositional processes in clastic
geologists, representing as it does the scale of many environments, and the ways by which various work-
stratigraphic petroleum reservoirs and their internal ers have systematized them in depositional hierar-
heterogeneities. chies, has revealed some common themes, although
Geomorphologists have devoted considerable at- it is premature to propose a unified hierarchy of
tention to the problem of time scales and their effects depositional elements for use in the description and
on analysis and prediction (Cullingford et al. 1980; classification of all clastic rocks. Most within-basin
Hickin 1983; Schumm 1985a). As Hickin (1983, p. 61) (autogenic) or external (allogenic) sedimentary con-
has stated, "time-scale selection largely determines troIs have durations of a rather constant and predict-
Time Scales and Physical Scales in Sedimentation 59
1:==---1 }Fourth-order cycle
lEe
...
g~~~~ } Third-ordercycle
=- ---------'----..
~
o
. ~-------------(-b-)---------
c
~
I
___ ~
.~ ~I~ Dynamic
equilibrium
I .~~ ;
Second -order eye Ie
Time-
tt)
Fig. 3.2a-c. The various time scales of geomorphic pro-

cesses. a The erosion cycle, as envisioned by W.M. Davis in
the nineteenth century. The lower line indicates the eleva-
tion of the valley floor, the upper line that of drainage
divides. Initial uplift is followed by degradationallowering
o Finer
_ Coarser
and episodic pulses of isostatic uplift in response to ero-
sional unroofing. Total elapsed time is in the order of 10 7- 8 - Threshold event
- - Complex response event
years for a major drainage basin, with minor uplift events
occurring on the scale of 106- 7 years (corresponding to the
tectonic cyclothems of Blair and Bilodeau 1988). Box la-
beled B is enlarged in b In detail, the valley floor shows an
episodicity on a smaller time scale (in the range of 102- 3
years) as a result of the periodic storage and flushing of
sediment from bars and floodplain deposits, for example
by avulsion events. Box labeled C is enlarged in c The
episodicity of b is shown here in greater detail. (Diagram
from Schumm 1977) Fig. 3.3. The hierarchy of cycles of sedimentation, based
on geomorphic concepts of the complex and episodic re-
sponse of fluvial systems to autogenic and allogenic forc-
ing. The primary cycle is the entire succession, reflecting
able order of magnitude, permitting a grouping of the gradual diminution of sediment grade following initial
sedimentary units according to the length of deposi- uplift (corresponding to the "erosion cycle" curve of Fig.
tional time that they represent. The sedimentary 3.2a; group 11 of Table 9.1). Second-order geomorphic
results vary from environment to environment be- cycles reflect isostatic adjustments (tectonic cyclothems)
cause of the various combinations of processes that or major climate change (the kinks in the curves of Fig.
3.2a; corresponding to group 10 of Tables 3.1, 4.2, 9.1).
occur. Thus, a comparison between subaqueous
Third-order geomorphic cycles are those exceeding geo-
unimodal flow (e.g., fluvial systems), subaqueous morphic thresholds, leading to periods of "metastable
oscillatory flow (tidal environments), and subaerial equilibrium" and periods of rapid change and adjustment
flow (eolian environments) is not straightforward, (the events shown in Fig. 3.2b). These processes occur over
however, the assignment of common groupings fa- various time scales (groups 8 and 9). Fourth-order cycles
are related to episodic erosion, and to the complex re-
cilitates clearer thinking about depositional con-
sponse of the fluvial system to any of the above changes
trols. As a simple example, the recognition of "super (group 7?). Fifth-order cycles are related to seasonal and
surfaces" in eolian systems enabled Kocurek (1988) other major hydrological events, such as the "lOO-year
to speculate about the potential mapping of migrat- flood" (groups 5, 6). (Schumm 1977)
ing ergs, the detailed correlation of ergs with marine
units, the application of sequence-stratigraphy con-
cepts to nonmarine systems, and the documentation
Table 3.1. Perceived importance of geologic events as related to size and age. (After Schumm 1985a)
Relative 1 day 1 year lOyr 102 yr 103 yr 105yr 106 yr 108 yr
magnitude
of event
Megaevent Local soil Gully Meander Volcanic Terrace Continental Major Mountain
slip or cutoff eruption formation glaciation folding, building
flow faulting
Mesoevent Rill Local soil Gully Meander Volcanic Terrace Continental Major
slip or cutoff eruption formation glaciation folding,
flow faulting
Microevent Sand Rill Local soil Gully Meander Volcanic Terrace Continental
grain slip or cutoff eruption formation glaciation
movement flow
Nonevent Sand Rill Local soil Gully Meander Volcanic Terrace
grain slip or cutoff eruption formation
movement flow
of regional climatic and tectonic controls on eolian signed a grouping based primarily on the time scale
deposits. The eventual availability of a common hi- represented by the deposit, or by the total time
erarchy of architectural units based on time scale elapsed during the formation of the deposit and its
and physical scale would facilitate more of this type bounding surfaces (which may be considerably
of correlation of events and cross-fertilization of greater). The grouping here is based primarily on the
ideas between contrasting environments. This is es- latter criterion - the recurrence interval of the depo-
pecially important in an era of compartmentaliza- sitional process. In some cases, such as sediment-
tion of knowledge, where sedimentary geologists gravity flows, there is a marked contrast between the
studying different aspects of a basin fIll might spe- duration of a depositional event and its recurrence
cialize in such diverse areas as (for example) fluid interval. The latter may vary by more than an order
hydraulics, facies analysis, geochemistry, paleocli- of magnitude within a basin or between basins, but
matology, regional tectonics, or geophysical subsid- quantifying such variations may be difficult or im-
ence models, yet have very little common language possible using present techniques. Because of such
to facilitate exchange of ideas and concepts. The variation, the grouping of some types of deposit in
classification and grouping discussed in this paper the discussion that follows is somewhat arbitrary.
are offered as a first step toward a "grand unified In a recent attempt to systematize our knowledge
theory" of architectural classification. of large-scale subaqueous bedforms, Ashley (1990)
Architectural scale concepts are also important in suggested that the scale of a bedform may simply be
the petroleum industry. Reservoir heterogeneities a reflection of the physical space, volume of sedi-
are present at at least four scales (Fig. 3.4) that can be ment, and elapsed time available for its formation
related to the scale hierarchy described here. The within a flowing system. Dunes, sand waves, and
analysis of each scale of heterogeneity requires dif- other two- and three-dimensional bedforms, which
ferent techniques, as described throughout the rest have gone by various names in the past, may all be
of this book. part of a genetically related continuum. Whereas this
argues against any attempt to impose an arbitrary
hierarchical size classification on hydrodynamic
3.2 The Grouping of Architectural Units sedimentary structures, the fact remains that the
in Clastic Rocks timing of depositional and erosional events within
trains of bedforms is strongly controlled by rhyth-
According to Depositional Time Scale
mic or episodic events on various time scales (diur-
nal, seasonal, etc.). Packages of cross-bedded strata
A tentative classification and grouping of architec- and their enclosing bounding surfaces are, therefore,
tural units into ten classes spanning at least 12 orders amenable to a hierarchical classification based on
of magnitude of time scale is shown in Table 3.2. depositional recurrence interval, as discussed in this
Deposits in most clastic environments can be as- book.
The Grouping of Architectural Units in Clastic Rocks According to Depositional Time Scale 61
Megascopic heterogeneity
t"ro[JUClll(] well
.
Macroscopic heterogeneity
EXPLANATION
D
Drained reservoir
0
Non·reservoir
Eillill
Untapped
~
Residual oil
~
t2::2.:8
Intra reservoir
compartment rocks mobile oil between grains seals
Fig. 3.4. Scales of heterogeneity in a meander-belt reser- heterogeneity corresponds to groups 9-11, macroscopic
voir sandstone. Untapped mobile oil is oil trapped by heterogeneity is equivalent to groups 7 and 8, mesoscopic
heterogeneities that require special mapping techniques to heterogeneity refers to groups 4-6 and microscopic hetero-
document (Sects. 9.5, IDA). In terms of the sediment geneity corresponds to groups 1-3. (Ambrose et al. 1991)
"groups" described in this book (Table 3.2), megascopic
The rankings that have been assigned to the types column 4. The basis on which these values have been
of bounding surface that enclose the deposits of a assigned is discussed in Sect. 3.4.
given group are shown in square brackets for the Group 1 deposits are those formed in a few sec-
fluvial/deltaic, eolian and coastal/estuarine columns onds, such as the lamination developed by the burst-
in Table 3.2. The bounding surfaces have typical and-sweep process in traction currents (Leeder
geometric shapes, areal extents and lithofacies asso- 1983), and the sediment layers accumulated by eo-
ciations, and can be used to define hierarchies of lian grain fall and grain flow.
depositional units or architectural elements. The Group 2 deposits are those formed in periods of a
rankings are tentative, as discussed in the next sec- few minutes to a few hours. Ripple trains are the
tion. In the submarine-fan column, the numbers in typical depositional unit in many clastic environ-
square brackets refer to proposed hierarchical clas- ments. As shown by Southard and others (1980), the
sifications of depositional units that have not been time taken for the migration of a waterlain ripple set
extended yet to include specific types of bounding by one wavelength typically varies from 20 min to 2
surface. The rankings are based on the work of h, with shorter times (20-60 min) more common.
Mutti and Normark (1987), and have been discussed Wind ripples may form and migrate in similar time
by Miall (1989), who showed how some of the periods.
architectural-element concepts derived from fluvial Group 3 deposits are those that form in periods of
sediments could be applied to coarse, deep-sea chan- a few hours to a day or two. Diurnal processes are the
nel-fill clastic systems. most important in this category of time scale, such as
The examples of sedimentary processes shown in tidal cycles, daily variations in spring-melt runoff
column three are not intended to comprise a com- reflecting temperature variations, and changes in
plete list, but to provide a flavor of the variation in wind strength and direction (which commonly occur
process with time scale. This list is amplified in the because of the different thermal properties of land
following discussion, and many examples of fluvial and adjacent water masses). These processes typi-
deposits classified into the various groups are dis- cally develop distinctive sediment bundles or cyclic
cussed throughout this book. The groups pertain to sequences on a small scale.
both the depositional elements and to their bound- In fluvial environments, group 3 deposits consist
ing surfaces, where shown. Order-of-magnitude val- of complete ripple sets and growth increments
ues for rates of sedimentation are indicated in (foreset bundles) of megaripples (dunes). In eolian
Table 3.2. Hierarchies of architectural units in clastic deposits

Group Time Examples Instan- Fluvial, Eolian Coastal, Shelf Submarine
scale of tan eo us deltaic estuarine fan
of processes sedimen-
process tation rate Miall Brookfield, Allen Dott and Mutti and
(years) (m/ka) Kocurek Bourgeois, Normark
Shurr
10-6 Burst-sweep Lamina Grain flow Lamina Llamina
cycle grain fall
2 10-5 Bedform 105 Ripple Ripple Ripple [3rd-order
_10-4 migration (microform) [E3 surface] surface
[1st-order in HCS]
surface]
3 10-' Diurnal 105 Diurnal Daily Tidal bundle [2nd-order
tidal cycle dune iner., cycle [E2 surface] surface
react. surf. [3rd-order in HCS]
[1st-order surface]
surface]
4 10-2 Neap-spring Dune Dune Neap-spring HCS
-10- 1 tidal cycle (mesoform) [3rd-order bundle [E2], sequence
[2nd-order surface] storm layer [1st-order
surface] surface]
5 10° Seasonal to Macroform Reactivation Sand wave, HCS
-10 1 10-year flood growth [2nd, [El], major sequence
increment 3rd-order] storm layer [1st-order
[3rd-order surface, surface]
surface] annual cycle
6 100-year 102-' Macroform, Dune, draa Sand wave [facies pack- macroform
flood e.g. point-bar [lst-, field, age (V)] [5]
levee, splay 2nd-order washover
[4th-order surfaces] fan
surface]
7 10' Long-term Channel, Draa, erg Sand-ridge, [Elongate Minor lobe,
-10 4 geomorphic delta lobe [ 1st -order, barrier island, lens (IV)] channel
processes [5th-order super tidal channel levee [4]
surface] surface]
8 104 5th-order 10-1 Channel belt Erg Sand-ridge [Regional Major
-105 (Milankovitch) [6th-order [super field, lentil (III)] lobe [turbo
cycles surface] surface] c-u cycle stage: 3]
9 105 4th-order 10- 1-10- 2 Depo. system, Erg c-u cycle [ss sheet Depo.
-106 (Milankovitch) alluvial fan, [super (II)] system
cycles major delta surface] [2]
10 3rd-order Basin-fill Basin-fill Coastal-plain [Litho some Fan
cycles complex complex complex (I) ] complex
[1]
11 2nd-order Basin-fill
cycle complex
Hierarchical subdivisions of other authors are given in square brackets; names of authors are at head of each column.
environments, wind ripples, depositing climbing in the supply of sand fed to the fore set crest by
translatent strata, form group 3 bundles (Hunter and superimposed megaripples (Dalrymple 1984). Vari-
Richmond 1988). In tidal environments, sand-wave ous types of tidal ripples were described by
cross-stratification may form bundles, reflecting Dalrymple and Makino (1989).
tidal-velocity asymmetry (Allen 1980) or variations
Group 4 deposits represent time periods of a few expansion (Jackson 1976b). Studies of the Atcha-
days to a few months. Many of the depositional falaya delta in Louisiana indicate that sedimentation
events that occur over such a time frame were during high-discharge events may add as much as
termed "dynamic events" by Jackson (1975). They 1000 km 2 of new land to the delta during the next 50
are events that move disproportionately large vol- years (Roberts et al. 1980).
umes of detritus in relatively short time periods. Many storm sequences containing hummocky
Many marine storm events fall into this category. cross-stratification, although developing in a few
Hummocky cross-stratification (HCS) sequences days or weeks, may also represent the action of vio-
are thought to form in a few days to a few weeks lent storms or storm groups that occur only every
during major storms. This time span also includes few years (Dott and Bourgeois 1982). Separation of
the lunar neap-spring tidal cycle. Periodicity in tidal such deposits from storm deposits of group 4 is
bundles was observed by Visser (1980) and in tidal arbitrary in this classification, because there is prob-
ripples by Dalrymple and Makino (1989) and ably a continuum of depositional variability between
equated with the lunar cycle. In fluvial systems, com- the sequences that represent individual storms dur-
plete trains of mesoforms, such as dunes, may be ing an annual storm season, and those that represent
deposited during such time spans. the "1O-year" or "100-year" event. As noted later,
Group 5 deposits are those that represent time work on hurricane deposits is beginning to suggest
spans of 1 year to a few tens of years. Seasonal some criteria for discriminating storm magnitude in
deposits, such as spring runoff events and glacial carbonate environments, but there are no such data
varves, are among the most important types of de- relating to hummocky cross-stratification or other
posit in this category. The deposits of occasional terrigenous clastic deposits.
major storms (shelf environment) and flash floods Group 6, in this classification, represents deposits
(fluvial environment), including those of strengths accumulating over time spans of hundreds to a
occurring only every few years (e.g., McKee et al. few thousands of years. Particularly violent dyna-
1967), are also group 5 deposits. Annual changes in mic events are thought to occur infrequently in
wind strength and direction lead to cyclic deposition many environments, and are commonly termed for
of eolian dunes, as has been recognized in several convenience (if not accuracy) the "IOO-year events".
ancient deposits (Hunter and Rubin 1983; Kocurek These would include violent hurricanes, cata-
et al. 1991). strophic flash floods, and sediment gravity-flow
Growth increments on macroforms in subaerial events in many environments (in some settings,
and subaqueous environments are typical deposi- sediment gravity-flows are infrequent enough to be
tional products, and may be separated by large-scale group 7 events). Many of the resulting depositional
reactivation surfaces showing low-angle truncations sequences are very similar to those of group 5, such
of underlying bedding (bounding surfaces are dis- as those described earlier, and could be distin-
cussed later). Examples are shown in Fig. 3.5. guished only by such clues as the large volume or
Coleman (1969) showed that in the Brahmaputra thickness of the deposit, or the scale of the resulting
River giant "sand waves" (possibly more accurately bedforms. For example, Wanless and others (1988)
classified as macro forms ) are active primarily dur- took advantage of the passage of a major hurri-
ing seasonal floods. Jackson's (1976) studies of the cane across the Caicos Platform in the British West
Wabash River showed that meander bends in that Indies to compare its effect on carbonate tidal-
river migrated up to 2 km in 50 years, generating an flat sediments with that of "normal" winter storms.
equivalent width of a new point bar. The Ucayali Hurricanes, which are group 5 or 6 events in
River, a tributary of the upper Amazon in Peru, this area, deposit centimeter-scale-thick beds of
shows comparable migration rates of 40-60 m/year peloidal grainstone, whereas winter storms (group
(Dumont and Fournier 1994). Iseya and Ikeda (1989) 5) generate millimeter-thick laminae. Much more
demonstrated that, in the ridge-and-swale topogra- work of this type is needed to provide criteria for the
phy of a point bar in a meandering river in Japan, the recognition of event magnitude in the sedimentary
recurrence interval of the ridges is estimated to be 25 record.
years. However, preservation of point-bar deposits Complete macro forms are group 6 deposits (Figs.
depends on the direction and style of meander mi- 3.5,3.6). Examples include point bars, sand flats, and
gration, which could include meander expansion, crevasse splays in fluvial systems, and similar ele-
translation, and rotation. Net sedimentation-accu- ments in submarine channels; major segments of
mulation rates are highest during phases of meander coastal barriers and spits, and large wash over fans in
0\
.,.
"~:"==",~'§:=-,,,,,:g,.,~~-"-~-::;'~~~:oo."., ~,, ,.
:
o 50m
\IE zet'O
......_---'-'....-=...;
-.."'-
:: .. ~::..~ -,.~-.....:.':.<:::.- .....'-'
®
r main boundinQ turfoc..
internal reactivation surfaces ©
SF S< . >:-:"::;;"--,~.· ~
cross .trotification
"'_'::f:~
Fig. 3.5. Examples of depositional units that accumulate in a few hundred to a few thousand years (group 6). Internal reactivation surfaces subdivide the deposits into
internally conformable assemblages. The reactivation surfaces may represent the result of erosion during seasonal or longer-term dynamic events, and the beds between
them typically accumulate in periods of a few years to a few tens of years (group 5). A Fluvial or estuarine point bar, Torrivio Member, Gallup Sandstone, near Gallup,
New Mexico (Miall1991 b). B Field of modern sand waves (dunes), Bay ofBourgneuf, France (Berne et a1. 1991). C Triassic eolian deposits, Holy Cross Mountains, Poland
(Gradzinski et al. 1979). Diagrams are drawn without vertical exaggeration
(")
o
(")
=
(1)
'0
...
'"
o
.....
en
(")
~
(1)
m
A sea head of posses
~
m
o ~--~m~~~/--------~=-~~~?-~~-.~.-.~.~.~.~.~
.-.-.~~~YO
',.', .
o 10 20 30 40 50km
Fig. 3.6. Sedimentation rates (both vertical and lateral) in about 1000 years. Time lines show the historic develop-
are extremely rapid in such major deltas as the Mississippi. ment of this channel-fIll and mouth-bar deposit, which is
This is an example of a group 6 deposit, a bar-finger sand about 7 km wide, 70 m thick and at least 40 km in length.
forming the core of a delta lobe (Southwest Pass). Such (Gould 1970)
lobes are formed and abandoned in the Mississippi system
coastal settings; and fields of sand waves in conti- dence from the Holocene record shows that con-
nental shelves. struction and abandonment of major delta lobes in
A few quantitative estimates of sedimentation the Mississippi system take about 1000 years (Kolb
rates are available, including the following (these and Van Lopik 1966; Frazier 1967). Individual major
data are used in a discussion of vertical accumula- distributary mouths in this system advance about 3
tion rates in a later section): Allen (1984) estimated km in 100 years, with the formation of levees and
that estuarine sand-wave fields take approximately mouth-bar deposits (Gould 1970; Fig. 3.6). It is not
1200 years to form. Data given by Hayes and Kana yet known whether separate growth increments cor-
(1976) show that coastal spits on the east coast of the responding to a group 6 (or any other) time scale can
United States grow laterally by average rates of about be distinguished in these fluvial and coastal deposits
200 m in 100 years, with local short-term growth on the basis of internal bounding surfaces.
rates of up to 90 m/year. Galveston Island, a regres- Group 7 deposits include complete major ele-
sive barrier system, has built seaward at an average ments of depositional systems, such as channels,
rate of about 100 m per 100 years (Bernard et al. major delta lobes and some delta complexes, draas
1962). Gagliano and van Beek (1970) documented or entire ergs, barrier islands, tidal channels, sand
the growth of major subdeltas in the Mississippi ridges on the continental shelf, and submarine-fan
system using old maps, and showed that deltaic channels and compensation cycles. They are formed
growth could fill an interdistributary bay by over periods of thousands to tens of thousands of
progradation from a single crevasse in about 100 to years. Examples are illustrated in Fig. 3.7. Most Ho-
150 years. Diversion of major distributaries in this locene depositional systems formed since the post-
river system is capable of generating major subaque- glacial sea-level rise, if considered in their entirety,
ous deposits in the space of less than 100 years. are group 7 deposits. Storm-generated sand ridges
Roberts and others (1980) documented the effects of "take many thousands of years to build" (Brenner et
the diversion of flow from the Mississippi into the al. 1985). The preserved deposits of the Galveston
Atchafalaya River since the early part of this century. Island Barrier system represent about 3500 years of
A major floodplain lake has been filled with sedi- accumulation (Bernard et al. 1962). Sandy Neck Spit,
ment, and a subaerial delta appeared at the mouth of Massachusetts, is 10 km long and dates back to about
the river in 1972. Complete abandonment of the 3.3 ka (Hayes and Kana 1976). Tidal-inlet fills
main course of the Mississippi River in favor of the formed by lateral migration near Sapelo Island,
Atchafalaya would probably already have occurred Georgia, are at least 3.4 ka old (Hoyt and Henry
but for river-control emplacements constructed by 1967).
the US Army Corps of Engineers (Sect. 10.3.1). Evi-
0'1
0'1
Well Bonk
m
Ooboy Sapelo Island
10
'f.> . ~;..~ Sound
-~;'i;ii~'~"" VE x50
Smith Knoll
® O+I----~--~----~
o 3km
Galveston Island sea

.,'u ...... -c: . F. r-T.,_~
®
Fig. 3.7. Examples of group 7 deposits. A, B Sand ridges in the North Sea. Stratification within the ridges has an actual dip of about 30 , and represents master bedding
surfaces or reactivation surfaces (see discussion in text; Houbolt 1968). C Channel-fill and spit deposit formed by lateral migration of a tidal channel. Dashed lines are time
lines. Section is parallel to axis of barrier island forming the coast of Sapelo Island, Georgia (Hoyt and Henry 1967). D The Galveston Island barrier system, Texas. Dashed
lines are time lines, with dates derived from 14C analyses (Bernard et al. 1962). Time lines in C and D may be preserved in the rock record as distinct stratification surfaces
comparable to the reactivation surfaces of Fig. 3.5
(')
o
()
=
.g....
tI)
o
...,
r/l
()
e.
tb
N S >-l
::r
(b
Lake BoroIario Gulf 01 MexQ1
Ponlchorfroin Boy C'J
....
m o
S«1 .g
O~
ItM!I 5'
CJQ
o
.....
10 >-
....
-- -- ------ - - -- ------ - - - - - - - - - - - - "
[-
- ------- - - -- - - - - - - - - - - - - - - - - - - - (b
"0..0 -
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-- "2....
4LLU~S"-GL4 N s eo.
Ste Marie Medffen'an«m c:::
'4L PILl. Cl4L de 10 Mer Sea ::l
, m ~.
sea til
J::.::/,:,<.:I sand CB coastal barrier ~I

'CJ 5'
MB mou1h bar soil o
CH distributary channel 6>
10 ~
t:::7.'
'\;d n'
progradational facies
~ :>0
(distributary-mouth and delta-front o
cloy, sill, silly sand) - 20 "~
[:=::J aggradational facies >-
(floodplain cloy. sill. peat; ""....o
levee silly cloy) P-
5'
CJQ
~ transgressive -facies
(boy silt, cloy) S
t:I
(b
o
'"~.
0'
::l
eo.
VE x 563 >-l
S'
(b
10 km en
"eo.
(b
Fig. 3.8. Two examples of postglacial delta complexes built during the last 6 to 8 ka. Above North-south cross section through the Mississippi complex, showing the thin,
overlapping lobes that developed on the continental shelf (numbered 1- 16). Lobe 3, the oldest seen here, is about 4.6 ka old (Frazier 1967). Below The Rh6ne delta. The
greater thickness of this delta probably reflects its location on the edge of the continental shelf. This delta did not build its deposit by lateral delta switching but by steady
seaward growth of barrier complexes, behind which a fluvial and lagoonal delta-plain aggraded. (Oomkens 1970)
0\
-..]
Turning to the alluvial environment, paleosols level change. For example, Suter et al. (1987) de-
require several thousand years to develop, and may scribed six sequences that have formed on the Loui-
take at least tens of thousands of years to reach siana continental shelf since about 150 ka, with an
"maturity" (Leeder 1975; Retallack 1984; Wright average duration of 25 000 years. There is increasing
1990; see Sects. 9.2,10.3.1 for additional discussion). evidence of astronomical forcing in the development
Most river valleys aggraded with fluvial and estua- of cycles of this duration - the so-called
rine deposits immediately following the postglacial Milankovitch cycles (Fischer 1986). For example,
sea-level rise. Following this, many rivers have had Olsen (1990, 1994) postulated a Milankovitch
6000 to 8000 years to build major delta complexes. cyclicity to explain variations in fluvial cycle thick-
The modern (postglacial) Mississippi delta complex ness and calculated stream discharge in some Devo-
is one example (Kolb and Van Lopik 1966; Frazier nian deposits of East Greenland (Sects. 9.2, 12.l3).
1967). The RhOne delta is another. Both are illus- Major bounding surfaces within group 8 deposits
trated in Fig. 3.8. may not be caused by autogenic mechanisms, such
Fluvial deposits of groups 8 to 11 are described as channel migration, but may be regional in extent
at length in Chap. 9. The sequence stratigraphy of and allogenic in nature. Such are the ravinement
these deposits is discussed in Chap. l3. The following surfaces and other types of diastem described by
are some brief notes on marine and nonmarine de- Nummedal and Swift (1987), and the sequence
posits. boundaries of Van Wagoner et al. (1990; see Chap.
Group 8 sedimentary bodies represent time peri- l3).
ods of tens to hundreds of thousands of years. Ex- Group 9 deposits represent major depositional
amples include major fluvial channel-belt deposits, systems, accumulating over hundreds of thousands
such as the "third-order" cycles of the Siwalik Group to a few millions of years. Some stratigraphic se-
in Pakistan (as demonstrated by detailed studies of quences, including many cyclothems (the "major"
the magnetostratigraphy by Johnson et al. 1985). The cycles of Heckel 1986), are of this group. These have
nine eolian "complexes" bounded by supersurfaces been termed "fourth-order cycles" in earlier synthe-
described by Kocurek et al. (1991) each represent an ses (Vail et al. 1977; Mialll984a, 1990), in a number-
average of about 700000 years. Other examples in- ing system that uses lower "order" numbers for
clude such major depositional elements as delta cycles of increasing duration (the reverse of the
complexes and alluvial fans, shelf sand-ridge fields "group" numbering used here).
(group III "regional lentils" of Shurr 1984), and ma- Group 10 deposits are basin-fill complexes (depo-
jor lobes of submarine fans (third-order "turbidite sitional systems tracts), representing time spans of
stages" of Mutti and Normark 1987). The distribu- millions to tens of millions of years. They corre-
tion in time and space of such elements is commonly spond to "third-order cycles" in earlier syntheses.
controlled by regional tectonism or by regional sea- There is ample evidence for such cycles in the strati-
level change which itself is commonly caused, in graphic record, including Cambrian grand cycles
turn, by tectonism, astronomical (Milankovitch) (Aitken 1966, 1978; Chow and James 1987), Carbon-
forcing, such as glacioeustasy, or other eustatic iferous megacyclothems (Heckel 1986) and meso-
causes. thems (Ramsbottom 1979), and the Cretaceous
Many stratigraphic sequences are of group 8, such mesothems of the Western Interior (Weimer 1960;
as the "minor" cyclothems of Heckel (1986), the Kauffman 1969, 1984). Regional or global sea-level
"punctuated aggradational cycles" of Goodwin and changes caused by regional tectonism, or by global
Anderson (1985), and the Cardium Sandstone cycles changes in seafloor spreading rates, may be the main
ofPlint et al. (1986). These have been termed "fifth- mechanism leading to the development of this type
order cycles" in earlier syntheses (Vail et al. 1977; of cyclicity. Another promising idea regarding the
Miall 1984a, 1990). The Cardium sequences may origin of third-order cycles is that of plate-margin
record tectonic thrust-loading events, as suggested tilting as a result of in-plane stress transmitted
by Swift and Rice (1984), who pointed out the com- across plate interiors from extensional or compres-
mon occurrence of repeated coarsening-upward sional plate margins (Cloetingh 1988). Miall (1990,
cycles in shelf sandstone bodies of the Cretaceous 1991c) provided reviews of this topic, and the ideas
Western Interior. This model has now been devel- are discussed further in Sects. 11.3.5 and l3.4.
oped by many workers, as discussed in Sect. l3.4. Second- and first-order cycles, in the terminology
Other fifth-order cycles (group 8 deposits of this of Vail et al. (1977) and Miall (1990), may be desig-
classification) may be caused by glacioeustatic sea- nated stratigraphic groups 11 and 12, if required.
Definition of Sediment Groups by Bounding Surfaces 69
3.3 Definition of Sediment Groups surfaces separating different assemblages of cross-

by Bounding Surfaces bed structures are coset boundaries (McKee and
Weir 1953), and were classified as second-order sur-
faces by Miall (1988a,b). Dynamic superimposition
For some clastic deposits, as noted earlier, attempts of bedforms, such as the migration of megaripples
have been made to recognize the types of facies over sand waves (Jackson 1975), may also lead to the
composition, vertical succession, geometry, scale, development of second-order surfaces.
and bounding surfaces that characterize the deposits In the tidal environment, the individual small-
formed over particular time scales. Where this can scale bedforms that make up class V and VI sand
be done it permits the deposits to be subdivided waves are group 2 structures of the present classifi-
empirically into a hierarchy of architectural units. cation. The boundaries between them were termed
Characteristics of the bounding surfaces that enclose E3 surfaces by Allen (l980).
these units may also include features useful for dis- In storm cycles containing hummocky cross-
tinguishing the depositional units themselves, such stratification, two types of internal bounding sur-
as the relationship of the surfaces to overlying and faces have been recognized by Dott and Bourgeois
underlying strata (erosional, gradational), their (1982). Individual laminae are separated by what
shape (flat, irregular, concave- or convex-up), areal they termed third-order surfaces, and are thought to
extent, and the nature of associated facies (e.g., they represent lamination produced by individual wave
overlie mud drapes, or are followed by poorly sorted oscillations or pulsations of wave trains. This pro-
lag deposits). However, sets of field criteria are not cess represents periods of seconds or minutes, and
yet available that can provide unambiguous inter- so the deposits are of group 1 or 2 in the present
pretations of the time scale of the sedimentary pro- classification.
cesses for every field case. Many deposits, therefore, Diurnal changes in bedform migration may de-
cannot readily be classified into the groups listed in velop discrete bundles of cross-strata, or rhythms,
Table 3.2, but could fall into anyone of two or three commonly separated by minor reactivation surfaces.
of the groups. For example, diurnal, seasonal, and Tidal bundles are particularly distinctive examples
random meteorological processes may produce of group 3 deposits. Boundaries between tidal
similar results in eolian (G. Kocurek, pers. comm., bundles are E2 surfaces, in Allen's (l980) terminol-
1989) and fluvial strata, particularly if preservation ogy, and are formed by erosion by the subordinate
of the products is incomplete. For these reasons, it is current in the tidal cycle, according to him. How-
not yet possible (and may never be) to establish an ever, Dalrymple (1984) showed that, at least in some
all-encompassing classification of the architectural cases, the spacing between E2 surfaces is several
subdivisions and bounding surfaces of all terrig- times the net bedform-migration distance per tidal
enous clastic deposits. A discussion of the progress cycle. He interpreted E2 surfaces and variation in
to date in this direction is the subject of the remain- tidal foreset thickness between the surfaces as a re-
der of this section, and the fluvial bounding-surface sponse to the arrival at the crest of the sand wave of
classification used in this book is described in Sect. the peaks and troughs of superimposed megaripples.
4.5. The equivalent structures in fluvial environments
Lamination and small-scale ripples develop rap- are simple reactivation surfaces, some of which
idly beneath traction currents in a wide range of probably reflect diurnal changes in stage and there-
sedimentary environments. These are group 1 and 2 fore separate units of group 3 rank. The variations in
deposits, respectively. The superimposition of trains tidal-bundle thickness related to neap-spring
of ripples may generate surfaces of bedform climb, changes that were described by Visser (1980) are also
or may be represented by virtually flat bounding bounded by E2 surfaces. In this case, the surfaces are
surfaces. These are set boundaries, in the termino- useful for defining sediment packages of group 4
logy of McKee and Weir (l953). Those that occur in rank.
fluvial sediments were classified as first-order sur- Pulses within individual storm events may occur
faces by Miall (l988a,b). This type of surface is one over periods of hours, days, or weeks, giving rise to
where little or no erosion is apparent, and the surface bundles of hummocky bedforms separated by sec-
records the virtually continuous sedimentation of ond-order bounding surfaces (Dott and Bourgeois
trains of similar bedforms. Changes in flow charac- classification), and representing group 3 deposits
teristics may lead to changes in the type of bedform, (present classification). Storm events character-
with resulting changes in cross-bed style. Bounding istically develop sequences that rest on scoured
surfaces, pass upward into hummocky cross- strati- surfaces in the ebb-tide deltas described by Yang and
fication, and are capped by plane lamination and Nio (1989, p. 187) are comparable. Houbolt (1968)
ripples. This is the idealized storm sequence of Dott termed those that occur in sand ridges "master bed-
and Bourgeois (1982). The first-order surfaces (their ding surfaces", and this term has also been applied to
terminology) that bound these sequences represent the large modern sand waves studied by Berne et al.
the completion of single storm events or storm sea- (1988, 1991). Berne et al. (1988) implied an analogy
sons, and therefore may be used to define deposits of between their master bedding surfaces and the E2
group 4 or 5 of the present work. At present, there are surfaces of Allen (1980). There may be a confusion in
no criteria for the distinction between the storm bedform terminology here, because what Berne et al.
sequences that might accumulate during individual (1988) termed sand waves are large structures 3.5 to
storms within an annual storm season, and those 7.5 m high, compared to an average height of 0.81 m
that represent rarer, and perhaps more violent for those described by Allen (1980) and Dalrymple
events, such as hurricanes. (1984; his height data).
In the eolian environment, dune cross-stratifica- These estuarine and fluvial macro forms vary
tion may include interbedded grain flow, grain fall, widely in physical scales and probably represent
and ripple cross-lamination. These can be separated sedimentary processes acting over a wide range of
in the field by their distinctive facies characteristics, time scales, although data relating to the migration
but determining the temporal significance of each and accumulation rates of sand ridges are sparse.
unit may be extremely difficult (G. Kocurek, pers. Because of this, it is premature to attempt to develop
comm., 1988). In coastal environments, differential a common architectural classification scheme for
heating and cooling of the sea and the land lead to these deposits. Fluvial third-order erosion surfaces
diurnal fluctuations in the strength and direction of (e.g., the reactivation surfaces of Fig. 3.5A) are con-
sea breezes, with a resultant distinctive bundling of sidered to be related to seasonal or longer term (e.g.,
eolian wind-ripple cross-lamination (Hunter and "10-year") floods. They may represent the internal
Richmond 1988). Such bundles are thought to have sedimentary breaks between individual ridges in
low preservation potential, but yearly cycles have point-bar ridge-and-swale topography. They there-
been tentatively identified in ancient deposits fore enclose packages of strata of group 5 rank. As
(Hunter and Rubin 1983). The eolian third-order noted above, the E2 surfaces in sand waves enclose
surfaces of Brookfield (1977) are dune reactivation packages of groups 3 and 4. Berne et al. (1988) sug-
surfaces, but these could represent daily, seasonal, gested that the master bedding surfaces in their
or longer term (e.g., 1DO-year wind storm) erosional bedforms may represent longer-term phenomena,
events (G. Kocurek, pers. comm., 1988). The deposi- longer than neap-spring tidal cycles. Possibly, they
tional units enclosed by these surfaces may therefore are related to storm scour on a yearly or longer-term
be of group 3, 4, or 5 of the present classification. scale, in which case the sediment packages between
Some of the eolian stratification surfaces shown in the surfaces are group 5 deposits of the present clas-
Fig. 3.5C may include third-order surfaces. sification. Berne et al. (1991) carried out seismic and
The architecture of estuarine and shelf sand coring investigations through fields of giant sand
waves (Allen 1980; Dalrymple 1984), and sand ridges waves, and illustrated a threefold hierarchy of
(Houbolt 1968; Harris 1988), and that of down- bounding surfaces. Their "second-order" surfaces
stream-accreted macroforms in fluvial and deltaic (equivalent to master bedding surfaces, labeled D2
deposits (Miall1988a,b; Yang and Nio 1989), is su- surfaces in core; their Figs. 10, 11) probably equate to
perficially similar (cf. Figs. 3.5B and 3.7A,B, although E2 surfaces and are of group 5 rank. The internal
these drawings are at different scales). In each case, structure and migration dynamics of sand ridges are
the deposit rests on a flat erosion surface and con- very poorly known. Harris (1988) suggested that
tains gently dipping « 10°) accretion surfaces that they consist of superimposed assemblages of sand
commonly truncate underlying bedding at a low waves, separated by master bedding surfaces. If this
angle (the reactivation surfaces of Fig. 3.5). Bedding, is the case, the master bedding surfaces may repre-
including cross-bedding, commonly downlaps onto sent major erosional events, such as the "lOO-year
these accretion surfaces. The deposits typically are event" of group 6, for example, the scour of rare
lens-shaped, with a convex-up upper surface. In the violent storms.
bounding-surface classification of Miall (1988a,b), The bounding surfaces that enclose entire fluvial
the internal accretion surfaces in fluvial macro forms macro forms, sand waves, and sand ridges typically
are classified as third-order surfaces. The C1 and C2 have convex-up shapes. This reflects accretionary
Sedimentation Rate and Its Relation to Depositional Recurrence Interval 71
growth terminated by a shift in the depositional monly by an order of magnitude. The reason for this
milieu, such as a channel-avulsion event. As noted is that sedimentation is rarely continuous for more
above, however, similarity of form does not neces- than a few weeks or months, at most, at anyone
sarily indicate similarity in origins. Fluvial point location, and typically is interrupted by erosion or
bars and downstream-accreted macroforms are nondeposition, so that net accumulation is almost
group 6 deposits. They are bounded by fourth-order always substantially less than total sedimentation.
surfaces (MiallI988a,b), although commonly, in the Areas where sedimentation might be expected to be
ancient record, the caps of these deposits are planed most nearly continuous include the deep oceans,
off during subsequent erosional events, and the top where very slow pelagic settling occurs, and in the
surface is not preserved. Fluvial macro forms are sea off the mouths of perennial rivers, especially
typically capped by the fifth-order surfaces that de- below the influence of waves. In this section, an
fine the base of succeeding channel-fill units. Little is attempt is made to evaluate variations in sedimenta-
known about the bounding surfaces that enclose tion rate with respect to the ten sediment groups
sand-wave and sand-ridge deposits in the ancient defined in this chapter.
record. An important consideration in evaluating sedi-
In many outcrops, the most prominent bounding mentation rates is the concept of recurrence interval.
surfaces are those between group 7 elements, such as As noted by Jackson (1975), disproportionate vol-
the fifth-order surfaces that define the base of fluvial umes of sediment are commonly moved by infre-
channels (Miall 1988a,b), the "first-order" eolian quent dynamic events. The sedimentation (and
surfaces of Brookfield (1977), and the "first-order" accumulation) rate during the event may be ex-
surfaces in the sand-wave deposits described by tremely high, but must be discounted for geological
Berne et al. (1991). These commonly have planar purposes by factoring in the (typically) much longer
geometry, except for the curvature at channel periods of time when sedimentation is slow, or might
cutbanks (rarely preserved), and may represent even be negative, as a result of erosion. For example,
hundreds to thousands of years of nondeposition, a magnetostratigraphic study of a record of fluvial
for example, between events of fluvial-channel or flash flood deposits in Argentina indicated that only
draa migration. There may be little to distinguish one flood event every 50 to 500 years is preserved in
these, at the outcrop level, from the surfaces that the rock record there (Beer 1990). Spasmodic sedi-
define sediment bodies of groups 8 through 10 (e.g., mentation also occurs in areas characterized by lat-
eolian supersurfaces, sequence boundaries), and eral accretion or progradation. A depositional
recognition and classification of such surfaces then element, such as a delta distributary mouth, or a
depend on careful mapping and regional strati- fluvial point bar, builds rapidly in a horizontal direc-
graphic interpretation. tion, giving rise to temporarily rapid local sedimen-
Lithostratigraphic and allostratigraphic subdivi- tation rates, but then the dispersal system switches
sions (formation and member boundaries) and se- elsewhere for an extended period of time, with re-
quence boundaries constitute a still larger scale of sultant abandonment and, perhaps, erosion of the
bounding surface. These include the sixth-order flu- new deposit.
vial surfaces of Miall (1988a,b) and the eolian super- These points can be illustrated by a discussion of
surfaces of Kocurek (1988). They are discussed some of the sedimentation rates that have been mea-
further in Sects. 4.5 and 9.6. sured for the various types and scales of deposits that
develop in rivers, tidal estuaries, and deltas. Esti-
mates of order-of-magnitude instantaneous sedi-
3.4 Sedimentation Rate and Its Relation mentation rates averaged over the time span of each
of the ten groups are given in Table 3.2. An "instan-
to Depositional Recurrence Interval
taneous sedimentation rate" is one calculated for a
short period of time and extrapolated to longer peri-
It is a common observation that measured or calcu- ods for purposes of comparison. Examples of the
lated sedimentation rates vary in inverse relation to type of calculation that can be made for some of the
the length of time over which the measurement is groups are given later.
made (Sadler 1981, provided detailed documenta- Among the smallest sedimentary structures are
tion). In general, the sedimentation rate of the depo- small-scale ripples (group 2 of this work). These
sitional units in each of the 11 groups described here typically migrate a distance equivalent to their own
differ markedly from those of the next group, com- wavelength in 20 to 60 min (Southard et al. 1980). A
5-cm-high ripple that forms in 30 min is equivalent its) accumulate at rates of 100 to 450 m/ka for peri-
to an instantaneous sedimentation rate of 876000 ods of about 20 years, before abandonment occurs.
m/ka. Clearly, this number bears no relationship to If we assume that a distributary will only build out
real long-term sedimentation rates, but it serves to across a given area of the delta front once during the
emphasize the extremes of sedimentation rate, to migration of one major delta lobe, we can calculate
compare with more geologically typical rates dis- the sedimentation rate of the mouth bar averaged
cussed later. Tidal sand waves (group 3) have simi- over the life of the lobe (group 7). In the postglacial
larly very high instantaneous rates. In the Bay of Mississippi delta, major lobes are formed and aban-
Fundy, Dalrymple (1984) demonstrated that in one doned in about 1000 years (Kolb and Van Lopik
tidal cycle sand waves migrate a distance about 1966; Frazier 1967; Fig. 3.8 above), giving an average
equivalent to their average height, which is 0.8 m. sedimentation rate for that period of 45 m/ka. The
Bay of Fundy tides are semidiurnal, and so this mi- recurrence interval of delta lobes themselves de-
gration is equivalent to a sedimentation rate of pends on subsidence rates, sediment supply, and the
584000 m/ka. Climbing-ripple stratification repre- configuration of the continental shelf. In the case of
sents intervals of extremely rapid sedimentation. the Mississippi complex, the river is attempting to
Data collected by Bristow (1993a) from large bars in switch discharge (and delta construction) to the
the modern Brahmaputra River suggested that for Atchafalaya River, where one of the earliest lobes
very short periods as much as 3.75 m of sediment developed about 6 to 8 ka before present (Sect.
may accumulate in the space of 1 h, an instantaneous 10.3.1). On this scale deposition of a 45-m-thick
sedimentation rate of 33 x 106 m/ka! mouth-bar deposit represents a sedimentation rate
The deposits formed by seasonal or more irregu- of 5.6 to 7.5 m/ka, although this calculation does not
lar runoff events in rivers (group 5) have extremely take into account the bay-fill and other facies
variable instantaneous sedimentation rates. The interbedded with the mouth bar deposit. This com-
flood deposit in Bijou Creek, Colorado, described by pares with typical values for Holocene sedimenta-
McKee et al. (1967) was formed by the most violent tion rates of 6 to 12 m/ka that are commonly quoted
flood in 30 years. It formed 1 to 4 m of sediment in for the Mississippi delta complex (e.g., Weimer
about 12 h, an instantaneous sedimentation rate of 1970). Other deltas do not accumulate so rapidly. For
730000-2920000 m/ka. Assuming no erosion, and a example the RhOne delta has a thickness of only 50
repetition of such floods every 30 years, this trans- m, accumulated since about 8.2 ka, indicating an
lates into a rate of 33-133 m/ka averaged over a few average sedimentation rate of 0.6 m/ka (Oomkens
hundred years. In fact, scour depths during the flood 1970).
ranged from 1.5 to 3 m, and true net preservation of A few sedimentation rates can be calculated for
anyone flood deposit over periods of hundreds or tidal-inlet and barrier deposits. The Galveston Is-
thousands of years may be negligible. Long-term land barrier is 12 m thick and 3.5 ka old at its base
rates measured over hundreds to thousands of years (Bernard et al. 1962; Fig. 3.7D), indicating an average
are likely to be an order of magnitude less, in the sedimentation rate of 3.4 m/ka. A tidal inlet at Fire
range of 10- 1 m/ka. Island, New York, has migrated 8 km in 115 years
The point bars (group 6) in the Wabash River are (Kumar and Sanders 1974). The depositional slope
on average about 5 m thick. The active area of each from spit crest to channel floor is about 500 m wide,
bar is about 200 m wide. If the bars migrate at a suggesting that at anyone point the entire tidal-inlet
maximum rate of 2 km in 50 years (Jackson 1976b) fill could form by lateral accretion in about 7 years.
they would take 5 years to migrate one point-bar The sequence is 12 m thick, indicating an instanta-
width, which is equivalent to an instantaneous sedi- neous sedimentation rate of 1714 m/ka. This migra-
mentation rate of 1000 m/ka. Comparable rates may tion rate is unusually rapid. Tidal inlets at Sapelo
be calculated from the migration of distributary Island, Georgia, appear to have migrated only about
mouth bars. A typical Mississippi distributary mi- 2.5 km since the postglacial sea-level rise (Hoyt and
grates 3 km in 100 years (Gould 1970; Fig. 3.6). The Henry 1967; Fig. 3.7C), indicating a sedimentation
mouth-bar deposits, from the mouth of the channel rate of 4.5 m/ka.
to the toe of the distal bar, are about 4 km wide (in a It is instructive to compare these rates with those
dip direction) and about 45 m thick. This lateral obtained by Oomkens (1970) and McKee et al. (1983)
migration is equivalent to an instantaneous vertical as a result of their studies of modern sedimentation
sedimentation rate of 340 m/ka. Data from modern rates on the continental shelf off river mouths. A
Dutch tidal deposits summarized by Yang and Nio short-term rate of 4.4 cm/month (528 m/ka) was
(1989) showed that ebb-tide deltas (group 6 depos- determined close to the mouth of the Yangtze River,
Sedimentation Rate and Its Relation to Depositional Recurrence Interval 73
China, by McKee et al. (1983), who studied the decay Bown (1993) also employed paleosols in an estima-
of short-lived radionucleides in the uppermost 15 tion of sediment accumulation rates, but used a dif-
cm of recent deposits, representing about 100 days of ferent technique that avoids problems of correlation
accumulation. The uppermost 200 cm of section, to Recent equivalents. In their well-studied Eocene
representing about 100 years of accumulation, Willwood Formation succession in Wyoming, they
yielded a rate an order of magnitude lower, 5.4 cm/ recorded numerous paleosols, and were able to con-
year (54 m/ka). Oomkens (1970) quoted sedimenta- struct an artifical composite section composed en-
tion rates of35 cm/year (350 m/ka) for the delta front tirely of paleosol units, by substituting channel and
of the modern RhOne River. The monthly rate at the floodplain clastics with a paleosol by detailed lateral
Yangtze mouth and that determined for the Rhone correlation. They were able to estimate the relative
compare with the migration rates calculated for Ho- time required for each type of paleosol to form by
locene point bars and distributary-mouth bars, this same process of lateral correlation. Thus, it was
whereas the yearly rate at the Yangtze Mouth is found that four vertically stacked stage 1 paleosols
similar to that of a delta lobe of the Mississippi. are stratigraphically, and therefore temporally, equi-
Of particular relevance to the subject of this book valent to a single stage 3 paleosol. Each paleosol
are the sedimentation rates of fluvial deposits at the could therefore be weighted according to its relative
groups 6 to 8 level, corresponding to deposits time of development, and a time-thickness plot
formed by long-term geomorphic processes, includ- developed. Given precise biostratigraphic and radio-
ing those related to geomorphic thresholds, and the metric brackets on the top and base of the succes-
largest scale of autogenic depositional feature (as sion, they could then convert this graph to absolute
discussed in detail in Chap. 10). Rates of sedimenta- accumulation rates, and correct this to a sediment
tion on modern alluvial fans and fluvial floodplains accumulation curve by correcting for compaction.
have been measured using 14C dates on plant mate- The results indicated a variation in sedimentation
rial, and tephrochronology. Available data were rate between 0.1 and 2 m/ka. An examinati-on of
summarized by Miall (1978d) and shown to encom- unsteady sedimentation processes by Friend et al.
pass a wide range, from 0.08 to 50 m/ka. However, (1989) suggested that in the Siwalik sediments of
these measurements were not correlated to specific Pakistan individual beds of a few meters in thick-
scales of architectural units, such as the depositional ness, corresponding to channel-fill increments
groups defined here. Bridge and Leeder (1979) also (group 6) and complete channel-fills (group 7) were
assembled data from modern fluvial systems in or- deposited at rates of up to 1 m/ka.
der to establish meaningful sedimentation rates for Badgley and Tauxe (1990) carried out a detailed
use in their computer simulation models. They used study of a transition interval between normal and
a range from 5 to 40 m/ka, with an average of 20 m/ reversed magnetic polarity within the interfingering
ka, for their models of channel-belt deposition based fluvial units previously described by Behrensmeyer
on autogenic avulsion processes (group 7). and Tauxe (1982). This interval, estimated to last
Leeder (1975) developed a quantitative model for between 4 and 10 ka, varied in thickness from 0.5 to
the estimation of floodplain accretion rates based on 2.7 m, indicating a fivefold variation in net-accumu-
the ages of paleosols in Recent sediments, particu- lation rate. Badgley and Tauxe (1990) were also able
larly those in the arid American southwest region. to demonstrate diachroneity of a major paleosol ho-
The maturity of the paleosol was assessed using vari- rizon and a channel-fill sandstone unit, both of
ous facies criteria (see Sect. 7.4.2), and this was re- which crossed this transitional polarity interval.
lated to the age of the deposit, using 14C dating. It was It is not immediately clear why the data of Bridge
suggested that rates of sedimentation in ancient allu- and Leeder (1979) should suggest sedimentation
vial deposits could be estimated using this model. rates an order of magnitude greater than those of
However, Wright (1990) indicated that the ages of Kraus and Bown (1993) and Friend et al. (1989). In
the original Recent American paleosols have been the case of Kraus and Bown (1993), the main control
substantially revised, and he adduced a considerable on absolute time was provided by radiometric dating
body of additional data to indicate that the relation- of bracketing volcanic units in an ancient succes-
ship between paleosol maturity and age varied by as sion, and in the study of Friend et al. (1989) absolute
much as an order of magnitude. He showed that time limits were derived by magnetostratigraphic
Leeder's (1975) mature "stage 4" paleosols in the correlation. This contrasts to the estimates from
Quaternary record require as little as 3 ka or as much modern floodplains that comprise the Bridge and
as 1 million years to develop. The reasons for these Leeder (1979) data base. It is possible that despite
variations are discussed in Sect. 7.4.2. Kraus and every effort to relate the calculations to specific beds
the estimates from ancient units incorporate signifi- basin subsidence, which is controlled by tectonic
cant intervals of a missing section of a type that are setting, and changes in base level, such as eustasy.
not present in the modern successions used by Miall (1978) showed that nonmarine basins, in vari-
Bridge and Leeder (1979). ous tectonic settings, have sedimentation rates aver-
Groups 8 and 9 of this book include the Quater- aged over millions of years of 0.03 to 1.5 m/ka. This
nary shelf-margin sequences of Suter et al. (1987) subject is reviewed at greater length by Miall (1990),
and the minor and major cyclothems of Heckel and in Chap. 11 of this book.
(1986). The sequences described from the Gulf Coast
by Suter et al. (1987) averaged 25000 years in dura-
tion and range in thickness from about 25 to 160 m,
3.S Application of Scale Concepts
indicating average accumulation rates of 1-6.4 m/ka.
Heckel (1986) documented the chronology of 55 to Basin Analysis and Petroleum Geology
cycles of Westphalian-Stephanian age in the US
midcontinent. Estimates of the length of this time The discussion in this chapter is intended to provide
span range from 8 to 12 million years. The thickness a framework for the analysis of clastic deposits at all
of the sucession varies from 260 m in Iowa to 550 m scales, from the individual core sample, through out-
in Kansas. These values indicate an average accumu- crop-scale analysis, to the mapping of reservoir het-
lation rate of between 0.02 and 0.07 m/ka. Many of erogeneities, depositional systems, stratigraphic
the cyles contain substantial fluvial-deltaic sand- sequences, and entire basin-flll complexes.
stone units and, according to Ramsbottom (1979), Research at each level of the hierarchy requires
who studied similar cyclothems in Europe, rates of different techniques. Depositional units in the rock
lateral deltaic growth must have been about as rapid record up to group 7 and, in exceptional cases, group
as that of the modern Mississippi; yet, the average 8, may be studied using outcrop data, especially if
sedimentation rate is two orders of magnitude less the lateral-proflling technique described by Miall
than that of the Holocene Mississipi delta complex (1988a,b) is employed (Chap. 4). Mapping and classi-
and its Pleistocene shelf-margin precursors on the fication of bounding surfaces between depositional
Louisiana Gulf Coast. Part of the explanation for this elements are a key to such analyses. Detailed subsur-
marked contrast is that the Carboniferous cyclo- face studies may be successful, if supported by de-
thems that were the subject of Heckel's study are tailed core analyses.
located in a cratonic region,where subsidence rates In the case of modern environments, side-scan
would be expected to be substantially lower than on sonar techniques coupled with high-resolution,
the continental margin of the Gulf Coast. In addi- shallow seismic profiling provide many critical new
tion, many of the cycles are separated by erosion data regarding depositional patterns, especially in
surfaces. Rapid sedimentation rates of an order-of- the study of estuarine and shelf bedforms and larger
magnitude more were measured by Johnson et al. features of groups 6 to 8 (e.g., Berne et al. 1988, 1991;
(1985) and Friend et al. (1989) in their study of the Harris 1988).
Siwalik sediments of Pakistan. Detailed documenta- Units as small as group 6 may be detailed on
tion of a nearly 2-km-thick section spanning more exceptionally high-quality petroleum exploration
than 9 million years yielded constant long-term sedi- seismic data, especially on three-dimensional seis-
mentation rates that increase from 0.1 m/ka at the mic surveys (e.g., Brown 1985, 1991). Larger features,
base to 0.3 m/ka at the top. Individual channel-fill including stratigraphic sequences of groups 8 to 10
cycles in this succession are estimated to have return (the third- to fifth-order stratigraphic cycles of Vail
periods of 1-4 x 105 years (group 8 deposits). et al. 1977; Miall 1984a, 1990), are now analyzed
Bentham et al. (1993) used magnetostratigraphic routinely on seismic records. Surface and subsurface
correlation to estimate accumulation rates of 0.17 to lithostratigraphic correlation, with the aid of
0.57 m/ka over a 4 million year time period, for a magnetostratigraphy, marker beds, and refined bio-
braided fluvial system of Eocene age in Spain. Beer stratigraphic indicators, is also employed (Miall
(1990) recorded a maximum accumulation rate of 1984a, 1990). The application of these various meth-
0.9 m/ka in a Miocene sheetflood deposit in Argen- ods to the study of fluvial stratigraphic architecture
tina. is described in Sect. 9.5, and examples of various
The long-term sedimentation rates of groups 8 to scales of heterogeneity in fluvial reservoirs (Fig. 3.4)
10 depend largely on rates of generation of sedimen- are described in Chaps. 14 and 15.
tary-accommodation space. This depends on both
Chapter 4
Methods of Architectural-Element Analysis
4.1 Introduction large to be viewed within one field of view, unless the
photographer is required to move so far back from
the outcrop that the scale of the photograph becomes
A wealth of information is contained in many out-
too small to be useful. The obvious alternative is to
crops of fluvial deposits, that is not captured by
construct a mosaic from overlapping frames. The
conventional methods of facies analysis, such as the
geologist moves several tens or hundreds of meters
measurement of vertical-profiles. In this chapter, the
back from the face, and carries out a traverse parallel
formal methods of field architectural-element analy-
to it, taking a series of overlapping photographs.
sis are described. Analysis and interpretation of the
Dipping beds may be viewed from an adjacent hill-
results are the subjects oflater chapters.
top, or from a low-flying aircraft. It is important to
Earlier work on the development of architectural
stay the same distance from the face, so that the
concepts for fluvial deposits is discussed in Sect. 2.4.
frames are all as nearly as possible the same scale.
This chapter describes current techniques, including
Ideally, this may be accomplished with the use of
the writer's contributions (Miall1985, 1988a,b), to-
surveying equipment, but the high degree of preci-
gether with the modifications, improvements, and
sion obtainable in this way is not really necessary for
additional suggestions contributed by other re-
sedimentological purposes. It is also useful for the
searchers, notably those of Soegaard (1990, 1992),
line of sight towards the outcrop to be situated in the
S.A. Smith (1990), and DeCelles et al. (1991), who
plane of the bedding, in order to minimize vertical
worked mainly with conglomerates.
distortions. For a road cut, which can be photo-
Methods of conventional facies analysis are not
graphed from the other side of the road, these re-
covered in this book. The reader is referred to Miall
quirements may be a simple matter to satisfy. For
(1990, Chaps. 2, 4) for a detailed discussion of the
many hillside exposures and river cliffs, it may be
methods of facies analysis, including the measure-
considerably more difficult to generate an adequate
ment and interpretation of vertical profiles.
suite of photographs, with variations in the terrain
and obscuring trees causing problems of position
and distance. In practice, the writer has found that
useful mosaics can almost always be constructed
4.2 Construction of Outcrop Profiles
with a little effort, although perspective problems,
and variations in the scale or orientation of different
Architectural methods are based on the two- and parts of the profile may remain problematic.
three-dimensional mapping oflarge outcrops, using Variations in scale from one frame to the next can
outcrop profiles. These are essentially vertical geo- be accomodated by varying the enlargment size of
logical maps, and are constructed in much the same the prints. This is particularly easily accomplished if
way. The geologist begins with a base map, and then each frame includes a scale, or some object whose
carries out traverses across the map, making obser- dimensions are known. Perspective problems are
vations and walking out key structures and contacts. less easily eliminated. They can be troublesome if the
The base map may be either a line-drawing of the outcrop is not flat, but includes projecting and re-
outcrop, made with or without the aid of surveying ceding regions, because such features will not over-
equipment, or it may be a photomosaic. The latter is lap exactly when adjacent frames are joined.
the easiest to construct and to use in the field. An almost inevitable element of vertical distor-
In the simplest case, an entire outcrop may be tion will be present in most profiles, because the
encompassed in a single photographic frame, but the camera cannot view all beds along the plane of their
largest and most interesting outcrops tend to be too bedding from a single viewpoint. For example, when
76 Methods of Architectural-Element Analysis
the line of sight is slanted upward toward a cliff face, graphic laboratories are equipped to produce stan-
the upper edge of the frame has a smaller scale than dard prints (such as those of 6 x4 inches, in the
the lower edge. Upper and lower edges cannot both United States) very quickly. Enlargements may take
be exactly overlapped, and some compromise match a few days. In any case, it is now possible to make a
must be achieved. All these perspective problems reconnaissance of an outcrop, and take the neces-
can be reduced by moving further away from the sary photographs, and then return with the results to
outcrop and taking the pictures through a telephoto carry out ground measurements within a few hours
lens. or days.
Use of advanced cameras with adjustable bodies The next step is to mount the photographic mo-
can eliminate some of the regular, geometric distor- saic on a lightweight board suitable to be carried into
tions. Wizevich (1992a) provided a useful discussion the field (thin plywood or Bristol board), and to
of equipment and techniques. attach an overlay of tracing paper or drafting film.
An alternative or supplement to photographic The geologist should then spend a considerable
methods is to construct a base map of the outcrop amount of time poring over the mosaic and drawing
using surveying techniques. Kocurek et al. (1991) in every visible bedding plane. For a large outcrop
carried out a very detailed architectural study of this may take several hours of careful work. The first
eolian sandstones using a laser theodolite. This round of drawing may be done in the office, but it is
method is, of course, complex and time-consuming. essential that the placement of these lines on the
It may be the only approach that can be used when overlay be checked against the actual outcrop. Com-
the topography of an outcrop is very complex. The monly, a complex corner of an outcrop, when re-
researcher should carefully assess, before beginning duced to the two dimensions of a photograph, can
such a survey, how accurate the outcrop base map yield confusing and conflicting bedding traces,
needs to be. Precision in the placement of bounding which can only be resolved by going back to the point
surfaces to within a few decimeters is claimed by from which the picture was taken. For very large or
Kocurek et al. (1991). While this is impressive, it may inaccessible outcrops, Dueholm and Olsen (1993)
not be necessary. Much depends on the measure- and Olsen (1993) recommended the use of photo-
ments that are to be made from the profile. Bearing grammetric techniques to analyze the outcrop pho-
in mind that most fluvial depositional elements have tographs. The construction of steroscopic overlaps
very irregular shapes, and change their thickness by between the photographs should be done to facilitate
several meters within a few tens of meters laterally, this. This technique will permit an analysis of the
striving for a high degree of base-map precision may channel-scale litho somes, but interpretation of the
be largely a wasted effort. A simple photographic smaller-scale features, such as individual macro-
mosaic, with all its shape and scale distortions, is forms, cannot be carried out reliably without ground
ideal for plotting data, and provides a powerful basis checking.
for visualizing and interpreting the geometry of the In the rocks, a bedding plane is a continuous
architectural elements. Vertical profiles do not need surface, but it may not be equally visible everywhere
to be documented directly in the field, but can be on an outcrop face, because of variable weathering
measured from the completed profile if required, for characteristics. For example, even major channel
example, to provide comparisons with subsurface scours may be virtually invisible where the basal
records. channel lag has been removed by erosion, and iden-
In most countries, development of field film tical sandstones are amalgamated across a clean
can now be done very quickly. "One-hour" com- bounding surface. Inexperienced interpreters tend
mercial photographic laboratories are available in to avoid tracing bounding surfaces through such
many small towns in North America and Europe, areas of faint bedding traces, leaving the lines "hang-
especially in tourist areas, and even in many Third- ing". It is an essential part of the interpretive process
World countries. Unless the services of a profes- to eventually complete and join up all these hanging
sional photographer are available it is recommended lines. It may be necessary, however, to leave the
that the mosaic be made from color-print film, as completion of this process until lithofacies and
this is the most popular type of mm and the one paleocurrent observations are underway, so that
which most commercial tourist laboratories are where there is ambiguity, alternative correlations
equipped to deal with. Instant cameras, such as can be evaluated in the light of the alternative fluvial
Polaroid, may also be usable, although print quality architectures they would create. Architectural-ele-
does not tend to be as good. Commerical photo- ment analysis necessarily includes this type of ongo-
Classification of Lithofacies 77
ing interpretation, and the geologist needs to be such as pedogenic calcretes, coal, and evaporites,
clear about this as the work proceeds. Eventually, as typically form minor components of most fluvial
discussed in later sections of this chapter, each re- systems, but need their own careful characterization.
sulting complete surface can be assigned an appro- As noted by Miall (1990, p. ISO), the scale of an
priate interpretation and rank, and can then be individual lithofacies unit depends on the level of
colored or coded accordingly on the final drawing. detail incorporated in its definition. Facies may be
Before leaving the outcrop, the geologist should defined very broadly to encompass mappable strati-
be sure to record two critical items of observation: graphic units, or they may be defined finely to
the orientation of the outcrop and its scale. Both may accomodate the level of detail obtainable in the cen-
vary along the profile, in which case this should be timeter-by-centimeter logging typically carried out
measured and carefully recorded. Scale should be on core. For the purpose of architectural-element
measured vertically to reduce possible distortions analysis, a relatively fine degree of descripion and
(very little of the face can be photographed from a subdivision is required.
point perpendicular to it in the horizontal plane). A It is good research practice to approach each new
surveying rod may be included in one or more of the rock unit afresh, with the aim of making complete,
photographs, or the size of a recognizable feature of unbiased observations of all important lithofacies
the outcrop may be measured and marked on the attributes. However, sedimentological research has
overlay. demonstrated that much of the apparent variability
The final, drafted profile may need to be pub- in sedimentary units disguises a limited range of
lished as a foldout, because even the A4 and quarto- basic lithofacies and biofacies types. The deposi-
sized paper that is becoming fashionable for tional processes which control the development of
geological journals, is too small to contain a large clastic fluvial lithofacies, such as traction-current
profile illustration without severe reduction. I rec- transportation, with its accompanying fluid turbu-
ommend publishing the photomosaic separately lence and its effects on beds of clastic grains, are
from the line drawing interpretation, preferably with common to all rivers and obey the same physical
the former positioned immediately above the latter. laws everywhere, with the production of similar
The photograph should be marked only with the suites of lithofacies. For example, hydrodynamic
necessary identifying information. This is analogous structures, such as ripples and cross-bedding, are
to the practice now adopted for the publication of formed by the migration of ripples and dunes. Con-
seismic sections. An uninterpreted line is given to- siderable experimental work has shown that in the
gether with the interpretation, so that the reader can development of these bedforms there are consistent
make an independent assessment of the raw data. empirical relationships between bedform size and
Alternatively, the overlay can be printed in place on shape and a limited suite of physical parameters, of
the outcrop photograph, or it can be published on its which the most important are the depth and velocity
own, without the photograph. Both methods have of the flow, and the grain-size of the sediment (e.g.,
the disadvantage of the loss or obscuring of informa- Harms et al. 1982). Repetition of similar conditions
tion. leads to repetition of similar depositional products,
which are then susceptible to a universal empirical
classification in the field.
Miall (1977) reviewed braided-river deposits, and
4.3 Classification of Lithofacies
demonstrated a consistency in the lithofacies assem-
blages occurring in a wide range of modern and
Observation and classification of lithofacies are now ancient sandy and gravelly sediments. However,
standard components of the facies-analysis metho- many of the researchers whose work was examined
dology for studying sedimentary rocks. Good sum- in this review had erected their own local classifica-
maries of the principles and methods are contained tion, which had obscured the similarities between
in Reading (1986), Miall (1990, Chap. 4), and Walker the deposits and prevented the recognition of com-
and James (1992). Beds are classified on the basis of mon depositional themes. Miall proposed a simple
their primary depositional attributes, notably (in the classification, making use of a two-letter code to
case of fluvial clastic deposits) bedding, grain size, facilitate quick field and laboratory identification
texture, and sedimentary structures. Biogenic struc- and documentation. Use of this lithofacies scheme
tures and fossils may be important locally as addi- by a number of workers led to the recognition that
tional descriptive attributes. Chemical sediments, the scheme could be applied to all fluvial deposits,
not just those of braided type, but also demonstrated purposes; how thick should a sand bed be before it is
that it was incomplete. In a later paper, Miall (1978c) separated out from Fl? Bed thickness cutoffs, and
expanded the classification with the definition of a accessory percentage limits should be established at
number of additional minor but significant litho- the outset of a project exercise in order to facilitate
facies types. Subsequently, the classification has consistent logging practices.
been used by dozens of researchers and has become The reader should be aware of the recent review of
a standard field methodology for the examination of bedform classifications undertaken by the Bedforms
fluvial deposits. and Bedding Structures Research Group of the Soci-
Not all professionals agree with the use of prede- ety for Sedimentary Geology (SEPM)(Ashley 1990).
termined lithofacies schemes, such as that described A synthesis of modern work on bedforms of all types,
here. Bridge (1993a) remains amongst the most scep- in fluvial, tidal, and other environments, resulted in
tical of their value. He is particularly concerned that a uniform approach to classification and a better
the use of such schemes by the inexperienced will understanding of the causes of morphological varia-
lead to their uncritical application, with a possible tion. This is discussed in Chap. 5.
loss of important detail where new observations
might have been made of deposits and structures
that differ from the "standard". This is an important
4.4 Principles of Paleocurrent Analysis
warning that should be borne in mind by all re-
searchers.
The scheme is presented in Table 4.1, and is used Paleocurrent data are essential for the purpose of
throughout this book. Modifications have been architectural-element analysis. For the typical, rela-
made to the scheme originally presented in Miall tively flat, two-dimensional cliff outcrop, these data
(1978c) in an attempt to eliminate some superfluous provide the essential third dimension. Orientation
categories and to better discriminate between oth- information on hydrodynamic sedimentary struc-
ers. The capital letter in the facies code indicates tures reflects the internal geometry of bar com-
dominant grain size (G = gravel, S = sand, F = fine- plexes, channels, and sand sheets. The dip and strike
grained facies, including very fine sand, silt, and of bounding surfaces reveal the orientation of accre-
mud). The lowercase letter serves as a mnemonic for tionary bar growth and meander migration (Fig.
the characteristic texture or structure of the litho- 4.1). Overall, the data will show the orientation of
facies (e.g., p = planar cross-bedding, ms = matrix- the outcrop relative to channel and bar trends, which
supported). The lithofacies are described in detail, is important to know when carrying out visual ex-
with illustrations, in Chap. 5. amination of the architecture in order to assess bar
It is recommended that the researcher use Table and channel evolution. For such purposes, even very
4.1 as a basis for field research, while remaining alive limited data can be extremely useful.
to the possibility that refinements are always pos- The tradition in basin-analysis practice has been
sible, based on detailed observations of new units. to collect very large numbers of readings without
For example, Leblanc Smith (1980) incorporated necessarily recording very precise data regarding
details on grain size into the coding scheme. the location of the measurements or, at best, keying
Haszeldine (1983a,b), who did not, in fact, use the such measurements to vertical profiles. While such
lithofacies scheme of Table 4.1, subdivided trough methods provide good estimates of local and re-
cross-bedding (lithofacies St) into two types based gional paleocurrent trends and of vertical changes in
on trough thickeness, and was able to demonstrate these trends, they do not help much in the elucida-
the utility of this refinement in interpreting the action of the internal architecture of the sand bodies.
cretionary development oflarge bar forms. A similar This approach also does not address the concerns
subdivision was used by Eberth and Miall (1991), regarding the architectural scale of depositional
who employed the codes Sti and Stii for large- and units (Chap. 3), and its effects on paleocurrent vari-
small-scale trough cross-bed sets, respectively. ability (Sect. 2.3.5; Fig. 2.10).
Some of the lithofacies classes are gradational The method recommended here is to use the pre-
with others. For example, there may be no clear liminary interpretations of the bounding surfaces
distinction in the field between lithofacies Sh and Sl; drawn up prior to detailed field examination of the
lithofacies Fl may contain minor coal streaks or car- outcrop as a guide to the architectural subdivision of
bonate nodules, leading to questions about a work- the rocks. Attempts should then be made to collect a
able definition of lithofacies C and P for logging suite of readings from each of the major elements
Principles of Paleocurrent Analysis 79
Table 4.1. Facies classification. (Modified from Mialll978c)

Facies code Facies Sedimentary structures Interpretation
Gmm Matrix-supported, Weak grading Plastic debris flow
massive gravel (high-strength, viscous)
Gmg Matrix-supported Inverse to Pseudoplastic debris flow
gravel normal grading (low strength, viscous)
Gci Clast -supported Inverse Clast-rich debris flow (high strength),
gravel grading or pseudoplastic debris flow (low
strength)
Gcm Clast-supported Pseudoplastic debris flow
massive gravel (inertial bedload, turbulent flow)
Gh Clast-supported, Horizontal bedding, Longitudinal bedforms,
crudely bedded imbrication lag deposits,
gravel sieve deposits
Gt Gravel, Trough cross-beds Minor channel fills
stratified
Gp Gravel, Planar cross-beds Transverse bedforms,
stratified deltaic growths from older
bar remnants
St Sand, fine Solitary or grouped Sinuous-crested and linguoid
to very coarse, trough cross-beds (3-D) dunes
may be pebbly
Sp Sand, fine Solitary or grouped Transverse and linguoid
to very coarse, planar cross-beds bedforms (2-D dunes)
may be pebbly
Sr Sand, very Ripple Ripples (lower flow regime)
fine to coarse cross-lamination
Sh Sand, very fine Horizontal lamination Plane-bed flow
to coarse, parting or streaming (critical flow)
may be pebbly lineation
Sl Sand, very fine Low-angle « 15°) Scour fills, humpback or
to coarse, cross-beds washed-out dunes,
may be pebbly antidunes
Ss Sand, fine to Broad, shallow scours Scour fill
very coarse, may
be pebbly
Sm Sand, fine Massive, or faint Sediment-gravity
to coarse lamination flow deposits
Fl Sand, silt, Fine lamination, Overbank, abandoned channel, or
mud very small ripples waning flood deposits
Fsm Silt, mud Massive Backswamp or abandoned channel
deposits
Fm Mud, silt Massive, desiccation Overbank, abandoned channel, or
cracks drape deposits
Fr Mud, silt Massive, roots, Root bed, incipient soil
bioturbation
C Coal, carbon- Plant, mud films Vegetated swamp deposits
aceous mud
P Paleosol carbonate Pedogenic features: Soil with
(calcite, siderite) nodules, filaments chemical precipitation
t.
accretion surfaces
within a single bar or bar complex, indicated direc-
tions can vary widely, sometimes over a range of
more than 180°. However, calculation of statistically
significant mean and variance data for each element
may be less valuable than the information provided
on the evolution of the bar with time. For example,
paleocurrent directions may show regular lateral
changes across the bar, indicating a change in accre-
orientation of tion directions as a result of shifts in meander posi-
cross-bedding
tion. Less than half a dozen readings may be
adequate to document this, especially if the data are
combined with information on the orientation of
bounding surfaces. Readings from successive ele-
ments show how the entire river system evolves
dip of through time.
accretion surfaces In order for such interpretations to be made, it is
obvious that the location of each paleocurrent read-
ing must be recorded very precisely. Each measure-
ment should be recorded by a numbered point on
orientation of the profile overlay. Normally, the location within the
cross-bedding
profile can readily be determined by comparing the
->- dip-strike of accretion surface shape of the outcrop and the location of vegetation
orientation of cross-bedding
around the observation point to the same details on
the photograph. Sometimes, however, the monotony
1>.:i:·:.::~:::J exposed active bars of the outcrop may make this difficult, or the geolo-
gist may be required by the steepness of the face to
Fig.4.1. Hypothetical examples of fluvial styles, indicat-
stand too close to the outcrop to be able to determine
ing the range of orientations of dipping accretion surfaces
and the variation in cross-bed orientations. A High-sinu- an exact location. In such cases, it is useful for
osity river, such as a typical meandering river. Note that records on the profile overlay to be made by an
although there is a regional overlap of cross-bed and accre- assistant, who locates him/herself some distance
tion-surface orientations, locally the two types of dip are away from the face. Information can be conveyed to
oriented nearly perpendicular to each other, and the rose
diagrams reflect this divergence. A counterpoint bar is the assistant orally, if necessary with the aid of two-
shown. Such bars can be distinguished from the more way radios.
common point bars by the fact that the accretion surfaces Details on what structures to measure and what
exhibit curvature in plan view that is concave in a downdip they indicate, methods to use for incompletely ex-
direction, contrasting with the convex curvature of point- posed structures, corrections for structural distur-
bar surfaces. Commonly, such bars are characterized by
finer grain sizes than other macroforms. B Low sinuosity bance, and some references to statistical and
river - a typical sandy-braided river. Such rivers may computer methods, are contained in Miall (1990,
exhibit relatively high local channel sinuosity, in which Sect. 5.9).
case macroforms accreting downstream may also be ori- Paleocurrent data can readily be displayed on the
ented at a high angle to the regional trend. In the example final profile that is prepared for publication. Indi-
shown here mean directions are skewed to the southeast,
vidual readings should be shown by arrows, with the
diverging from the overall channel orientation of south-
southeast because of a major channel and bar complex head of the arrow located at the measurement point,
(near the center of the reach) oriented in an easterly direc- and the tail of the arrow oriented to indicate direc-
tion. (Miall1994) tion. Symbols added to the tail, such as varying num-
bers of ticks, or variations in the design of the arrow
head, or some other device, may be used to indicate
exposed in the face. It is essential to key each mea- the type of structure indicated by the arrow. Orienta-
surement to structure type and lithofacies, which tion measurements on bounding surfaces should be
can readily be done using the lithofacies code clearly differentiated, for example, with differently
scheme discussed in Sect. 4.3. The usual rules about sized arrows. The orientation of the arrow tail may
the statistical validity of small samples apply at the be plotted in one of two ways to suit the needs of the
level of the individual architectural element. Even particular study. The more conventional way is to
Classification of Bounding Surfaces 81
define paleocurrent north as the top of the diagram, sion of the differences between the schemes was
so that, for example, arrows pointing horizontally to given by Miall (1995a).
the left indicate flow due west (azimuth 270°). In
some cases, it renders the profile more easy to read if Bridge Miall
the paleocurrent data are oriented with respect to the
outcrop face. Readings parallel to the face then ap- microscale set (e.g., ripple) 1st-order unit
pear as horizontal arrows pointing left or right, flow mesoscale set (e.g., dune) 1st-order unit
directly into the face is shown as a vertical arrow micro/mesoscale coset 2nd-order unit
pointing upward, and so on. The second method has macro scale inclined 3rd-order unit (macro-
the advantage that the reader can more easily inte- stratum form increment)
grate the architecture of the face and the corre- macro scale inclined 4th-order unit
sponding paleocurrent readings when assessing the strata set (macroform)
profile visually. The disadvantage comes when it is group of macroscale sets 5th-order unit
necessary to compare profiles taken from outcrops (channel)
that have different orientations. Comparisons be- group of macro scale sets 6th-order unit (e.g.,
tween paleocurrent directions in different outcrops channel-belt)
are then not as obvious visually.
Summary paleocurrent rose diagrams may use- A question of scale arises in the use of such classi-
fully be added to the profile, if space permits. A rose fications, because rivers and their deposits vary
for each element is a particularly useful addition. enormously in their size. However, Bristow and Best
Summary statistics can be prepared as a table for (1993) have pointed out that rivers show a consider-
inclusion in the text. able degree of self-similarity over the complete range
The use of paleocurrent analysis as a mapping of scales (see also Sect. 8.3), and the classification
tool is discussed in Sect. 9.5.7. described here can be applied to rivers of all scales
providing due attention is paid to river size. For
example, Fig. 4.4 illustrates the range of depositional
scales in the Mississippi, one of the world's largest
4.5 Classification of Bounding Surfaces rivers. It is instructive to compare this with the sub-
divisions of alluvial fans shown in Fig. 4.3. One of the
The principles of the hierarchical classification of challenges in the study of ancient rocks is to deter-
depositional units are detailed in Chap. 3. In this mine the scale of the rivers responsible for forming
section, it is shown how the ideas can be applied the deposits, based on architectural studies of the
more specifically to fluvial deposits. A version of preserved record.
Table 3.2 that focuses on fluvial deposits is included First- and second-order surfaces record bound-
here as Table 4.2, and incorporates a column sum- aries within microform and meso form deposits. The
marizing the nine major types of bedding and definition of first-order surfaces is unchanged from
bounding surface that characterize fluvial deposits. Allen (1983a). They represent cross-bed set bound-
Most of these types are illustrated in Figs. 4.2 and 4.3. ing surfaces (Fig. 4.5). Little or no internal erosion is
The method was developed initially for sandstone apparent at these boundaries, and they represent the
deposits, but further work by S.A. Smith (1990) and virtually continuous sedimentation of trains of
Soegaard (1990,1992) has demonstrated the applica- bedforms of similar type. Subtle modifications in
bility of the classification to conglomerates as well. attitude, with minor erosion, may be caused by reac-
DeCelles et al. (1991) used the system, with some tivation following stage changes (Collinson 1970), or
modifications, to study alluvial-fan deposits. As S.A. may be the result of changes in bedform orientation
Smith (1990) pointed out, the textural and structural (Haszeldine 1983b). In core, these surfaces may not
monotony of conglomerates may make recognition be very prominent, but the presence of a bounding
and correlation of bounding surfaces more difficult surface can be recognized by truncation or wedge-
in these types of deposit. out of cross-bed foresets.
An alternative classification of depositional units Second-order surfaces are simple coset bounding
and bounding surfaces was erected by Bridge surfaces, in the sense of McKee and Weir (1953).
(1993a), but based on very similar principles to those These indicate changes in flow conditions, or a
discussed here. A comparison between the two no- change in flow direction, but no significant time
menclature schemes is provided below, and a discus- break (Fig. 4.6). Lithofacies above and below the
Table 4.2. Hierarchy of depositional units in alluvial deposits. (Modified from from Miall1991b)
Grp Time Examples Instan- Fluvial, Rank and
scale of taneous deltaic characteristics
of processes sedimentation depositional of bounding
process rate (m/ka) units surfaces
(a)
10-6 Burst-sweep Lamina Oth-order,
cycle lamination surface
2 10- 5 Bedform 105 Ripple 1st-order,
-10- 4 migration (microform) set bounding surface
3 10-3 Bedform 105 Diurnal 1st-order,
migration dune increment, set bounding surface
reactivation surface
4 10-2 Bedform 104 Dune 2nd-order,
-10- 1 migration (mesoform) coset bounding surface
5 10° Seasonal events, 102- 3 Macroform 3rd-order,
-10 1 lO-year flood growth dipping 5-20° in direction
increment of accretion
6 102 100-year flood, 102-3 Macroform, 4th-order, convex-up
-10 3 channel and e.g., point bar, macroform top, minor channel
bar migration levee, splay scour, flat surface bounding
immature paleosol floodplain elements
7 103 Long-term 10°-10 1 Channel, 5th-order,
-10 4 geomorphic delta lobe, flat to concave-up
processes, e.g. mature paleosol channel base
channel avulsion
8 104 5th-order 10- 1 Channel belt, 6th-order,
-105 (Milankovitch) alluvial fan, flat, regionally extensive,
cycles, response minor sequence or base of incised valley
to fault pulse
9 105 4th-order 10- L 10- 2 Major dep. system, 7th-order, sequence
-10 6 (Milankovitch) fan tract, boundary; flat, regionally
cycles, response sequence extensive, or base of incised
to fault pulse valley
10 106 3rd-order 10- 1-10- 2 Basin-fill 8th-order,
-10 7 cycles. complex regional
Tectonic and disconformity
eustatic processes
surface are different, but the surface is usually not Third- and fourth-order surfaces are defined
marked by significant bedding truncations or other when architectural reconstruction indicates the pre-
evidence of erosion, except for the same kinds of sence of macroforms, including lateral-accretion
minor modification which occur on first-order sur- deposits (LA; see next section for definition of archi-
faces, as noted above. In core, these surfaces may be tectural elements and the element codes) and down-
distinguished from first -order surfaces by a change stream-accreting macroforms (DA). Individual
in lithofacies. Decelles et al. (1991), who studied a depositional units ("storeys" or "architectural ele-
Cenozoic alluvial-fan complex, defined the contacts ments") are bounded by surfaces of fourth-order or
enclosing individual sediment-gravity-flow deposits higher rank.
as first-order surfaces. However, by the definitions Third-order surfaces are cross-cutting erosion
used here such contacts should be ranked as second- surfaces within macro forms that dip at a low angle
order, as they indicate a significant change in flow (normally < 15°) and may truncate underlying cross-
conditions. bedding at a low angle (Figs. 4.7-4.10). They may cut
- - -- - - - - - 1O-3000m - -- - - --+
® bounding surface
heirarchy
Fig. 4.2. The scales of depositional elements in a fluvial being built up by migrating "sand waves". Foreset termi-
system, showing the bounding-surface hierarchy (Table nations of these are shown at top of diagram, but the
4.2). Circled numbers indicate the ranks of the bounding internal cross-bedding that results has been omitted for
surfaces. In c, the two-letter codes are for architectural clarity
element types (Table 4.3). In d, the sand flat is shown as
through more than one cross-bed set. They are com- tics are readily recognized in core. These surfaces
monly draped with mudstones and followed by an indicate stage changes, but no significant change
intraclast breccia. Facies assemblages above and be- in sedimentary style or bedform orientation within
low the surface are similar. Such surfaces are equiva- the macro form. They indicate a form of large-scale
lent to the "epsilon" cross-bedding of Allen (1963a). "reactivation", to adapt Collinson's (1970) term.
Third-order surfaces may also develop at the top of DeCelles et al. (1991) assigned a third-order rank
minor bar or bedform sequences, and are draped by to large-scale cross-strata in fanglomerates, which
mudstone or siltstone, indicating the falling stage. he also interpreted as accretion surfaces on macro-
Succeeding strata commonly contain a basal forms.
intraclast breccia composed of rip-up clasts of the Fourth-order surfaces represent the upper
draping fine-grained sediments. These characteris- bounding surfaces of macroforms, in the original
Fig. 4.3. The major geomorphological

features of alluvial fans, showing the
7 (FORMATION BOUNDARY) rank numbering of lithosomes and
bounding surfaces used by DeCelles et
al. (1991). The cycle of dispersion, en-
trenchment and backfllling is adapted
from Schumm et al. (1987) and is dis-
cussed in Chap. 10
5(CHANNEL COMPLEX)
6 (FAN SURFACE)
ENTRENCHMENT
BACKFILLING: 5th-ORDER L1THOSOMES (LOBES) BACK UP

ONTO FAN & BURY 5e AND 5a SURFACES
scheme of Miall (1988a,b). However, several re- such as chute channels, would be included in this
searchers have used this rank in different ways, and category. Where major channels are present these
there is potential for confusion if a consistent usage would be bounded by higher order surfaces. Indi-
is not observed. In the braided-fluvial systems stud- vidual crevasse-splay lithosomes within the flood-
ied by the writer, fourth-order surfaces are typically plain are bounded by fourth-order surfaces, and
flat to convex-upward (Figs. 4.8, 4.9). Underlying may contain separate growth increments bounded
bedding surfaces and first- to third-order bounding by third-order surfaces.
surfaces are truncated at a low angle or may be In the work of DeCelles et a1. (1991) two types of
locally parallel to the upper bounding surface, indi- fourth -order surface were recognized, erosional (4e)
cating that they are surfaces oflateral or downstream and accretionary (4a). Macroform bounding sur-
accretion. The form of the accretionary surface is faces, analogous to those defined by Miall (1988a,b)
commonly echoed by that of internal third-order are designated type 4a. Type 4e is defined for the
surfaces within the underlying macroform element. "erosional bases of large lenticular lithosomes."
Mud drapes on the element underlying the surface DeCelles et a1. (1991) interpreted these as channels,
are common. which is a potential source of confusion with the
Macroforms are "group 6" deposits in the classifi- writer's scheme (but see below). DeCelles et a1.
cation of Table 3.2. They represent sedimentary (1991) stated that fourth-order surfaces are by far the
events with a duration or recurrence interval of most easily recognizable surfaces in his rocks,
1OL 103 years. It is appropriate to define other types whereas Cowan (1991) found it difficult to distin-
of fourth-order surface for other lithosomes of this guish surfaces of this type in his braided-fluvial
rank. The basal scour surface of minor channels, sandstones, and concluded that accretionary upper
LEVELS OF RESERVOIR HETEROGENEITY

LEVEL l-MeANDER BELT LEVEL 2-MEANDER SCROLL
.1
Fig. 4.4. Hierarchies of depositional units in the modern third-order surfaces. The entire point-bar deposit is en-
Mississippi River. Although this river and its deposits are closed by a fourth-order surface. Level 3 A point-bar incre-
at a much greater scale than most fluvial deposits encoun- ment, bounded by a third-order surface. Bedding units of
tered in the rock record, the same principles of hierarch i- different type are in contact across second-order surfaces.
cal classification apply. Compare this figure with Fig. 4.3. Level 4 Lobe sheets consist in part of stacked bedforms
Levell Bounded at the base by a fifth-order surface. Level of comparable type, separated by first-order surfaces.
2 Individual point-bar increments are defined by several (Jordan and Pryor 1992, reprinted by permission)
bounding surfaces of macroforms are rarely pre- well spacing, as the units defined by third-order
served in his particular fluvial system. surfaces are estimated to have dimensions of less
Third- and fourth-order surfaces may be recog- than about 10 ha (Miall 1988a; Doyle and Sweet
nized in the subsurface by their low depositional 1995).
dips (Figs. 4.8-4.1O). The dip should be apparent in Second-, third-, and fourth-order surfaces, in this
core and may also be revealed by dip meter data and classification, were all included in the second-order
such tools as the Schlumberger MicroScanner device category of Allen (1983a; see Fig. 2.30). Third- and
(Sect. 9.5.8). However, third- and fourth-order sur- fourth-order surfaces correspond to the "minor sur-
faces may be very difficult to distinguish in indi- faces" of Bridge and Diemer (1983; see Fig. 2.29).
vidual wells. The low dips and draping breccias or Fifth-order surfaces are those bounding major
mud lenses are similar. The best way to distinguish sand sheets, such as broad channels and channel-fill
these surfaces from each other is if the lithofacies complexes (Figs. 4.7-4.12). They are generally flat to
assemblages above and below the surface are differ- slightly concave-upward, but may be marked by lo-
ent, indicating a change in element type. Correlation cal cut-and-fill relief and by basal lag gravels. These
between cores would be possible only with very close are the third-order surfaces of Allen (1983a) and the
Fig.4.5. Coset of planar cross-bedding (lithofacies Sp) showing first-order bounding surfaces. Cretaceous, Banks
Island, Arctic Canada
Fig. 4.6. Two second-order surfaces (arrows) enclose a bed of plane-laminated sandstone (lithofacies Sh), which shows
gradational contacts above and below with ripple cross-lamination (Sr). Eocene, Axel Heiberg Island, Arctic Canada
Fig. 4.7. A third-order surface (arrows) within a homo- unit Numbers indicate ranks of bounding surfaces.
geneous, sandstone-dominated lateral-accretion deposit Eocene Eureka Sound Group, Axel Heiberg Island, Arctic
that accreted to the left. The macro form rests on a channel Canada
scour (fifth-order) surface (arrows) above a siltstone
"major surfaces" of Bridge and Diemer (1983; see the accretion of gravel as discrete macro forms. This
Figs. 2.29, 2.30). Many paleosol horizons may be creates horizons of laterally continuous, but geneti-
correlated with specific channels, indicating con- cally distinct macro forms which are separated by
temporaneous development. These horizons within high-angle contacts." Such high-angle contacts
floodplain sequences may therefore be classified as would include the fifth-order contacts between frag-
fifth-order surfaces. ments of separate channel-fills, the fourth-order
As first pointed out by Williams and Rust (1969), contacts separating discrete macro forms, and the
there is a hierarchy of channels in many multiple- third-order contacts developed by macroform reac-
channel rivers (Fig. 2.28). The major channel scour is tivation. All might appear similar in outcrop. Very
defined as the fifth-order surface, and the minor careful textural and clast-imbrication studies may
channel fills within the complex are then bounded by assist in identifying the correct rank of each surface.
fourth-order surfaces. In the work of DeCelles et al. Soegaard (1990, 1992) proposed slightly different
(1991), fifth-order surfaces are considered to be terminology for the definition of the various ranks,
"mosaics of fourth-order surfaces", and represent in his study of a Pennsylvanian fanglomerate, but in
the basal bounding surfaces of fan trenches and practice the geometrical and genetic characteristics
lobes. They defined erosional (Se) and accretionary of the units he defined do not differ markedly from
(Sa) surfaces. Type Se is comparable with the fifth- those discussed here.
order surfaces bounding channel belts, as defined Sixth-order surfaces define groups of channtHs or
here. paleovalleys. Mappable stratigraphic units such as
In conglomerates, S.A. Smith (1990) noted that members or submembers are bounded by sixth-
"abrupt lateral transitions between distinct macro- order surfaces (Fig. 4.12). Surfaces of sixth-order
form types occur in the abandoned reaches of some and higher rank were not defined by Allen (1983a).
modern gravel-bed rivers. These gravel macroforms DeCelles et al. (1991) defined the major bounding
are juxtaposed as a result of channel migration and surfaces enclosing entire alluvial fans as sixth-order
D~
=----- G~ _---------;--_ _
-4 Sh
1-1 5
------
o 4m Sh
b ...._
1 ..............--'''"--...
Fig. 4.8. A downstream-accreted macroform (element indicated, and the macroform rests on a channel-floor
DA) bounded at the top by a convex-up fourth-order sur- fifth-order surface. Kayenta Formation (Jurassic), Colo-
face. One of several internal third-order surfaces is also rado (see Miall1988c)
surfaces. Where fan deposits are distinguished by Seventh-order surfaces enclose the major litho-
different clast lithologies or textures, such surfaces somes representing discrete allogenic events.
may be prominent and easily mapped, but fan DeCelles et al. (1991) recognized that such a surface
interfingering may generate ragged contacts corre- encloses the discrete, but laterally equivalent fan
sponding to the surfaces oflower rank defining com- bodies assigned to the Beartooth Conglomerate
ponent channels and macro forms, and these may be stratigraphic unit, all of which represent the sedi-
difficult to trace. mentary response to a pulse of tectonic activity along
Fourth-, fifth-, and sixth-order surfaces may ap- the basin-margin fault. Many sequence boundaries
pear very similar to third-order surfaces in core. also constitute seventh-order surfaces. Lithologic
They are best differentiated by careful stratigraphic contrasts across surfaces of this type are likely to be
correlation between closely spaced cores, an objec- significant. As discussed in Chaps. 11 and 13, distin-
tive that is best achieved in the most intensively guishing between the various allogenic mechanisms
developed field, where well spacing may be a few that can generate seventh-order lithosomes and
hundred meters, or less. Considerable potential also bounding surfaces (such as tectonism and eustasy) is
exists to identify and map surfaces of this rank using one of the current major research tasks in strati-
3-D seismic data. Brown (1991) provided several graphic geology.
examples of channels mapped using seismic data Eighth-order surfaces are regional dis conformity
(fifth-order surfaces), and it may well prove possible surfaces that develop in response to continental-
to map individual components of a channel fill if the scale or global geologic events. The boundaries of
interpreter knows what to look for. third-order sequences are of this rank.
Classification of Architectural Elements 89
Fig. 4.9. A lateral-accretion deposit 4 m thick, resting on a quence. Several internal third-order erosion surfaces are
major fifth-order surface and capped by a gradational present. All surfaces are indicated with arrows showing
fourth-order surface, above which is the floodplain se- rankings. Pennsylvanian, Alabama
Correct identification and correlation (where order surface in the floor of an adjacent channel.
possible) of these first- to eighth-order bounding Third-order surfaces may downlap onto channel-
surfaces are clearly essential if the various scales of scour surfaces of sixth order or higher rank where
reservoir heterogeneity are to be identified and dealt these define sequence boundaries.
with appropriately in the design of production mod-
els (Chap. 14). Even in excellent outcrop the correct
classification of bounding surfaces is not always
easy. Three useful rules may make the task easier: 4.6 Classification of Architectural Elements
1. A surface of any given order may be truncated by
a surface of equal or higher order, but not by one Early attempts at the classification of architectural
oflower order. In this way, the rank of the surface elements are discussed in Chap. 2 (see Figs. 2.39,
always defines the major process that generated 2.40). The nomenclature proposed by Miall (1985),
it. as revised in later work (Miall 1988a,b) is reviewed
2. Accretionary surfaces may be removed by erosion here. Recent work in which the method has been
preceding the deposition of the next unit. In such used is listed by Miall (1994, 1995a) and is briefly
cases, the top of a lithosome may be more logi- discussed by Fielding (1993b). A discussion of some
cally defined on the basis of what follows it. For of the problems with the architectural approach
example, the top of a macro form is defined by a (Bridge 1993a; Miall 1995a) indicates the need for
fourth-order surface, except where it has been cut caution in the definition of architectural elements
into by a major channel, the base of which typi- and the use of the technique in field research.
cally constitutes a fifth-order surface. A brief note on the meaning of the terms architec-
3. Minor surfaces may change rank laterally. For tural element and lithosome is in order. In this book,
example, the upper fourth-order bounding sur- the term lithosome is used to refer to a discrete
face of a macro form may merge into a second- depositional unit bounded by a distinct bounding
Fig. 4.10. A macroform, probably oflateral-accretion type, with several internal third-order erosion surfaces indicating
changes in accretion direction. These are indicated by arrows. Cretaceous, Gamtoos Basin, South Africa
Fig.4.11. Closeup of a fifth-order surface at the base of a from others of third- to sixth-order in small outcrops or
major sandstone sheet. Note minor downcutting and core. Permian, Poland
intraclast breccia. There is little to distinguish this surface
Fig. 4.12. Two major sixth-order surfaces (arrows) divide hollow (element HO: see Sect. 6.9) about 200 m wide,
this unit into three submembers, which can be traced for floored by a concave-up fifth-order surface. Morrison
about 200 km. The middle submember contains a scour Formation (Jurassic), New Mexico
surface, of any rank. The word is thus a useful, de- Architectural elements are amenable to descrip-
scriptive, nongenetic term. tive and genetic classification, as are their compo-
When a river is viewed from the air, it can be seen nent lithofacies. Miall (1985) proposed the following
to consist of various straight and curved channel components of a descriptive classification:
reaches, and large areas of exposed gravel, sand, or
1. Nature of lower and upper bounding surface: ero-
mud, termed bars. In many rivers the shape and
sional or gradational; planar, irregular, curved
distribution of these various features might seem
(concave-up or convex-up).
chaotic, but most such units have distinctive surface
2. External geometry: sheet, lens, wedge, scoop, U-
forms, the terms for which, such as point bar, side
shaped fill.
bar, sand flat, chute channel, crevasse splay, etc.,
3. Scale: thickness, lateral extent parallel and per-
constitute the familiar lexicography of fluvial geo-
pendicular to flow direction.
morphology. Their development follows certain
4. Lithology: lithofacies assemblage and vertical se-
relatively predictable patterns (Bridge 1985, 1993b)
quence.
that leave their record in the resulting deposits. The
5. Internal geometry: nature and disposition of in-
channels and bars are the component depositional
ternal bounding surfaces; relationship of bedding
elements of the river, and the sediments that com-
and first- to second-order surfaces to these sur-
prise them are termed architectural elements. An
faces (parallel, truncated, onlap, downlap)
architectural element may be defined as a compo-
6. Paleocurrent patterns: orientation of flow indica-
nent of a depositional system equivalent in size to, or
tors relative to internal bounding surfaces and
smaller than a channel fill, and larger than an indi-
external form of element.
vidual facies unit, characterized by a distinctive fa-
cies assemblage, internal geometry, external form. Miall (1985) suggested that there are eight basic
and (in some instances) vertical profile. The term is architectural elements in fluvial deposits, based on a
used here for units enclosed by bounding surfaces of review of studies available up to the early 1980s (Fig.
third- to fifth-order rank. 4.13). Subsequent work has provided a number of
elaborations of this original classification and, as Element Interpretation Element

Fielding (1993b) pointed out, researchers may need code
to erect their own classification to better reflect the
Top stratum floodplain deposits FF
characteristics of the fluvial unit being investigated.
(A) sand wedge levee LV
Thus, Cowan (1991) defined a completely new ele-
(B) proximal sand marginal overbank CS
ment, the scour hollow, based on his work in the
sheet sheet
Morrison Formation of New Mexico, and Eberth and
(C) distal sand sheet distal overbank sheet CS
Miall (1991) showed how detailed work on a specific
unit, in this case the Cutler Formation of New Ridge accretionary deposit CH, LA,
Mexico, could yield a useful subdivision of some of associated with DA
the more important elements. Miall's (1985) original channel shifting
element classification, as modified by recent work, is Chute channel cutoff channel CH
given in Table 4.3. A classification of the architec- Bar platform coarse, in -channel GB, SB,
tural elements in the Cutler Group is given in Table deposits etc.
4.4. A standard classification of architectural ele-
Basal channel channel floor lag GB
ments in the overbank environment is proposed in
gravels
Table 7.1 (details of these various element types are
given in Chaps. 6 and 7).
For the purpose of interpretive annotation of out-
Other element classifications have been used by
crop profiles and efficent note taking, each element
other workers. For example, Lang (1993) subdivided
type may be assigned a two- or three-letter symbol.
element GB into four element types, and developed
Most of these are self-explanatory. The three-letter
his own subdivisions of sandy channel-fill and flood-
codes given in Table 4.4 are designed to permit
plain deposits. Brierley (1991) developed the follow-
subdivision of some of the elements. For example,
ing classification based on surfaces studies of a
channel-fill sandstone bodies (CH) can be readily
modern river floodplain. Corresponding element
subdivided on geometrical grounds into sheets
codes of this book are given in the third column.
(CHS) and ribbons (CHR). Crevasse channels (CR)
are distinguished from the main channels by their
small size and fine-grained fill. They include those
Sp
filled primarily by sandstone (CRS) and those with a
CH
Channei
~CH 51
Sr
mixed, and largely fine-grained fill (CRM). Flood-
plain fines (FF) include two distinct assemblages,
ii!5~~ Gm
those containing sheets of ripple-marked sandstone,
Sh which were probably deposited relatively proximal
FI to the channels (FFP), and those lacking this litho-
Si facies, which represent distal flood-basin deposits
(FFD).
LA Lateral Accretion In employing architectural-element classifica-
tions in the field, three considerations should be
borne in mind:
SG Sediment Grav ity Flow

1. Scale: The question of scale arises because some
__ ..oCI. .-. _ _ ~ .". --Sr elements occur on several different scales within
fZf & 5t
the same deposit, and the researcher needs to
SB Sand Bed Form
adjust his/her sense of scale to the problems to be
addressed. Thus, many deposits could simply be
DA Downstream Accretion
subdivided into channels (CH) and floodplain
LS Laminated Sand
, fines (FF). This is not very useful. Most of the
important detail is derived from a subdivision of
:C <' 4 c e :x 5 ¢ FI
[0.2 - 2.0m FF Overbank Fines these elements into the smaller channels and bars,
different types of flood-basin deposits, and so on.
Fig. 4.13. The eight basic architectu.ral elements in fluvial At a small scale of observation, limited outcrop
deposits. (Miali1988a,b; modified from Miali 1985)
may lead to the classification of an assemblage of
Table 4.3. Architectural elements in fluvial deposits. (Modified from Miall1985)

Element Symbol Principal Geometry and relationships
facies
assemblage
Channels CH Any Finger, lens or sheet; concave-up erosional
combination base; scale and shape highly variable;
internal concave-up 3rd-order erosion
surfaces common
Gravel bars and bedforms GB Gm, Gp, Gt Lens, blanket; usually tabular bodies;
commonly interbedded with SB
Sandy bedforms SB St, Sp, Sh, SI, Lens, sheet, blanket, wedge, occurs as
Sr, Se, Ss channel mIs, crevasse splays, minor bars
Downstream-accretion macroform DA St, Sp, Sh, SI, Lens resting on flat or channeled base,
Sr, Se, Ss with convex-up 3rd-order internal erosion
surfaces and upper 4th-order bounding surface
Lateral-accretion macroform LA St, Sp, Sh, SI, Wedge, sheet, lobe; characterized by
Se, Ss, less internal lateral-accretion 3rd-order surfaces
commonly Gm,
Gt, Gp
Scour hollows HO Gh, Gt, St, SI Scoop-shaped hollow with asymmetric fill
Sediment gravity flows SG Gmm,Gmg, Lobe, sheet, typically interbedded with GB
Gci, Gcm
Laminated sand sheet LS Sh, SI; minor Sheet, blanket
Sp, Sr
Overbank fines FF Fm,Fl Thin to thick blankets; commonly
interbedded with SB; may fill abandoned
channels
Facies classification from Miall (1978b).
Table 4.4. Architectural elements in the Cutler Formation, New Mexico. (Adapted from Eberth and Miall1991)
Element Symbol Lithology Geometry Interpretation
Major ss. CHS Gm, Se, St, wId> 15, with 'wings', Braided fluvial channel
sheet Sh, Sl, storeys 2-5 m thick,
rare Sp units < 1 km wide
Major ss. CHR Gm, Se, St, w/d< 15, with 'wings', Anastomosed fluvial channel
ribbon Sh, SI, storeys <5 m thick
rare Sp
Minor ss CS Se, Sh, SI, Tabular, flat-based, Sheet splay
sheet rare St, Sr < 1 m thick, wing of
CHS sheet
Minor ss CRS Se, St, Sl, w/d< 10, Minor crevasse channel
lens Sh,Sr < 1 m thick
V-shaped CRM Se, Sr, Fl, 1-10 m thick Deep, stable crevasse channel
mixed fill P 1-30 m wide
Interlam. FFP Sr, Fl, P Tabular, Proximal floodplain deposits
mixed < 15 m thick
sheets
Siltstone- FFD Fl, P Tabular Distal floodplain deposits
mudstone < 5 m thick
sheets
cross-bedded sandstone or conglomerate litho- Friend et al. (1979) and Friend (1983). According to
facies as element SB or GB, respectively, whereas these authors, there are essentially three types of
better exposure of the internal geometry of the fifth-order lithosomes: sheet sandstones, formed by
beds may lead to the recognition that the unit is nonchannelized flow, for example during ephemeral
actually part of a macro form, such as DA or LA. flash floods; fixed channels, where the channel is
2. Interbedding: Elements may be interbedded at laterally stable; and mobile-channel belts, where a
the scale of the individual lithofacies set or coset. broad channel or channel complex occupies a fixed
For example, alluvial fanglomerates, comprising position for a lengthy period of time, and develops
stacked GB elements, commonly contain thin minor channels and bars within a broadly defined
stringers of cross-bedded sandstone (element basal scour surface. Mobile channels may evolve by
SB). Many sandstone macroforms are draped by lateral migration or in-channel avulsion, or a combi-
thin mudstone lenses, representing abandon- nation of the two processes. Fixed channels give rise
ment or preservation of FF deposits. At what to ribbon sandstones, and Friend et al. (1979) pro-
point is element subdivision cut off and general- posed that these be defined as those linear sandstone
ized? This is a common problem in facies defini- bodies having a width/depth ratio of less than 15.
tion, and the solution depends only on practical Sheet sandstones and mobile-channel-belt sand-
criteria of data availability and descriptive con- stones have ratios greater than 15, and are com-
venience. monly in excess of 100.
3. Intergradation: Elements tend to grade laterally Sheet sandstones formed by sheet floods com-
into each other. For example, a mid-channel sand monly have flat bounding surfaces showing little
flat in a braided system may accrete laterally evidence of erosion. This type of sheet sandstone,
along its flanks and in a downstream direction at and ribbon sandstones formed by aggradation of
its downstream end. Therefore, a DA unit may fixed channels, commonly have simple fills,.in which
pass laterally along strike into an LA unit, or the litho somes and bounding surfaces of third- and
same transition may occur down depositional fourth-order are absent. By contrast, sheet sand-
dip, indicating an evolution of macroform geo- stones formed as the fill of mobile channels typically
metry with time (Fig. 6.24c). have complex fills composed of the deposits of bar
complexes and minor channels.
It is not suggested that the classifications offered Channel stacking patterns depend on such factors
in Tables 4.3 and 4.4 should be rigidly followed in all as total sediment load, channel migration, and
future fluvial research. Some other classifications switching rates, and basin subsidence rate (Chap.
that have been published are noted above. However, 10). Friend (1983) suggested that where the sediment
it is claimed that the method of systematic observa- load is dominated by coarse, bed-load materials,
tion and classification required to construct such overbank areas would be poorly developed and
descriptive frameworks should be an integral part of channel fills would be superimposed directly, with
future research endeavours. Other researchers may floodplain deposits thin or absent. By contrast, a
wish to construct their own classifications and significant suspension load may lead to significant
codes. However, it is incumbent on all researchers, deposition of floodplain successions, and the
where they modify an existing nomenclature for isolation of channel fills from each other. These
their own purposes, to explain and justify such revi- variations led Friend (1983) to propose a sixfold
sions. Use of the same lithofacies and element codes classification of fifth-order lithosomes (Fig. 2.35),
and bounding-surface rank numbers in ways differ- although no examples of stacked fixed channels
ent from those proposed by earlier workers can lead had been identified at the time he proposed his
to unnecessary confusion. classification (nor has such a depositional pattern
been identified since, according to the writer's
knowledge).
4.7 Classification of Channels The description and classification of the channels
and their component elements are part of the basis
and Larger Bodies for the erection of local "summaries of the environ-
ment", or overall facies models for particular fluvial
Lithosomes bounded by fifth-order surfaces, corre- styles. This process, and the results, are the subject of
sponding to individual channel fills and channel-fill Chap. 8. However, it is worth pointing out, at this
complexes, may be classified using the proposals by stage, that such interpretations are by no means
Annotation of Outcrop Profiles 95
straightforward, and the researcher should be wary beginning and end of sets of transgressive and retro-
of falling into the simplistic traps set in earlier fluvial gressive processes that are repeated in the rock
literature, for example, that lateral-accretion depos- record. Regardless of the terminology used, the ge-
its indicate point bars in meandering channels. As netic concepts that underlie sequence-stratigraphic
noted in Chap. 8, LA deposits occur in a wide range principles are important aids for the regional
of fluvial environments. Brierley and Hickin (1991) stratigrapher (Chap. 13). Employing such principles
carried out a detailed element analysis of a modern can facilitate the understanding of regional strati-
gravel-sand river that showed braided, wandering, graphic processes more readily than the classic
and meandering reaches. The three fluvial channel methods of lithostratigraphy, which classify similar
styles could not be distinguished from their deposits facies together regardless of their age range, and
on the basis of either lithofacies assemblages or ar- separate dissimilar facies even where they are inter-
chitectural-element types. In Chap. 8, it is shown bedded and clearly contemporaneous. Although ba-
that most element types occur in several different sic geological mapping is more easily carried out
fluvial settings. In order to distinguish them and to using lithostratigraphic methods, this older ap-
arrive at reliable interpretations of channel style it is proach is much less useful as a basis for interpreta-
necessary to take into account all available data on tion. Formal allostratigraphic terminology is now
element geometry and orientation. Detailed paleo- being proposed for some such successions. For ex-
current analysis is a very important part of this inter- ample, Plint (1990) subdivided a suite of shoreface
pretive process. deposits (that included some fluvial units formed
Lithosomes of sixth-order and higher rank re- during times of low base level) into a series of
quire classification and documentation using the allomembers defined by regional disconformities
principles of lithostratigraphy or allostratigraphy and cycle boundaries. In another study, referred to
(Sect. 9.6). Increasingly, sequence concepts are now at length in Chap. 13, Van Wagoner et al. (1990)
being used to classify and describe such deposits attempted to adapt existing lithostratigraphic ter-
(Chap. 13). Large-scale channel belts and other dis- minology to a new sequence framework. A unit
crete facies tracts may be mappable as individual originally called the Grassy Member was found to
stratigraphic units and assigned member or forma- contain a major regional dis conformity as a result of
tion rank. For example, DeCelles et al. (1991) defined their sequence analysis, and was split into upper and
the surface enclosing individual fanglomerate bo- lower parts, each of which comprised part of two
dies as sixth-order in rank, and designated the entire separate stratigraphic sequences. The result is a
assemblage of contemporaneous conglomerate clumsy stratigraphic nomenclature that needs a
units as a formation-rank unit, enclosed by a sev- complete recasting in sequence terminology.
enth-order surface. Research on nonmarine sequences is actively un-
Deposits whose large-scale architecture is con- derway (Chap. 13), and a descriptive architectural
trolled by tectonic pulses or repeated base-level classification of sequence geometries and bounding
changes may be suited to a stratigraphic approach surfaces will likely be an eventual outcome. Terms
that employs allostratigraphic principles. As dis- derived from the Exxon work on sequence stratigra-
cussed in Chaps. 11 and 13, fluvial depositional sys- phy, such as low-stand and high-stand systems tracts
tems tracts in the rock record are commonly (Posamentier and Vail 1988; Posamentier et al. 1988;
bounded by major erosion surfaces (of sixth-order Van Wagoner et al. 1990), are interpretive in nature,
and higher rank), and may contain marine flooding and only to be employed following thorough analy-
surfaces. These surfaces are typically regional in ex- sis, as discussed later. They cannot, in any case, be
tent, and indicate major allogenic controls on sedi- applied without qualifications to rocks that are en-
mentation. They provide the basis for subdividing tirely nonmarine.
stratigraphic successions using regional contem-
poraneous or near-contemporaneous erosional or
flooding events. This is the formal basis for
allostratigraphy, as defined in recent stratigraphic
4.8 Annotation of Outcrop Profiles
codes (see Miall1990, Sect. 3.6; Sect. 9.6, this Vol.).
Both regional erosion surfaces and flooding surfaces A completed profile is analogous to a geological
have been proposed as sequence boundaries by dif- map. It is basically a descriptive diagram of the ex-
ferent authors (e.g., Galloway 1989a; Van Wagoner posed rocks. However, like a geological map, it may
et al. 1990), that is, as surfaces used to define the contain a degree of interpretation. On geological
maps uncertain contacts and structures are indi- considerable amount of internal evidence to suggest
cated by dashed or dotted lines, and geological units such correlations.
that cannot be mapped separately may be shown Additional details can be added to the annota-
unsubdivided. The same illustrative devices should tion, where this is deemed useful. Bounding surfaces
be used on outcrop profiles. It is important to re- of third- and fourth-order rank may be designated
member, however, that the classification of bound- by primes or some other form of superscript added
ing surfaces and litho somes, according to the to the surface lettering. Architectural elements so
systems described in this chapter, is an interpretive subdivided may be designated by the letters A, B, C,
process. The classification is firmly grounded on etc. added to the element number. The element code
observable attributes, such as the shape of the may be added after this number, where a reliable
bounding surfaces, and the facies associations of the interpretation can be arrived at. A complete profile
deposits immediately above and below, but a degree annotation may then look something like this:
of interpretation is commonly required to produce a
fully annotated profile. 4-SB
It is recommended that the major bounding sur- D (5th order)
faces in each profile, of fifth-order rank and higher, 3C-DA
be lettered in sequence from the base of the outcrop C2 (4th order)
upwards, using capital letters. The architectural ele- 3B-DA
ments bounded by these surfaces are then numbered Cl (4th order)
in the same order, using Arabic numerals. This sys- 3A-LA
tem provides a basic reference framework for de- C (5th order)
scribing the profile. 2-LA
Each profile should be lettered separately, using a B (5th order)
local lettering and numbering sequence, unless pro- I-SB
files can be correlated laterally. If specific bounding A (5th order)
surfaces can be traced between outcrops, it may be
useful to use the same lettering and numbering se- Some sandstones sheets are built by a complex
quence throughout, but it must be remembered that process of cutting and filling, so that only fragments
fluvial deposits tend to be characterized by rapid of the bars and minor channel fills are preserved in
lateral changes in facies and architecture, and it can- the final deposit, many of which may not extend
not be assumed that surfaces and elements can be completely across the outcrop. Numbering of the
correlated across unexposed areas, unless there is a elements and subelements must then be carried out
..
DEDUCTIVE REASONING
3 4 5
LEVELS I 2
CLIMATIC VARIABLES
Precipitation
/
(Mean • Intensity.
Spalial Variation)
-
LOCAL FLUVIAL REGIME REGIONAL HYDROLOGIC E vaperation
FIELD EVIDENCE
REGIMEN
-
HYDRAULIC Gradient
Paleochannel
Morphology
CONDITIONS
I--
Channel Pallern
Hydraulic Geometr
Rlii10ff
ond =1 Drainage Densit
~ GroLO.C ''''''''_ES
Velocity (width, depth, Sediment Ruggedness
and
Sedimentology Sediment Transport hydraulic rod ius) Yields Soil Thickness
I
Prevailing Lithology.
I
Structure, and
Initial Relief
FEEDBACK
Fig.4.14. Flow diagram showing the relationships bet- attempt to reconstruct these parameters by inductive rea-
ween geological and climatic variables to hydraulic condi- soning, moving from left to right, except for independent
tions and the resulting fluvial style and sedimentary evidence relating to regional geological variables. (Baker
record. The controls, in nature, act from right to left, as 1978a)
shown by the boxes connecting the arrow,. Geologists
::>
::l
::l
o
ao·
::l
o
o'""
a(')
o
...
'0
...o'"
;:n
::;--
en
G=---
F -----:2£ IS G element
n' 16 3 ~_
,.E -~ ~= ~==----::=:::;; _ ~ F, E
B C---'-<~t-
• ~
------ __ .. Sp
r ---= ~ -=--::;
~t::==:::: - 4-581s _~
_
0
- - - __ A _ _
'-=:L-:"---
~..4-.S.!:DA/LA ~A
_~
----- -g - c'
A~~ '=' ;" - - - - - ¥ _ -;F -" -=3.. e la..
SOLDIER CREEK paleocurrent A ~ element I
~ ~~ n t 7
-.025" north
r-
Fig.4.15. Example of a completed, annotated photomosaic through a fluvial deposit. Castlegate Sandstone (Upper Cretaceous) at Soldier Creek, Utah. Major
bounding surfaces are indicated by capital letters, architectural elements are indicated by numbers with element code, where known. Paleocurrent data are
oriented such that flow parallel to the outcrop face is given by horizontal arrows. A detailed interpretation of one of the elements in this profile is provided in
Fig. 6.32. (MialI1994)
\0
....,
with care, and no single vertical profile across the outcrop from the same place as the photographs
outcrop will contain a complete sequence of surface were taken. It is particularly useful to distin-
and element notations. Profiles of this type can pro- guish the major litho somes by defining fifth-
vide a graphic illustration of the lengthy process of and higher-order surfaces wherever possible.
channel migration and bar accretion that can occur 5. Commence traverses of the outcrop, making
within a single mobile channel before it is finally detailed observations of lithofacies (Chap. 5)
abandoned. and paleocurrent directions. Mark observation
points precisely on the profile overlay, if neces-
sary with the assistance of a second worker lo-
cated at the photo site (who has a better overview
4.9 Summary of Methods
of the outcrop). Attempt to complete the tracing
of bounding surfaces left "hanging" following
The field methods described in this chapter are the preliminary photo interpretation. Attempt to
designed to lead to detailed descriptions of strati- obtain a representative suite of readings of cur-
graphic architecture for purposes of resource ex- rent indicators and dipping bounding surfaces
ploration and production, and to facilitate the for at least each fifth-order lithosome, and pre-
interpretation of sedimentary controls. Observation ferably for third- and fourth-order units as well.
leads to interpretation by inductive reasoning. Baker 6. Carry out element interpretation using the ar-
(l978a), in an examination of the methods of fluvial chitectural details obtained from facies and ori-
geomorphology, illustrated the links between sedi- entation data (Chaps. 6, 7).
mentary controls and the real-world outcome by 7. Annotate elements and bounding surfaces.
using a flow chart (Fig. 4.14). Architectural methods 8. Correlate adjacent outcrops. If possible, corre-
for detailed outcrop study, which form the basis for late specific major lithosomes and bounding
inductive reasoning, contain the following compo- surfaces by walking them out or by visual tracing
nents, to be carried out in essentially the order given. from a distance, but be particularly cautious of
References are made to the chapters where the rel- attempting to correlate units and surfaces lower
evant detail is provided. A typical example of a com- than fifth-order in rank (Chap. 9).
pleted, annotated photomosaic is illustrated in Fig. 9. Synthesize lithofacies, element, and paleocur-
4.15. rent data to arrive at interpretations of fluvial
style and channel pattern (Chaps. 8, 9).
1. Photograph the outcrop and construct a photo-
10. Where appropriate, develop a regional allo-
mosaic.
stratigraphic and sequence framework from
2. Document the scale and orientation of the pro-
such regional correlations (Chap. 13).
file.
11. Where required, measure the scale and geom-
3. Enter all visible bounding surfaces on a profile
etry oflithosomes at each level of the hierarchy,
overlay.
for the purpose of documenting reservoir het-
4. Check the correctness of the surface drawings in
erogeneities (Chap. 1O).
the field, and carry out preliminary element in-
terpretation. This can be done by viewing the
Chapter 5
Lithofacies
5.1 Introduction processes, but individual flows may evolve from one
set of conditions to another, so that the final deposit
may reveal a range of textures and structures within
Most fluvial deposits are clastic. A simple tripartite
one lithosome. Figure 5.1 illustrates a model that
subdivision into gravel, sand, and fine-grained
attempts to summarize the relationships between
(mud, silt, and very fine-grained sand) lithofacies is
the various processes and their results.
a useful first step in description and classification.
The main differentiation is between two types of
Most beds may be classified readily into one or the
process: (1) flows that are turbulent and have low
other of these groups, reflecting a natural separation
sediment concentrations, in which transport and
of processes and sorting of the sediment load. Some
deposition take place by traction and very 'limited
mixed lithofacies, such as pebbly sandstones, do oc-
suspension, and (2) flows in which the higher vis-
cur but can normally be classified according to their
cosities associated with high sediment concentra-
dominant grain-size class.
tions dampen turbulence, leading to laminar flow
This chapter is intended primarily as descriptive.
and grain support by buoyancy. This second class of
Other texts, such as Middleton and Southard (1977),
event is given the general term sediment-gravity
Collinson and Thompson (1982), and Allen (1984),
flow.
deal with the mechanics of sediment transport and
Useful work on gravel sedimentation by traction
the theory of bedform generation in some depth, and
currents and the resulting lithofacies has been car-
it is not the purpose of the present book to duplicate
ried out in modern rivers by Williams and Rust
this material. Some of the major recent advances in
(1969), Rust (1972), N.D. Smith (1974), Boothroyd
this area are touched on in Chap. 2. Recent work on
and Ashley (1975), Lewin (1976), Hein and Walker
bedform genesis and classification (especially
(1977), Allen (1983c), Southard et al. (1984), and
Ashley 1990) is discussed in this chapter only where
Carling (1990). Ashmore (1991) carried out a very
it provides a useful genetic underpinning for the
useful series of model experiments. Sediment-grav-
lithofacies classification. The major lithofacies types
ity flows are locally important, particularly in allu-
are listed in Table 4.1. Most were initially proposed
vial fan settings, and have been studied by
by Miall (1977, 1978c); references to other defini-
Blissenbach (1954), Bluck (1967), Johnson (1970),
tions are given where appropriate.
Larsen and Steel (1978), Schultz (1984), Blair (1987),
Maizels (1989,1993), Hubert and Filipov (1989), and
Blair and McPherson (1992, 1994).
5.2 Gravel Facies
S.2.1 Depositional Processes in Gravel-Bed Rivers 5.2.7.2 Traction Currents, Fluid Flows
5.2.1.7 Introduction The processes of gravel transport and gravel bed-

form, channel and bar formation are less well under-
Gravel transport takes place under a wide spectrum stood than those of sand-bed rivers. The reasons for
of physical conditions and leads to a range of tex- this are partly observational. It is much more diffi-
tural and structural variations in the ensuing depo- cult to observe gravel-bed rivers during active bed-
sits. The major controlling factors are the condition load transport because the high transport energies
of the flow, either laminar or turbulent, and the make in-place measurement and sampling hazard-
sediment concentration. Specific depositional re- ous, because direct observations may be prevented
sponses (lithofacies) are associated with particular by the opaqueness of the water, which commonly
100 Lithofacies
CHARACTERISTICS OF FLUID-SEDIMENT MIX
LAMINAR .... . . TURBULENT

• HIGH~(70%) SEDIMENT CONCENTRATION (40 %) -'LOW -.(0%)
NON-NEWTONIAN ..... BINGHAM NEWTONIAN • FLUIDAL
COHESIVE .... . . COHESIONLESS
VISCOUS,HIGH YIELD STRENGTH III( . . NON -VISCOUS, LOW YIELD STRENGTH
FLOW TYPE DEBRIS FLOWS . . . HYPERCONCENTRATED GRAIN FLOWS"'FLUID FLOWS
DOMINANT
CLAST-SUPPORT
MECHANISM
LOBES TYPE G TYPE G
I
I
MAssivE TYPE A
:
I
C I
I
STRlJCTURELESS ____ ~rM-A-SS-IV-E---' TYPE C 3/4
rn I NON GRADED ___ ~ STRUCTURE LESS
oQ.. I
I
I
POORLY SORTED ~ NON GRADED ~ CRUDELY BEDDED
1&.1 WELL SORTED ~ IMBRICATED,
o SHEETS INVERSELY
...o +
CHANNELS
GRADED
>- ,
I
I
<!I
o
-J
:
I
o
•
I
I
J: I
Q..
II:
o
::IE
I--"'~INCREASED EROSIONAL CAPACITY - .....~ '-_ _---'--'-'
Fig. 5.1. Relationships between the physical characteris- stages (in the direction shown by the arrows), resulting in
tics of a fluid-sediment mixture, the resulting flow type, vertical and downstream gradations in texture and struc-
and the corresponding deposit. Note that an individual ture. (Maizels 1989)
flow event may evolve through more than one of these
bears a heavy suspended-sediment load, and be- tural elements in the final deposit. In addition to
cause trenching of coarse gravels is difficult. It has changes in conditions imposed by external factors,
been found that there are multiple processes affect- such as diurnal and seasonal variations in discharge
ing the morphology and stratification of gravel de- and temperature, deposition and erosion at the bar
posits, and that these interact and change very and bedform scale have their own continuous effects
quickly, so that relating hydraulics to bedforms and on water velocities, depths, and sediment loads
stratification is a complex problem. Flume modeling immediately over, around, and downstream from
of gravel-bed rivers has been slow to generate useful each feature, resulting in corresponding localized
results, because of equipment limitations. However, changes in the conditions nearby. For example,
it has now been demonstrated that modeling of deposition of a gravel lobe locally raises the water
gravel transport can be done in the flume by using level, and can lead to small-scale flooding and chan-
sand, and by careful scaling down of all appropriate nel diversion. Bank and bar erosion feeds slugs of
physical parameters. bed load into the system, which may be deposited as
Analysis of gravel-bed rivers and their deposits is a lobe in an area of shallowing or flow expansion a
carried out at two scales. At the sediment-water in- short distance downstream. In this way, momentary
terface, variations in turbulence patterns, sediment changes in the bed are propagated downstream and
transport rate, flow speed, and depth control bed the river is maintained in a continually changing
textures and structures, and are reflected in the re- state of dynamic disequilibrium. Preservation of any
sulting stratification patterns and lithofacies charac- given depositional unit is due to chance abandon-
teristics. These processes, in turn, are part of the ment and burial. It is rare for complete macroforms
larger-scale evolutionary development of channels to be preserved in a gravel deposit.
and bars that governs the development of architec-
Gravel Facies 101
In this section, we consider the develoment of This is an orientation of each clast such that the
textures and structures and the resulting lithofacies plane formed by the long and intermediate axis of
variations. Gravel macro forms are described and the clast dips upstream, and is the most stable posi-
interpreted in Chap. 6. tion for a clast to take up. It is a characteristic feature
Variations in the texture and structure of gravel of gravel-bed rivers (Fig. 5.3) and is commonly pre-
deposited from traction currents reflect the extreme served in clast-supported conglomerates, where it
unsteadiness of flow and transport rates in natural provides a useful paleocurrent indicator. Suspen-
gravel-bed rivers. Gravel tends to be transported in sion of clasts during floods may generate a fabric
repeated pulses as loosely defined 'slugs' or 'lobes'. with the long axis parallel to flow.
This reflects turbulence patterns in the river, and On a bed-by-bed scale the most important varia-
also reflects the way in which gravel is released spas- tion in texture and structure is that between hori-
modically into the river from bedform migration zontally bedded gravel showing crude stratification,
and cutbank erosion. Under conditions of high bed- and cross-bedded gravel. Hein and Walker (1977)
shear stress only the coarsest clasts are deposited, provided the most convincing explanation for the
leading to open-work, clast-supported gravels. Finer relationship between these two major types of gravel
clasts, and sand, infiltrate the clast framework at lithofacies (Fig. 5.4). A slug of gravel, released from
lower flow velocities. Changes in flow velocity and an eroded bank or bar upstream, is transported as a
shear stress can occur over very short periods, re- lag in an active channel. It comes to rest at a point of
flecting changes in river discharge and responses to shallowing or flow expansion, where it constitutes a
channel migration and avulsion patterns immedi- channel-floor lag a few clasts thick, in the form of a
ately upstream. As a result, the texture of the de- diffuse gravel sheet. The sheet accretes into a
posits may show rapid vertical changes (Fig. 5.2). bedform by vertical and downstream growth. The
Clasts deposited from traction currents com- balance between these two directions of growth
monly have a regular fabric, termed imbrication. seems to determine the final structure. Under condi-
Fig. 5.2. Exposure of alluvial gravel deposit, showing the variation in texture resulting from constant changes in
discharge and sediment-transport patterns. Quaternary glaciofluvial deposits, near Banff, Alberta, Canada
102 Lithofacies
Fig. 5.3. Imbrication in a modern gravel-bed river, Lake Hazen area, Arctic Canada. Flow was towards the right
DIFFUSE GRAVEL
SHEET 1(,\ 1
lUI
''N, Fig. 5.4. Model for the development of

the major types of bedform structure in
gravels, depending on flow symmetry
~ and fluid (QF) and sediment (Qs) dis-
I I \
charges. (Hein and Walker 1977)
tions of rapid gravel transport the sheet lengthens bedform may lead to decreased shear stress and a
downstream faster than it aggrades, and the result is corresponding upward reduction in clast size.
a low-relief sheet elongated downstream, with flat or In some deposits gravel and sand are segregated
very gently inclined stratification (Fig. 5.5). This within the same bedform. Topset beds are composed
structure was termed a longitudinal bar in some of well-sorted sand, commonly with parting linea-
older literature. Rein and Walker (1977) suggested tion, and bottomset beds consist of cross-bedded
that where gravel transport is slower, vertical accre- gravel. Cross-bedding continues from the gravel up
tion takes place. An avalanche face develops, and a into the sand, where it undergoes a reduction in
separation eddy may result, which serves to amplify slope (Fig. 5.7). Form sets, where preserved, have a
the bedform-forming process. A transverse, cross- convex-up form and have been referred to as hump-
bedded bedform is generated (Fig. 5.6). In both back dunes (Allen 1983c). Segregation occurs where
cases, decreasing water depths over the growing the sediment mixture is bimodal, and where the
Gravel Facies 103
Fig. 5.5. Active growth of a longitudinal gravel bedform. Flow is from bottom right to top left. Field of view is
approximately 20 m wide. Banff National Park, Canada. (Photo courtesy of D.G. Smith)
gravel clasts are well-rounded, facilitating sorting by gravel. This dynamic model dispenses with the need
rolling on the bed. Dunes develop, in which the to invoke separate depositional processes for each
topsets are under critical flow conditions, equivalent textural variation.
to the plane-bed condition of flow-regime theory Bluck (1979, 1980) showed that in some cases
(see Sect. 5.3.1). Gravel clasts are readily rolled over meso forms are capped by coarse gravels, which may
this smooth surface, and accumulate in the lee of the interfinger with finer gravel or pebbly sand foresets,
dune, as a forward-accreting apron. The process has resulting in small coarsening-upward sequences.
been termed "gravel overpassing" (Allen 1983c). Forbes (1983) referred to this as surface armoring.
Modeling experiments by Carling and Glaister Such an arrangement may develop in several ways,
(1987) and Carling (1990) have shown that the over- such as the sweeping of gravel sheets across the bar
passing process can also explain cases where gravel tops at high stage, and development of sandy scour
cross-bed sets show a considerable internal variation fills at the toe of the foreset during low-water stages
in texture, as illustrated in Fig. 5.8. Gravel clasts (Massari 1983). Crowley (1983) showed that similar
show an upward decrease in grain-size, as a result of coarsening-upward textures occur in some large
sorting on the avalanche slope. The matrix-filled and sandy bedforms, and are the product of changing
matrix-supported gravel forming the base of the sets water velocity and depth over the bar crest during
results from the simultaneous deposition of gravel active bar growth (element DA). Coarsening-up-
clasts that roll down the foreset slope, and sand ward, therefore, is probably a dynamic component
settling from suspension in the separation eddy. of many large gravel elements.
This sand component is absent on the foreset slope, Gravel sheets building into deeper water or areas
so that the gravel retains an open-work texture. At of flow expansion, or those covered by gradually
the top of the set infiltration occurs from the inter- waning floods may develop lee-side separation ed-
mittent-suspension sand load beneath the separat- dies. This is accompanied by and encourages the
ing flow, resulting in a matrix-filled, fine-grained growth of foresets, leading to the development of
104 Lithofacies
Fig. 5.6. Development of a transverse gravel bedform. has been abandoned by falling water. Height of bedform
Cross-bedding formed where the bedform prograded to- above water is about 1 m. Near Mito, Japan
ward the right into a pool of deeper water. This bedform
Erosional base
8 Crest
I
Overpassing gravel
_Curren!
-
Load of mixed
sand and gravel
Fig. 5.7A,B. Typical form of stratification developed in segregated gravel-sand mixtures. B Form sets have been termed
humpback dunes. (Allen 1983c)
Gravel Facies 105
waves that can develop during sub critical flow con-

Plane-bedded sandy fine ditions. These structures have rarely been identified
granules. in the ancient record. Rust and Gostin (1981) de-
~'.
...... ".----:::--" (4) Matrix-filled fine
-
.;:
scribed Holocene examples.
l
r--~::--,:!...:--:,::~ conglomerate. ~
••••••••••• ..:.,........ (2) Open-work coarse 5.2. 7.3 Sediment Gravity Flows
•••••• ••••• ". conglomerate. .0
I--~~--~T--~~ ~ In flows with a high sediment concentration, grain
••••••• ". '. (3) Matrix -filled coarse ~
1-------------, conglomerate. 0
transport is a result of buoyancy or matrix strength.
At modest levels of sediment concentration the flow
Plane-bedded sandy fine is cohesionless and may be internally turbulent.
....._ _ _ _ _ _ _--' granules. Crude lamination, with imbrication of clasts, may
result. These flows are erosive, and may produce
Fig.5.8. Cross-bedded gravel set showing variation in tex-
tures. Foresets are indicated by dotted lines. Simultaneous
flutes and other forms of basal scour. At higher levels
deposition of gravels with different textures is explained of sediment concentration (> 40%) the flow is said to
by the overpassing process, as discussed in the text. be hyperconcentrated. Shear stress is transmitted
(Carling 1990) through the flow by a dispersive pressure resulting
from intergranular collisions. Coarser grains move
to regions of lower shear, at the edge of the flow,
transverse bedforms (lithofacies Gp), as described resulting in inverse grading. Such flows commonly
by Hein and Walker (1977). Some examples of contain isolated large clasts that have been rafted on
lithofacies Gt also represent migration of transverse the top or edge of the flow. At still higher sediment
bedforms with curved crests, while others represent concentrations the flow has pseudoplastic character-
the fill of minor channels. Where such channels de- istics. The matrix is cohesive, and has a strength
bouch into pools they develop cross-bedded chute adequate to support large boulders. Little internal
bars (lithofacies Gp) (Massari 1983; Ramos and sorting takes place, although larger clasts gradually
Sopena 1983). Large-scale Gp sets (> 2 m thick) drop out, so that vertically and downstream there
suggest deposition in deep, confined channels may be a crude clast grading. The upper part of the
(Kraus 1984; Middleton and Trujillo 1984). S.A. flow undergoes little internal shear, and may be
Smith (1990) recorded sets of cross-bedded gravel transported as a semi-rigid 'plug'. Events of this type
up to 3 m thick with foreset dips of24-35°, indicating are termed debris flows, and are common in some
deposition of bedforms in water at least 3 m deep. In alluvial-fan settings. Their high viscosity leads to
rare cases, lateral-accretion sets can be recognized. lobate flow outlines, with convex-up margins. The
Such deposits are defined as part of the LA element size of clast that can be transported in debris flows
because of their implication for relatively long-term depends on the yield strength of the flow, which is
lateral migration of a channel-bar complex, result- related to the viscosity and thickness of the flow. For
ing in a distinctive architecture (Sect. 6.7). this reason, the maximum clast size in a debris-flow
Southard et al. (1984) reported a gravel transport deposit commonly varies with flow thickness.
process that they observed in shallow channels, Individual flows may evolve through the three
which they described as the formation of chutes and stages of increasing concentration as a result of
lobes. The lobes are distinctive, coarse gravel accu- downstream loss of water, through infiltration into
mulations, but their preservation potential appears the substrate. Downstream reduction in stream
to be low. slope results in a loss of momentum and transport
It has been observed that on the bed of many energy. Eventually, the increase in internal friction
gravel rivers there is a distinctive arrangement of the resulting from a loss of momentum and of internal
large clasts into ribs or stripes perpendicular to flow. fluid lubrication leads to "freezing" of the flow. As
Koster (1978) demonstrated that these structures, this occurs internal shear may develop shear lamina-
conveniently termed "transverse ribs", form at hy- tion and crude clast fabrics. The relationships be-
draulic jumps between supercritical and subcritical tween the various flow types and their lithofacies are
flow. They also occur beneath kinematic standing shown in Fig. 5.9.
106 Lithofacies
c c
v
Fig. 5.9. Left Conceptual classification of debris flows. I plastic debris flow, II clast-rich debris flow, III
Poles represent cohesive-plastic behavior with increasing pseudoplastic debris flow, and IV pseudoplastic debris
clay content (C), viscous-fluid behavior with increasing flow with inertial bed load. Right The four main lithofacies
water content (V), and granular-collisional behavior with classes and their intergradation. Facies codes are dis-
increasing clast content and shear rate (G) . Flow types are: cussed in the text. (Schultz 1984)
5.2.2 Gravel Lithofacies Gmm and Gmg. Lithofacies Gm of Miall (1978c) has
been replaced by Gh to avoid confusion in the use of
The seven major gravel lithofacies that occur in flu- the lowercase letter m.
vial systems are differentiated initially on the basis of
texture, and secondly, on the basis of their internal Lithofacies Gmm and Gmg: Matrix-Supported
structure. It is useful to record the average grain size Gravel. These lithofacies were recognized by Rust
of the clasts, and any vertical changes in clast grain (1978b), based on his work in some Carboniferous
size. Such changes within an individual lithofacies conglomerates. He used the code Gms (for matrix-
unit reflect changes in shear stress over the bedform, supported gravel), which does not permit a distinc-
while larger-scale changes, over several or many tion between massive and graded units. The most
lithofacies units, indicate progressive changes in the distinctive attribute of Gmm and Gmg is the absence
channel system. A close relationship between bed of a clast framework. The clasts are poorly sorted and
thickness and maximum clast size is common in are supported by a poorly sorted matrix of sand, silt,
sediment-gravity-flow deposits (Bluck 1967). A con- and mud. The beds maybe massive (Gmm: Fig. 5.10),
venient field method for rapidly recording maxi- or may show grading of clasts and/or matrix (Gmg:
mum clast size is to measure the ten largest clasts Fig. 5.11). Imbrication is normally absent, but tabu-
present within a small area of exposure, and to take lar clasts may assume approximate horizontal orien-
the mean of these readings. tations. Beds of this lithofacies have sharp but
The lithofacies codes have been modified from nonerosional relationships with underlying beds.
those presented by Miall (1978c) in order to incorpo- They commonly have sharp lateral terminations.
rate additional observations on debris flows in an These characteristics reflect the formation of the
ancient fan deposit by Schultz (1984), the classifica- lithofacies by the process of high-strength debris
tion of which, in turn, was based on work on sub- flows. Flows passively occupy preexisting alluvial
aqueous debris flows by Eyles et al. (1983). A code topography - they will, for example, occupy chan-
scheme for poorly sorted, subaqueous, glacigenic nels and assume a channelized form. Flows are lo-
diamicts was proposed by Eyles et al. (1983), using bate in plan view and, because they have internal
the capital letter D for diamict, followed by the low- strength, they develop lobate, convex-up margins.
ercase letters m or c for matrix-supported or clast- This form is preserved when the flows stop forward
supported, respectively. A second lowercase letter, movement as a result of the development of internal
m, i, or g, was added for massive, inverse, or nor- friction due to water loss.
mally graded textures. Schultz (1984) showed that
the same scheme could be applied to subaerial debris Lithofacies Gci: Clast-Supported, Inverse-Graded
flows in alluvial fans. This scheme has been adapted Gravel. This lithofacies can occur in two ways, as a
for use here, retaining the capital letter G to indicate clast-rich, high-strength debris flow, or as a low-
the dominant grain size. Lithofacies Gms, as origi- strength flow with an inertial bed load transported
nallyproposed by Rust (1978b), has been replaced by by laminar to turbulent flow.
Fig. 5.10. Debris-flow deposit; lithofacies Gmm. Note the poor sorting and matrix-supported framework. Carbonifer-
ous, Poland
Fig. 5.11. Debris-flow deposit; lithofacies Gmg. Clasts are logs and other plant debris, which show a crude upward size
decrease and which also become somewhat more dispersed in the matrix. Tertiary, northern Kyushu, Japan
108 Lithofacies
Fig. 5.12. Clast-supported gravel with

crude horizontal stratification (litho-
facies Gh), and minor interbedded
sand lenses. Note upward decrease in
clast size. Devonian, Prince of Wales
Island, Arctic Canada
Lithofacies Gcm: Clast-Supported, Massive Gravel. Lithofacies Gt: Trough-Crass-Bedded Gravel. This
This lithofacies represents low-strength, pseudo- lithofacies is distinguished by the presence of broad,
plastic debris flows, deposited from viscous, laminar shallow scoop-shaped bodies, typically 0.2-3 m deep
or turbulent flows. and 1-12 m wide. These units cut into each other
both laterally and vertically. The erosional base may
Lithofacies Gh: Clast-Supported, Horizontally Strati- be followed by a lag deposit of coarser grain size than
fied Gravel. This lithofacies (Fig. 5.12; first defined the trough fill. Foreset dip is steepest in the smaller
as Gm in Miall 1978c) consists of clast-supported troughs, reaching a maximum of about 30°.
pebble and cobble gravel with crude horizontal
stratification. Most deposits have a clast framework Lithofacies Gp: Planar-Crass-Bedded Gravel. Planar
and an abundant sand matrix. Individual beds are cross-bed sets in gravel reach 4 m in thickness, but
typically a few decimeters thick, with multistory are typically less than 1 m. Textural variation within
units reaching several meters in thickness. Bed con- these sets may be considerable, because of variations
tacts may be obscure because of the absence of well- in sorting resulting from changing hydraulic condi-
defined bedding. Clasts are commonly imbricated. tions and gravel-clast overpassing (Fig. 5.13).
Sand Facies 109
Fig. 5.13. Example of cross-bedded gravel deposit (lithofacies Gp), with variable textures, controlled by the overpassing
process. Pleistocene, near Mito, Japan
5.3 Sand Facies some years it has been known that the main types of
bedform are stable in flowing systems only under
5.3.1 Sand Bedform Genesis and Classification limited ranges of these three main parameters, and
that these stability ranges, for the most part, do not
Sand lithofacies in fluvial systems result from the overlap significantly. Figure 5.14 illustrates the sta-
transport of sand by traction currents, as bed load bility fields for the principal bedform types. Each of
and in intermittent suspension ("saltation"). The these classes gives rise to a characteristic lithofacies,
principles of fluid turbulence that control bedform as described in the next section.
generation are discussed briefly in Sect. 2.4.1. Em- One of the problems with the application of this
pirical observation of bedforms in flumes and in a type of classification to natural systems is that bed-
variety of modern environments, including rivers forms occurring in the dune field have a wide range
and estuaries, have led over the years to a prolifera- of morphologies and sizes (height, wavelength), re-
tion of morphological terms for the various types sulting in a corresponding variability in sedimentary
that have served to confuse rather than elucidate structures. There has been considerable debate re-
their origins. However, a recent synthesis and review garding the genetic significance of these variations,
of this work have cut through much of the confusion including whether they indicate a corresponding
and resulted in some useful simplifications (Ashley range of distinct depositional processes.
1990). The major and most obvious subdivision of
The morphology of sand bedforms depends pri- dunes is into those of two-dimensional (2-D) type
marily on three parameters, sand grain size, flow and those of three-dimensional (3-D) type. Two-
depth, and flow velocity. Other parameters, such as dimensonal forms tend to occur at lower flow speeds
fluid viscosity and temperature, exert a minor con- for a given grain size. They have simple, prismatic
trol over bedform morphology, but can be safely cross sections and give rise to planar-tabular cross-
ignored for the study of most natural systems. For bedding, bounded by simple, planar first-order
110 Lithofacies
't-
~Q) 2.0 +
... + +
+ ++
::l i------.tr:-.+- Fr =1.0 ---+t--+---'-----::=-O~--------_rl
- + + + + ~~~--
~ ~ ;~- Fr =:.84 -~;:---~~;:---::::::'-:=-''"''---------------,rl
E ~ ~ 6u~\·
~ ! 1.0 UPPER PLANE P. ~~ ~l~.':.
E Q; ~ ~ • I
..
abrlJpt '" • •••
:>: iU 0.8 • ,, " , i ..
::::
u 3: • ••• DUNES
\ ~
~
~ ~
P
0.6
o
•• I
....• •
00';>6 ~ • ~~
o '
3: 0 o o "'<:0... ••: :-tl~~
o -
;: '0 0.4 8 0 o 0 ,,0
o 6> 0 x
00 NO MOVEMENT
c: ~ 0
:ll:c 0 0 RIPPLES x
:a '0:a 0
0 0 o
-
0 o
c: 0 0
111
~ 0.2 Mean flow depth

0.25-0.40 m
Median sediment size, mm

(standardized to 10°C water temperature)
Fig. 5.14. Plot of mean flow velocity against median sedi- cated by the names in the center of each field. Note the
ment size, showing stability fields of bed phases in sand. overlap between some of the fields. (Ashley 1990)
Different symbols indicate the various bed phases, as indi-
surfaces (Fig. 5.15). Three-dimensional dunes are layer. A review of height-spacing data for flow-trans-
characterized by curved lee faces and by deep, circu- verse bedforms over a range of spacing from 0.01 to
lar or oval scour pits at the foot of the lee slopes. The > 1000 m (Fig. 5.18) reveals a continuum, except for
typical depositional product is trough cross-bed- a single discontinuity at a spacing of 0.5 to 1 m. This
ding. In most deposits the troughs are stacked into corresponds to the break that has commonly been
overlapping, mutually erosional sets bounded at the observed between ripples and the larger bedforms,
base by curved first-order surfaces (Fig. 5.16). An the cause of which was discussed by Leeder (1983; see
alternative term for dune is megaripple; 2-D forms Fig. 2.22). The lack of any natural grouping in the
have been referred to as transverse bars and sand larger forms is significant, and suggests that they
waves; 3-D forms include linguoid and lobate bars. comprise a single genetic population. Ashley (1990)
The use of the term "bar" for these features is now reported a consensus view (of a discussion panel
regarded as incorrect. It is retained for macro forms, which examined this question) that the distinction
which are long-term products of river systems (pri- between the various large-scale 3-D forms relates
marily the "group-6" deposits of Chap. 3). simply to the space, time, and volume of sediment
Ashley (1990) argued that the apparent distinc- available for their formation. The morphology re-
tions between the various morphological forms en- sponds to fluctuating water levels and velocity, and
compassed by these different terms are largely to the different turbulence patterns of unsteady and
artifacts of our limited data. A review of available reversing flow. Dunes of different scales that are
data from fluvial and tidal environments indicates dynamically superimposed on each other indicate
that 3-D forms always occur at higher flow speeds the development of nested boundary layers wher-
than 2-D forms where other parameters, such as ever space is sufficient. Large bedforms develop a
grain size and depth, are equal (Fig. 5.17). This dif- boundary layer in which the smaller forms are stable.
ferentiation relates to the growth of three-dimen- The smaller forms may remain active even where the
sional separation vortices in the lee of the dunes as large forms upon which they rest are inactive, but
shear stress increases within the turbulent outer this simply reflects the limited availability of sedi-
Sand Facies III
Fig. 5.15. Two-dimensional dunes. Above Block diagram showing the typical surface form, and the resulting stratifica-
tion (Harms et al. 1982); Below Modem example, Platte River, Nebraska. (Miall and Smith 1989)
ment, or inadequate time, for the large forms to As noted in Sect. 2.4.1, experiments on the dune-
migrate. In other words, no genetic distinction needs upper flat bed transition by Saunderson and Lockett
to be made between the various sizes of 3-D dunes (1983) documented the existence of modified dune
(and the resulting cross-bedding) in a given fluvial forms which preserve distinctive styles of stratifica-
system. However, it may well be useful to make a tion. Most characteristic are the humpback dunes
descriptive subdivision of these forms, in order to that form by the sweeping of sand over a dune crest
clarify the structure of the deposits and the mecha- without the generation of a separation eddy (Fig.
nisms of channel and bar construction. Ashley 5.19). The dune crest is not coincident with the top of
(Table 6, 1990) suggested the terminology for such the foreset, but is located behind it. The continued
descriptions. development of humpback forms leads to low-angle
112 Lithofacies
Fig.5.16. Three-dimensional dunes. Above Block diagram showing the typical surface form, and the resulting stratifica-
tion (Harms et aI. 1982); Below Modern example. (Miall and Smith 1989)
« 15°) sigmoidal cross-bedding. The crests of the 5.3.2 Sand Lithofacies

humpbacks may be capped by plane bedding. The
combination of flat bedding and low-angle cross- Sand lithofacies are classified on the basis of the
bedding is characteristic of the stratification that is dominant primary sedimentary structure. Detailed
preserved from this transitional phase. measurements of grain size are not required. Deter-
Antidunes form under conditions of supercritical mination of modal grain size in the field, preferably
flow, but are rarely preserved in nature as they tend with the aid of a hand-held grain-size comparator, is
to be readily modified by the changing turbulence sufficient for the purposes of facies analysis. Qualifi-
patterns that develop during decreasing flow speeds. cations should be observed wherever appropriate as
Langford and Bracken (1987) described examples to the degree of sorting, and the presence of acces-
from modern stream deposits. sory components such as intra- or extraclasts of
pebble or larger size.
Sand Facies 113
20.0 Fig. 5.17. Depth-velocity diagram, showing the

1
stability ranges of 2-D and 3-D dunes in medium
to coarse sand. (Compiled by Ashley 1990)
~
10.0 r-
• •• •
-
r.9
'0 0
° •• °
'·0
° .!e•
5.0
° _
Oaf):;
° •
• :rf.
° .. ~.
° °
•
E 6 6
6
.. ......
6 " ..
I 66
l-
e..
w 1.0 ' - -
o 6
3: 6
....
o
....J
lJ... 0.50
00 00 CO
6 ..
OCO~
~ ~~
;---
20 3D
0 • Dalrymple el 01.11978)
0.10 r- 6. ... Boolhroyd B Hubbard (1974)
1-
o ~. <:7 ... Coslello B Soul hard

0 • Prall (1971)
0.05 I
0.1 0.2 0.4 0.6 0.8 1.0 2.0 3.0
FLOW VELOCITY (m/sec)
It is suggested that lithofacies Se, which was pro- (15-35°), with sharp, angular, upper and lower ter-
posed by Rust (1 978b) and incorporated in the minations, indicating avalanching of sand on
scheme of Miall (l978c), can be abandoned as a foresets (Fig. 4.5). Upper and lower bounding sur-
separate category, as its recognition is now consi- faces are typically flat, with little evidence of scour-
dered superfluous (see discussion below). No other ing. Sand is typically sorted by the process of ripple
changes from the original suite of seven sand litho- migration up the stoss side of the dune, resulting in
facies are proposed, although some additional re- fore sets in which the modal sand grain size may
finements in definitions are now possible, and one differ by several size classes.
code has been modified. The basic form of the cross-bedding is modified
under different flow conditions (Fig. 5.20). At high
Lithofacies Sp: Planar-Crass-Bedded Sand. This flow speeds, approaching the transition to the plane-
lithofacies forms by the migration of 2-D dunes (Fig. bed condition, separation eddies become smaller,
5.15). Sand is transported up the flank of the and the foreset flattens out. Curved toesets may de-
bedform by traction and intermittent suspension velop at this time. Minor sand transport by the
(commonly forming a carpet of small-scale current backflow beneath the separation eddy may deposit
ripples), and deposited at the crest, where bed-shear small current ripples at the toe of the avalanche
stress drops at the point of flow separation. Cross- slope, with foresets oriented in the opposite direc-
bedding is typically at or near the angle of repose tion. Falling water may lead to the abandonment and
114 Lithofacies
log L Fig. 5.18. Plot of height versus spacing of 1491

-2 -1 0 1 2 3 flow-transverse subaqueous bedforms. Note the
100 2 discontinuity at 0.5 to 1 m spacing. (Ashley 1990;
H= 0.0677 Lo.809S after Flemming 1988)
log H
n = 1491
10 r = 0.98
E 0
....
J:
~
w 0.1 -1
J:
0.01 -2
0.001 .. -3
0.01 0.1 10 100 1000
SPACING (m)
subaerial erosion of the dune, producing a curved Lithofacies St: Trough-Cross-Bedded Sand. Troughs
upper dune surface. If this is covered again by rising develop by the migration of 3-D dunes (Fig. 5.16).
water and renewed sand transport, the dune may be They occur in fine- to very coarse-grained sand.
reactivated, so that the erosion surface is preserved Pebbles may be present, and there is commonly a lag
as a cross-cutting surface within the dune, termed a of poorly sorted sand with intraclasts of siltstone or
reactivation surface (Collinson 1970). mudstone at the base of the trough. Cross-stratifica-
Planar cross-bed sets form in sand of very fine to tion consists of curved sets, with an angle of dip that
very coarse grain size. The lower limit of set thick- rarely reaches the angle of repose. The cross-stratifi-
ness is 5 cm. This corresponds to the natural size cation normally curves out at the base of the trough,
break in the height of transverse bedforms deter- which invariably shows an erosional relationship to
mined by the size of separation eddies in the turbu- underlying stratification. The curved foresets and
lent outer layer (Sect. 2.4.1). Set thicknesses of 0.5 to the asymptotic downlap at the base are the features
1.5 m are typical of most fluvial sandstones. Sets that distinguish trough cross-bedding from planar
several meters thick are abundant in the Hawkes- cross-bedding in small outcrops and in core.
bury Sandstone, near Sydney, Australia (Conaghan Troughs occur as solitary or grouped forms. Soli-
and Jones 1975; Rust and Jones 1987), reaching a tary sets are commonly observable on bedding
maximum thickness of 7.5 m. Coleman (1969) re- planes, where the distinctive basal scour surface and
corded dunes 1.5 to 8 m in height in the modern the curved trace of the infilling cross-bedding may
Brahmaputra River of Bangladesh, and what he be commonly traced for distances down paleoflow of
termed sand waves that had a height range of 8 to 15 up to 6 m (Fig. 5.22). Grouped sets (termed "festoon
m. However, some of the sand waves may be cross-bedding" in older literature) may make up
macro forms rather than mesoforms. An example of thicknesses of several meters of sand (Fig. 5.23).
a solitary large-scale set is illustrated in Fig. 5.21. The lower limit of set thickness is 5 cm, as in the
Cosets comprising thicknesses of several meters of case of 2-D dunes. In most fluvial sandstones, trough
sand are common. Individual sets may be traced for sets are rarely larger than 1 m in thickness, but larger
tens of meters parallel to bedding, although com- forms have been observed, indicating correspond-
plete form sets are rarely preserved because of the ingly greater water depths and dune heights, as in the
continual channel shifting and scour occurring in case of the 2-D dunes discussed above.
most fluvial systems. In rare cases, dunes may be abandoned by rapidly
falling water, and the dune scours subsequently
Sand Facies 115
ASYMMETRICAL DUNE IV HUMPBACK DUNE

STRONG SEPARATION
~CON_~X BEDS
.. ... .
. .~ . .
30·
. . . .
INITIAL
. .. ............. ~~ 30"
..
ACCUMULATION
II ASYMMETRICAL DUNE V DUNE TRANSITION
y BURSTING WEAK SEPARATION
/~~ · \~*?o>/--~
30"
III DUNE TRANSITION VI ASYMMETRICAL DUNES
Fig. 5.19. Schematic diagram showing the transition from tachment point (III) leads to the development of a hump-
asymmetrical dune to humpback dune and back to asym- back dune (IV). These have small foreset slopes reflecting
metrical dune. A mound of sand forms a symmetrical, the small scale of the separation eddies. Draping of the
convex feature (I, point W) with convex bedding, which humpback form leads to low-angle, sigmoidal cross-bed-
then develops as an asymmetrical dune (II). Bursting takes ding (V). Continued vortex development (Z) may lead to
place near the reattachment point (X). Erosion of the dissection of the humpback form and regeneration of
topset and progradation of low-angle sets from the reat- asymmetrical forms (VI). (Saunderson and Locket 1983)
filled with other bedforms, including lithofacies Sl and in other settings where the river carries a large
and Sr. This is an example of bedform "lag", where load of very fine sand and silt.
the form of the structure indicates that it was not in Ripples develop at low flow speeds « 1 m/s), and
equilbrium with flow conditions throughout its gen- are very sensitive to changes in flow conditions.
eration (Allen 1984). Ripple trains with double crests may develop where
an older ripple set is superimposed by a younger set.
Lithofacies Sr: Ripple Cross-Laminated Sand. A vari- Ladderback ripples occur where flow directions
ety of asymmetric ripple types characterize this change, as in a partially abandoned pool subject to
lithofacies. Sand grain size ranges from coarse to fluctuating vortices entering from a main channel,
very fine, but fine- to medium-grained sand is most or where temporary currents in very shallow water
typical. A wide variety of internal structures may be are driven by wind. Examples of these variations in
generated from ripple migration, depending on flow ripple morphology are illustrated in Fig. 5.26.
velocity and the rate of sediment supply (Jopling and Ripples are, by definition, less than 5 cm in height,
Walker 1968; Allen 1984). Migration of trains of many are less than 2 cm. Solitary ripple trains are
ripples with a low rate of sedimentation from sus- common, and may be well exposed on bedding-
pension leads to ripples that are mutually erosive plane surfaces. Cosets form thicknesses of decime-
(Fig. 5.24), termed "type-A" ripples by Jopling and ters to a few meters.
Walker (l968). Where sediment is added from sus-
pension during ripple migration, mutual erosion of Lithofacies Sh: Horizontally Bedded Sand. This
the ripples is incomplete, leading topartial preserva- lithofacies occurs under two quite different condi-
tion of the stoss sides of the ripple, and" climbing" of tions. The most important is that which represents
the ripple train ("type-B" ripples; Fig. 5.25). The the upper plane bed condition, at the transition from
latter condition is common in fluvioglacial outwash, sub critical to supercritical flow (Fig. 5.14). This
116 Lithofacies
High water stage
Intermediate water stage
Low water stage

(a)
Risen water stage

(a)
Fig. 5.20. Changes in morphology and internal structure commonly with different indicated flow directions. Ero-
associated with changes in stage over a 2-D dune. At high sion by minor waves and surface runoff as the dune
stage a strong separation eddy gives rise to asyinptotic emerges leads to a rounding of the dune crest, and this is
foresets and countercurrent ripples. The separation eddy then preserved as an internal reactivation surface when
is reduced in strength during falling water, resulting in stage rises and the dune commences advancing again.
angular foreset terminations. At low stage the dune may (Collinson 1970)
become inactive, with minor structures draped over it,
Fig. 5.21. Solitary set of large-scale planar cross-bedding, approximately 4 m thick. Despite the size of this unit, it is a
mesoform rather than a macroform, and is assigned to lithofacies Sp. Devonian, Scotland
Fig. 5.22. Solitary set of trough cross-bedding (lithofacies St) exposed on a bedding-plane surface. Proterozoic
Matinenda Formation, near Elliott Lake, Ontario, Canada
Fig. 5.23. Coset of trough cross-bedding (lithofacies St) viewed in cross section. Tertiary Eureka Sound Group, Axel
Heiberg Island, Arctic Canada
Fig. 5.24. Type-A ripple cross-lamination. Devonian Old Red Sandstone, Scotland
Fig. 5.25. Type-B ripple cross-lamination, Quaternary fluvioglacial deposits, near Ottawa, Canada
Sand Facies 119
Fig. 5.26. Examples of varying ripple morphology, reflect- B Double-crested ripples, showing flat-topped crests re-
ing small changes in flow conditions. A Ladderback sulting from erosion. Triassic, Spain
ripples, Beaufort Group, near Beaufort West, South Africa.
120 Lithofacies
Fig. 5.27. Lithofacies Sh, with parting lineation. Note crescent scours around pebbles. The wings point downstream (to
the left). Ordovician, near Toronto, Canada
phase is most stable in very fine- to medium -grained Plane beds also develop in coarse to very coarse
sandstone at velocities of around 1 mls and water sand at low flow speeds (-0.4 m/s), during initial
depths of 0.25 to 0.5 m, but also occurs at lower sand bed movement, but this condition is rarely
velocities at shallower depths. Coarser grains, in- preserved in nature.
cluding pebbles, are rare, and are emplaced by being
rolled along in the sand traction carpet. Units of Sh Lithofacies Sl: Low-Angle Cross-Bedded Sand. This
up to several meters thick may be deposited during lithofacies was proposed by Rust (1978b), following
single dynamic events, such as flash floods, when descriptions by Miall (1977) and Cant and Walker
flow conditions may remain in the critical stage for (1976, their facies G). Grain size and bedding charac-
periods of many hours. teristics of this lithofacies are similar to that of
Lithofacies Sh is distinguished by flat, parallel lithofacies Sh, with which it is commonly associated.
lamination, with parting lineation (otherwise The main distinguishing characteristic is the pre-
termed streaming lineation or primary-current lin- sence of low-angle cross-bedding, dipping at < 15°,
eation) occurring on bedding planes (Fig. 5.27). This and normally < 10° (Fig. 5.28). The cross-bedding is
structure is generated by small longitudinal vortices commonly asymptotic with reference to upper and
at the base of the inner turbulent layer, and provides lower set bounding surfaces. Parting lineation may
an excellent paleocurrent indicator. The orientation be present, with an orientation parallel or oblique to
of the lineation is parallel to flow, but does not the dip of the cross-bedding.
indicate which of the two opposing directions is the Some occurrences of lithofacies SI represent
correct one. This information can be obtained by plane beds deposited on initially dipping surfaces,
reference to interbedded cross-bedding structures, such as scour hollows. In other cases, the lithofacies
or by examining contained pebbles, around which represents distinct bedform geometries. The
small erosional shadows develop in response to the washed-out dunes and humpback dunes that occur
locally enhanced shear stress. The wings of the cres- at the transition between subcritical and supercriti-
cent scours point downstream (Fig. 5.27). cal flow typically form this lithofacies (Fig. 5.19).
Sand Facies 121
Fig. S.28. Succession of sandstone beds deposited under bed set of lithofacies SI downlaps onto this unit. Jurassic
near-critical flow conditions. Beds behind notebook are Sh Morrison Formation, near Gallup, New Mexico
showing gentle structural dip to the Left. A low-angle cross-
Antidunes also give rise to a similar lithofacies, but Lithofacies Se: Erosional Scours. This lithofacies was
are rarely preserved in the ancient record. proposed by Rust (1978b), who compared the facies
The distinction between plane beds on dipping characteristics to lithofacies SS of Cant and Walker
surfaces and bedforms with low-angle internal (1976). It comprises an erosional scour surface fol-
stratification is best made with reference to the over- lowed by up to 0.25 m of massive, coarse-grained
all associations of the lithofacies. Interbedding of sandstone containing intraclasts of mudstone and
lithofacies Sh and Sl, with low-angle cross-bedding siltstone.
downlapping onto Sh bedding surfaces, indicates The original definition of this lithofacies was
deposition close to the upper plane-bed condition, made by Cant and Walker (1976) in order to permit
with the generation of a variety of dune-like struc- them to incorporate bounding surface information
tures. Where possible, the two types of Sl may be into a Markov chain analysis of their sandstone suc-
distinguished by additional lithofacies coding. For cession. However, this writer now considers that it is
example, Sli may be used for inclined plane bedding, a superfluous category. Bounding surfaces may now
and SIb for low-angle cross-bedding representing be analyzed and described using a separate termi-
bedform generation. Cowan (1991) used the codes Si nology, as described in Sect. 4.5. This must be carried
and Sl for these contrasting cases. out before Markov chain analysis is attempted and,
Cotter and Graham (1991) pointed out some simi- as shown in Chap. 8, such analytical methods can
1arities between beds of Sl, which they observed in an now be seen to have a severely limited usage in
ephemeral-fluvial sandstone, and hummocky cross- fluvial deposits. The beds overlying the scour surface
stratification, but concluded that the beds were may be classified into any of the other lithofacies
formed under unidirectional, transitional to upper classes. Lithofacies Ss, as originally proposed by
flow regime current conditions. Miall (1977), is suitable in many cases.
.....
N
N
Fig. 5.29. Lithofacies Ss forming overlapping scour-fill sets. Basal scour surfaces are indicated by arrows. Jurassic Morrison Formation, near Gallup, New Mexico
t""
s:
o
S'
n
(D"
en
Fine-Grained Clastic Facies 123
Lithofacies Ss: Scour-Fill Sand. Above fluvial erosion 5.4 Fine-Grained Clastic Facies
surfaces there commonly are a few decimeters to a
meter of coarse- to very coarse-grained, poorly
sorted sand, which may show poorly defined cross- Fine-grained clastic sediments are deposited pri-
bedding and contain abundant intraclasts and other marily from the suspension load of rivers. This can-
lag materials (e.g., waterlogged plant debris) (Fig. not take place within active channels, because the
5.29). The bedding may drape over the irregular shear stress and turbulence are such that the load
shape of the basal bounding surface. Sole markings, remains in suspension. Deposits of mud, silt, and
including flute marks and tool marks, are common very fine sand therefore indicate deposition in flood-
on these basal bounding surfaces. plain areas, in abandoned channels, and in aban-
The characteristics oflithofacies Ss indicate rapid doned areas of normally active channels, for
deposition of poorly sorted, coarse bed load. While example during seasonal low-water stages. The
the process is clear from the facies characteristics, thickness of depositional units ranges from the few
the details of the environmental interpretation of the millimeters of mud drape formed in small aban-
lithofacies depend on its facies associations. For ex- doned channels, to thicknesses of tens to even hun-
ample, what is the type of bounding surface on which dreds of meters in the floodplains of major,
it rests? Questions of this type relate to the architec- suspension-load streams.
tural-element level of the analysis. For example, The original suite of two fine-grained litho-
lithofacies Ss may comprise the basal fill of elements facies erected by Miall (1977) was modified by
CH (channels) or HO (scour hollows). The same type Rust (1978b), and McLean and Jerzykiewicz (1978)
of deposit may also occur at the base of macroforms, proposed two additional categories, one of which
or draping the third-order surfaces that define their is now considered to be superfluous, as discussed
growth increments. A description of the geometry of below.
the basal bounding surface should not, therefore, Lithofacies FI, Fsm, and Fm represent a gradation
comprise part of the lithofacies definition, as origi- in grain size and bedding characteristics from rela-
nally proposal by Miall (1977), but should be part of tively coarse, proximal floodplain deposits to more
the architectural description of elements and distal deposits.
bounding surfaces.
Distinction between Ss and large, solitary St sets Lithofacies Fi: Laminated Sand, Silt and Mud. Inter-
migrating along a channel floor may be difficult and lamination of mud, silt, and very fine-grained sand is
even, perhaps, arbitrary. However, possible confu- common in overbank areas, and represents deposi-
sion about a lithofacies classification is less impor- tion from suspension and from weak traction cur-
tant than what the observations tell us about rents (Fig. 5.30). Very small-scale ripples may be
sedimentary processes during deposition of a unit of present in the sand and silt beds. Undulating bed-
interest. ding, scattered bioturbation (including animal foot-
prints), desiccation cracks, plant roots, coal streaks,
Lithofacies Sm: Massive Sand. Sandstone beds in and scattered pedogenic nodules may be present.
outcrop may appear massive if weathering does not Typical thicknesses of continuous FI deposits range
pick out the lamination. However, true massive from a few centimeters to many meters, depending
sandstones do exist. They may show grading, or on sediment supply, fluvial style, and basin subsid-
reveal very faint, patchy lamination. Such beds are ence rates. Individual beds may be as thin as a few
the deposits of sediment gravity flows. A characteris- millimeters. Careful observation may indicate that
tic occurrence of this facies is in small channels units are organized into thin fining- or coarsening-
resulting from bank collapse (Sect. 6.2). Massive upward successions.
texture may also be produced by postdepositional Discrete sand beds thicker than a few centimeters
modification, for example by dewatering and bio- may be separately logged as lithofacies Sr, St etc., and
turbation. Faint residual sedimentary structures may represent more energetic floodplain traction
may reveal such origins. currents, such as those that deposit crevasse splays.
124 Lithofacies
B
Fine-Grained Clastic Facies 125
c
Fig. 5.30. Lithofacies Fl. A Channel sandstone overlying Formation, Triassic, near Winslow, Arizona. C Interlami-
dark floodplain siltstones containir).g thin interbedded nated shales and thin sandstones. The cast of a dinosaur
silty sandstone beds. Devonian Old Red Sandstone, Scot- footprint filled with sandstone extends downward from an
land. B Thinly laminated sandstone-shale succession with overhanging sandstone bed at center of picture. Triassic,
flow rolls resulting from syndepositional loading. Chinle Spain
Similarly, beds dominated by or consisting of more the absence of thin sand beds in a heterogeneous,
than a few centimeters of continuous coal, root beds, fine-grained succession may be difficult to observe,
or pedogenic nodules should also be logged as sepa- and the distinction between FI and Fsm is then diffi-
rate lithofacies. The reader is referred to comments cult or arbitrary. I have not used this lithofacies class
in Sect. 4.3 regarding the need to establish limits in any of my field programs.
between lithofacies classes, where a possible overlap
exists, for example between lithofacies FI containing Lithofacies Pm: Massive Mud, Silt. Miall (l977) pro-
coal streaks, and separate beds of coal. posed this lithofacies for the mud drapes that com-
monly occur within gravelly and sandy braided
Lithofacies Psm: Siltstone, Claystone. A lithofacies sediments, where they represent the deposits from
coded Fsc was proposed by McLean and standing pools of water during low-stage channel
Jerzykiewicz (1978), and appears as such in Miall abandonment. Thicknesses of a few millimeters to a
(l978c). The lowercase letters stand for silt and clay, few centimeters are typical. Carbonaceous streaks,
respectively. For consistency, this is relabeled here as plant rootlets, and desiccation cracks are common.
Fsm, with the m standing for mud. The main feature The lower bounding surface of the bed may conform
of this lithofacies, which distinguishes it from FI, is to the shape of any underlying bedforms, such as a
the absence of sand beds. In other features, such as train of ripples. Rust (l978b) suggested expanding
bedding and presence of accessories, this lithofacies this lithofacies to include any bed of massive mud-
compares with Fl. It probably represents a floodplain stone. As such, it may represent the most distal
deposit somewhat more distal relative to clastic floodplain facies, including deposition in floodplain
sources such as nearby fluvial channels. In practice, ponds.
126 Lithpfacies
Fig.5.31. Lithofacies Fr. A Seat earth beneath lignitic coal bed. Tertiary Eureka Sound Group, Banks Isand, Arctic
Canada. B Rootlets in overbank siltstones, Triassic Buntsandstein, Spain
Nonelastic Facies 127
Lithofacies Fcf: Claystone with Freshwater Molluscs. adopted by sedimentologists for paleosols. Calcitic
This lithofacies class was proposed by McLean and paleosols are most typical of arid to semiarid, oxidiz-
Jerzykiewicz (1978). However, it is regarded here as ing climates, whereas siderite nodules occur in wa-
superfluous. Lithologically, the beds are identical to terlogged, reducing settings, and are commo'nly
Fm, and the molluscs constitute a lithologic acces- associated with coals.
sory, which is insufficent to require definition of a The stratigraphy and facies of paleosols consti-
different lithofacies class. tutes a specialized study that is touched on further in
Sect. 7.4.2, where additional petrographic and facies
Lithofacies Pr: Root Bed. These are common in veg- classifications are provided.
etated floodplains (Fig. 5.31). This lithofacies repre- Paleosols constitute useful local marker beds,
sents soil development in a humid climate. and can therefore yield valuable information on
Carbonate nodules are typically absent, but silica basin stratigraphy (e.g., Allen and Williams 1982;
cementation may have occurred as a result ofleach- Behrensmeyer 1987; Eberth and Miall 1991).
ing. The host lithology may be sand,· silt, or mud, Paleosol maturity can be deduced from color, soil-
with any primary stratification, such as cross-bed- layer differentiation, and nodule morphology, pro-
ding, ob'scured or destroyed by root emplacement viding data on floodplain exposure times and clastic
and bioturbation, giving the beds a mottled appear- influxes (Allen 1974b; Leeder 1975; Bown and Kraus
ance. The roots may retain their original carbon, or 1987). These deposits are therefore of considerable
may be replaced by calcite or siderite. value in the study of large-scale basin architecture, a
Where associated with overlying coals (Fig. subject that is considered at length in Chaps. 9 and
5.31A), these beds have been termed seat earths or 10.
underclays. As an alternative to the clastic coding Soil texture and geochemistry are sophisticated
discussed here, seat earths may be discussed and subjects, yielding much information on local bio-
classified under the heading of paleosols, as de- chemical processes, local climate, etc. (Johnson
scribed in Chap. 7. 1977; Bown and Kraus 1981, 1987; Retallack 1981; see
also books referenced in Sect. 7.4.2). However, much
of this is beyond the purpose and scope of this book,
and is not discussed in detail.
5.5 Nonelastic Facies
Lithofacies C: Coal. Coal is typically associated with
A limited range of chemical sediments occurs in deltaic and fluvial floodplain environments (Fig.
fluvial deposits, mainly but not exclusively in the 5.33). McCabe (1984) has suggested that the presence
floodplain. The most important of these are the soils, of thick coals, as opposed to carbonaceous mud-
which are best developed in semiarid climates. In stones, indicates the presence of raised peat swamps
humid, tropical settings coal may occur, while in arid undergoing rapid plant accumulation under humid-
climates minor evaporites may develop in inland tropical conditions.
sabkhas.
Other Facies. Carbonates occur as tufas, and as bed-
Lithofacies P: Pedogenic Carbonates. A wide variety ded deposits, with stromatolites, oolites and
of textures and structures develops where flood- oncolites, in freshwater ponds and in some fluvial
plains are exposed to surface weathering processes channels (Ordonez and Garcia del Cura 1983; Eberth
for extended periods (thousands of years). Rain and Miall 1991). Algae form encrustations and re-
infiltration leaches dissolvable ions downward, placements wherever suitable substrate material is
whereas evaporation and capillary groundwater flow present, such as dead plants or shells, and clasts.
during arid periods concentrate the same ions near Oolites and oncolites in the form of grains are moved
the surface. The result is the gradual development of by fluvial transport (Fig. 5.34). Eberth and Miall
carbonate cements that coalesce into nodules and (1991) reported lacustrine marls interbedded with
these, in turn, into more or less continuous carbon- fine-grained floodplain clastic units. They consisted
ate substrates, commonly with a blocky fracturing oflaminated micrite, microspar, siltstone, and ferric
pattern (Fig. 5.32). Modern calcitic soils are referred oxide.
to as caliche or calcrete, and the terms have been
128 Lithofacies
B
Fig. 5.32. Lithofacies P. A Isolated caliche nodules (behind and below hammer) in overbank deposits below a channel
sandstone. Devonian Old Red Sandstone, Scotland. B Mature caliche. Triassic, near Siguenza, Spain
Nonelastic Facies 129
Fig. 5.33. Exposure of coal in a fluvial setting. Tertiary Eureka Sound Group, Ellesmere Island, Arctic Canada
.. .',. ' ...

~
' .. "-
- - O O LIT ES - - -
- -- - - - - - · - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - ONCOllTES " IN SITU·
- - - - - TR.6NSPORTEO O N C O l I T E S - - - - - -
- - - - - - $ T R O M A T O L ITE S - - - -
- - - - - ftJF"A ON NIGHER ~I.ANTS TUFA ON 8A:YOPtnT'E s
CALCu.t()APHIC SOILS
I,·
Fig. 5.34. Idealized sedimentological model of fluvial carbonate formation in the Tertiary Madrid Basin. (Ordonez and
Garcia del Cura 1983)
130 Lithofacies
5.6 Associated Facies limestones (Sect. 12.9.5), and shales containing

lacustrine faunas, such as pelecypods, ostracods, or
fish (e.g., Gersib and McCabe 1981).
Fluvial deposits are commonly intimately interbed- Estuarine environments may contain fluvial and
ded with the deposits of other environments, notably tidal deposits in close association (Zaitlin et al.
lakes, eolian deserts, estuaries, glaciated environ- 1994). Tidal influence may be indicated by the pre-
ments, and volcanic environments. It is not the pur- sence of typical tidal sedimentary structures, such as
pose of this book to describe such environments or tidal rhythmites, flaser bedding, and inclined
their deposits in detail. The following are some brief heterolithic stratification (Thomas et al. 1987; Figs.
notes on these deposits, and some additional refer- 2.38, 13.18) in otherwise typical fluvial successions
ence is made to some of the related environments in (e.g., Shanley et al. 1992). The fluvial-tidal relation-
Sect. 12.9, which is a discussion of the climatic infor- ship is a common one where coastlines undergo
mation that can be gleaned from such deposits. repeated changes in base level, and has therefore
In arid fluvial settings, windblown sediment is received considerable attention from sequence
common, consisting of individual ventifacts, iso- stratigraphers (Sect. 13.3.3).
lated eolian dunes, or discrete laminae or beds of In glacial environments, fluvial sediments occur
eolian dust (loess) or sand. Good descriptions of on outwash plains (sandurs) in front of continental
interbedded fluvial and eolian strata were provided glaciers and ice sheets. There is, therefore, a com-
by Langford (1989) and Langford and Chan (1989), mon association with till and moraine deposits, and
to which additional reference is made, with a brief the deposits of glacial lakes (Eyles et al. 1983; Miall
description of other examples, in Sect. 12.9.4. 1983b; Eyles and Eyles 1992).
Fluvial floodplains commonly contain ponds, Fluvial sedimentation is also an important com-
and the distinction between ponds and lakes may ponent of the volcanic environment. The rapid pro-
become somewhat arbitrary. Many fluvial systems duction of abundant volcanic debris and the violent
also drain into and out of lakes. Fluvial and lake rainfall that commonly accompanies volcanic erup-
sediments are therefore intimately interbedded in tions typically leads to catastrophic debris-flow ac-
many settings. Fluvial crevasse splays may be diffi- tivity (e.g., Vessell and Davies 1981; Lajoie and Stix
cult to distinguish from lacustrine deltas. Associated 1992). The sedimentary apron around volcanos may
lake sediments may include oil shales, palustrine be interbedded with lava flows.
Chapter 6
Architectural Elements Formed Within Channels
6.1 Introduction Bristow and Best (1993, p. 2) summarized the

complexities in channel-fill sedimentation that may
occur during changing stage:
A basic set of eight elements was recognized by Miall
(1985), and has been updated, as noted below. The "Where bars exist for periods of time in excess of a single
flood event they will experience a complex history of ero-
purpose of this chapter is to provide a brief summary
sional and depositional modifications related to changes
description and key illustrations of the main element in stage. At higher flow stages when the largest volumes of
types. Further descriptions are provided in the case- sediment are tranported, the channels are often scoured,
studies that illustrate the discussion of fluvial style in bars may be reduced in height or in some cases completely
Chap.8. eroded. However, during falling stage maximum deposi-
Most of these elements are macro forms which, in tion occurs as discharge and flow competence are reduced.
Channel beds aggrade, the high stage bedforms may be
the definition ofJackson (1975), are the product of modified and new bars may be formed or enlarged as
the cumulative effect of sedimentation over periods sediment is deposited. As discharge continues to fall, bars
of tens to thousands of years. They include major may become emergent and dissected by low stage chan-
channels and bars (such as point bars, side bars, sand nels. Additionally, the nature of the falling limb recession
flats, and islands), flood sheets, and sediment- (rate and length of recession) will be important not only in
the reworking of higher stage sediments, but also in the
gravity-flow lobes. It is the plan view of these macro- deposition and spatial distribution of the finer grained
form elements that generates the familiar channel sediments (silts and clays) ... "
styles, so commonly illustrated by low-level aerial
Modifications in the original element scheme of
photographs of modern rivers.
Miall (1985) are as follows: The proposed code for
Macroforms are "group 6" deposits, in the classi-
midchannel macro forms that accrete downstream
fication of Table 3.2. In principle, they are bounded
was changed from FM (foreset macroform) to DA
by fourth-order surfaces (Table 4.2). However,
(downstream accretion) by Miall (1988a) because
where an element rests on the floor of a channel its
the old code seemed to lead to confusion, and be-
base is defined by the base of the channel, which is a
cause DA is comparable to LA, in terms of the ele-
surface of fifth-order rank or higher (e.g., sequence
ment code and the element it defines. A new
boundary of seventh order). In addition, elements
in-channel element, HO, is introduced here, bring-
are commonly truncated by overlying channels, in
ing the total to nine. These are listed in Chap. 4
which case their upper bounding surface is the fifth-
(Table 4.3). Figure 4.13 is a summary illustration of
order surface at the base of that channel. Fourth-
the eight in-channel gravel and sand elements that
order surfaces are, for this reason, commonly not
are the focus of this chapter. Examples of a typical
preserved. Where they are present, they may com-
range of macro forms and other architectural ele-
monly be recognized by their convex-up shape
ments in a sheet-braided fluvial deposit are illus-
which, in part, parallels stratification in the underly-
trated in Fig. 6.l.
ing beds, and indicates that they are surfaces of
accretion. Large channel-fill sheets may consist of
stacked macro forms separated by fourth-order sur-
6.2 Channels (Element CH)
faces. All the elements may contain internal growth
increments, defined by third-order surfaces. Typi-
cally, these show evidence of cut-and-fill erosional Most coarse (gravel, sand) deposits in fluvial systems
relief. They may rest on mud drapes capping the are deposited in channels, therefore designation of a
underlying element, and they may be followed by channel element in an ancient unit is not a particu-
poorly sorted beds forming the base of the next larly significant step forward. On outcrop profiles,
element, such as gravel lags. use of the CH coding is useful only where the chan-
132 Architectural Elements Formed Within Channels
--====~~~--=-~--~~=-~==~===---
~:::::::===========~=-------=-
element CH: channel (side-filled. w/d=12)
element CH: channel (filled with increments of element 58)
element CH: channel (as above) channel floor scours
o......---''--...J'10m
element DA: downstream accretion macroform
Fig.6.1. Examples of a typical range of architectural ele- vertical scale. Bounding surfaces are numbered according
ments in a sheet-braided sandstone, the Westwater Can- to rank. Unnumbered surfaces are of first- and second-
yon Member of the Morrison Formation, San Juan Basin, order. Individual lithofacies consist mainly of St, Sp, Sh,
New Mexico. The drawings are at the same horizontal and and S1. (MialI1988a)
nel-fill is simple, and cannot be further broken down main basal bounding surface may be labeled as a
into components such as DA or LA. This section major fifth-order surface, with component fifth-or-
describes and illustrates some of the architectural der surfaces indicated by letters with superscripts,
characteristics of channel bounding surfaces and using the annotation system described in Sect. 4.8.
simple aggradational fills. Minor channels include partially to completely
Major channels are bounded by fifth-order sur- abandoned channels, chute channels cutting across
faces. However, there may be a hierarchy of channels point bars, other channels crossing sandflats that are
(and bars) in a fluvial system, with smaller channels generated during falling water, channels generated
within the larger, major channel (Fig. 6.2). Examples by slumping and mass-flow processes at the cutbank
of this were described by Williams and Rust (1969) of the main channel, and crevasse channels that feed
and Bristow (1987). Where components of a channel crevasse splays on the floodplain. These are bounded
hierarchy can be mapped in the ancient record, the by fourth-order surfaces.
Channels (Element CH) 133
<r-- 10 km --t> <:1-- 2 50 m --i> Fig. 6.2. Channel hierarchies in A the Brahmaputra River
and B the Donjek River (after Williams and Rust 1969).
Numbers in circles refer to bars, other numbers refer to
channels. The first-order channel comprises the whole river,
which includes several second-order channels. Bars scale
within the channels in which they occur. In the Brahmaputra
River, third-order channels modify higher-order bars but
still have bars within them, which cannot be shown at this
scale. (Bristow 1987)
3.5
~A
13
19
'=
E
Fig. 6.3. Diagram to show the lack of relationship between complexes formed by lateral channel migration or switch-
the geometry of an individual active channel and the ge- ing with little contemporaneous subsidence; C stacked
ometry of resulting channel-flll sand bodies. Numbers channel complex formed by vertical aggradation. (Miall
above each channel are the width/depth ratios of the sand 1985)
bodies. A,D Simple channels; B, E, F broad channel-fill
Channel geometry is conveniently defined by channel types are illustrated in Figs. 6.4-6.12, to
depth, width/depth ratio, and sinuosity. Sand body which the following discussion is keyed.
thickness is commonly employed as a guide to origi- Channels have concave-up erosional bases (Figs.
nal channel depth, based on reasoning explained in 6.5-6.12). The top of the channel may be erosional
Sect. 10.4.1. Methods of analysis of channels were (Fig. 6.5) or gradational (Fig. 6.6). Channels com-
discussed by Rust (1978a) and Friend (1983), as sum- monly have multistory fills, with each story bounded
marized in Chap. 2 (Figs. 2.33, 2.34, 2.35). The result- by an erosion surface (Figs. 6.4, 6.7). Channel mar-
ing classification is further discussed in Chap. 4. gins become gentler in slope with increasing channel
Studies of the relationships between width, depth, width. Cutbank slopes in excess of 45°, possibly even
and fluvial style, and their investigation in the sub- vertical or undercut, are not uncommon bordering
surface, are discussed in Sect. 10.4. There may not be narrow channels, especially those developed by
a simple relationship between sand body width and rapid erosion in arid regions ("wadi"; Figs. 6.11,
channel dimensions. As shown in Fig. 6.3, channels 6.12). Sheet-like channels may have practically im-
can aggrade both vertically and laterally, leading to perceptible channel margins, sloping at a few de-
sand bodies having very different width/depth ratios grees or less (Fig. 6.4). These variations reflect bank
than the channel which deposited them. Various stability. Channels cut into mud-dominated fines,
....
U>
, 64 wells / section ""
Jmwu 52 ~~~:~;;:' ::======~~~~~~==::======::::::=====------- ____________

Jmwm 6
- --===:::::
- ~
~===---
:::: ===-:s:~ =::: ~
~~- = ~~--c: a
=- Jm~~~ ---::::: ~ _< ' -=--==:-
o 100m
'-325° -
Fig. 6.4. Example of braided sheet sandstone, Westwater Canyon Member, Morrison Formation, near Fort Defiance, Arizona. Note the absence of clearly defined channel
margins. Numbers indicate rank ofboundjng surfaces. Kd Dakota Sandstone; Jmw Westwater Canyon Member; u upper; m middle; I lower. (Miall 1988a)
?;
::r
"
::..
<t>
"2..,
~
~
<t>
:3<t>
...::s
en
'r:t
o
..,
:3<t>
p..
~
~
s·
(")
::r
~
::s
~
en
Channels (Element CH) 135
Fig.6.5. Ribbon sandstone body 8 m thick, deposited in a into thick floodplain fines containing pedogenic carbon-
fixed channel of anastomosed type. The fill consists of ate units. Cutler Group (Upper Paleozoic), El Cobre Can-
simple superimposed cross-bed sets. The channel is cut yon, New Mexico (see Eberth and Miall1991)
particularly where the banks are stabilized by a preexisting sediment into which the channel is cut
dense root network, offer considerable resistance to (Carson 1984a-c; Church and Rood 1983; Crowley
erosion (D.G. Smith 1976) and tend to be steep (Figs. 1983). This is one reason why attempts to define
6.5, 6.10). Peat is particularly difficult to erode fluvial facies models on channel geometry have not
(McCabe 1984), and may result in steep channel always been successful.
margins developed by sudden bank failure (Fig. Recognition of the channel-fill element in a flu-
7.16). Cemented paleosols, such as calcretes, also vial deposit depends on the ability to define the
offer considerable resistance to erosion and may sloping channel margins. Ribbon sandstones, being
result in steep channel margins (Allen and Williams narrow, are commonly completely exposed (e.g., Fig.
1982; see also discussion of this point by Gibling and 6.5) but, apart from this, cutbanks are rarely pre-
Rust 1990). By contrast, those channels which cut served in outcrop (examples are illustrated in Figs.
into unconsolidated sand and gravel are easily 6.6-6.8,6.10). Examples of complete, or nearly-com-
eroded and may retreat rapidly, giving rise to lower plete major channels exposed at the surface were
channel-margin slopes, or stepped margins with described by Okolo (1983) and Hopkins (1985).
steep cutbank sections alternating with flat terraces Channel reconstruction is commonly attempted by
formed by bar complexes and partly filled minor correlation of closely spaced outcrop or subsurface
channels. Where the sediment load of a river is sections (e.g., Campbell 1976; Hopkins et al. 1982;
dominated by sand or gravel, a braid plain may Putnam 1982a,b) but, because of the presence in
develop, with almost unconfined, sheet-like chan- most deposits of a hierarchy of channels of different
nels (Fig. 6.4). The channel cross section geometry is scales, such correlation may be difficult or impos-
therefore not necessarily an inherent property of a sible (e.g., Wightman et al. 1981; see discussion in
river with a particular slope, discharge, or sediment Sect. 9.5.3 of Mannville Sandstone in Lloydminster
load, but at least partly reflects the nature of the area, Alberta). As shown in Sect. 10.3.2, the well-
Fig. 6.6. Cutbank of channel in sand-dominated system. The channel is about 5 m deep and is overlain by a thick
floodplain succession. Eureka Sound Group (Tertiary), Bylot Island, Arctic Canada
known channel reconstruction of the Morrison For- width and depth for the purpose of paleohydraulic
mation in New Mexico by Campbell (1976), which reconstruction are likely to result in large errors.
was constructed by this process of outcrop mapping, If the channel margins cannot be defined, field
has been shown to be unreliable. Large channel-fill analysis is likely to result in a classification of the fill
complexes may be better termed paleovalleys, and in terms of one or more of the other architectural
contain the accumulated deposits of many of elements. For example, the fills of ephemeral chan-
the other types of element described later in this nels on arid braid plains, particularly on lake mar-
chapter. They are bounded by sixth-order bedding gins, are typically sheet-like and may consist mainly
contacts. Good examples up to 8 km across and 90 m of elements SB: sandy bedforms, and LS: laminated
deep were described by Blakey and Gubitosa (1984). sand sheets.
Many small to medium-sized petroleum fields are Channels filled by simple vertical aggradation
contained in fluvial paleovalley fills and paleochan- commonly show fining-upward successions, reflect-
nels (e.g., Mannville Sandstone, Alberta; Muddy ing one of two processes, progressive abandonment
sandstone, Wyoming; "J" and "D" sandstone, Colo- as a result of upstream avulsion, or the plugging
rado; Red Fork Sandstone, Oklahoma; see Chaps. 14, action of a few dynamic events (e.g., flash floods).
15). Typical cycles include:
Where the channel is of broad mobile or sheet
GB~ DA~SB~ OF
type, defining the channel margins may be difficult
LS ~SB~OF
or impossible. Large channels filled by continually
shifting minor channels (the familiar braided pat- The thickness of such cycles cannot exceed the
tern) may contain evidence of several or many tem- depth of the channel, and is likely to be much less
porary channel margins (Fig. 6.4), and the overall where the dynamic event strips away earlier deposits
channel-fill geometry then means little in terms of before depositing their sediment load. As shown by
conventional channel classifications (Schumm 1963; Godin (1991) cycles in channel fills may represent
see Fig. 6.3). Attempts to determine the channel flood events, or the accretion of individual macro-
Fig. 6.7. Rare preservation of a cutbank in multistory lithofacies Sh with numerous low-angle internal third-
sandstone sheets formed by flash-flood sedimentation in order erosion surfaces. Kayenta Formation (Jurassic), Ari-
an arid environment. The channel fill consists mainly of zona
Fig.6.8. Close-up of cutbank illustrated in Fig. 6.7. Note blocks of siltstone on the cutbank that have been eroded from
the siltstone bed at the upper left. Kayenta Formation (Jurassic), Arizona
Fig. 6.9. Abandoned channel filled with mudstone, within a delta-plain succession. Note sandy levee deposits high on
the flanks of the channel. Carboniferous, Kentucky
Fig. 6.10. Channel cutbank that became a slide plane, along which the channel fill slumped down to the right. Note the
tilted sandstone block at right. Carboniferous, Kentucky
Fig. 6.11. The person in the center is standing within the good example of a fossil "box canyon", a steep-sided wadi,
fill of a narrow, steep-sided channel eroded in the Canyon typical of erosional valleys in arid regions. Canyon de
de Chelly Sandstone, and filled with Shinarump Formation Chelly National Monument, Arizona
(Triassic). Edges of channel are shown by arrows. This is a
forms, and so caution should be taken in the ex- Some deposits occur as sheets, bounded by irregular
trapolation from cycle thicknesses to channel di- channel bases with linear scours; other deposits oc-
mensions. cur in narrow channels with steep margins. Both
Channels, particularly in high-sinuosity streams, types originate as channel-bank failures, with slum-
may be abandoned by chute or neck cutoff, in which ped bank materials evolving from landslip masses
case they will be filled by fine-grained deposits [ele- (e.g., Fig. 6.10) into sediment-gravity flows. The ori-
ment CH(FF); see Table 7.1] showing a channelized, entation of the channels and their basal scours is
concave base (Fig. 6.9). Hopkins (1985) described typically at a high angle to the main channel trend.
three well-exposed channels that contain concave-
up fills deposited mainly by vertical aggradation
during progressive or sudden abandonment. Okolo
6.3 Gravel Bars and Bedforms (Element GB)
(1983) described channels filled obliquely from the
side during gradual abandonment. Components of
these channel fills may be interpreted as alternate Lithofacies Gh, Gp, and Gt define three main kinds of
bars (Sect. 6.7). meso forms. The formation of these is described in
Minor crevasse-, chute-, and bar-top channels, Sect. 5.2.1.2. In preserved gravel deposits, the vari-
bounded by fourth-order contacts, typically contain ous mesoforms and constituent lithofacies are com-
assemblages of lithofacies Ss, Se, and Sl, with gravel monly complexly interbedded (Fig. 5.2), and it is
lags and thin units of flow-regime bedforms (ele- therefore important, in a description of the GB ele-
ment SB: sandy bedforms; lithofacies St, Sp, Sh, Sr), ment, to understand how varying water and sedi-
showing no particular cyclic order. ment discharge may lead to rapid lateral and
A distinctive type of minor channel containing temporal changes in lithofacies and architecture.
massive sandstones (lithofacies Sm) was described Mesoform and macro form development in
by Jones and Rust (l983), Rust and Jones (1987), gravel-bed rivers is strongly affected by the nature of
Turner and Munro (1987), and Wizevich (1992b). gravel transport. Gomez et al. (l989) and Hoey
Fig. 6.12. Close-up of the vertical

margin of the wadi fill shown in Fig.
6.11
(1992) showed that transport is not regular at any sodes of high water and sediment discharge, these
point in a braided system, but occurs as a series of sheets grow upward and downstream by the addition
pulses or sediment "slugs" at different spatial and of clasts to form horizontally stratified gravel sheets
temporal scales. Stochastic variations in entrain- (Gh), the structure termed longitudinal bars in some
ment are responsible for some pulses, and bar ero- older literature (Rust 1972). These mesoforms reach
sion, channel avulsion, and tributary supply may about 1 m in height, and may show either an upward
feed gravel into the system at irregular rates. In increase or decrease in clast size, depending on their
natural systems, these sediment pulses are several mode of accretion. Clast accumulation in place tends
hours in duration, and have a significant effect on to result in upward fining, as the deposit builds to
channel and bar development, as summarized be- shallower water levels. However, these mesoforms
low. tend to fine downstream, and they may also migrate
The simplest deposits are the thin, diffuse gravel downstream. In such cases, the coarser topmost de-
sheets of Hein and Walker (1977), which are a few posits migrate over the finer deposits that form the
clasts thick, have diffuse, lobate margins, and move base of the meso form (Gustavson 1978). Processes
only during peak flow (lithofacies Gh). During epi- that lead to gravel-sand segregation and to upward
Gravel Bars and Bedforms (Element GB) 141
fining and coarsening within the gravel deposits are S.A. Smith (1987) described a minor type of mid-
discussed in Sect. 5.2.l.2. Gustavson (1978) and channel bar, termed a counterpoint bar, that devel-
Massari (1983) noted the close relationship between ops on the outside of meander bends as a result of
the three main gravel lithofacies and confirmed the turbulence and shoaling near the bend entrance
general model developed by Hein and Walker (fine-grained equivalents, termed concave benches,
(1977). were described earlier by Nanson and Page 1983).
These gravel mesoforms are amalgamated into Each bar type may evolve into the other, as a result of
macro forms (bars) in several ways. Our present un- flow diversions resulting from meander migration
derstanding of gravel macro forms follows from the and channel avulsion.
classic work of Leopold and Wolman (1957), who Large-scale turbulence leads to the development
modeled the formation of midchannel bars in a of alternate bars in initially straight channels, as a
flume. Observations in modern rivers by Hickin straight thalweg develops a meandering trace
(1969), Bluck (197l), Lewin (1976), Gustavson (Hickin 1969; Lewin 1976). These evolve into fully
(1978), Forbes (1983), Ferguson and Werritty (1983), developed point bars (element LA), the development
Hooke (1986), and Carson (1986) were synthesized of which is discussed briefly in Sect. 6.7.
by Ashmore (1991), who carried out a series of flume Ashmore (1991) recognized three main methods
runs to demonstrate the major processes. Ferguson of midchannel gravel bar construction. These are
(1993) also provided a useful review of braiding pro- illustrated in Figs. 6.l3-6.15 using diagrams from his
cesses in gravel-bed rivers. flume runs. The first method is a process whereby
Bar types include midchannel and bank-attached point and alternate bars are separated from the bank
types. The latter are termed point, alternate, or by the development of a chute channel (Fig. 6.l3).
lateral bars depending on their shape and position. This occurs when sediment is eroded rapidly from a
Transverse bar maroin (with slip face)
..
Bed· load sheet margin (no slip face)
Row direction (a )
~ El<Oosed or inactive bar surfaces
~ Cutbank
• ",~
- -.. Scour hole or pool
J ! I I I ! , !
Scour chute
, iii i , , , i
( b)
Fig.6.13. Evolution of mid channel bar

in a gravel-bed river by process of chute
cutoff, as modeled in a flume. An initial
point bar A in diagram a is separated
from the bank by the development of a
scour pool, as shown in h. A new point
bar develops on the opposite bank, but is
removed by erosion as the new
midchannel bar enlarges and stabilizes o 1m
~I_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _- ",
(c). (Ashmore 1991)

Fig. 6.14. The central-bar braiding me·

chanism in a gravel-bed river. A, B, and C
" are successive bed-load sheets. Eventually,
...i
--.:......:..:;~"-----
one of these stalls (the stippled area in
'~_:.:.::
.......
~:*,;i ( 0 ) b), and becomes the locus of gravel accu-
~-"'7""..,...",~o::rT:
mulation. Drawn from flume model by
~
Ashmore (1991). Symbols as in Fig. 6.13
~;:::;:=~~~;!.~~~~u ..
: - - -•• ••.•.
...' ~ .<'
;lI<, ....
,
•
~~··~~(C)
_._--'"'--.
~ .... """' -
1m
'------~'
---".--
.
.:.,.
..-
....... ... . ~ ..........--.:... ... ,
B'
'. .. ,
....:... - A~
~.-~=:::. -. -~.,
D ...:=::-- ~ Fig. 6.15. Development of midchannel bar
~~~. (" ~ in gravel-bed river by the migration and
coalescence of transverse dunes. From
~
'; ~,....:
flume experiment by Ashmore (1991).
1m
Symbols as in Fig. 6.13
~--------------~I
bar upstream. The arrival of this slug of sediment at process involves the migration of thin bed-load
the head of the point bar causes additional flow to be sheets that become stalled in midchannel positions
directed over the bar. The second process, central because of slight changes in shear stress (Fig. 6.14),
bar initiation, is one that was first modeled by for example at places where the channel widens or
Leopold and Wolman (1957), in an experiment that shallows. Under different conditions of bed-shear
has been widely quoted in the fluvial literature. The stress transverse bedforms with slip faces develop.
Gravel Bars and Bedforms (Element GB) 143
The migration and convergence of these bedforms ges of gravel organized between third- and fourth-
also lead to midchannel bar formation, the third of order bounding surfaces. However, at present, we
the main processes of bar development (Fig. 6.15). can only speculate on the geometry and relative ori-
Leddy et al. (1993) also modeled braid bar devel- entation of the surfaces and current structures.
opment in the laboratory, and proposed three Some comments follow on what is known about
slightly diferent mechanisms of bar development. ancient GB deposits. An example is illustrated in Fig.
"Choking avulsion" occurs when a sediment lobe 6.16.
migrates through an active channel into an area of Element GB typically forms multistory sheets tens
flow expansion, where deceleration and flow diver- to hundreds of meters thick. Flat or irregular erosion
gence occur. Part of the flow is diverted into a differ- surfaces between meso forms are common. Steeply
ent course. Typically, this will involve diversion dipping channel margins are rarely seen, partly be-
across an older bar deposit, and a remnant of the cause they tend to be minor parts of a gravelly fluvial
sediment lobe is left as the nucleus of a new landscape. Actively migrating channels may under-
midchannel bar. "Constriction avulsion" occurs cut older bar gravels, producing cutbanks 1-2 m
when a sediment lobe migrates into an area of high, but when filled with later bar gravels of similar
channel harrowing, as at a channel junction. Block- composition and texture, the cutbanks may be diffi-
ing of the channel leads to overbank flooding, cult to identify (Fig. 6.17).
crevassing, and diversion of part of the flow into S.A. Smith (1990) defined three styles of GB ac-
a different course, leaving the sediment lobe as a cretion in the Triassic Budleigh Salterton Pebble
new bar nucleus. "Apex avulsion" occurs as a result Beds of southern England. Couplets of Gh and St
of the lateral and downstream growth of macro- separated by fourth-order surfaces represent depo-
forms, accompanied by corresponding cutbank ero- sition in topographically segregated parts of a deep
sion on the opposite side of the channel. Eventually, braided channel (Fig. 6.1Sa). Units of Gp alternating
this erosion may lead to breakthrough into lower laterally with Gh and sandstone lithofacies (Fig.
areas in the channel complex, and partial flow diver- 6.1Sb) represent the variations in gravel sheet
sion. growth described by Hein and Walker (1977), that
Germanoski and Schumm (1993) modeled the ef- S.A. Smith (1990) postulated took place in a rela-
fects of adding and removing sediment load from tively deep channel, as evidenced by the height of
experimental braided streams, and documented the some of the Gp sets (3 m). Superimposed Gt wedges
subtle but significant changes in fluvial style that bounded by fourth-order surfaces represent scour
result. In the case of gravel-bed rivers, addition of fills (Fig. 6.1Sc).
sediment load leads to increased braid-bar develop- Element GB may be interbedded with minor to
ment and aggradation, whereas removal of load predominant sheets or lenses of element SG: sedi-
leads to amalgamation of braid bars and to channel ment gravity flows. Element SB typically comprises
incision. Changes in sediment load could be very 5-10% of even the coarsest gravel succession, and
local effects resulting from changes in the materials represents slack-water deposits, such as abandoned
introduced into the channel by bank erosion, or channel fills (minor CH element, where identifiable),
longer-term, regional changes brought about by tec- bar-edge sand wedges, and micro deltas (Rust 1972;
tonic or climatic causes. Base-level changes will also Miall 1977), or the deposits of topographically el-
modify sediment load, as discussed in Sect. 11.2.2. evated parts of a deep gravel river (Fig. 6.1Sa), such
The different conditions that lead to the forma- as parts of the Donjek River, Yukon (Williams and
tion of gravel sheets and slip-face bedforms have Rust 1969; Rust 1972; S.A. Smith 1990). These ele-
been discussed earlier (Sect. 5.2.1.2), drawing on the ment associations are discussed further and illus-
work of Hein and Walker (1977). In actual streams, trated in Chap. S. The development of fourth- and
all these processes may be expected to occur simulta- fifth-order lithosomes in gravelly alluvial fans is dis-
neously, or in succession, as the organization of the cussed in Sect. 10.2.
channel and bar complex changes, and conditions of It is now known that channel confluences are sites
water and sediment discharge vary. Unfortunately, of significant scour in gravel-bed rivers (Ashmore
there is virtually no information on the internal ar- 1993; Bridge 1993b; Ferguson 1993; Siegenthaler and
chitecture of the different types of bar deposits Huggenberger 1993), and that large scoop-shaped
formed by these three processes. Lithofacies types deposits may occur there. As Ashmore (1993, p. 130)
are as described in Chap. 5, and it is anticipated that, stated, "They are one of the few places in gravel
internally, the bars would consist oflenses and wed- braided stream where flow is sufficiently deep to
~----------~5~--------------------------~
GB
__------------~4~--------------
LA
.: .
-----
I==::::=--___ ~ ........ ~
~.. LA
.. . ~ .. 5 ...... ;...: SB~ ...
·····~···4·········· · ....
o 5m
I
Fig.6.16. Examples of LA, GB, and other elements in gravels. Buntsandstein (Permian-Triassic), central Spain. Interpre-
tation is adapted from Ramos and Sopena (1983)
Sediment-Gravity-Flow Deposits (Element SG) 145
Fig.6.17. Beds of Gh draping the floors and margins of outcrop (arrows). Enon Conglomerate, Cretaceous, near
minor channels. At least three channel-floor bounding Oudtshoorn, South Africa
surfaces, offourth- or fifth-order rank, are present in this
a allow the development of avalanche faces at unit bar

margins." A distinctive element termed the "hollow"
~
(HO) is the result. This is described in Sect. 6.9 .
.. : .l1li: •• : •• :: •••• : •• ;: •. : ••. .' ••. : ••:: ••• :.~• .-....:. :i;:!;'*;-"
! •..• : ••.• : •••• ; •••• ; •••• : •••• : ••.••.•••.••••••••••~...... -.- ~.
...:.:... •• : .. ! ••. , .•...•• i • •~
6.4 5ediment-Gravity-Flow Deposits

n ...... • ••• • ••• • ••• • •••
(Element 5G)
b
This element occurs as narrow, elongate lobes or
multistory sheets, and is typically intimately inter-
bedded with element GB or SB (Fig. 6.19). Lithofacies
Gmm, Gmg, Gci, and Gcm are predominant. The
element forms by debris flows and related mecha-
c nisms, as described in Sect. 5.2.1.3. Individual beds
average 0.5-3 m in thickness, rarely exceeding 3 m.
Flow units may be lobate in plan view, with widths of
up to about 20 m, and downstream lengths of several
kilometers (Hooke 1967; Wasson 1977a,b; Vessell
Fig. 6.18. Three styles of element GB in the Budleigh and Davies 1981; Nemec and Muszynski 1982; Blair
Salterton Pebble Beds (Triassic), England. a Couplets of and McPherson 1992). Amalgamated flows with total
trough cross-beded sandstone (lithofacies St, element SB) thicknesses of several meters are common. A single,
and horizontally bedded conglomerate (lithofacies Gh, exceptionally extensive SG unit, triggered by a cata-
element GB). b Planar cross-bedded conglomerate (Gp)
interbedded with horizontally bedded conglomerate and strophic rockfall, was described by Biirgisser (1984).
sandstone (Gh, St). c Large-scale trough cross-bedded Flow units typically have irregular, nonerosive
conglomerate wedges (Gt) (S.A. Smith 1990) bases. Flow events passively occupy existing ero-
Fig. 6.19. Typical outcrop of element SG, showing various sandstone lenses (S). Cutler Formation, Colorado.
sediment-gravity-flow deposits (substitute G for D for ap- (Schultz 1984)
propriate facies codes used in this book) and interbedded
sional channels or the irregular topography formed In some cases, large exposures show that short
by earlier sediment-gravity-flow and sheet-flood sequences of bedforms, resting on dipping bounding
events. Internally, they may show a wide range of surfaces, are interbedded with each other over wide
textures and fabrics. Disorganized textures are areas below major (fourth-order) convex-up bed-
typical of rigid plugs that are rafted at the center of ding contacts, indicating that they were dynamically
some debris flows (Bull 1977). Grading and in- related and formed simultaneously. This type of ar-
verse grading are common. Nemec and Muszynski chitecture is the key diagnostic characteristic of ele-
(1982) described an upward transition in some ments LA and DA (lateral and downstream accretion
flow types (their facies C) from graded to low- units), described in the next sections. Where these
angle cross-stratified gravels, which they inter- architectural features can be conclusively ruled out,
preted as a transition from debris-flow to traction- the deposits probably represent fields or trains of
transport mechanisms. Buck (1983) described a individual bedforms that accumulated predomi-
sand-dominated diamictite facies interpreted as nantly by vertical aggradation. In some cases, evi-
mud-flow deposits. Shultz (1984) proposed a four- dence of both vertical and lateral accretion may be
fold classification of sediment gravity-flow depo- present in a given element. Gibling and Rust (1990)
sits based on matrix content, packing charac- proposed the calculation of an "aggradation index"
teristics, and grading. His lithofacies scheme has to quantify the relative importance of the two differ-
been adapted for use in this book (Sect. 5.2.1.3; Table ent styles of accumulation, but not enough measure-
4.1). ments of this have been made to determine where
numerical cutoffs should be placed between domi-
nantly aggradational (e.g., SB) and accretionary (LA,
DA) geometries.
6.S Sandy Bedforms (Element SB)
Vertical stacking of different bedform types indi-
cates long- or short-term changes in flow regime.
The familiar flow-regime bedforms that form in Short-term changes occur during stage changes
sand-dominated river systems have been described (flash floods, seasonal fluctuations; group 5 deposits
by many writers (e.g., Allen 1968, 1984; Southard of Table 4.2). Longer term changes reflect aggrada-
1971; Harms et al. 1975, 1982; MiallI977; Collinson tion and reduction in water depth over periods of
and Thompson 1982; Ashley 1990). Dunes (3-D several to many years (groups 6, 7 deposits of Table
dunes of Ashley 1990; lithofacies St), sand waves, and 4.2). Both can result in similar lithofacies assem-
transverse bedforms (2-D dunes of Ashley 1990; blages and successions (which is one of the problems
lithofacies Sp), linguoid bedforms (large 3-D dunes, with vertical profile analysis; see Godin 1991),
lithofacies Sp in small outcrops), upper flow-regime requiring examination of the architecture and
plane beds (Sh), washed-out and humpback dunes overall context of the deposits in order to arrive at
(Sl), and ripple marks (Sr) occur in a wide variety of correct interpretations. Such deposits contain first-,
fluvial settings and show a range of assemblages and second-, and third-order contacts. A brief discussion
vertical sequences. of some typical examples of the SB element follows.
_ 330 r.n
A ~
::t
~
to
It>
p..
0'
B 330 S
en
:::::;--~ --------===--- -- - ---- -- ---- ----- --- -----------
'@
It>
'S"
::1.
r.n
~
C _ 330
Fig. 6.20A-C. Lateral profiles through sheet sandstones of element SB. Association 2, Battery Point Formation (Lower Devonian), Gaspe, Quebec, Canada.
(Lawrence and Williams 1987)
,....
....,
""
Channel-Floor Dune Fields. Fields of 3-D dunes (litho- Shallow Channel-Fill Assemblages. Miall (1977) defined a
facies St), characteristically occupy the deeper por- distinctive lithofacies assemblage dominated by Sp
tions of active channels wherever the bed load is and showing little or no internal cyclicity (Fig. 5.20).
predominantly sand (e.g., Fig. 6.20). Lenses or lobes This was named the Platte type of braided river
of trough cross-bedded sand a few meters thick deposit, after the Platte River, Nebraska. This assem-
and tens to hundreds of meters wide may result. blage represents the migration of fields of linguoid
They may be cut by broad, shallow scours and or transverse bedforms, many of which are capped
erosion surfaces of second and third order, indicat- or draped by Sr or FI during falling water (N.D.
ing stage fluctuations (Harms et al. 1963; Harms Smith 1970, 1971, 1972; Blodgett and Stanley 1980).
and Fahnestock 1965; McGowen and Garner 1970; The "cross-bedded simple bars" of Allen (1983a)
Jackson 1976a; Cant and Walker 1978; Nijman are similar. Many ancient examples of this assem-
and Puigdefabregas 1978; Plint 1983; Buck 1983; blage occur, for example, in the Cretaceous and
Stear 1983; Tyler and Ethridge 1983; Lawrence and Lower Cenozoic deposits of Arctic Canada (Miall
Williams 1987). In the shallower parts of channels, 1976, 1984b) and the Nubian Sandstone (Jurassic-
including the tops and flanks of macroform ele- Cretaceous) of North Africa (Harms et al. 1982;
ments (such as point bars and sand flats), transverse Bhattacharyya and Lorenz 1983).
bedforms and sand waves (2-D dunes) are common. Crowley (1983) suggested that in at least some
These generate sheets of planar-tabular cross- cases these bedforms form part of much larger
bedding (lithofacies Sp) (N.D. Smith 1970, 1971, macro form structures 200-400 m in length and
1972; Levey 1978; Jackson 1976a; Nijman and 0.7 -1.5 m high. The architecture of these structures
Puigdefabregas 1978; Cant and Walker 1978; is discussed in the next section. One of their diagnos-
Blodgett and Stanley 1980; D.G. Smith 1983). tic characteristics may be an upward-coarsening
Fig.6.21. Sandstone sheet (element SB) resting on reces- nantly by vertical aggradation. It is interpreted as a cre-
sive coal bed, at base of outcrop. The sandstone unit con- vasse-splay deposit in an upper delta-plain setting.
tains gently dipping accretion surfaces at left, indicating Pocahontas Basin, Carboniferous, Kentucky. (Horne and
that it developed in part by lateral progradation, but domi- Ferm 1976)
Sandy Bedforms (Element SB) 149
succession generated during high-stage flow condi- Crevasse Channels and Crevasse-Splay Deposits. These are
tions. Not alliinguoid bedforms fields seem to be typically composed of element SB. Channels should
interpretable in terms of Crowley's (l983) macro- be identifiable by the concave-up channel floor (a
form hypothesis - the distinctive meandering main fourth-order bounding surface), and proximity to
channel of Crowley (1983; his Fig. IDA) is not present the main channel. Splays are sheet-like bodies tens
in much of the Platte River (Blodgett and Stanley to hundreds of meters across and typically 1-2 m
1980; their Figs. 2, 3), and even where the macro- thick that thin and pass laterally into element FF:
forms are present the complete sequence may not be overbank fines. Both upward-coarsening and up-
preserved. Miall's (l977) original Platte model may ward-fining successions may be present, indicating
therefore still have its uses, but clearly the search for progradation or gradual abandonment, respectively.
the macroform architecture should be pursued Plant litter and vertebrate bones are common
wherever possible. (Collinson 1978; Ethridge et al. 1981; Gersib and
McCabe 1981; D.G. Smith 1983; Bridge 1984; Eberth
Bar-Top Assemblages. Many workers have described the and MiaIl1991). Where architectural description of
characteristic small-scale cross-bedding that occurs these deposits is possible, they may be assigned their
in shallow areas of active channels, particularly on own element codes, as described in Sect. 7.2. Figure
bar tops. Various types of ripple cross-lamination 6.21 illustrates an example of a broad, sheet-like
(lithofacies Sr) are the result. These small-scale splay deposit in which local lateral accretion can be
structures are typically deposited during falling wa- observed.
ter and, where preserved, their capping of larger
bedforms or bars produces local fining-upward se- Distal Braid-Plain Sandstone Sheets. On distal braid plains,
quences. Such sequences are almost ubiquitous in such as those bordering playa lakes, fluvial deposits
fluvial environments and their occurrence has little may be entirely composed of element SB. Sheets of
diagnostic value in terms of fluvial style. sand develop in broad, virtually unconfined chan-
Fig. 6.22. Superimposed sheets of element SB, bounded mainly of St, Sh, and Fl. Peel Sound Formation (Lower
by fifth-order surfaces, and constituting successive flood Devonian), Somerset Island, Arctic Canada. (Miall and
cycles in a distal braid-plain setting. Lithofacies consist Gibling 1978)
nels (Fig. 6.22). Aggradation and progressive aban- Particularly vigorous flood events in ephemeral
donment of these channels occur slowly or during channels may produce a distinctive type of litho-
single flood events. In either case, fining-upward facies assemblage and sand body geometry, de-
cycles are commonly the typical result. Williams scribed below under the heading of element LS.
(1971) and Hardie et al. (1978) described modern A few workers have documented the ocurrence of
examples, and Miall and Gibling (l978) documented giant cross-bedding, indicating the former existence
an ancient example. In the latter case, cycles are of deep channels which migrated down very large
mostly between 1 and 3 m in thickness, and show bedforms. Such bedforms are mesoforms. They are
an upward transition from a scoured base through internally of simple structure, with steep foreset dips
Sh, Sp, or St, to Sr and Fl, or directly to Fl omitting (e.g., Fig. 5.21), and are not to be confused with the
Sr. Similar deposits characterize the arid "terminal more complex macro forms described in the next
fan" deposits of northern India (Parkash et al. 1983) section. Giant bedforms are particularly common in
and elsewhere, a depositional style that is gaining the giant sandy rivers draining the Himalayas, where
increasing recognition in the ancient record (Sect. sand waves up to 15 m high have been recorded
8.3). (Coleman 1969; Singh and Kumar 1974). However, as
noted below, at least some of these sand waves are
Other Occurrences OfElement SB. In the pre-Devonian, the probably macroforms. Possible ancient analogs were
lack of vegetation is thought to have resulted in a described by Conaghan and Jones (l975). McCabe
predominance of weakly channelized bed load (l977) and Jones and McCabe (l980) analyzed cross-
streams (Schumm 1968a; Cotter 1978). The architec- bed sets up to 40 m thick and 1 km wide that they
ture and composition of the resulting fluvial depos- interpreted as the deposits oflarge prograding alter-
its were probably in many cases similar to the distal nate bars in a major delta distributary (Fig. 2.27).
braid-plain sand sheets described here. Long (l978) Reactivation surfaces (third-order surfaces) were in-
discussed some Proterozoic examples. terpreted in terms of fluctuating water depth. If, as in
Fig. 6.23. Example of a midchannel sand flat in a modern bounding surfaces of first- to third-order dipping in
sandy-braided river. These compound bars develop by the direction of accretion. (Photograph courtesy of D.G.
lateral and downstream accretion, and contain internal Smith)
Downstream-Accretion Macroforms (Element DA) 151
this case, the deposits can be interpreted as alternate chitectural analysis is essential for complete descrip-
bars, they should be assigned to a category of tion of these deposits because, unlike elements GB,
macroform rather than to element SB (see Sect. 6.7). SG and SB, they contain significant internal three-
dimensional geometrical complexities. Large com-
pound bar forms have been described from many
6.6 Downstream-Accretion Macroforms modern rivers (Fig. 6.23), including the side bars of
the Tana (Collinson 1970), the sand flats of the South
(Element DA)
Saskatchewan (Cant and Walker 1978), and the sand
waves of the Brahmaputra, as reinterpreted by
Downstream-accretion and lateral-accretion depo- Bristow (1987, 1993). The bars are characteristically
sits are the principal products of accretion within the 1-15 m high and 10-1000 ill long. The maximum
bar complexes of major sand-bed channels. Up- recorded height and length refer to the bars of the
stream accretion also may occur locally (e.g., Bristow Brahmaputra. It is only in recent years that such
1987, 1993). The bars are scaled to the size of the large-scale deposits have been described in ancient
containing channel, and their height is a rough guide deposits. Research by the author indicates that DA
to minimum channel depth. Three-dimensional ar- and LA elements are very common in braided sheet
St 0
(a)
/...J
s m 000m
A Fig. 6.24. Generalized models of down-

stream-accretion (DA) architectural ele-
ments. a Loosely based on Allen (1983a)
and Kirk (1983); b loosely based on Cant
and Walker (1978) and Haszeldine
(1983a,b). a and b from Miall (1985).
c Model developed from a synthesis of
macro forms in the Castlegate Sandstone,
Utah, showing the system of annotation of
bounding surfaces used in this book. Note
5-100m gradation between LA and DA architecture
within the same element (Miall 1994).
0,2) cross-bed set
Scales are approximate. Internal geo-
(3,4) minor bounding surface metry varies considerably depending on
channel depth, grain size, discharge
-- direction of local flow (5) channel floor (e) amount and variability
la ~ge Sp St set small Sp. St sets
A.
o 10m
'---'----'
St
c. 5
D.
Fig. 625. Ancient examples of DA elements showing the example, England (Haszeldine 1983a,b); B Poison Canyon
distinctive, internal, downstream-dipping bounding sur- Sandstone, Morrison Formation, New Mexico (Miall
faces, down which trains of medium-scale bedforms mi- 1988a); C Example from Kayenta Formation, Colorado
grated. Letters are lithofacies codes (Chap. 5), numerals (Miall 1988c); D part of a "sand wave" in the modern
are rank of bounding surfaces (Chap. 4). A Carboniferous Brahmaputra River (Coleman 1969). (Miall1988a)
sandstones (e.g., Figs. 4.8, 4.15), a fact that appeared obvious, and forms the main basis for the definition
to be missed by those focusing on vertical-profile of the element, but in some cases the accretionary
studies. geometry may be very subtle and difficult to define
The essential characteristic of a downstream- in small outcrops. Figure 6.21 illustrates a crevasse-
accretion element are that it consists of several splay sandstone sheet that contains some evidence of
(possibly many) cosets of downstream-oriented lateral progradation, yet is classified here in the SB
flow-regime bedforms dynamically related to each element because of the dominance of vertical aggra-
other by a hierarchy of internal downstream-dip- dation. For more precise work, use of Gibling and
ping bounding surfaces (Figs. 6.24, 6.25). These re- Rust's (1990) "aggradation index" may be useful to
veal the former existence of an active, nonperiodic, assist in discriminating betwen the two styles of
possibly irregularly shaped bar form comparable in accumulation.
height and width to the channel in which it formed. It has been shown that midchannel macroforms
The bars contain second- and third-order surfaces may accrete downstream at their downstream ends
that generally dip gently « 10°) downstream (Banks and laterally along their flanks (Bristow 1987, 1993).
1973; Allen 1983a; Haszeldine 1983a,b; Kirk 1983; Upstream accretion on the upstream flank has also
Miall 1988a,c, 1992a, 1994), although oblique or been recorded (Bristow 1987, 1993). Preserved
gentle upstream dips, around and over low-relief bar macroforms may therefore grade laterally from DA
cores (sand shoal of Allen 1983a) may also be pre- to LA architecture (Allen 1983a; Miall1993, 1994), as
sent. Between these surfaces are sets or cosets of Sp, suggested in the accompanying model diagram (Fig.
St, Sh, Sl, or Sr. The Sh and Sllaminae are organized 6.24c). Care must be taken in the interpretation of
parallel or subparallel to the internal bounding sur- two-dimensional outcrops to distinguish these ge-
faces. Detailed paleocurrent studies show that the ometries, using all available paleocurrent informa-
flow-regime bedforms advance generally down the tion. It is suggested that where a gradation between
slopes defined by the second- and third-order sur- the LA and DA end members can be documented a
faces (Haszeldine 1983a,b) or oblique to the surfaces cutoffbe employed as follows (Fig. 6.24c). Where the
draping the bar cores (Allen 1983a; Miall 1994). orientation of the accretion surface and that of the
These data reveal a picture of fields of bedforms cross-bedding within the same element are within
driving across, around and down the bar forms (Fig. about 60° of each other, it indicates that the element
6.24c). grew by accretion in a direction parallel or oblique to
Most sandstone elements are accumulated by local flow, and the element is designated a DA unit,
both vertical aggradation and lateral accretion. In even if local flow is oriented at a high angle to the
DA and LA elements, the evidence of accretion is regional trend. Where the orientations of the accre-
Downstream-Accretion Macroforms (Element DA) 153
SYMMETRICAL dunes arranged in an en echelon pattern. Internally,

they are composed of a single, large-scale Sp set
resting on an apron of fines and draped by coarser-
grained St or Sr sets. The upward coarsening reflects
varying shear stress in relation to increasing water
depth from top to bottom of the advancing foresets
(Fig. 6.27). Crowley (1983) suggested that these
macro forms are comparable to, and grade into, the
alternate bars of moderately sinuous rivers.
Many of the variations in composition and geom-
etry between described macroforms probably reflect
fluctuations in stage and local changes in sediment
supply (Germano ski and Schumm 1993). Many of
the second- and third-order surfaces have the char-
EXAMPLES OF COMPLEX SAND FLATS acter of reactivation surfaces (Collinson 1970). The
"sand flat" macro forms of Cant and Walker (1978)
are cut by numerous erosional channels during fall-
ing water. Kirk (1983) described a distinctive low-
stage lithofacies assemblage draping the macroform,
distinguished from the body of the structure by di-
Fig.6.26. Development of sand flats, based on the surface- vergent paleocurrents that reflect falling-water sur-
process studies ofeant and Walker (1978) face runoff and bar-top channel orientation.
Bridge (1993b) developed simple models for the
development of midchannel macroforms such as DA
tionary surfaces and the cross-bedding are more units, based on his dynamic models of flow in river
nearly perpendicular (> 60° difference), the element bends, and observations of modern rivers. Figure
is designated an LA unit. Where adequate data can- 6.28 provides a general architectural model, and Fig.
not be gleaned from the outcrop, the macro forms 6.29 suggests how patterns of deposition and erosion
may be designated DAlLA. change during changes in flow stage.
Cant and Walker (1978) referred to the large, Wizevich (1992b) developed a model for a typical
midchannel macro forms in the South Saskatchewan DA unit in his Carboniferous deposits (Fig. 6.30).
River as sand flats. They showed that these evolve Resting on a fourth-order surface is a thin unit of
from large, simple (2-D), flow-transverse bedforms, pebbly sandstone with poorly defined cross-beds
termed cross-channel "bars" by Allen (1983a) and (lithofacies Ss). This is followed by a large Sp set,
Cant and Walker (1978) Elevated parts of these bed- above which are various assemblages of smaller-
forms (which may be emergent at low water) become scale St and Sp. The element is capped by small
the nuclei of new sand flats, which anchor part of the scoops or channels filled with asymmetric Sp sets.
bar in the middle of the channel. Sediment is added The large Sp set at the base of this succession com-
to the cross-channel bedform by the migration of pares with the similar set in Haszeldine's (1983a,b)
fields of dunes and ripples. These may move more model (Fig. 6.25A), and probably corresponds to the
slowly over the crest of the bar, which may become cross-channel "bar" that initiates DA development
anchored completely if an emergent nucleus is in Cant and Walker's (1978) model.
present. The opposite end of the crest line, in deeper Descriptions of macro forms in modern rivers
water, continues to advance more rapidly, so that the typically suffer from the lack of three-dimensional
entire bedform swings around oblique to the chan- control. Thus, Cant and Walker's (1978, Fig. 14) sand
nel direction (Cant and Walker 1978, Fig. 9; Allen flat model predicts a simple tabular sheet of Sp
1983a, Fig. 19). The macroforms accrete sediment cosets, and Crowley's (1983, Fig. 1O) macroform
partly by this process of bedform capture on the model similarly predicts simple superimposed sets
upstream or flanks, and partly by rapid burial and of Sp and Sr. Missing from these descriptions are
preservation of superimposed bedforms on the ad- any indications of dipping second or third-order
vancing downstream face (Fig. 6.26). internal bounding surfaces. Crowley'S (1983) model
Crowley (1983) described a Platte-type macro- shows reactivation surfaces that could be third-or-
form consisting of suites of large, linguoid (3-D) der surfaces, but their distinctive characteristics
Pla.tte-type macroform
A B
~'_'IJor chat"".\ Hig/1 Walur T
slipface
channel
South Platte-type macroform

Majo< 81 ip lace
------.~----~-----------~~-----------
MinOt channel
Lobe
"'-~~~- Aproo
- Major channul
Upper South Platte-type macroform
~Lo 2S
SCALE
50 m
plan views nol 10 scale
- Major slipface
Fig. 6.27. The macro form models of Crowley (1983). The macroform is comparable to the alternate bars described
Platte-type macroform is comparable to the downstream- in Sect. 6.7
accretion element described here. The South Platte-type
(notably their low depositional dip) are not em- scour can lead to changes in channel orientation,
phasized, and they could be mistaken for bedform resulting in complex accretionary geometries, com-
reactivation surfaces. Bristow (1993) provided a monly with superimposed minor channels scoured
three-dimensional facies model for the bars of the out during stage changes. Miall (1994) documented a
Brahmaputra River, based on natural cutbank out- suite of LA and DA units in the Upper Cretaceous
crops along the modern channel, but the model only Castlegate Formation of Utah, which illustrate some
relates to the upper few meters of these very large bar of the resulting complexities. An example is illus-
forms (Fig. 6.31). trated in Fig. 6.32.
In the ancient record, detailed architectural stud- In some cross sectional configurations certain
ies rarely reveal simple "textbook" examples of types of lateral-acretion deposit could be confused
macroforms, such as those shown in the model dia- with downstream-accretion ele~ents, as noted in
grams in this section. Bar accretion and channel the next section.
Lateral-Accretion Deposits (Element LA) 155
Fig. 6.28. Theoretical depositional mo-

® del for a simple braided channel pattern
(above) in which bars and channels
show oblique to downstream migration,
and showing locations of cross sections
1-5 (below). Lines in plan and cross sec-
tions indicate configuration of first-to
third-order bounding surfaces. Arrows
indicate orientation of bedforms during
deposition of uppermost units in the
® cross sections. (Bridge 1993b)
CD
t t
~V.E.-2
o
o ~_ _ _ _ _ _~ V.E.-2
®
,/
6.7 Lateral-Accretion Deposits (Element LA) being driven inward, up onto the upper point-bar
surface, until they are stranded by a loss of depth and
Where the main flow of the channel is directed away shear strength (J ackson 1976b). Flow is then diverted
from the bank, as on the inside of a meander, surface around them to the lower part of the point bar, and
flow impinges on the outer bank, leading to a "setup" the mesoform becomes part of the point-bar deposit,
(raised water level as a result of pressure against the enclosing a swale against earlier bar sediments. On
bank), and active cutbank erosion (Fig. 6.33). The tight bends separation zones may be set up, with
flow turns downward, developing a helical overturn complex vortex flow causing the development of
pattern. The return flow, at depth, passes obliquely banks and benches on the point-bar surface (Nanson
up the bed of the inner bank. Because of the reduced 1980; Fig. 2.37).
shear stress associated with this current, significant Meanders evolve in several different ways, in-
sedimentation takes place, and the bank accretes cluding cross-channel and down-channel migration,
laterally at a high angle to the principal flow direc- rotation or expansion (Willis 1989; Fig. 6.34) . Sedi-
tion (Fig. 2.20). The helical flow pattern decays as the ment is added to the inner bank at a rate comparable
flow emerges from the bend, and is replaced by a to that at which it is removed by erosion along the
helical overturn in the opposite direction as the flow outer bend of the meander. The result is the develop-
impinges on the cutbank of the next bend down- ment, by lateral growth, of a compound, bank-at-
stream. Sediment removed from the cutbank is in- tached macro form, termed a point bar. A distinctive
corporated in the overall sediment load of the river. architectural element results, characterized by large-
Large slump blocks may accumulate in the deepest scale, gently dipping third-order bounding surfaces
part of the channel as a lag deposit, whereas material that correspond to the successive increments of
broken down into individual grains is incorporated lateral growth (Figs. 4.7, 4.9). These surfaces have
into the bed load and suspension load and is swept traditionally been termed epsilon cross-bedding,
downstream, much of it becoming deposited in following the classification of bedforms by Allen
bedforms and bars. (1963a). They usually show offlapped upper termi-
Depending on the grain size of the sediment load, nations, followed by fine-grained facies of the FF
and the meander geometry, transverse flow across element. Their lower terminations downlap onto the
the accreting point bar surface may take several channel floor. The height or thickness of the element
forms. Large or small gravel or sand bedforms mi- approximates the bankfull depth of the channel, and
grate subparallel to the strike of the accretion sur- can be used to estimate channel scale, as discussed in
face, with larger mesoforms ("transverse bars") Sect. lOA.
(A)
Potent;a! rising stage 0'05;00 , lallillg slage deposilion
o Potent;a! risiog stage deposilion , lalling stage Olosioo
_ POIeoliaJ ClOSS-l>ar chano" ditectioos at high stage
(8) DOWNSTREAM PART Of CURVED CHANNEL OR CONF LUENCE
falhng stage erosion
CURVED CHANNEL OR CONFLUENCE ENTRANCE
faJl~ stage deposit
rising stage deposit flood stage deposit
Fig. 6.29. Theoretical model of erosion and deposition flood stage geometry following cutbank erosion and bar
during flow stage in a simple braided channel. Cross sec- deposition, 4 subsequent low stage geometry. (Bridge
tions show 1 low stage geometry, 2 flood stage geometry, 3 1993b)
Fig. 6.30. a Typical succession oflithofacies within a DA macroform. Numbers 3 and 4 within arrows indicate loca-
unit in the Carboniferous Lee Formation, central Appala- tions and ranks of bounding surfaces. (Wizevich 1992b)
chian basin. b Model of the development of the DA
Fig. 6.31. Model for bar-top sedimentation in the which may be imperceptible in small outcrops. Significant
Brahmaputra River, based on observations of natural bank sedimentation takes place during falling stage, when the
exposures in the modern river. Upstream accretion occurs bar top is dissected by minor channels and large amounts
in the upper portion of the bar, lateral accretion in the of suspended sediments are deposited on bar margins.
middle portion, and downstream accretion in the lower (Bristow 1993a)
portion. Accretion surfaces all dip at very low angles,
O-Sm
3-7 m
t 1-1.Sm
(a) I~ t 0-1 m
Fig. 6.30
;,(,,:t(/l1;':;]f3'('!~~~';?{K;!'i;;}CN>'YY'J',,>n';i:" """ ; ~ ";""""" ~""

:.:. .. .
I [{8
I': ,'::1
dun • •
ripples
~ plane bed lafT)ination
1...:- -=:'" I ripple cross lamination
~ trough cr oSS ~traliflcatlon
~ Planar cross stratification
!Lo 1 2 3 • SOOm
E!!!!
~·::i~:.~l~
sill
intraformational conglomerate
Fig. 6.31
Crowley (1983) suggested that LA deposits of

high-sinuosity channels are dynamically compa-
rable to the DA deposits oflower sinuosity channels
and that both reflect the long-term behavior oflarge-
scale vortices affecting the entire turbulent bound-
ary layer (Fig. 6.27). Carson (1986) showed that in
many gravel-bed rivers much of the point-bar devel-
opment takes place as a result of flow expansion
where the flow enters the bend. LA elements may
also occur as midchannel bars in multiple-channel
rivers, such as the Brahmaputra (Bristow 1987; Fig.
6.36).
The internal geometry and lithofacies composi-
tion of LA elements is highly variable, and depends
on channel geometry and sediment load, but the
presence of lateral accretion sets is the common
theme (Fig. 6.35). The width averages two-thirds of
Fig.6.32. Use of orientation data to reconstruct bar archi- the channel width (Sect. 1004.1), at least in single-
tectures in an ancient braided stream deposit, Castlegate channel rivers, so that the dip of the lateral-accretion
Sandstone (Upper Cretaceous), Utah. Observations of ac- surface varies according to the width/depth ratio
cretion surfaces in the outcrop profile (B) are used to (Leeder 1973). Dips of up to about 25° have been
construct a map of the top surface of the element (A).
These data, combined with paleocurrent information, are recorded in some fine-grained point bars (e.g., Miall
combined in a reconstruction of bar geometries (C). In this 1979a), in rivers corresponding to model 7 of Miall
case, element 1A is interpreted as a DA unit, crosscut at its (1985; see Chap. 8). In wide channels, particularly
downstream end by a minor channel (Miall 1994). The where the sediment load is gravelly, the accretion
entire profIle from which this example was selected is surface is covered by bedforms and by minor bars
shown in Fig. 4.15
Fig. 6.33. Examples of river meanders, with point bars forming on the insides of the bends. Channel is about 20 m wide.
Milk River, Alberta, Canada. (Photo courtesy of D.G. Smith)
(a)
3¥=-g- ~
(b)
~
(f) (a-e)
~
(t)
Fig. 6.35. Examples of lateral accretion elements. No ver-

tical exaggeration. Fluvial-style model numbers of Miall
(1985) are indicated. a Conglomerate point bar (lithofacies
Gm), with chute channels (lithofacies Gt), model 4 (Ori
1979); b element composed of medium-grained sandstone,
with abundant internal planar-tabular cross-bedding
(lithofacies Sp), model 6 (Beutner et al. 1967); c fine- to
very-coarse sandstone and pebbly sandstone with cobble
to boulder conglomerae lag. Abundant internal cross-bed-
ding (lithofacies Sp, St, Sh, and Sl), modelS (Allen 1983a);
d small sandy point bar with abundant dune and ripple
Fig. 6.34. Models of meander migration. Arrows show cross-bedding (lithofacies St, Sr), model 6 (Puigdefabregas
migration paths. A Translation; B expansion; C rotation; D 1973); e point bar composed mainly of fine sandstone and
wavelength variation; E complex evolution. (Willis 1989) siltstone (lithofacies Sl) with minor medium- to coarse-
grained, cross-bedded sandstone (lithofacies St) at base,
model 7 (Nanson 1980); f giant point bar with thick, fme-
grained trough cross-bedded sandstone at base
and channels, and this obscures the simple geometry (lithofacies St) passing up into accretionary sets of alter-
of the LA element (e.g., Schwartz 1978). Lateral- nating fine sandstone and argillaceous siltstone showing
accretion surfaces may be difficult to identify in such evidence of tidal bundling (lithofacies Se), model 7
(Mossop and Flach 1983). (Diagram from Miall1985)
deposits.
Lithofacies assemblages within lateral-accretion
deposits vary markedly, depending on the caliber of
the sediment load, and on discharge variability. LA the LA element is subordinate to element GB. Ex-
elements can be classified into four groups, accord- amples are illustrated in Figs. 6.16, 6.35a, and 6.37.
ing to grain size and facies. Earlier classifications by Modern examples are e.g.: Bluck (1971), Lewin
Jackson (1981) and Smith (1987) have been incorpo- (1976), Gustavson (1978), Forbes (1983), Hooke
rated into the present subdivision. Gradations be- 1986; ancient examples: Ori (1979, 1982), Arche
tween the four groups are to be expected even within (1983), Ramos and Sopena (1983), Cavazza (1989).
a single fluvial deposit, as a result of variations in the
energy of discharge events, meander evolution, etc. Gravel-Sand Rivers (ModelS). Deposits consisting of sand
(some of these variations are discussed in Chap. 8), or pebbly sand contain a wide variety of lithofacies
plus longer-term changes in tectonic and climatic reflecting vigorous bedform and bar progradation
control. Examples of descriptions of these groups are and chute development. Bedding within this type of
as follows, with the fluvial-style model numbers of LA element is complex and may obscure the under-
Miall (1985) indicated: lying laterally accreted geometry. Indicated flow di-
rections in these cross-bedded deposits are parallel
Gravel Rivers (ModeI4). Gravel-dominated deposits are or subparallel to the strike of the accretion surfaces.
relatively rare, and in most gravelly fluvial deposits Modern examples are: Bernard and Major (1963),
Architectural Elements Formed Within Channels
Htlg h l in Ie-el
> 2e IZ:1l McGowen and Garner (1970), Bridge and Jarvis

24 -2e Eilli (1976), Jackson (1976b), Levey (1978), Campbell and
Hendry (1987); ancient examples: Nijman and
20-24
Puigdefabregas (1978), Allen (1983a), Campbell and
<20 .
Hendry (1987), Diaz-Molina (1993).
rzZl
Sand Rivers (Model 6). Figure 6.38 illustrates a modern
example and Figs. 4.7, 4.9, 4.10, 6.35b,d and 6.39
are ancient examples. For modern rivers, see:
Sundborg (1956), Davies (1966); for ancient ex-
amples: Beutner et al. (1967), Puigdefabregas (1973),
Shelton and Noble (1974), Nami and Leeder (1978),
Puigdefabregas and Van Vliet (1978), Ethridge et al.
(1981), Hobday et al. (1981), Plint (1983), Stear
(1983), Mossop and Flach (1983), Bridge and Diemer
(1983), Link (1984), Diemer and Belt (1991), Munoz
et al. (1992), Willis (1993a), Alexander and
Gawthorpe (1993).
Fig. 6.36. Midchannel macro form, Brahmaputra River. Sand-Silt-Mud Rivers (Modell). The simplest LA elements
Arrow at right shows flow direction. The bar is elongated
downstream and is enlarging by accretion on the west and
are those composed of fine sand, silt, and mud. Sec-
south sides. Surface traces of the growth are shown by the ondary bedforms on the accretionary surfaces are
dashed lines, and active scroll bars (SB) are also indicated. rare and small in scale, and the accretion surfaces are
This macroform is in part an LA unit (in its central part and steeply dipping and more readily identified in out-
on its west side), and in part a DA unit (along the east flank crop. Benches may be generated by horizontal sepa-
and at the downstream end). (Bristow 1987)
Fig. 6.37. GB and LA elements in Pleistocene gravels. Bounding surfaces are also indicated by arrows with rank numbers
Near Mito, Japan. Note person at lower right for scale
Fig. 6.38. View across a modern point-bar surface, look- crest lines of these dunes are oriented down the dip of the
ing upstream. A field of 2-D dunes that was migrating point-bar accretion surface. (Photo courtesy of M.O.
toward the viewer has been abandoned by falling water. Hayes and R. Levey)
The point bar is accreting toward the left. Note that the
ration eddies at sharp bends (Figs. 2.37, 6.40). For there form significant deposits of fine sand, silt, and
modern examples, see: Nanson (1980), Jackson mud in "concave bench complexes" (Figs. 2.37, 6.40).
(1981); for ancient examples: Miall (1979a), Stewart In some muddy rivers, it has been found that accre-
(1983), D.G. Smith (1987), Wood (1989). tionary development of mud, silt, and fine sand may
LA deposits do not retain a constant geometry or take place on both sides of the channel. The deposits
composition around any given meander bend. As a on the inside of the meander bend are referred to as
result, the classic fining-upward profile (Allen 1964, an "inner accretionary bank." This element type is
1965a,b) may not be present. In gravelly rivers, Bluck common in some of the muddy, ephemeral rivers of
(1971), Lewin (1976), and Bridge and Jarvis (1976) interior Australia (Taylor and Woodyer 1978; M.R.
showed that the coarsest part of the point bar is Gibling, pers. comm. 1995). A detailed study of the
located at the upstream end of the bar (bar head) ridge-and-swale topography that develops in the up-
and may migrate downstream over sandy bar tail permost levels of point bars was provided by Gibling
deposits. Wood (1989) interpreted facies variations and Rust (1993). They described variations in some
in some ancient point-bar deposits on this basis. exhumed Carboniferous examples, and related the
Jackson (1976b) found that in the Wabash River variations to slight differences in the style of scroll-
(sand and pebbly sand) the helical flow patterns bar accretion.
responsible for the fining-upward point bar profile Variations in the style of meander 'evolution (Fig.
tend to develop only in the downstream part of a 6.34) lead to subtle but important differences in the
meander bend (Fig. 6.41). Carson (1986) made a internal architecture of the resulting LA deposit.
similar observation on gravel-bed rivers in New These variations were described by Willis (1989),
Zealand. Nanson and Page (1983) showed that who used a computer model to simulate various
within tight meanders flow separation may occur at meander configurations. Figure 6.42 illustrates his
the downstream end of a point bar. Eddy currents cross sections through a meander that evolves by
Fig. 6.39. View of a point bar in the Chinle Formation to the right. The accretion surfaces are covered in trough
(Triassic), central Arizona. View is downstream, along the cross-bed sets (lithofacies St), the axes of which are ori-
strike of the bar, the accretion surfaces of which dip gently ented away from the viewer
down-valley translation. The channel orientation at straight deltaic distributary channels, such as the
each stage of point-bar growth (arrows in each cross modern Mahakam delta, Indonesia (Verdier et al.
section) should be indicated by paleocurrent mea- 1980), and the Carboniferous fluvial-deltaic deposits
surements. These cross sections are therefore useful of northern England (McCabe 1975, 1977; Okolo
as templates for the interpretation of randomly ori- 1983). It has also been observed in fluvial settings,
ented outcrops through point-bar deposits. Note such as in straight reaches of modern anastomosed
that obliquely downstream-oriented cross sections rivers (D.G. Smith 1983), and the Carboniferous
(Fig. 6.42C) reveal an architecture similar to that sandstone deposits of the Appalachian basin (ele-
of downstream-accretion deposits, indicating the ment type DA2 of Wizevich 1992b). The modern
need for caution in the interpretation of individual, gravel bars described by Lewin (1976) are compa-
isolated outcrops. The application of these ideas rable. The bars consist of macro forms developed by
to actual ancient examples is discussed by Willis accretion oblique to the channel trend (Fig. 6.43).
(1993a). Detailed studies of some ancient lateral- They may consist of simple, large-scale cross-bed
accretion deposit, revealing complexities in internal sets, in which the cross-bed foresets are the accretion
accretionary geometry, are described by Diaz- surfaces (McCabe 1975, 1977). In some cases minor,
Molina (1993) and Miall (1994). bedforms develop by secondary flow oblique to the
As noted in Sect. 6.6, Crowley (1983) has descri- slip face, as in the DA2 elements ofWizevich (1992b).
bed a gradation between the midchannel macro- McCabe (1977) described large-scale internal reacti-
forms of low-sinuosity, braided channels (element vation surfaces (third-order, in the classification of
DA), and the laterally accreted bars of high-sinuosity this book) that indicate major stage changes during
rivers (element LA). An intermediate form is the the generation of the bar. It is a matter of conve-
alternate bar, which develops on the insides of the nience how alternate bar elements are classified and
meandering thalweg in rivers of low to moderate coded. McCabe (1977) treated them as large-scale
sinuosity (Fig. 6.27). This element type is common in bedforms. Okolo (1983) described them as channel
Hollows (Element HO) 163
Fig. 6.40. LA element composed of very fine-grained sand tion eddies at the downstream ends of tight bends (Nanson
and silt. Note the steep accretionary dip, and the flat and Page 1983; see Chap. 8). Eureka Sound Formation
benches formed by the development of horizontal separa- (Tertiary), Ellesmere Island, Arctic Canada. (Mia1l1979a)
fills. Wizevich (1992b) designated them as DA2 - a 0.4-2.5 m thick, and rest on flat to slightly scoured
distinctive type of downstream-accretion element. erosion surfaces. They may be capped, gradation-
ally, by Sp, St, or Sr, indicating waning flow condi-
tions at the end of a flood event. Individual sheets
may be traced laterally for more than 100 m. At the
6.8 Laminated Sand Sheets (Element LS)
edges they thin and split into thinner units domi-
nated by finer-grained sands and silts of lithofacies
Sheets of laminated sand (lithofacies Sh, Sl) with Sr. These beds probably represent the margins of
minor Sp, St, or Sr commonly dominate in some individual flood sheets. Channel cutbanks are rare to
ancient rock sequences, and have been interpreted absent. Stacked sequences may reach tens of meters
as the product of flash floods depositing sand under in thickness.
upper flow-regime plane bed conditions (MiallI977,
1984b; Tunbridge 1981, 1984; Sneh 1983). The flood
deposits of Bijou Creek, Colorado (McKee et al.
6.9 Hollows (Element HO)
1967), are invariably quoted as a close modern ana-
log; they provided the basis for the Bijou Creek
model of Miall (1977). Ephemeral streams of the Research by Cowan (1991) in the Westwater Canyon
Lake Eyre Basin also contain local accumulations of Member of the Morrison Formation has indicated
this assemblage (Williams 1971). that some architectural units previously classified as
The characteristic architecture of this element has channels require a quite different interpretation,
been best described by Tunbridge (1981), Sneh and justify the erection of a new type of architectural
(1983), and Stear (1985). Individual sand sheets are element. Although the features to be described here
FULLY DEVELOPED TRANSITIONAL DEPOSITIONAL FACIES

.j
DEPOSITIONAL FACIES
----'HIGH' FLOW
o::~
'" ·-·-·-'HIGH' FLOW
+ '+·T·+·LW FLDW - - - 'HIGH' FLOW
CURRENT
CURRENT + + +- -LOW· FLOW
DISPERSION
DUNE HEIGHT (CM)
0255015
10
8-
9
1-
e
--MLWP
-MLWP
~
... 4
z
Q
~
;:! 3
'"
.j
'"
1- .i 2
.: . '\
0-
-3 0 3 6
MEAN SIZE (PHI)
o I--r"-'r-..,.....<D
- - 'HIGH' FLOW -3 0 3 6 o 50 100 150
- - 'LOW' FLOW MEAN SIZE (PHil MEAN VEL. (CMJSEC)
- - 'HIGH' FLOW
t+++i 'LOW' FLOW
Fig,6.41. Point bar successions in a gravel-sand river, the applies to upper part of meander bend where helical flow is
Wabash River (modelS of Miall198S). Numbers in circles not well developed. Fully developed model applies to lower
at right are subfacies: 1 channel lag, 2 lower point bar, 3 part of bend, where helical flow is established. (J ackson
upper point bar, 4 levee, 5 overbank. Transitional model 1976b)
8 A s'
200
0.2'5" ...... ... ..:. ..... .' ........... ". ... .. ........ .... ..... ...... ....... 0:50
0.50
Fig.6.42. Cross sections of simulated point-bar deposits lines, with values in mm. Arrows show channel orientation,
formed by down-valley meander translation. Grain-size relative to cross section (up is into the cross-section
contours, in one-phi increments, are shown by dotted plane). (Willis 1989)
__ Medium scale cross - beds

(dunes)
.....-J
Giani foresels ....-
(alternate bor)
,;.----
Undula lory beds
(spurs)
Fig. 6.43. Model for the development of a straight distributary channel with alternate bars, the latter comprising large-
scale, simple cross-bed sets up to 40 m thick. (McCabe 1975, 1977)
Fig. 6.44. Architectural diagram for the Westwater Canyon Member of the Morrison Formation, northern New Mexico,
showing the predominance of element HO (Cowan 1991). A typical outcrop example ofHO is illustrated in Fig. 4.12
are common in the Westwater Canyon Member, on- fourth-order surfaces (Figs. 4.12, 6.44). The dip on
going research is indicating that they are present in these basal surfaces reaches 26°, but is normally less.
other gravel-braided and sand-braided systems. In They are not cylindrical in shape, as are channels,
the Westwater Canyon Member element HO re- but are scoop-shaped. The fill typically consists of
sembles the "smaller channels" of Campbell (1976). lithofacies Sh and Sl, commonly with the dip of the Sl
They are up to about 20 m deep and 250 m wide, and bedding planes parallel to the lower bounding
are bounded at the base by curved, concave-up surface, indicating that deposition took place on a
KEY
bed elevation contours (cm)
avalanche face edges
• • line of section
'" base of scour (0 cm elevation)
10 20 motr ..
I t
c d
scale for
cross-sections
501 ~~~tical
o~ggeratjon )(4
a LOa
.
~
g a ~~~~n~1 :V~~QnChe tace maximum scour b
!~~~l~
o *
Fig. 6.45. Confluence of two channels, River Ure, Yorkshire, England. Note the scour, up to 2.5 m deep and 40 m long
with avalanche faces at the mouth of each channel. (Best 1987)
sloping surface. The hollows fill by lateral or vertical more, the fact that the hollows are most likely
accretion or combinations of both. They may be trough-shaped in three dimensions suggests an ori-
symmetrical or asymmetrical in cross section, and gin other than by elongated channelization. They
the orientation of dipping Sl sets may be perpendicu- may be ancient analogs of large, elongate flute-like
lar or oblique to the axis of the hollow. This axis, scours formed at channel bases, as documented by
itself, is commonly oriented at a high angle to re- Coleman (1969) from the Brahmaputra River. The
gional paleoflow. Siegenthaler and Huggenberger physical processes involved in the formation of these
(1993) reported very similar scour-fill structures in flute-like scours are not yet known. Similar but
Pleistocene gravel deposits in Switzerland. The much smaller-scale structures have been docu-
scours are up to 100 m wide and 6 m deep. mented from an ephemeral stream deposit by Olsen
These hollows may be interpreted in several ways. (1989), and were interpreted to have been formed by
They are not channels. Channel cutbanks in sand are the erosional action of spiral vortices during a sheet
rarely as steep as observed in these hollows. Further- flooding event.
Cowan (1991) proposed that these hollows form- obliquely, or vertically by an avalanche deposit in a
ed by a process of deep scouring at points of channel short period of time during channel switching or a
convergence. Scour depths of up to six times the flood event. The scours are, therefore, envisaged to
mean channel depth have been documented in rivers form as clusters or isolated features, depending on
and laboratory flumes by Mosley (1976), Mosley and the density and spacing of the channels within a
Schumm (1976), Best and Brayshaw (1985), Best braided channel belt.
(1987, 1988), Best et al. (1989), and Salter (1993). Cant (1976, p. 125) interpreted deep scours,
Bristow et al. (1993) proposed a facies model for the reaching nearly three times the mean braid-channel
scour and fill processes at channel confluences, and depth, to have formed upstream of a large emergent
Ashmore (1993) modeled their development in bar. This process is analogous to scouring on the
gravel-bed rivers. These studies suggest that a upstream margin of an obstacle clast in a flow (cf.
mechanism of deep scouring may be a significant Best and Brayshaw 1985), with the emergent bar
process at channel junctions (Fig. 6.45) and below acting as an obstacle within the channel.
mid-channel macro forms in braided streams, where Because these scours result in the deposition of
stream junctions and bar forms abound (Best 1987, sediment below mean channel depth, they have a
p. 34; Best et al. 1989). Salter (1993) documented high preservation potential. In the deposits studied
scour where flow is deflected against banks or is- by Cowan (1991), they are common, whereas other
lands. Best (1987) showed that avalanche faces can types of macro forms, such as DA and LA units, are
develop on the upstream end of these scours (Fig. preserved only in fragmentary form.
6.45). This allows the scours to be filled laterally,
Chapter 7
Architectural Elements of the Overbank Environment
7.1 Introduction 1. The relatively coarse deposits formed by over-

bank flow of channel bed load, constituting levee,
crevasse-channel and crevasse-splay deposits.
In contrast to the considerable amount of research
2. Fine-grained deposits formed in low-energy en-
devoted to the study of channel-fill sediments,
vironments, including ephemeral sheet floods
overban~ deposits have received much less atten-
and more permanent floodplain ponds.
tion, as has been noted by several researchers (e.g.,
3. Biochemical sediments formed by pedogenesis,
Farrell 1987; Kraus 1987; Kraus and Bown 1988). The
evaporation, or organic activity.
main exception to this is a recent focus on paleosols,
stimulated by their use as lithostratigraphic and time Geomorphologists recognize a relatively simple
markers (Sects. 7.4.2, 9.5.1, 10.4.3) and as climatic subdivision of floodplain landforms and processes.
indicators (Sect. 12.9.2). In the original architec- For example, Brierley (1991) erected the following
tural-element classification (Miall 1985), overbank classification:
deposits were assigned to a single element, desig- Top stratum (floodplain deposits), consisting of:
nated OF. This is clearly inadequate as many re-
searchers have pointed out. In fact, deposits formed (A) sand wedge (levee)
outside the main fluvial channels can be classified (B) proximal sand sheet (marginal overbank sheet)
into three broad classes. (C) distal sand sheet (distal overbank sheet).
His classification of in-channel deposits is given in

Sect. 4.6.
Nanson and Croke (1992) listed six processes of
floodplain sedimentation. Four of these, lateral
point-bar accretion, braid-channel accretion, coun-
terpoint accretion, and oblique accretion refer to
channel aggradation, although the latter two pro-
cesses may deposit fine-grained sediment that could
be confused with floodplain deposits in small out-
crops. These processes and their deposits are de-
scribed in Chap. 6. The other processes of floodplain
construction are overbank vertical accretion and
abandoned-channel accretion.
Figure 7.1 illustrates the overbank elements de-
veloped adjacent to a reach of the Mississippi River,
and Fig. 7.2 is a cross section showing the distribu-
tion and architecture of a coal-bearing fluvial unit in
which overbank deposits constitute about 55% of the
• LEVEE D BACKSWAMP 0 UPPER POINT BAR
total rock volume. Figures 7.3 and 7.4 illustrate an
element classification and depositional model devel-
~ CREVASSE SPLAY D ABANOONED CHANNeL oped for the Miocene Lower Freshwater Molasse of
Switzerland. Another example of ~n element classifi-
Fig. 7.1. The distribution of overbank subenvironments cation of a fluvial unit that includes significant thick-
adjacent to a reach of the Mississippi River, False River nesses of overbank deposits is given in Table 4.4.
cutoff, Louisiana. (Farrell 1987) Standardized element classification and coding for
170 Architectural Elements of the Overbank Environment
SOUTHWEST
, NORTHEAST
II I I I I t I II I I , I , ,
l00FTT3DM CACH?
I~KM
_ COAL AND CARBONACEOUS SHALE ~ LIMESTONE
o 2MI
~ SANDSTONE (CHANNEL) MOLLUSK FOSSILS

o SANDSTONE (CREVASSE) "V" CLINKER
VA.
r-·::=I SILTSTONE AND SHALE STRATIGRAPHIC SECTION
Fig. 7.2. Stratigraphic cross section showing the architec- deposits. The latter include overbank fines, crevasse sand-
ture of a fluvial depositional system made up of isolated stones, and coal. Paleocene, Wyoming-Montana. (Flores
and multistory channel sandstones and thick overbank 1981)
CS
LV
CH,LA
OJm[IImJIrr:rrT~~:J FF
Fig. 7.3. Stratigraphic model of the ar-

chitectural elements within the Lower
CH,LA Freshwater Molasse (Miocene) of Swit-
zerland, showing the thickness of the
major sandstone units. Element codes (in
capital letters) are those given in Tables
4.3 and 7.l. Oblique lines are faults.
approx. 10 m (Modified from Platt and Keller 1992)
Introduction 17l
errace
meander belt
sandstt)ne
overbank
and palaeoso ls
Fig. 7.4. Summary facies model for the Lower Freshwater Molasse. (Platt and Keller 1992)
Table 7.1. Clastic architectural elements of the overbank environment

Element Symbol Lithology Geometry Interpretation
Levee LV FI Wedge up to 10 m thick, 3 km wide Overbank flooding
Crevasse CR St, Sr, Ss Ribbon up to a few hundred m wide, Break in main channel margin
channel 5 m deep, 10 km long
Crevasse CS St, Sr, Lens up to 10 x 10 km across, Delta-like progradation from
splay Fl 2-6 m thick crevasse channel into floodplain
Floodplain FF Fsm, FI, Sheet, may be many km in lateral Deposits of overbank sheet flow,
fines Fm,Fr dimensions, lOs of m thick floodplain ponds and swamps
Abandoned CH(FF) Fsm, FI, Ribbon comparable in scale to Product of chute or neck cutoff
channel Fm,Fr active channel
clastic deposits are suggested in Table 7.1 and are in the Canadian Arctic Islands (Figs. 8.44, 8.45),
discussed below. This should avoid any confusion maps of two large splays in the modern Magdalena
between different authors For example, Miall (1985) River, Colombia (Fig. 8.52), and cross sections
used the code LS for laminated sand sheets formed in through two splay deposits in the Cumberland
ephemeral channels, whereas Platt and Keller (1992) Marshes, Saskatchewan (Fig. 8.53). Lithofacies are
employed this code for levee sandstones. described and illustrated in Chap. 5 (Figs. 5.30-5.34).
Overbank deposits are particularly important Gravel-bed and sand-bed braided rivers have
components of the deposits of rivers bearing a channel systems that tend to occupy most of the river
fine-grained sediment load, including fine-grained floodplain, so that overbank environments are small
meandering rivers (Sect. 8.2.9) and anastomosed in area and their deposits constitute a minor compo-
rivers (Sect. 8.2.10). These are the fluvial styles nent of the fluvial assemblage (e.g., Reinfelds and
that are characterized by broad floodplains. Several Nanson 1993). A significant exception to this gener-
maps and photographs of rivers and their deposits alization was described by Bentham et al. (1993). The
that fall into these two classes of fluvial style are Eocene Escanilla Formation of the Spanish Pyrenees
provided in Chap. 8. These include a profile and consists of more than 40% by volume of fine-grained
interpretation of an outcrop that exposes sheet-like bioturbated, mottled, and pedogenically modified
bodies of crevasse-splay sandstone in a Tertiary unit fine siltstones and silty mudstones. Bentham et al.
(1993) attribute the preservation of such a large vol- Switzerland). In addition, there is a considerable
ume of overbank material to rapid subsidence and body of literature dealing with coal deposits that is
sedimentation and frequent overbank flooding. beyond the scope of this book (e.g., see Rahmani and
Flores 1984; DiesselI992).
7.2 Levee and Crevasse Deposits

7.2.1 Levee Deposits (Element LV)
The lithofacies and architecture of these deposits Adjacent to the lower Mississippi River natural
have been described by several workers studying levees are up to 3 km wide and 9 m high (Farrell 1987;
modern fluvial environments, including Coleman Fig. 7.1). Similar widths were encountered in the
(1969: Brahmaputra River), D.G. Smith (1986: Magdalena Valley by D.G. Smith (1986). Along the
Magdalena River), Farrell (1987: Mississippi River), Brahmaputra River of Bangladesh, levees extend 700
N.D. Smith et al. (1989: Cumberland Marshes, m from the channel margins and are built up of
Saskatchewan) and Brierley (1991: Squamish River). overlappping lenses up to 1 km in width, parallel to
Good descriptions of examples in the ancient the channel margin (Coleman 1969). Adjacent to
record have been given by Gersib and McCabe (1981: smaller rivers, such as reaches of the Saskatchewan
Carboniferous, Nova Scotia), Allen and Williams in the Cumberland Marshes, levees are less than
(1982: Devonian, southwest Wales), Flores and 100 m wide (N.D. Smith et al. 1989). Depositional
Hanley (1984: Paleocene, Wyoming), Rust et al. dips away from the channel margins of up to 10 0
(1984: Carboniferous, Nova Scotia), Bridge (1984: were observed in some Eocene levee deposits by
Devonian, Ireland), Fielding (1986: Carboniferous, Bown and Kraus (1987), but dips of 2-4 0 are more
England), Eberth and Miall (1991: late Paleozoic, common in these beds.
New Mexico), and Platt and Keller (1992: Miocene,
PROXIMAL DISTAL PROXIMAL DISTAL

SPLAY ~ SPLAY LEVEE LEVEE
.~ 0
-I ~.E:
(J <I)'"
~
6
UNITS
I:!:::J SILTY SAND (SURFICIAL SpLAY)
7 ~'F;·)(i SILTY SAND (SURFICIAL LEVEE)

_ RO O TED CLAY (BACKSWAMP)
c:::J SANDY SILT Fig. 7.5. Stratigraphic successions through

8
~ BLUE CLAY overbank deposits, modern floodplain of the
Mississippi River. (Farrell 1987)
Levee and Crevasse Deposits 173
Levees typically consist of rhythmically bedded sippi River are up to 5 m deep (Farrell 1987). Fielding
units of silty, ripple-laminated sand a few decimeters (1986) recorded crevasse channel fills up to 6 m
thick (lithofacies FI; Fig. 7.5). Bioturbation or root thick. Even in smaller fluvial systems such as the
development commonly obscures lamination. Cumberland Marshes of Saskatchewan crevasse
Abundant vertebrate remains may be present (Bown channels reach 4.5 m depth (N.D. Smith et al. 1989).
and Kraus 1987). Each sedimentary rhythm repre- Rhee et al. (1993) reported small channels less than 1
sents a flood event. The levee element may reach 10 m in depth in a coarse-grained terminal fan deposit.
m in thickness, and constitutes a tapering wedge of Crevasse channels range from a few tens of meters to
deposit, thinning and fining away from the channel a few hundred meters in width. These channels in-
margin. Paleocurrent measurements made on con- cise levee and other backswamp deposits, and typi-
tained ripple cross-lamination indicate flow perpen- cally form small, delta-like distributary systems,
dicular to the channel or obliquely downstream, becoming shallower away from the main channel
away from the channel margin (e.g., Coleman 1969). (Fig. 8.52). In the case of the Magdalena River, Co-
lombia, the network of crevasse channels extends up
to 10 km away from the channel margin (Fig. 8.52),
7.2.2 Crevasse-Channel Deposits (Element CR) but this seems to be exceptional. Those flanking the
Mississippi River are up to 5 km long (Fig. 7.1), and
The scale of crevasse channels depends on the scale those flanking the Brahmaputra River rarely exceed
of the river. Those associated with the lower Missis- 500 m in length (Coleman 1969). The bounding
Fig. 7.6. Overbank deposits of the Moenkopi Formation environment. Note the channel-shaped unit near the base,
(Triassic), Arizona. Resistant units are distal crevasse filled with fine-grained deposits. This may be the fill of an
splay deposits. light-colored units are caliche. Dark, reces- abandoned channel: element FF(CH), or a distal crevasse
sive units are mudstones and siltstones of the overbank channel (element CR). Outcrop is about 10 m high
Fig. 7.7. Floodplain deposits of the Chinle Formation interpreted as crevasse splays (element CS). Massive, ir-
(Triassic), near Ghost Ranch, New Mexico. Fine-grained regular-weathering unit at top is the Jurassic Entrada For-
deposits are floodplain fines (element FF). Sandstones are mation, an eolian unit. Outcrop is about 10 m high
surfaces of crevasse channels are classified as fourth- Crevasse channels are particularly important in
order surfaces (Table 4.2). anastomosing systems, because it is by the process of
Crevasse channel-fills are ribbon-like bodies typi- crevassing that the river compensates for the aggra-
cally consisting mainly of fine- to medium-grained dation or plugging of individual channels. Flow from
sandstone with trough cross-bedding and ripple crevasse channels may eventually rejoin the main
cross-lamination (lithofacies St, Sr). Internal scours channel downstream, or be diverted into the course
(lithofacies Ss) are common. The coarsest units are of a paleochannel, and may, in this way bring about a
equivalent in grain size to the deposits of the main significant diversion of flow (D.G. Smith 1983; N.D.
channel, but in average finer-grained facies are to be Smith et al. 1989; Richards et al. 1993). Such avulsive
expected, as these channels represent a "stripping" processes are discussed further in Sect. 10.3.
of the top of the flow in the main channel as a result
of overbank flow and incision.
Coarse lags containing fossil remains, such as 7.2.3 Crevasse-Splay Deposits (Element CS)
abraded bones, are a common component of cre-
vasse channels (e.g., Bown and Kraus 1987). Eberth Crevasse splays are the delta-like deposits that form
and Miall (1991) described a type of crevasse channel adjacent to the margins of main channels (Figs. 7.1,
filled with fine-grained deposits which they termed 7.2,7.4,8.52). In fact, they may be virtually indistin-
"V-shaped mixed-fill units." These contained thin- guishable from lacustrine deltas. Typical examples
bedded, fine-grained facies (lithofacies FI), com- of crevasse splays in outcrop are illustrated in Figs.
monly with a pebbly-sandstone lag at the base, and 7.6 to 7.9 and 8.44. The sediment is introduced into
abundant disarticulated vertebrate remains. A simi- the backswamp environment by crevasse channels,
lar deposit is shown in Fig. 7.6. and is deposited as a result of flow expansion and
Levee and Crevasse Deposits 175
Fig. 7.8. Overbank deposits of a fluvial delta plain, show- internal, gently dipping accretion surfaces. Carboniferous,
ing laterally extensive, fine-grained deposits (several units Kentucky
of element FF) and two crevasse-splay lenses containing
loss of flow power as discharge leaves the confines of cross-lamination are common. Interbedded laminae
the channels and spreads out as sheet floods. These of siltstone and mudstone constituting lithofacies Fl
deposits are particularly important components of are also common. The assemblages are characterized
the anastomosed fluvial environment, where the for- by thin bedding and abundant surfaces of non-
mation of crevasse channels and splays forms an deposition and small-scale erosion (third-order sur-
intermediate step in the shifting of main channels faces), reflecting the origin of the splays by periodic
into new positions on the floodplain (N.D. Smith et or irregular sheet flooding. Grain size decreases
al. 1989; see Sects. 8.2.10,10.3.3). away from the main channel toward the fringes of
Crevasse splays form lens-shaped bodies up to 10 the splay. Internally, splay deposits may exhibit low-
km long and 5 km wide. They are typically 2-6 m angle accretion surfaces recording growth by lateral
thick. They are cut by their feeder crevasse channels, progradation (Figs. 6.21, 7.8). The same process
and interfinger at their margins with fine-grained commonly leads to upward coarsening through the
floodplain deposits. They overlap, and may pass im- splay deposit (Fig. 7.10). The top of the splay may
perceptibly into levees along the flanks of the main show upward fining as abandonment takes place.
channels. The bounding surfaces of splay deposits This upper part of the succession may show a15un-
are classified as fourth-order surfaces (Table 4.2). dant bioturbation.
Bown and Kraus (1987) reported cumulative levee- N.D. Smith et al. (1989) examined the evolution
crevasse splay successions 15-20 m thick and ex- of crevasse channels and splays in the Cumberland
tending for up to 10 km away from channel margins. Marshes anastomosed fluvial system of Saskat-
The deposits of crevasse splays typically consist of chewan. They subdivided splays into three classes
fine- to medium-grained sandstone with abundant (Fig. 7.11), and demonstrated an evolutionary devel-
hydrodynamic sedimentary structures, plant roots, opment from the first to the third class, as channels
and bioturbation. Trough cross-bedding and ripple lengthen and deepen, and flow becomes concen-
Fig.7.9. A crevasse-splay sandstone overlying a coal seam. Carboniferous, Kentucky
...Ig~~s trated in a few of the channels. The third stage is that

u::!:cn >u.
0 of a few larger channels isolated in floodplain fines,
BEGINNING OF and is analogous to the channel pattern of the anas-
ABANDONMENT tomosed fluvial style.
FINING UPWARD-
WORMY FABRIC
7.3 Fine-Grained Clastic Deposits
a:
~
FULLY DEVELOPED
w
~
--=-<- SPLAY
c(.cc:,c;.
~
W ---_=:=c::: This category of deposits includes sheet-like units
III:
THICK CROSS
::E cc:.,cc:c. formed by settling from sheet floods, deposits
L.AMINATED UNIT
~~'ic- formed in floodplain ponds, and the fill of aban-
~
doned channels. Stratification is tabular, and on
INCIPIENT floodplain surfaces individual units extend for hun-
CREVASSING
COARSENING UPWARD
dreds of meters, or even for distances of kilometers
CROSS L.AMINATED (e.g., Willis and Behrensmeyer 1994). In some fluvial
1 settings, particularly anastomosed rivers, and in any
~
floodplains distant from the meander belt, these de-
ROOTED CL.AY
posits may comprise the bulk of the total fluvial
depositional thickness.
Fig. 7.10. Generalized stratigraphic section through a cre-
Lithologically these deposits are rather monoto-
vasse-splay deposit, based on coring of the splays of the nous, although important variations in depositional
lower Mississippi River. (Farrell 1987) processes lead to subtle variations in color, texture,
Fine-Grained Clastic Deposits 177
Fig. 7.11. The three types of crevasse splay in the lowing an avulsive event, and progressively evolve into
Cumberland Marshes anastomosed fluvial system, type 3. (N.D. Smith et al. 1989)
Saskatchewan. Splays initially form type-1 systems, fol-
and accessory components. For example, in her extent (e.g., Willis and Behrensmeyer 1994). There
work on the Mississippi River floodplain, Farrell may be considerable vertical lithologic variability,
(1987) distinguished between blue clays deposited in reflecting the fact that the depositional surface is flat
poorly drained swamps, and rooted clays deposited and readily susceptible to small changes in deposi-
from suspension following overbank flooding tional processes. Sedimentation may take place in
events, and subsequently colonized by vegetation. separate increments representing individual flood
This second facies is interpreted as the deposit of a events, or by continual slow settling of fine-grained
well-drained swamp, possibly a levee. sediment from suspension in permanent swamps or
In general, the presence of lamination indicates ponds (Sect. 10.3.1). Where seasonal or longer-term
deposition from suspension or from very low-energy drying-out of the floodplain takes place, desiccation
underflows. Such stratification is commonly de- and pedogenic processes may be important, and this
stroyed by the presence of organisms, the activity of element may be interbedded with paleosols.
which leads to bioturbation, producing trace fossils A typical profile through floodplain deposits is
and, ultimately, a churned and mottled texture in illustrated in Fig. 7.12. Architectural diagrams show-
which stratification may be rare or absent altogether. ing the scale of typical floodplain deposits are shown
Load structures may represent footprints of land in Fig. 7.13. Typical outcrops of this element are seen
vertebrates. Pedogenic processes and colonization in Figs. 7.6-7.8.
by plants also disturb lamination. The common
occurrence of invertebrate fossils suggests that the
deposits did not dry out during deposition, and may 7.3.2 Abandoned Channel Fills (Element FF(CH))
indicate the presence of permanently saturated
swamps or ponds. Variously termed oxbows, bayous (Mississippi), or
In areas distal to channel margins, floodplain billabongs (Australia), abandoned channels are
fines are commonly intercalated with distal splay common components of many fluvial styles, particu-
deposits (element CS). Good examples of this are larly the finer-grained systems, including sandy
illustrated in Figs. 7.6 to 7.8. It is a matter of conve- meandering, fine-grained meandering, and anasto-
nience whether sand units a few decimeters or less in mosed systems. An example in the Mississipi valley
thickness are given a separate element classification, is illustrated in Fig. 7.1. A cross section through an
or are grouped with the floodplain fines. ancient example is illustrated in Fig. 8.33. An out-
crop photograph of an ancient example on a delta
plain is illustrated in Fig. 6.9. A fluvial example is
7.3.1 Floodplain Fines (Element FF) shown in Fig. 7.14.
Commonly, these channels remain as lakes or
This element consists of sheet-like units many hun- ponds for a considerable length of time. The mouth
dreds of meters or even several kilometers in lateral of the channel, at the cutoff, may gradually become
... SAND .. i~ by loosely consolidated Quaternary glacial deposits.

J: I' 0 ~+~:
... .
Tributaries rising within the Amazon lowlands carry
0'" ..... ~f4ilJ° (Jo:- FACIES ".
0 much smaller sediment and water loads, and are
dammed at their junctions with the main tributaries
to form lakes. The sediment entering these lakes
1
from the lowland tributaries forms small deltas, and
ROOTED CLA V UNIT the remainder of the lake may be expected to slowly
2 fill with mud. In the second case, Kuenzi et al. (1979)
provided an interesting study of fluvial-deltaic sedi-
(91l KE Y
mentation on the Guatemalan coastal plain (Fig.
~~
3 ~ CONGLOMERATIC MuoaALLS
7.15). Fluvial styles here seem unusual, but may, in
0..
~
§
CROSS LAMINATIONS
"it< fact, be typical of nonmarine forearc (or backarc)
PELL ET AL FABAIC
regions characterized by active intermediate volcan-
4 '"0
~p.
~ WORMY FABRIC
W
ism. Explosive volcanic eruptions coupled with very
fI)
~ LARGE ROOT TRACES z high rainfall lead to the catastrophic transportation
a: C
~ a: and sedimentation of large quantities of coarse
w 5 ~8
ROOT HAlAS
8 0
~
w C9
I
..J volcaniclastic debris. This may be confined to the
::E ..J
W
it main trunk rivers, and the result is that tributaries
6 are dammed at their junction with the main river and
back up to form lakes. These then fill with muds and
CONGLOMERATIC small Gilbertian deltas. A possible ancient example
7 MUOBALL S of this volcaniclastic fluvial style with tributary lakes
was described by Ballance (1988).
8 The clay plugs that commonly fill abandoned
SIL TV SAND UNIT
channels may be of considerable importance in sub-
dividing channel-sand reservoir bodies into com-
9 partments, or providing seals on the updip flanks of
BLUE CLA V UNIT ;'00..
a:~<
..J "W channel sands. Several examples of this are de-
o<;t scribed in Chaps. 14 and 15.
oa:",
10 ". KRINITZSKY AND SMITH (1989) 0.. 0
Fig. 7.12. Typical vertical prome thro ugh a backswamp

deposit. From the modern Mississippi floodplain, Louisi- 7.4 Biochemical Sediments
ana. (Farrell 1987)
There is a considerable body of literature dealing

silted up, leaving the abandoned channel to become with the sedimentary geochemistry of these rocks,
progressively more and more undisturbed by turbu- which is beyond the scope of this chapter. The main
lent eddies entering from the main channel. The interest in fluvial biochemical sediments from the
waters of this elongate pond may therefore become point of view of this book is as architectural compo-
gradually more static, and it slowly fills with clays nents of a stratigraphic assemblage (this section), as
deposited from suspension. Fine-grained lithofacies indicators of original depositional environments
are therefore typical. Peat and other organic remains (this section and Chap. 8), as stratigraphic marker
may also be common. beds (Chap. 9), and as indicators of changes in auto-
A variant of the common abandoned meander genic and allogenic sedimentary controls (Chaps. 10,
model is where the sediment load in main channels is 11, 12). Coals and paleosols have a particular signifi-
so large that the mouths of tributary streams are cance in the study of nonmarine sequence stratigra-
blocked, causing the tributaries to back up as small phy because of what they tell us about the balance
lakes. Two examples of this, in very different envi- between subsidence and sedimentation. In many
ronments, were described by Baker (l978a) and cases, their widespread extent is an indication of
Kuenzi et al. (1979). In the first case, Baker (l978a) regional changes in sedimentation patterns, and has
shows that the sediment load in the main tributaries led to their designation as markers of significant
of the Amazon system is very high, as these rivers stratigraphic surfaces, such as sequence boundaries
have their headwaters in Andean regions underlain (Chap. 13).
Biochemical Sediments 179
(A)
o 400 3600 4000

METERS
A
(8)
44~~1!II
40
36
32
500 1000 1500 2000 2500
B
16
12
8
4~~MOO~.
500 1000 1500 2000 2500 3000
Fig. 7.13. A Architecture of major channel and overbank deposits by blank areas. B Detail of areas A and B shown at
bodies in the Chinji Formation (Siwalik Group), near enlarged scale. Stipple Coarse-grained units (crevasse
Chinji, Pakistan. Stippled units are major channel fills, splays); vertical ruling disruption, e.g., by bioturbation;
showing accretionary geometries and direction of cross-hatching carbonate-leached horizons of well-devel-
paleoflow (down is north, which is directly into the outrop oped paleosols; blank areas undisturbed lamination.
plane). Paleosols are shown by vertical ruling, fine-grained (Willis and Behrensmeyer 1994)
Paleosols may be subdivided into various layers (1984) noted, modern depositional models for coal
and have been classified into many different types, evolved from the work of Fisk (1960), who empha-
with appropriate codings for field description. There sized the importance of peats in the overbank re-
seems to be no particular need to assign element gions of the modern Mississippi river and delta, and
codings to coals or evaporites. mapped the association between peat beds, crevasse,
levee, and channel deposits. Most of these modern
peats are high in clastic ("ash") content, and would
7.4.1 Coal not constitute true coals when lithified. McCabe
(1984) demonstrated that in most clastic settings the
Coal occurs in a wide range of depositional environ- influx of detritus from channels and crevasse splays
ments (DiesseI1992), but is most commonly associ- during overbank flooding dilutes the organic con-
ated with fluvial and delta-plain settings. As McCabe tent and leads to the formation of carbonaceous
Fig. 7.14. Clay-filled abandoned channel [CH(FF)) above a point-bar deposit (LA), Scarborough Formation (Jurassic),
Yorkshire
shales rather than true coals. Low-ash coals probably The architecture of a coal seam is not necessarily
require raised swamps to form. These develop in that of a simple sheet. Commonly, clastic facies
areas where rainfall exceeds evaporation, and or- interfinger with the seam forming "splits". Examples
ganic growth is rapid (McCabe 1984; Moore 1987). are shown in Figs. 7.2 and 7.17. On a large scale, splits
Peat undergoes considerable compaction during the are caused by differential subsidence of the deposi-
formation of coal, as a result of water loss, etc. Ryer tional basin (Diessel 1992). However, on a smaller
and Langer (1980) compiled data showing peat:coal scale (the examples illustrated here), the causes are
thickness ratios ranging between 1.4:1 and 30:1, with autogenic, and are related to channel migration and
a median of7:1. A further discussion of coal develop- avulsion and to the growth and abandonment of
ment, and its climatic significance, is given in Sect. crevasse splays. Detailed mapping of coal seams in
12.9.1. mines may reveal the pattern of channels and cre-
Coals may also form by allochthonous processes. vasses, where the coal has been "washed out" by
Degraded peat mats may be washed into lakes, and fluvial channel erosion. An example of such a map is
accumulate as sapropels. Cannel coals form from the illustrated in Fig. 7.18. The ribbons are several kilo-
accumulation of windblown spores. meters wide, and consist of a main fluvial channel
Coal seams are typically interbedded with fine- plus amalgamated levee and crevasse-splay deposits.
grained overbank sediments of element FF (Figs. The coal seam itself may be subjected to detailed
7.2, 8.44, 8.45). They may also overlie or underlie stratigraphic and facies analysis. Hacquebard and
crevasse-splay deposits (Figs. 6.21, 7.9) and fluvial Donaldson (1969) carried out a detailed analysis of
channel-fill deposits, including point-bar deposits seams in the Sydney and Pictou coal fields in Nova
(Fig. 4.9). Figure 7.16 illustrates a case where a peat Scotia. Based on maceral content they recognized
bog was incised by a fluvial channel, and rapid un- three broad types of coal-swamp depositional envi-
dercutting of the peat by lateral erosion formed a ronments that reflect original water-table levels:
steep step in the base of the resulting channel sand-
stone.
1. Forest moor: well-drained swamp, characterized
by the presence of trees, and yielding vitrinite-
and fusinite-rich deposits.
o Z 3 4KM
2. Reed moor: poorly drained swamp, resulting in
spore-rich coal.
3. Open moor: subaquatic, with sporinite, cannel
and boghead coals and carbonaceous shales.
The distribution of these environments within
one seam is shown in Fig. 7.19. This cross section
illustrates the Harbour Seam, which is estimated to
have taken 2800 years to accumulate. The seam is
characterized by forest-moor depo~its, with occa-
sional flooding producing the other coal types. An
increase in reed moor intervals in the upper half of
the seam indicates a gradual rise in the water table.
Figure 7.20 is a detailed section through this seam,
showing details of coal petrography and palynology.
Note the dominance of Lycospora in the lower half of
the seam. This spore type was derived from the tree
Lepidodendron, and confirms the forested nature of
the swamp at that time. Punctatosporites, a herba-
ceous plant, is characteristic of reed-moor coals, and
occurs near the top of the seam.
Diessel {1992} used maceral-derived ratios for in-
Fig. 7.15. Formation of lakes on a fluvial floodplain by
damming of the mouths of tributaries. Guatemala coast. vestigating vegetation variations and mire type.
(Kuenzi et al. 1979)
Fig. 7.16. A coal seam cut by a channel sandstone in a fluvial delta-plain setting. Note the steep cutbank, probably formed
by the sudden breakup and undercutting of a peat mat by channel erosion. Carboniferous, Kentucky
I,
EXPLANATION
I': ,." :!
Fig.7.17. Examples of splits in a coal seam. The Stockton seam, east Virginia. Lower diagram shows the splits at an
enlarged vertical scale. (Ferm and Staub 1984)
I
N ii •Terre Haute
o ,
20 40 , km
60
~
I
k mi
I
o
/
"3
\/11
L-~ /'
Folsomville / Dykersburg
.1r.J
Indiana
Evansville
-:;;:JJIZ .-
(~/'-"""
( ., ..........., .) "
<7,-\ ' ,)
Coal Missing or Split (
\.- l.- , -
~ Coal Missing ,r Kentuck,y

I
Fig. 7.18. Map of three major sandstone ribbons in Carboniferous coals of Illinois and Indiana. (Compiled by McCabe
1984)
\
A TLANTIC OCEAN
5km
I
SCALE
o
FM
I+e ~====::;::=~A+O RM
o
OM
LEGEND
A: SIII'ORE.·CLARITE T DUROCLARIT(
•= fUSITO·CI.A.RIT[
C :: VITIIIO·CLARIT( + CUTICLE. CURln
0= CLARODUItIT[+ DURIl[ + CARIUQllIT[
f,M = fORlST TERRlSTRI"L MOOR
f .. = fORlST l1li001'
lIt .. = RUO MOOR
OM:::: OptN MOOR
•- 7 = ~ROGRAPHIC INTER.,ALS
Fig. 7.19. Stratigraphy of a coal seam, the Lloyd Cove seam, Sydney, Nova Scotia. See text for discussion. (Hacquebard
and Donaldson 1969)
7.4.2 Paleosols The three main field features of paleosols are root
traces, soil horizons, and soil structures (Retallack
Fossil soils have received a considerable amount of 1988). The original carbonaceous root material may
attention from sedimentologists, geochemists, and be preserved, but commonly this is replaced by or
Quaternary geologists because of the information encrusted with carbonate or siliceous material.
they yield regarding climates and rates of sedimenta- Movement of soluble material by groundwaters and
tion, plus information on the evolutionary patterns evaporation (translocation) develops complex crys-
of alluvial floodplains. Several research syntheses on talline textures. Commonly, there are networks of
these topics have now been published (Wright 1986; irregular, mineralized planes within the soil, termed
Reinhardt and Sigleo 1988; Martini and Chesworth cutans. These may consist of clay skins, formed by
1992). (Additional publications dealing with mod- leaching (eluviation). Cutans may also consist of
ern soils and their geomorphology are referenced in ferruginized planes, manganese encrustations, or
these texts.) In this section, some basic descriptive cracks filled with clastic material (clastic dikes).
data regarding paleosols are provided. Their use as Sheets of calcite, barite, or gypsum may also occur.
stratigraphic markers is discussed in Chap. 9, and Aggregates of soil material are termed peds. They are
the information they can provide on autogenic and bounded by cutans or voids. Compaction of peds
allogenic sedimentary processes is considered in and voids may develop slickensided surfaces. Min-
Chaps. 10 to 13. eral concentrations, termed glaebules, are common
Some paleosol types are so distinctive that they in soils. They are typically nodular or concretionary,
have been given specific geologic names. These in- consisting of calcareous, ferruginous, or sideritic
clude cornstone, a nodular calcrete, and ganister, a material. Clay glaebules also occur. Diffuse mineral
silicified sandy soil horizon (Retallack 1988). aggregation leads to mottling.
Miospore Percent
Variations in soil components, color, and layer-
.
I I I
0 50 100 ing relate to climate, source terranes, and rates of
subsidence and sedimentation (which control the
i -c•
•
!! rate of dilution of the soil by clastic material) (Fig.
i~ uf
·c
0
l!
.ic 7.21). In addition, a very important control fro in the
...
0
;; 8. l!
• point of view of the stratigrapher is the length of time
Legend •
1i
U
c
IL
::J
1 0
...• ...~
.5!
u
~ over which the soil matures before burial. The lateral
DVitrita variability (stratigraphy) of soils within an alluvial
~Clarita basin reflects differences in relief and drainage. A
~ Carbargilita soil unit showing such lateral variations is termed a
o Ouroclarita
catena (Milne 1935; Bates and Jackson 1987). Bown
~ Clarodurita
llliWlI Ourita E and Kraus (1987) pointed out that the concept of the
mOurital catena was introduced primarily to encompass
_Fusite variations in modern soils related to their position
on hillslopes, whereas most paleosols of interest to
sedimentologists occur within the floodplain, and
o 0 differences within and between individual soils de-
pend on maturation time and the amount of dilution
of the soil with clastic sediment. They found that
these factors depend primarily on proximity to ac-
1ft
tive fluvial channels, and suggested that a different
terminology is required to accomodate such obser-
O.5m vational and interpretive differences. They intro-
duced the term pedofacies, which they defined to
encompass the
"laterally contiguous bodies of sedimentary rock that dif-
fer in their contained laterally contiguous paleosols as a
result of their distance (during formation) from areas of
high sediment accumulation" (Bown and Kraus 1987, p.
Fig. 7.20. Palynology and petrography of a typical vertical 599).
profile through the Harbour Seam, Nova Scotia. (McCabe
1984; modified from Hacquebard and Donaldson 1969)
III
t-
~';r~
~~
-
c z
~~
"""
"
!ii 015T ANCE FROM CHANNel

'"
~Alluvial Ridge~ 1-1-----------Floodplain--------.o..j"ITerrace
mature soils
locally poorly drained moderalely malure
soils· climatit sOits.
controlled. Immature
mature.
~CATENARY RELATIONSHIP ~
coarser sediment,
( PEDOFACIES RELATIONSHIP >
weil·drained. immature ~ CHRONOSEOUENCE .....-...;.
soils. -..;--- RELATIONSHIP
Fig.7.21. Soil landscape relationships on a floodplain. (Wright 1992)

A B c
7
4-6
7 'Ir_--,----,----~--_.----r_--._--_.----r_--._--_r--_,r__
1 I
Fig. 7.22. The complex of soil stratigraphies that can be tonism and sea-level change (Chap. 11). 1-7 Geomorphic
caused by a single phase of valley incision and filling, such surfaces with corresponding paleosols; A -C stratigraphic
as characterizes the base-level cydes associated with tec- sections. (Wright 1992)
Table 7.2. Descriptive codes for dassification of paleosol horizons. (After Retallack 1988)
Code Description
o Surface accumulation of organic materials (peat, lignite, coal), overlying dayey or sandy part of soil
A Usually has roots and a mixture of organic and mineral matter; forms the surface of those paleosols lacking
an 0 horizon
E Underlies an 0 or A horizon and appears bleached because it is lighter colored, less organic, less sesqui-
oxidic, or less dayey than underlying material
B Underlies an A or E horizon and appears enriched in some material compared to both underlying and
overlying horizons (because it is darker colored, more organic, more sesquioxidic, or more dayey) or
more weathered than other horizons
K Subsurface so impregnated with carbonate that it forms a massive layer
C Subsurface horizon, slightly more weathered than fresh bedrock; lacks properties of other horizons, but
shows mild mineral oxidation, limited accumulation of silica carbonates, soluble salts, or moderate gleying
R Consolidated or unweathered bedrock
Adapted from Retallack (1988).
Internal erosion caused by channel shifting and cation be based in the first place on careful field-
base-level changes can develop erosional terraces work, which should include detailing the thickness
within alluvial plains, leading to variations in matu- and vertical relationships of the various layers, as
rity of individual soils, lateral erosion of soils, and well as the taking of samples for subsequent labora-
superimposition of younger upon older soil hori- tory analysis. Retallack (1988) also provided a code
zons (Wright 1992; Fig. 7.22). These complexities all system for gradational layering and for subordinate
require a rigorous field observational methodology, descriptive characteristics, but the reader is referred
which Retallack (1988) set out to provide. Some of to his original work for these details. Many different
the basic descriptive techniques and classifications types of soil have been recognized, based on agricul-
summarized by Retallack (1988) and Bown and tural and other work on modern soils. Descriptions
Kraus (1987) are provided here, but the reader is of these types, adapted for the study of paleosols, are
referred to these papers for a more complete discus- provided in Table 7.3. The use of these names re-
sion of field methods, and to the other books and quires considerable experience. More amenable to
papers referred to here for additional detail. field use by the nonspecialist clastic sedimentologist
Paleosols typically consist of several distinct lay- are the paleosol classifications that reflect simple
ers that can be defined on the basis of their color, maturation processes. Two alternative schemes are
texture, or mineralogy. Retallack (1988) provided a provided here. That in Table 7.4 is a generalized
descriptive classification, which is given in Table 7.2. classification provided by Retallack (1988). The al-
He recommended that the application of the classifi- ternative scheme of Table 7.5 summarizes the
Table 7.3. Paleosol types. (Adapted by Retallack 1988 from US Department of Agriculture classification)
Type Features
Vertisol Abundant swelling clay (mainly smectite) to a presumed uncompacted depth of 1 m or
more to a bedrock contact, together with hummock and swale structure, especially
prominent slickensides or clastic dikes
Entisol No horizons diagnostic of other types; very weak development
Inceptisol No horizons diagnostic of other types; weak development
Aridisol Light coloration, thin calcareous layer close to surface of profile, or evidence of pedogenic gypsum
or other evaporite minerals
Mollisol Organic (but not carbonaceous or coaly), well-structured (usually granular) surface (A) horizon,
usually with evidence of copious biological activity (root traces, burrows) and with subsurface
horizons often enriched in carbonate or clay
Histosol Surface organic (0) horizon of carbonaceous shale, peat, lignite, or coal, precompaction thickness
of at least 40 cm
Alfisol Thick, well-differentiated (A, B, C horizons) profile, with subsurface B horizon appreciably
enriched in clay and often red sesquioxides, or dark with humus, and also high cation content
(Ca2+, Mg2+, Na+, K+)
Ultisol Similar to Alfisol but with sparse cation concentration
Spodosol Thick, well-differentiated (A, B, C horizons) profile, with sandy subsurface B horizon
cemented with iron or aluminum oxyhydrates or organic matter; little or no clay or
carbonate
Oxisol Thick, well-differentiated to uniform profile, clayey texture, subsurface horizons
highly oxidized and red, and almost entirely depleted of weatherable minerals
Table 7.4. Stages of paleosol development. (Retallack 1988)

Stage Features
Very weakly developed Little evidence of soil development apart from root traces; abundant primary textures of
parent material
Weakly developed Surface rooted zone (A horizon), incipient subsurface clayey, calcareous, sesquioxidic,
or humic, or surface organic horizons
Moderately developed Surface rooted zone and subsurface clayey, sesquioxidic, humic or calcareous or surface
organic horizons
Strongly developed Thick, red, clayey, or humic subsurface (B) horizons, or surface organic horizons
(coal or lignite), or especially well-developed soil structure, or calcic horizon
Very strongly developed Unusually thick subsurface (B) horizon, or surface organic horizons (coal or lignite)
or calcic horizon; mostly developed at major geologic unconformities
Table 7.5. Stages of paleosol development. (Bown and Kraus 1987)

Stage Description
No horizon formation. Yellow or orange mottling, occasional development of yellow or orange beds with
reddish streaks; color contacts diffuse. Carbonate accumulation initiated. Entisol
2 First incipient horizon formation. A horizon thin. B horizon red, sometimes darker red at top. Color
contacts diffuse. Significant clay and plasma translocation initiated. Carbonate accumulation rapid. Alfisol
3 Marked horizon formation. Thick A horizon. B horizon purplish at top, dark and/or light red in middle,
yellow or orange at base. Incipient, impermanent spodic zones sometimes present at top of B horizon. Color
contacts diffuse. Significant plasma translocation, much concentrated as glaebules. Clay translocation near
peak, carbonate translocation at peak. Alfisol-Spodosol
4 Profound horizonation. A horizon differentiation initiated, with beginning of clay leaching. B horizon thick.
Aluminum and iron translocation at peak. Decrease in color mottling. Colour contacts more distinct. Well-
defined spodic horizons. Spodosol
5 Profound horizon formation. Thick leached A horizon. A and B horizons much thickened relative to stage 4.
Mottling rare. Color contacts sharp. Thick spodic horizons. Spodosol
A B
Fig. 7.23. Stratigraphic sections of two pedofa-
Cr." •••• Spl.y cies successions. Letters on left indicate soil hori-
crn . . . . Splay
zons. Willwood Formation (Eocene), Wyoming.
(Kraus 1987)
P.I.o.ol
Cr." •••• Spl.y
m . . ,. ".
!II.:.....
f
1IIMI •• lene
Cr.y •••• Splay
[-,. Gr.y
Orang.
Light R.d
1 P.I.o.ola
Dark R.d
'i' l" Burrow.

<it Ii!. ~ Rhl zollth.
0 0 a Motll ••
~ Ero.lon Surl.c.
with Sh.1I Lag m........
I !M~' .1II.toilll •
. . d.lone
Irragul.r .011 contact
Levee
Floodplain
Distal
5 4 3
Fig. 7.24. The pedofacies model ofBown and Kraus (1987) and Kraus (1987,1992). Numbers refer to the paleosol stages
of Table 7.5
paleosol types identified in the Eocene Willwood graphic techniques are described by Retallack
Formation of Wyoming, which formed the basis for (1988). The pedofacies model, relating paleosol
the pedofacies concepts of Bown and Kraus (1987). "stage" to distance from the alluvial channel, is
Figure 7.23 shows two stratigraphic sections shown in Fig. 7.24. Examples of detailed profiles
through two simple pedofacies successions, illus- through the five types of paleosol stage are shown in
trating the graphic plotting methods of Bown and Fig. 7.25. Caution is required in the application of
Kraus (1987) and Kraus (1987). More complex such models, because it has been shown that other
A
2
EXPLANATION
DGrey
uare"ge (/)
fZ:Zl Light red !I:

w
Iii
~
_Oerkred
Purple
~I Cley~ich zones
• .' . Fel AI glaebules
•• Mn gleebules
. 0. ;> CeCQl glaebules
'>.~ Rhizoliths
"U"" Burrows
0
STAGE 5
Fig.7.25. A The five basic paleosol stages identified in the ment; Cea calcrete; Cg gleyed; Eb eluviated zone in A
Willwood Formation of Wyoming. A, B, and C are horizon horizon. B Distribution of important soil components in a
types. Subtypes are as follows: Ao epipedon; Bca calcrete; stage-5 paleosoL (Bown and Kraus 1987)
Bg gleyed horizon; Bt clay enrichment; Bfe iron enrich-
-
Organic Matter Content Horizonation Redox Conditions
coal .... HISTOSOL poor -PROTOSOL
r- no coal Ir poor due to
low GLEYSOL
high or fluctuating ~
pedoturbation _VERTISOL
good
- -
In Situ Mineral 1I1uviation of Insoluble Accumulation of
Alteration Soluble Minerals
Minerals/Compounds
OXISOL~extensive
ARG I LLiSOL _ clay CALCISOL-CaCO 3 -rich
poor
organic matter GYPSISDL-CaS04 -rich
SPODOSOL_and iron poor
poor
Fig. 7.26. Main Criteria used in the paleosol classification of Mack et al. (1993), presented as a flowchart showing how
individual paleosols may be identified
Fig. 7.27. General view of Death Valley, California, showing the pattern of ephemeral stream channels and gypsum-
encrusted playas on the valley floor,liiIld the marginal alluvial fans . (Photo courtesy of S. Prosser)
Fig. 7.28. The lowest point in Death Valley, showing a pool of hypersaline water and part of the playa flats occupying the
valiey center
factors may affect paleosol maturity and pedofacies. The implications of pedofacies studies for inter-
For example, Atkinson (1986) showed that varia- pretation of autogenic channel processes and flood-
tions in paleosols in an Eocene succession in Spain plain evolution are discussed in Chaps. 9 and 10. The
could be related to regional changes in fluvial calculation of sedimentation rates using paleosol
style and sedimentation rate which, in turn, reflected maturity and stratigraphy is discussed in Sect. 3.4.
tectonically driven subsidence rates. Theriault and The presence of widespread paleosols may indicate
Desrochers (1993), who studied a Carboniferous low rates of generation of accommodation space in a
succession in the Canadian Arctic, related areal sedimentary basin, and such an interval may corre-
changes in calcrete type in part to variations in late with a sequence boundary, as discussed in Sect.
substrate lithology. Wright (1990) demonstrated 13.3.1.
that variations in biogenic activity and the supply of
the calcium ion markedly affected rates of pedogen-
esis. 7.4.3 Evaporites
One of the problems with most paleosol classifica-
tions currently in use is that they are based on classi- Evaporite deposits, including bedded gypsum, anhy-
fications developed for agricultural purposes, which drite, halite and other facies, and evaporite crystals
originally included features such as moisture con- developed within clastic facies, are a characteristic
tent that are not relevant for geological purposes. A deposit of ephemeral lakes, or playas, occurring at
simpler classification of paleosols, emphasizing pre- the distal fringes of sheetflood fluvial units, or at the
servable criteria, was developed by Mack et al. center of basins with inland (centripetal) drainage
(1993). Their flowchart for identifying paleosol type systems. Their sedimentology is described well by
is illustrated in Fig. 7.26. In a later paper, Mack and Hardie et al. (1978) and will not be repeated here.
James (1994) discussed the uses of this classification Figures 7.27 and 7.28 illustrate the playa ephem-
in studies of paleoclimate (Sect. 12.9.2). eral fluvial systems of Death Valley, California.
Chapter 8
Fluvial Styles and Facies Models
8.1 Controls on Channel Style down smooth surfaces of glacial ice and tilted glass
plates in a laboratory also develop meanders. Details
of the mathematical and hydraulic models that have
The development of end-member facies models for a
been developed to quantify the meandering process
few distinctive fluvial styles has obscured the con-
in rivers are beyond the scope of this book. Weare
tinuum 1;hat exists between fluvial channel patterns.
concerned here with what can be determined about
The fourfold classification of channel styles dis-
channel style from the geological record, because
cussed in Sect. 2.3.6.5 (Fig. 2.1S) has served sedimen-
this provides information about the architecture of
tologists well, but should be replaced by a more
the beds, and about the basin controls that were
sophisticated approach to channel classification.
operating during deposition.
This needs to be based on an understanding of the
Observations and measurements of modern riv-
controls that determine channel style.
ers, and experiments with flumes, have shown that
Considerable theoretical and experimental work
~he ch~nnnel pattern in alluvial rivers (those flowing
has been carried out to explore the causes of mean-
III theIr own sediment) is primarily dependent on
dering and braiding. The natural meandering of
discharge, sediment load, and slope. Lane (1957) and
channels, in particular, has received much attention.
Leopold and Wolman (1957) demonstrated a natural
It is now known that meandering occurs in all fluid
transition between braided and meandering styles,
systems as a result of turbulence, internal shear, and
dependent upon channel slope and discharge (Fig.
bank and bed friction. Straight channels, including
S.1). Thus, for a given slope, a river changes from
canals and channels in flumes, commonly develop
meandering to braided as discharge is increased.
meanders because of these controls. Air and ocean
Schumm (196Sa) proposed the following general
currents meander despite the absence of banks to
equations:
impose a frictional shear. Threads of water flowing
0.05
Braided
0.01
•c.. 0.005
.2
(/)
G
c
c 0.001
0
.c
U
0.0005 Meandering
0.0001
0.00005
102 10~ 10 4 105 10·
Bankfull Discharge (ctsl
F~g. 8.1. The relationship between channel slope, bankfull Annual Review of Earth and Planetary Sciences, v. 13,
dlscharge, and channel style. (Leopold and Wolman 1957; © 1985, by Annual Reviews Inc.)
Schumm 1985b; Reproduced, with permission, from the
192 Fluvial Styles and Facies Models
w,d,l etated banks are highly resistant to erosion, as dem-

Qw oc - - - onstrated in an experiment by D.G. Smith (1976).
s
This limits the supply of sediment to the river, and
w,l,s
Qsoc also reduces the bank widening that commonly ac-
d,P companies braiding. In vegetated areas, therefore,
where Qw=discharge, Qs=sediment load, w=width, because of the effects on discharge variability, bank
d=depth, l=meander wavelength, s=slope, and P= stability and sediment supply, braiding is inhibited.
sinuosity. For a few sets of rivers in specific climatic Baker (1978a) quoted examples in tropical Africa
ranges, some of these relationships have been quan- and South America where dense bank vegetation
tified, as summarized by Ethridge and Schumm inhibits braiding, even where abundant coarse sedi-
(1978). However, the data are inadequate to permit ment is present in the banks. Stanistreet et al. (1993)
generalizations that can be applied to all geological described the channels of the Okavango fan in
conditions, a fact that considerably reduces the use- Botswana, which are low-sinuosity to meandering
fulness of quantitative geomorphology for geologi- rivers, in which banks are stabilized by peat levees.
cal reconstructions. Additional details on the The changes in river behavior when forests in upland
relationship between discharge and slope and the catchment areas are destroyed by fire or deforesta-
braided-meandering threshold are given by Bridge tion are marked and well known. Runoff becomes
(1993b). more flashy, the sediment load increases, and a ten-
One of the primary prerequisites for braiding is a dency may develop for debris flows to occur.
high coarse-grained sediment load (bed load). Rivers are sensitive to longitudinal changes in the
Leopold and Wolman (1957, p. 50) stated: controlling variables and to temporal changes in
these variables. Carson (1984a,b,c) showed from his
"Braiding is developed by sorting as the stream leaves studies of rivers in the Canterbury Plains area ofN ew
behind those sizes of the load which it is incompetent to Zealand that the transition from meandering to
handle ... if the stream is competent to move all sizes braided takes place at higher slopes as the caliber of
comprising the load but is unable to move the total quan-
tity provided to it, then aggradation may take place with- the sediment load is increased. He also showed that
out braiding." the development of the braided pattern is very sensi-
tive to the local rate of supply of bed load to the river.
In addition to the relationship between discharge A large bed load, supplied, for example, from easily
and channel style shown in Fig. 8.1, Miall (1977) and erodible banks, leads to channel shoaling and local
Rust (1978a) noted that braided rivers are com- flooding, bank incision, and avulsion. N.D. Smith
monly characterized by high discharge variability; and D.G. Smith (1984) described the William River
for example, rivers in alpine and arctic areas, with in northern Saskatchewan, a river that changes
highly seasonal discharge variations, tend to be downstream from a single-channel to a multiple-
braided, as are ephemeral rivers in arid regions. channel, braided pattern where a large sand bed load
Miall (1977, p. 7) argued from these and other obser- is introduced by eolian processes. Stollhofen and
vations that discharge variability is a primary con- Stanistreet (1994) demonstrated a change in fluvial
trol on the braiding pattern: style from meandering to braided when volcanoes
erupted in a fluvial plain and provided a large supply
"In rivers of highly variable discharge competency will be of pyroclastic debris. Friend and Sinha (1993) re-
similarly variable, and there will be long periods of time corded much local variablity in sinuosity and braid-
throughout which the river will be unable to move at least
the coarsest part of its bed-load. The incidence of bar ing character in the large rivers of India. They
initiation, flow diversion and the creation of new channels confirmed the conclusions of Carson (1984a,b,c),
(braiding) will thus be high." and noted that the nature of the local alluvial sub-
strate, and the sediment input from tributaries, will
On the scale of the alluvial basin the nature of the also contribute to continuing downstream adjust-
vegetation cover has an important effect on dis- ment and consequent change in the configuration of
charge characteristics, sediment load, and fluvial the river. Important experimental work that con-
style. In vegetated areas presently, runoff following firms these trends has been reported by Ashmore
major rainfalls is rarely catastrophic because the (1991), Leddy et al. (1993), and Germanoski and
precipitation is absorbed by soil and plants and re- Schumm (1993), as noted in Sects. 6.3 and 11.2.2.
leased slowly. Similarly, sediment is stabilized by Temporal changes may be exemplified by several
roots, and sediment yields are relatively low. Veg- studies of rivers in the United States. Nadler and
Controls on Channel Style 193
Schumm (1981) and Schumm (l985b) described ex- processes in the headwaters, would have changed
amples of two rivers, the South Platte and Arkansas, little (Fig. 8.2, curve 2). Schumm (1968a) noted that
that evolved from braided to meandering patterns primitive flowering plants appeared by Permian
within historic times as a result of damming and time, but would not likely have occurred outside
irrigation that resulted in lower discharge variabi- tropical rain forests. Conifers were abundant by the
lity, lessening of flood peaks, and stabilizing of Jurassic and modern deciduous forests appeared by
banks by the growth of vegetation. Conversely, the Mid-Cretaceous. The Devonian-Cretaceous pe-
Schumm and Lichty (1963) described the case of the riod therefore marked a period of increasing stabili-
Cimarron River, Kansas, that changed from a sus- zation of land surfaces by vegetation, but it was
pension-load, meandering morphology to a broad, probably not until the early Cenozoic that interfluve
shallow, bed-load, braided morphology during a and upland areas were colonized by plants capable of
single major flood in 1914. Channel widening con- surviving severe weather and climatic fluctuations
tinued until 1942, during a period of below-normal (Fig. 8.2, curve 3). Grasses appeared in the Miocene,
precipitation, which inhibited vegetation growth. and since that time runoff and sediment yields
Changes in regional slope may to some extent be would have been much as they are today (Fig. 8.2,
accomodated in the river system by changes in chan- curve 4).
nel pattern (Schumm 1993; Wescott 1993). As noted Developing these ideas further, Schumm (l968a)
by Wescott (1993), channel slope SC' valley slope Sy deduced that bed-load streams would have been pre-
and sinuosity P are related: dominant in early geologic time. Cotter (1978) car-
Given constant discharge and sediment load the 4

river will tend to maintain a constant channel slope.
Therefore, an increase in regional slope (Sy) as a
result of tectonic activity will tend to be compen-
sated by an increase in sinuosity. This ability of the
fluvial system to respond to regional changes in the 3
major external controlling variables must be taken "t:)
into account in interpretations of the fluvial re- t

sponse to tectonic activity and changes in base level. ~
Some of the early models of sequence stratigraphy ~
did not recognize this, and are overly simplistic 51
(Miall1991c), as discussed in Sect. 11.2.2. ...._" 2
o
Turning to long-term temporal changes, it has ~
been suggested that the evolution of vegetation has
had a major effect on fluvial styles through geologic
time, because of the implications for bank stability
and erodibility, and the rate of supply of sediment
into river systems. Schumm (l968a) developed a
suite of hypothetical curves for the relationship bet- , /40
ween precipitation and sediment yield through time "","'"

(Fig. 8.2), in which he developed this idea. Prior to
----::.:.....--4
the Devonian, there was little or no land vegetation,
and the land surface probably appeared much as arid
Mean onnuo I precipitation (inches)
areas do today, even where rainfall was high. Runoff
would have been flashy and sediment yield large Fig.8.2. Hypothetical suite of curves illustrating the rela-
(Fig. 8.2, curve 1). From the Devonian to the end of tionship between precipitation and sediment yield during
the Paleozoic, vegetation was probably confined to geologic time. 1 Before the appearance of land vegetation;
2 after the appearance of primitive vegetation; 3 after the
nearshore and coastal plain areas, so that bank stabi- appearance of flowering plants and conifers; 4 after the
lization would have begun, although discharge and appearance of grasses; 4a increase in sediment yield for
sediment-yield characteristics, controlled mainly by tropical monsoonal climates. (Schumm 1968a)
ried out a test of this proposal, using the Paleozoic morphosis, suggests that it is average peak discharge
fluvial sedimentary record of the Appalachian basin and the rare violent flood that may determine chan-
as his data base. He was able, from a limited number nel patterns. Thus, it may not be discharge variabi-
of case studies, to confirm that pre-Devonian fluvial lity, as such, that determines channel style, but the
deposits in this basin were formed in braided fluvial fact that the channel-forming high-discharge events
environments, whereas Devonian and Carboni- that lead to rapid bank erosion and macro form evo-
ferous deposits recorded the appearance of mean- lution have the highest stream power in rivers with
dering fluvial styles. Long (1978) confirmed that the most variable discharge.
most Proterozoic fluvial deposits were formed in A useful idealized model for the evolution of dif-
braided environments. ferent channel styles was developed by Bridge
Bridge (1985, 1993b) disagreed with the idea that (1985), based on much experimental and theoretical
discharge variability is a primary control on braid- work (Fig. 8.3). It has been known since the early
ing, pointing out that "most channel patterns can be 1960s that flow around a bend leads to a pressure
formed in laboratory flumes at constant discharge, setup at the cutbank, and a helical overturn pattern
and many rivers with a given discharge regime show (Sect. 6.7; Figs. 2.15, 2.20). Flow velocity and shear
downstream changes in pattern." However, down- stress increase transversely across the convex side of
stream changes in channel pattern may be caused by the bend, toward the thalweg, which, downstream,
variations in bank erodibility and consequent local progressively shifts to the outer side of the bend at
changes in bed load, as suggested by the work of the point of maximum curvature. The transverse
Carson (1984a,b,c). Carson's work, together with variation in velocity leads to an obliquity in the
that of Schumm (1969, 1981, 1985b) on river meta- advance of bedforms, as shown in Fig. 8.4. Essen-
Fig.8.3. Idealized model for the early evolution of differ- (macroforms). Sinuous thalwegs lead to the development
ent channel patterns (right) from straight channels of dif- of rows of alternate bars. Narrow channels with single rows
ferent widths (left). Heavy curved lines show location of of bars evolve into wandering, and then into meandering
thalweg, lines with ticks show original position of alternate channels with point bars. Wider channels develop mid-
bars, and these are shown as remnants by dashed lines in channel bars and become braided. (Bridge 1985)
diagrams at right. Stippled areas are topographic highs
Fig. 8.4. Idealized patterns of flow velocity and

bedform orientation in fluvial channels. Symbols
as in Fig. 8.3, with the addition of short curved
lines transverse to flow that show schematically
the orientation of dune crest lines. This model
mimics the development of scroll "bars" on point
bars (above) and cross-channel "bars" in sandy
braided systems (below). (Bridge 1985)
tially, the same model may be applied to the develop- channel macro form. As noted by Bridge (1985),
ment of midchannel braid bars, such as in the sand- "channel bends with chute channels are braided
flat model of Cant and Walker (1978; Fig. 6.26). The channels."
basic similarities in flow patterns, bedform configu- The relationships between sediment load and
rations, and bar development at channel bends in all channel form were explored by Schumm (1963, and
rivers (Fig. 8.4) indicate that studies of the geological reproduced in much of his later work, e.g., 1981,
results of these processes in small outcrops cannot 1985b) in a widely used channel classification (Table
hope to lead to useful conclusions about fluvial style, 8.1, Fig. 8.5). This classification summarizes many of
a point now made by many writers (e.g., Miall1980; the most important trends, for example, that rivers
Bridge 1985; see also Sect. 2.4.2). For example, the transporting a coarse sediment load are more likely
identification of DA or LA units and the description to be of low-sinuosity, multiple-channel type, and
of their internal architecture cannot, by itself, indi- that they are more unstable and more prone to
cate channel sinuosity or channel multiplicity. channel avulsion than are rivers with a fine-grained,
Bridge's conceptual models are developed further in suspended load. However, the classification is sim-
a later paper (Bridge 1993b), in which detailed plified, as shown in Fig. 8.6. Straight channels can
speculations are provided regarding the lithofacies occur where the sediment load is dominantly of bed-
successions and architectural construction to be load, mixed-load, or suspended load type. Braided
expected in a set of basic evolutionary cases for channels are of bed-load or mixed-load type.
simple channel-macroform configurations. Devel- Rust (1978a) showed that two simple parameters
opment of a complete model of a given ancient river could be used to define channel style for most rivers.
depends on a thorough architectural analysis of the These are channel sinuosity, P, and a braiding pa-
sediments, incorporating local and regional strati- rameter. The latter expresses the number of bars or
graphic, facies, architectural-element, and paleo- islands per meander wavelength of the river. Rust
current data (Chap. 4). (1978a) adapted this second parameter from the
The unifying models of Bridge (1985, 1993b) pro- "braiding index" of Brice (1964). He pointed out that
vide a useful counterweight to the increasing com- simply counting bars and islands is highly depen-
plexities introduced by the facies-models approach dent on the stage of the river. During full flood, most
to fluvial sedimentology. To a considerable extent topographically elevated areas of the river may be
we are prisoners of our terminology. The different under water, resulting in a braiding index of near
names that have evolved for each fluvial style tend zero. He suggested that braids be defined by map-
to emphasize their differences rather than their ping the channels that surround them at low water.
similarities. The flume runs of Ashmore (1991) With these two measures Rust (1978a) erected a
demonstrated the arbitrary nature of some of the fourfold classification (Table 8.2) that corresponds
terminology. As shown in Fig. 6.13 gravel-bed to the subdivision referred to at the beginning of this
"braided" rivers can evolve from "meandering" section (Fig. 2.18). Friend and Sinha (1993) proposed
rivers by the development of simple chute channel a modified braiding index, and suggested natural
across the inside of a point bar. The outer part of ranges for braiding and sinuosity, based on mea-
the bar, once isolated, evolves into a typical mid- surements in Indian rivers.
Table 8.1. Classification of alluvial channels by sediment load. (Schumm 1963, 1985b)
Type of Bedload Single-channel systems Multiple-channel systems
channel (% of total load)
Suspended load <3 Suspended-load channel Anastomosing system
W /D ratio < 10,
sinuosity> 2.0,
gradient relatively gentle
Mixed load 3-11 Mixed-load channel Delta distributaries
WID ratio 10-40, Alluvial plain distributaries
sinuosity < 2.0,
gradient moderate, may be braided
Bedload >11 Bed-load channel Alluvial fan distributaries
WID ratio> 40,
sinuosity < 1.3, gradient relatively steep,
may be braided
W/D=width/depth ratio.
CHANNEL TYPE
{..
Suspended Load Mixed Load Bed Load
..,;;
~
~ .3.3 l:
(!) 0
.
c:t E
~ tt l:
! r.
z
U)
t - ;;;

UJ (!)
.~
"0 3a~ ~
Ul "2
~
:::i "0W U"
- .,e
Z
a:
.. ....
~ Il:
c:t .<:C III c:
0
Q. UJ c. .- c:t u
0 .,"0 ~ :;: z
...J Z a VI
UJ c:t ,~
UJ
Z UJ .<:
Z :::Ii: ;:; > r. "0
c:t ~ i= Ul :;
l: c:t ; U
."
U ...J ."
UJ c:
0
!
a:: :;:
.<:
U
Legend
_____ Channel Boundary I r.
Ul
~ "'c:
~-- .. Flow 0 ; .~
- -
...J
.<: ,.
'0';
..... <l
HIGH RELATIVE STABILITY LOW
-
(3%» Low- Bed Load ITotol Load Ratio -High (>11%)
- --
Small - Sed imlnt Size Large
Small Sediment Load - Large
-
Law - Flow Velocity High
Low Stream Power High
Fig. 8.5. A simple channel classification based on sedi- from the Annual Review of Earth and Planetary Sciences,
ment load. Variations in some other variables are shown. v. 13, © 1985, Annual Reviews Inc.)
(Schumm 1981, 1985b; Reproduced, with permission,
The cutoffs used in Table 8.2 (braiding para- type example. Study of his maps indicates that the
meter=l, sinuosity=1.5) were suggested by Rust river has a braiding parameter of 2.8. A wandering
(l978a) based on a review of modern rivers and some reach of the Squamish River, British Columbia, has a
earlier literature discussion to which he made refer- braiding parameter of 2.5 (Brierley and Hickin
ence. It is suggested here that they need revision. In 1991). The braided rivers illustrated by Rust (l978a)
particular, a type of gravel-bed river termed "wan- all have braiding parameters greater than 8. It is
dering" has now been recognized that is intermedi- suggested here that a range of 1.0-3.0 be used to
ate between braided and meandering. Church (1983) define the "wandering" category. Similarly, it is sug-
used the Bella Coola River, British Columbia, as his gested that a range of P= 1.2-1.5 be used to define the
(a) a source area many hundreds of kilometers distant

2 3 4 5 (Potter 1978). For example, the Ganges, Brahma-
putra, and Mekong, all of which are partly braided,
\~ (J
~~l
l@JQJ have headwaters located in the Himalayan Moun-
'J ~Q
r (;
tains. Rivers in the Amazon Basin show widely vary-
/.~I
~~~ II ing fluvial style, reflecting the location of their
JDII \°11I,
J.
catchment areas. Those heading in the high Amazon
are characterized by strongly seasonal discharge and
II~ derive sediment from relict Pleistocene fluvioglacial
(b) deposits. They are of moderate sinuosity and par-
7 8 9 10 tially braided. Rivers sourced within the lowland
6 areas carry much smaller sediment loads and are
~~)
~
highly sinuous streams (Baker 1978a). Climatic con-
trols on channel patterns and fluvial style are dis-
cussed at greater length in Chap. 12.
fq~ Several channel patterns illustrated in Fig. 8.6 are
~l not amongst the classic suite of fluvial styles (the

four that became most well known in the 1970s;
Table 8.2), but are receiving greater attention as
(e) 14
13 sedimentological techniques improve. For example,
11 12 the "wandering" style (pattern 4 in Fig. 8.5, pattern 3
in Fig. 8.6) is intermediate between braided and
~
meandering (Church 1983). Such rivers have a
single, relatively stable, dominant channel of
intermediate to high sinuosity, but also contain
I short braided reaches with bars and islands. This
style is shown by some gravel-bed rivers, as dis-
cussed later in this chapter. The causes of the wan-
dering style are not clear, although Brierley and
Hickin (1991) described a downstream change
within a single river from braided to wandering
Fig. 8.6. The range of alluvial channel patterns. a bed load to meandering as slope and sediment grain-size
channels; b mixed-load channels; c suspended-load chan- decrease. It may, therefore, simply represent a tran-
nels. (Schumm 1981) sitional condition.
Pattern 2 in Fig. 8.6 shows a straight channel with
a sinuous thalweg, with alternate bars developing on
intermediate sinuosity range. Lowest sinuosities are the insides of the meanders. This pattern occurs in
those of "straight" rivers. Schumm and Khan (1972) delta distributaries oflow slope. Crowley (1983) also
showed that this condition develops at very low suggested that it develops as an intermediate form
slopes (Fig. 8.7), such as near the mouths of some between low-sinuosity braided and meandering sys-
delta distributaries. tems in some sand-bed rivers. The major deposi-
For a variety of reasons, the geometric and hy- tional forms are alternate bars, the characteristics of
draulic characters of a river may not precisely reflect which are discussed in Sect. 6.7. The overall fluvial
local conditions. For example, local bedrock con- style is described later in this chapter.
trols may slow down erosion and set back the re- Pattern 14 in Fig. 8.6 is that of the anastomosed
sponse of a river to climate change. Many rivers are river, as described in the definitive work of D.G.
"underfit", that is, their discharge is too small to be Smith (1973, 1983) and D.G. Smith and N.D. Smith
correlated with existing channel characteristics (1980). These rivers are characterized by a network
(Dury 1964). For example, many rivers have yet to of relatively stable, interconnected channels of low
adjust to the reduced discharge that accompanied to high sinuosity. They commonly show low slopes
the postglacial climate changes of the Holocene. In and high aggradation rates. D.G. Smith's work sug-
the case of many of the world's very large rivers, the gests that the anastomosed pattern is favored by high
sediment load and discharge may partially reflect the rates of aggradation. This can occur where the river
climate, relief, and sediment-yield characteristics of flows across a basin that is undergoing rapid subsi-
Table 8.2. Classification of alluvial channels by geometric characteristics. (Rust 1978a)

Single-channel Multichannel
(braiding parameter < 1) (braiding parameter> 1)
Low sinuosity Straight Braided
(P < 1.5)
High sinuosity Meandering Anastomosing
(P> 1.5)
1.3
)0-
I- 1.2
en
0
:::I
Z 1.1
en
MEANDERING BRAIDED
1.0 0
°o~-------n~------~~------~~------~~--------~-J
SLOPE (%)
Fig. 8.7. Relationship between slope and sinuosity at consi~n, from the Annual Review of Earth and Planetary
stant discharge, based on flume experiments. (Schumm SCIences, v. 13, © 1985, by Annual Reviews Inc.)
and Khan 1972; Schumm 1985b; Reproduced, with permis-
dence, but with a downstream "damming" effect, 8.2 Facies Models

caused by local bedrock control.
As pointed out by Bristow and Best (1993) studies
of channel deposits and dynamics have ranged Most of the various channel styles shown in Fig. 8.6
across at least five orders of magnitude of channel have now received considerable attention by sedi-
scale, from laboratory models to the largest rivers, mentologists, with data collected from both modern
such as the Brahmaputra and Mississippi. Channels and ancient examples. In Miall's (1985) review, flu-
show a considerable degree of self-similarity across vial styles were discussed with reference to 12 char-
this range of scales, which is why laboratory models acteristic examples, which were given model
are a useful way to explore fluvial processes. Ashley numbers. Additional work has demonstrated the
(1990) noted a continuum in the sizes offlow-trans- need to recognize a few more categories, for ex-
verse bedforms, and suggested that bedforms also ample, in order to distinguish some useful interme-
adjust their scale and geometry to the space and diate categories (e.g., the "wandering" style), and in
sediment available. Scaled flume models are there- order to clarify some differences in depositional be-
fore a useful way to explore fluvial processes, but havior that reflect climatic controls. In this chapter,
much work remains to be done to clarify scale con- 16 fluvial styles are described.
siderations in ancient deposits. The same architec- As stated in the earlier review (MiallI985, p. 289):
tural principles can be applied at all field scales, as "It must be emphasized that this section does NOT repre-
demonstrated in Chap. 4 (cf. Figs. 4.2, 4.3, and 4.4). sent an attempt to provide a comprehensive suite of fluvial
Part of the task of architectural analysis of an ancient models, to replace or add to those already in existence. The
purp.ose is ~olely to illustrate some of the variablity in style
deposit is to determine the scale of the channels, bars that IS pOSSIble m natural environments, much of it barely
and islands that formed the deposit under consider- appreciated by sedimentologists. .... Every conceivable
ation. This can be approached by documenting such gradation between any two of the models illustrated here is
attributes as the thickness and lateral extent of chan- to be expected ... the writer accepts no responsibility for
nel and bar deposits, following the quantitative ap- any uncritical pigeonholing or force-fitting of field ex-
amples into any of these models."
proaches described in Sect. 10.4.
Facies Models 199
Table 8.3. Architectural characteristics of some common fluvial styles. (After Mial11985)
Style No.' Sinuosityb Braiding Sediment Characteristic Modern examples and
name parameter' type elements d interpreted ancient examples
Part 1: Gravel-dominated rivers
Gravel braided 1 Low High Gravel, SG (GB, SB) Hooke (1967), Bull (1972), Wasson
with sediment- minor sand (1977), Nemec and Muszynski (1982),
gravity flows Schultz (1984), Blair and McPherson
(1992)
Shallow 2 Low High Gravel, GB (SB) Boothroyd and Ashley (1975),
gravel braided minor sand Boothroyd and Nummedal (1978), Miall
"Scott type" and Gibling (1978), Vos and Tankard
(1981), Ramos and Sopena (1983),
Dawson and Bryant (1987), Munoz et
al. (1992)
Deep gravel 3 Low to Int. to Gravel, minor GB, SB, DA Williams and Rust (1969), Rust (1972,
braided into high sand, fines (FF) 1978b), Steel (1974), Minter (1978),
"Donjek type" Massari (1983), Miall (1984b), Morison
and Hein (1987), S.A. Smith (1990),
Reinfelds and Nanson (1993)
Gravel Int. to Int. Gravel, minor GB, DA, LA Church (1983), Ferguson and Werrity
wandering high sand, fines (SB, FF) (1983), Desloges and Church (1987),
Billi et al. (1987), S.A. Smith (1990),
Brierley and Hickin (1991)
Gravel 4 High Low Gravel, minor GB, LA, FF Bluck (1971), Lewin (1976), Gustavson
meandering to into sand, fines (SB) (1978), Schwartz (1978), Ori (1979,
1982), Arche (1983), Ramos and Sopena
(1983), Forbes (1983), Hooke (1986)
Part 2: Sand-dominated, high-sinuosity rivers

Gravel-sand 5 Int. to Low to Sand, SB, LA, FF Bernard and Major (1963), McGowen
meandering high into pebbly (GB) and Garner (1970), Bridge and Jarvis
(the "coarse- (1976), Jackson (1976a,b,d), Levey
grained (1978), Nijman and Puigdefabregas
meandering" (1978), Crowley (1983), Allen (1983a),
model) Campbell and Hendry (1987), Todd and
Went (1991)
Sandy 6 High Low Sand, LA, SB, FF Sundborg (1956), Allen (1963b,
meandering minor fines LV, CR, CS 1970b), Davies (1966), Beutner et al.
(the "classic" FF(CH) (1967), Puigdefabregas (1973), Nami
meandering and Leeder (1978), Puigdefabregas and
model) Van Vliet (1978), Ethridge et al. (1981),
Hobdayet al. (1981), Plint (1983),
Bridge and Diemer (1983), Link (1984),
Farrell (1987), Alexander (1992),
Jordan and Pryor (1992), Diemer and
Belt (1991), Munoz et al. (1992),
Alexander and Gawthorpe (1993)
Ephemeral sandy- High Low Sand, LA, SB (FF) Stear (1983), Shepherd (1987),
meandering minor fines Lawrence and Williams (1987)
Fine-grained 7 High Low Fine sand, LA, SB, FF Taylor and Woodyer (1978), Miall
meandering silt, mud LV,CR,CS (1979a), Nanson (1980), Jackson (1981),
FF(CH) Stewart (1983), Mossop and Flach
(1983), D.G. Smith (1987), Wood
(1989)
Table 8.3 (Contd.)

Style No." Sinuosityb Braiding Sediment Characteristic Modern examples and
name parameter' type elements d interpreted ancient examples
Anastomosed 8 Low to High Sand, fines SB, CH (LA) Rust (1981), D.G. Smith (1983, 1986),
high FF, LV, CR, Rust and Legun (1983), Rust et al.
CS (1984), Flores and Hanley (1984),
N.D. Smith et al. (1989), Shuster and
Steidtmann (1987), Eberth and Miall
(1991), Kirschbaum and McCabe (1992)
Part 3: Sand-dominated, low-sinuosity rivers

Low sinuosity Low Low Sand DA-LA, SB, FF McCabe (1977), Galloway (1981),
braided- Crowley (1983), Okolo (1983),
meandering Bridge et al. (1986), Olsen (1988),
with alternate bars Wizevich (1992b)
Shallow 9 Low to High Sand SB (FF) Miall (1976, 1984b), Blodgett and
perennial into Stanley (1980), Crowley (1983), Allen
braided (1983a), Smith and Smith (1984), R0e
"Platte type" and Hermansen (1993)
Deep perennial 10 Low to Int. to Sand, DA,LA, SB Cant and Walker (1978), Kirk (1983),
braided int. high minor fines (FF) Haszeldine (1983a,b), Allen (1983a),
"S. Saskatchewan Ramos et al. (1986), Lawrence and
type" Williams (1987), Miall (1988c, 1994),
Bristow (1987,1988, 1993a,b),
Wizevich (1992b), Luttrell (1993)
Willis (1993a,b)
Highcenergy, Low to Int. to Sand, DA,SB,HO, Miall (1988a), Cowan (1991)
sand-bed braided into high minor fines (FF)
Sheetflood 11 Low High Sand, SB (FF) Williams (1971), Miall and Gibling
distal minor fines (1978), Parkash et al. (1983), Sneh
braided (1983), Lawrence and Williams (1987),
Olsen (1989), Cotter and Graham
(1991), Munoz et al. (1992)
Flashy, 12 Low High Sand LS (FF) McKee et al. (1967), Miall and
ephemeral minor fines Gibling (1978), Rust (I978b), Tunbridge
sheetflood (1981, 1984), Sneh (1983), Miall
"Bijou Creek type" (1984b), Lawrence and Williams (1987),
Langford and Bracken (1987), Mertz
and Hubert (1990), Cotter and
Graham (1991), Bromley (1991a)
a Model numbers from Miall (1985).
b Suggested sinuosity (P) values: low < 1.2, intermediate 1.2-1.5, high> 1.5.
, Suggested braiding parameter values: low < 1, intermediate 1-3, high> 3.
d Elements shown in brackets are minor components. Element CH is shown only where it forms a significant proportion
of the assemblage, and cannot be subdivided into component elements.
A summary of the models and the data base from mentation, but the convergence of ideas between
which they have been built is given in Table 8.3. sedimentology and geomorphology that seems to be
Nanson and Croke (1992) provided a review evolving is reassuring. Their classification is given in
of fluvial floodplains from a geomorphological Table 8.4, omitting all the descriptive detail, which
perspective (although they quoted much of the sedi- would largely repeat what is presented in the rest of
mentologicalliterature), and developed a classifica- this chapter. Class A in this classification deals
tion that is very similar to that used here. Their focus mainly with upland streams, which do not deposit
was on surface processes, and their emphasis was on large, permanent alluvial accumulations. This class
floodplain development rather than channel sedi- of river is therefore of minor importance to sedimen-
Facies Models 201
Table 8.4. A classification of floodplains. (Nanson and Croke 1992)

Class A: High-energy, noncohesive floodplains
Disequilibrium floodplains which erode in response to extreme events, typically located in steep headwater areas
where channel migration is prevented by valley confinement
Al Confined, coarse-textured floodplains
A2 Confined, vertical-accretion floodplains
A3 Unconfined, vertical-accretion sandy floodplains
A4 CutcandcfIll floodplains
Class B: Medium-energy, noncohesive floodplains

Equilibrium floodplains formed by regular flow events in relatively unconfined valleys
Bl Braided river floodplains
B2 Wandering gravel-bed river floodplains
B3 Meandering river, lateral-migration floodplains
B3a Lateral-migration, non scrolled floodplains
B3b Lateral-migration, scrolled floodplains
B3c Lateral-migration/backswamp floodplains
B3d Lateral migration, counterpoint floodplains
Class C: Low-energy, cohesive floodplains

Floodplains formed by regular flow events along laterally stable, single-thread or anastomosing low-gradient
channels
Cl Laterally stable, single-channel floodplains
C2 Anastomosing river floodplains
C2a Anastomosing river, organic-rich floodplains
C2b Anastomosing river, inorganic floodplains
tologists. Classes Band C encompass rivers that de- now be made of ever more refined observational
velop broad floodplains underlain by significant al- details to reconstruct local variations in fluvial style,
luvial accumulations, and are therefore the rivers of and the subtleties of their dependency on slight
interest to sedimentologists. The subdivision of changes in depositional controls, including tec-
these two classes corresponds closely to the classifi- tonism, climate, base-level change, source-area geo-
cation used to subdivide fluvial styles in the follow- logy, vegetation, the effects of human interference,
ing sections of this chapter. The subdivisions of class and so on. Walker (1990) suggested that the retreat
B3, the floodplains of meandering rivers, reflect the from a limited suite of end-member facies models to
style of floodplain sedimentation, which in part re- a multiplicity of models based on shifting assem-
flects variations in sediment grain size. These sub- blages of architectural elements would lead to sedi-
classes are therefore similar to the subclasses of mentological anarchy. However, as attempts have
meandering river defined in Table 8.3 (part 2). Other been made to demonstrate in earlier chapters of this
floodplain detail is discussed in Chap. 7. book, facies and element analysis serve to simplify
Sixteen examples of fluvial architectural style are and codify the study of complex real-life fluvial as-
reviewed in the following sections. Each of these is a semblages.
"facies model" in the sense used by Walker (1984) Earlier workers expressed concern that the accre-
and Reading (1986). That is, it is a summary of a tionary geometries oflarge bars would be difficult to
particular environment, in which local details have identify in the geological record. They were con-
been distilled away, leaving the "pure essence" of the cerned mainly with the study of point bars, but the
environment, its facies, and its architecture (see same concerns apply to all macroforms. Thus,
Miall 1990; Chap. 4, for a complete discussion of Collinson (1986) suggested that the scour and-fill
facies analysis methods). The resulting summary associated with the transport of coarse bed load, and
acts as a norm for purposes of comparison, and as a the presence of numerous bedforms on the point -bar
framework and guide for future observation. As Dott surface, would obscure the presence oflateral-accre-
and Bourgeois (1983) remarked, fluvial facies tion surfaces and render point bars difficult to iden-
models have "multiplied like rabbits", and this un- tify. It has also been argued that because the dip of
doubtedly makes facies studies more difficult. How- these surfaces is dependent on the width/depth ratio
ever, it also makes them more realistic, as use can of the river, the low accretionary dips would make it
difficult to recognize accretionary geometries in sinuosity and the range of channel and bar orienta-
large, wide rivers (Leeder 1973). As the discussion in tions - two of the key differences between meander-
the subsequent sections should make clear, recent ing, wandering, and braided gravel-bed rivers. As
work has resulted in numerous documented ex- noted elsewhere in this book, fluvial style should not
amples of macroforms in a wide range of fluvial be interpreted from single exposures of one channel-
settings. Nevertheless, the concerns quoted above fill succession, the method tested by Brierley and
are real. The best architectural methods may not be Hickin (1991), but should be based on three-dimen-
able to identify accretionary geometries where expo- sional data from several successions. Such a data
sures are inadequate, for example, where an outcrop base would provide a more quantitative measure of
is oriented parallel to the strike of gently dipping the relative importance of the various facies and the
accretion surfaces. It remains very difficult to docu- types of channel and bar present in the river.
ment macro forms in the subsurface. Several example of intermediate fluvial styles are
It is important to note that almost all the develop- referred to in the succeeding sections. The "wander-
ments that have taken place in the documentation of ing gravel-bed" river and the "low-sinuosity
the two- and three-dimensional architecture of flu- braided-meandering river with alternate bars" are
vial systems have relied on studies of the ancient by definition intermediate fluvial styles. Reference is
record, and several of the fluvial styles described in also made to transitions between the "high-energy
this section are based almost entirely on studies of sand-bed braided" and the "flashy, ephemeral
the ancient. There is a pressing need for research sheetflood" styles, and between the latter and the
into the internal architecture of modern river depo- "sheetflood distal braided style". Differences bet-
sits, preferably using high-resolution seismic or ween some of these fluvial styles are small, and it
ground-penetrating radar methods, such as are de- could be argued that no purpose is served by propos-
scribed in Sect. 9.5. Only work of this type can pro- ing so many "models". The point in doing so is
vide the undisputed link between surface processes simply to recognize the real-life variability of natural
and the preserved deposits that is necessary for the systems and to encourage flexibility in their inter-
reliable definition of process-response models. pretation. Many actual examples of fluvial succes-
What is reported here is therefore work in progress sions will display characteristics of more than one of
toward that goal. the models described here. This is not "sedimento-
One important test of the architectural method in logical anarchy", as suggested by Walker (1990), but
a modern river was carried out by Brierley and a recognition of the complex reality of fluvial sys-
Hickin (1991). They documented the facies se- tems. Brierley (1993) refers to element analysis as a
quences and architectural elements in bank expo- "constructivist" approach to facies studies. Observa-
sures of a modern gravel-sand river in British tion and measurement of two- and three-dimen-
Columbia that undergoes a downstream style change sional outcrops using the architectural-element
from braided to wandering to meandering. Facies approach provide a data base from which measure-
successions and architectural cross sections were ments and calculations can be made concerning
similar for all three fluvial styles, and the authors channel dimensions, flow characteristics, etc. In this
concluded that these kinds of data would not be able way, each river and each fluvial deposit are treated as
to provide predictions of fluvial style in other com- an individual, possibly unique entity. End-member
parable fluvial settings. It is possible to question force-fitting, which is what tends to occur when few
their results because the river is undergoing rapid models are available, is therefore avoided.
evolution, and the surface channel style may not be An excellent example of an ancient fluvial system
an accurate representation of the fluvial style at the that does not fit any of the "standard" models (in-
time the sampled channel and floodplain assem- cluding those described below), and yet contains
blages were deposited. Nevertheless, their conclu- many features tantalizingly similar to several of
sion might seem disturbing. In fact, their results bear them, is the Escanilla Formation (Eocene) of the
out the point this writer has been trying to make, that Spanish Pyrenees. A regional architectural recon-
it is only by carrying out a detailed architectural struction is provided by Bentham et al. (1993), and
documentation that the researcher can hope to ar- additional details of the gravel and sand bodies are
rive at precise and accurate interpretations. In the described by Dreyer et al. (1993). This unit consists
case of Brierley and Hickin's (1991) study, their work of poorly interconnected sheet sandstones and con-
did not include paleocurrent analysis, which would glomerates, separated by fine-grained sediments
provide vital data for the interpretation of channel constituting more than 40% by volume of the total
'Tj
Shallow Deep Gravel Gravel Gravel-sand I»
n
gravel-bed gravel-bed wandering meandering meandering n;'
braided braided '"
~
o
0..
Sr los~ ....?l. !!.
'"
Gmm
minor
channels
Gh crevasse
splay
G~~~..,
St major Fm
St
channel Sr channel channel
with St @I fill fill
debris-flow A macrofonns with with
Sf
deposit Gh Gh
. ~
point Sp point
Sp bar bar
Gp . Sp
Gh
Gh St
Gh Gh
Gh Gh
Gp t5>
~~~~...,.,..~~~
Gt ~ .. q,,~~~
channel
fill
with
alternate
bar
Gh
Fig.8.8A-P. Typical fluvial lithofacies assemblages and vertical profLIes for the 16 fluvial styles discussed in Chap. 8. Vertical arrows indicate cyclic successions of various
types, showing direction of fining and bed thinning. Numbers in circles indicate rank of bounding surfaces. Lithofacies codes are given at left of column. (Adapted from
Miall1977,1978b)
N
o
'"
tv
o
Sandy Ephemeral Fine-grained Anastomosed Low-sinuosity Shallow "'"
meandering sandy meandering meandering braided with perennial
alternate bars braided
+ Sr
i ffi~ eolian dune St superimposed
2-D
Sp dunes
channel
+ crevasse p fill
+ crevasse with
r'pl., Fm I simple
alternate S
Sp bar r
Sp :
SI Q) eolian lens Sp
F1
G> Sl '
•
r-
t
SI
Sf '
:~~' Se (3)
- (3)j St .
Sh
@.t Sp
@
channel @+ -crevasse
fill : -'.. Sp @ splay
Fl Gh
@
,h_ & : . @~ wi.
~~ ~"' I
wIth bar -- @ .Sp
~int
bar
Sr
SI
II _ _ Sr __ .-. _ fill ~~ ~(3)11 chute
channel
<.1>
fill
with
~i nt St .
r,~~I
bar
(3)i truncated longitudinal
(3) channel fill
li.G> ~r:;~1 5
"rl
1 [~ [
Fig.8.8G-L e.
U>
(1)
-s.
§'"
p..
"rl
~
(;i'
'"a:
o
p..
:!.
'"
Facies Models 205
Deep High-energy Sheetflood Flashy

perennial sandy braided distal braided ephemeral
braided sheetflood
i SI Sp
r
SI Sh @i
Sh
i
Sl
SI G>i Sr -"--"
superimposed
® Sh flood
(j)
d) major scour
hollow
@I @ cycles
channel SI
(3)
element
r
(3)1 superimposed
(3)i
@
i
J
Se Sr flood
Sr @ Sh (3) cycles
SI Sp
<Zl major
channel
SI (3)1
51
Se
with
macroforms
5p
5t t @t
Fm
~ St 5h t G)I
i
d)
Sh
Sb crevasse
t
i
splay
~r
i
.
1
i
1 ®
t
i [
Fig.8.8M-P
formation. Bentham et al. (1993) pointed out the Some characteristic fluvial assemblages are illus-
superficial similarity of this broad architectural style trated in Fig. 8.8. Some of these profiles have been
to that of anastomosing rivers, but they concluded adapted from the original model descriptions of
that the geometry and internal features of the coarse Miall (1977, 1978b). Others are new. Each represents
members indicated a braided style. Many of these an "ideal" profile, and should therefore not be relied
consist of simple conglomerate and sandstone sheets upon as a template for interpretation. The local vari-
(elements GB, SB), comparable to those forming in ability in each deposit, and the similarities between
shallow, gravel-bed braided rivers (Sect. 8.2.2) and many of the vertical sections given in Fig. 8.8, should
shallow, perennial, sand-bed braided rivers (Sect. serve to emphasize the need for two- and three-
8.2.12). Ribbon bodies, and sheets with lateral accre- dimensional architectural analysis in sedimento-
tion bedding, indicating sinuous channel reaches logical studies.
(compare with the sand-bed meandering river of The descriptive sections that comprise the re-
Sect. 8.2.7), are also present. This unusual mix of mainder of this chapter follow the organization of
coarse-member architectures is contained within Table 8.3. Gravel-bed rivers, the first to be described,
thick floodplain deposits, itself an unusual feature of are characteristic of proximal, basin-margin set-
a dominantly braided system. Bentham et al. (1993) tings, including alluvial fans. Sand-bed rivers are
attributed this last feature to rapid sedimentation typical of alluvial plains, including more distal set-
and frequent channel avulsion in an unconfined ba- tings. They are conveniently subdivided into those of
sin. high- and low-sinuosity.
8.2.1 Gravel-Bed Braided River one (described in Sect. 8.2.2) in which sediment-
with Sediment-Gravity-Flow Deposits gravity flows are rare or absent, with sedimentation
dominated by element GB (the role of depositional
Many basin margins are characterized by wedges of processes and their deposits in the classification of
gravel deposits. Commonly, these are deposited by alluvial fans is discussed in Sect. 8.3). In both cases,
distributary fluvial systems that form the cone- minor thicknesses of sand and fine-grained deposits
shaped deposits of alluvial fans. Two fluvial styles (elements SB, OF) represent low-water and over-
are typical, one the subject of this section, in which bank sedimentation, respectively. Both types of flu-
sediment-gravity-flow deposits (element SG), re- vial style may occur in the same basin, with the
lated deposits of high-energy stream flow, and their frequency of debris flows and other sediment-gra-
winnowed derivatives make up the successions, and vity flows dependent on the geology and climate of
AGE orMATER IAL

EXPLANATION Pc Pruenl chonn,l
MATERI AL SIZE. (maa.. siu pr.unl) Ovc Overflow chORn,1
~oboble debt-II-flow lobcl Ore Old., channel mat.rial
: .' Pebbae· to c:obblt· siztd MOltrloJ
,.: :. ~. Cla), - to pebble-sized material Ote Old....' chonn,l ma,.-la'
Otbtil-flow leven
All other ma'tflal II cobbl.- yt YOUnt" fan lurfoce
10 boulder-sized Of Old", 'Qn , ... ,fo"
_ _'l-
AOe coni act" o~proalmol'. seMmOIiC'} Hochuru toward old.r Contoured part of map by plan.
loble melhodl. Co",'rol on
CCH'I'ocl b,t.un unit. di"ltujlui,hed b,. "!Oltrlot Wi'll" Q,
topo~roph,. or mollf iol sin. dlltlnct ove rlo pp ln 9 r.malnder b~ oir PholoQfoptll,
oppro.llmOI. , rela tionship .xI,I' Conlour Int,rvol 30 f•• 1.
Eros lanai bonks
Fig. 8.9. Surface of the Trollheim fan, Death Valley, Cali- flow processes (Blair and McPherson 1992), but other fans
fornia, showing the distribution of the major types of where this type of flow is important may also be strongly
fluvial deposit, and the position of the modern fan channel influenced by traction currents and their deposits
(Hooke 1967). This fan is dominated by sediment -gravity-
Facies Models 207
Fig. 8.10. Model of gravel-bed braided river showing dissected lobes of sediment-gravity-flow deposits (element SG).
Modell of Miall (1985)
.... ' t, ~ • .- . ' ~ .
3m
Fig. 8.11. Cross section through postglacial fan deposits, Tasmania, showing interbedding of elements GB and SG.
(Wasson 1977)
the source area. Even adjacent fans may show differ- typical ancient deposit. Note the tabular shape of the
ent fluvial styles because of differences in the source- SG units, their nonerosional bases, and their curved,
area geology, or because of differences in forest convex-up margins. Units composed of element GB
cover, leading to different weathering rates, and con- typically rest on irregular or channelized, erosional
sequent differences in the rate at which loose sedi- bases.
ment is supplied to the fluvial system, so that one fan Blair and Mcpherson (1992) reexamined the
may be formed entirely by traction-current sedi- Trollheim fan, the basis for the debris flow-domi-
mentation, with the next fan significantly affected by nated fan model of Miall (1978c) and Collinson
sediment-gravity flows (Hooke 1967). Debris flows, (1986). They concluded from the field evidence that
and other types of sediment-gravity flow, are par- the Trollheim fan was formed exclusively from the
ticularly common in volcanic areas (e.g., Nemec and products of debris flows, with postflow winnow-
Muszynski 1982; G.A. Smith 1987), but the distinc- ing and reworking processes being responsible
tive characteristics of volcaniclastic deposition are for the modification of some units prior to burial
not addressed in this book. and preservation. They suggested that the vertical
Figure 8.9 illustrates the modern Trollheim fan in profile model of Miall (1978c) is in error in in-
Death Valley, California, which Miall (1978b) sug- cluding traction-current-deposits, including thick-
gested be used as the type example of this style of nesses of Gh and various sandy facies, because they
sedimentation. On the surface of the Trollheim fan did not find these facies in the deposits of the
there are numerous remnants of debris flow depo- Trollheim fan. However, other descriptions of fan
sits. These are initially lobate in shape, immediately deposits by Bull (1972), Wasson (1977), and Schultz
after deposition, but are subject to incision by subse- (1984) clearly indicated that these other facies do
quent steam-flow processes. The edge of each flow occur interbedded with debris-flow deposits in mod-
tends to be abrupt. They have convex-up margins, ern and ancient alluvial fans. Therefore, the facies
somewhat like viscous lava flows. Characteristic model may be incorrect only in the choice of the
lithofacies are Gmm, Gmg, Gci, and Gcm. Other Trollheim fan as the type example. Figure 8.8A illus-
deposits in the Trollheim fan consist of element GB. trates a vertical profile containing these accessory
Figure 8.10 illustrates a fluvial model for this type of facies.
channel, and Fig. 8.11 is a cross section through a
Rust and Koster (1984) suggested that the pre- is deposited (Fig. 8.12; Sect. 5.2). Channel depths
sence of debris-flow deposits should be used as a on the order of 1 m are typical. Channel margins are
primary criterion in the recognition of the alluvial rarely identifiable in outcrop (Fig. 6.17). Element GB
fan environment, a proposal favored by Blair and predominates (Sect. 6.3), and consists of tabular
McPherson (1994). However, the work of Hooke bodies with numerous minor internal erosion sur-
(1967), quoted above, on the controls of sediment- faces, and varying assemblages of gravel traction-
gravity-flow processes, suggests that this criterion current deposits (predominantly lithofacies Gh, Gp,
could produce misleading results. Also, it would ex- Gt). Channels may be abandoned at low stage, in
clude other types of depositional systems that have which case thin lenses aiId wedges of sand may be
traditionally been included in the fan category. deposited (Fig. 8.13), comprising element SB. Thin,
Stanistreet and McCarthy (1993) presented a more sandy, bar-top sheets may be formed, and small
inclusive classification of alluvial fans, as discussed delta-like wedges of sand are formed at bar margins
further in Sect. 8.3. by surface runoff. Typically, element SB comprises
Large-scale autocyclic processes that control the about 5% of most fluvial successions formed in this
development of fourth- and fifth-order litho somes type of river (Fig. 8.8B).
in alluvial fans are discussed in Sect. 10.2. An architectural model for this type of river is
shown in Fig. 8.14. Miall (1977) suggested that this
fluvial style be named the "Scott-type" after the Scott
8.2.2 Shallow, Gravel-Bed Braided River fluvioglacial outwash river in Alaska (Boothroyd
and Ashley 1975; Boothroyd and Nummedal 1978).
Proximal gravel-bed rivers and braid deltas, in Figures 8.15 and 8.16 are outcrop examples of the
which sediment-gravity flows are rare to absent, resulting deposits. Figure 8.15 is most typical of this
consist of a shifting network of unstable, low-sinuo- fluvial style, showing the characteristic thick, multi-
sity channels in which a variety of gravel bedforms story conglomerate deposit formed in gravel-domi-
Fig. 8.12. View of a typical gravel-bed braided river. Athabasca River, Alberta, shown at low-water stage in early spring,
prior to main spring runoff
Facies Models 209
Fig. 8.13. An abandoned channel in a gravel-bed braided river, Athabasca River, Alberta. (Photo courtesy of T.
Jerzykiewicz)
Fig.8.14. Model of gravel-bed river dominated by traction-current deposits (element GB). Model 2 of Miall (1985)
nated alluvial-fan distributaries. In this case, the 8.2.3 Deep, Gravel-Bed Braided River
deposits were probably formed in an arid tropical
environment (Miall and Gibling 1978), but compa- In some braided rivers, several distinct topographic
rable deposits are formed in fluvioglacial settings levels may be recognized, including major and
(e.g., Fig. 5.2; see Chap. 12 for a discussion of the minor channels, bar surfaces, and the floodplain,
climatic significance of the various gravel assem- spaced over a vertical range of several meters. These
blages). Figure 8.16 is an example in which unusual levels may represent stages of progressive down-
thicknesses of overbank deposits have been pre- cutting by the river, and are most recognizable in
served. It seems likely that subsidence was rapid in areas undergoing active degradation. The Donjek
this foreland basin setting, leading to high rates of River, Yukon Territory, Canada, is an example of
deposition and preservation of overbank deposits. this type (Fig. 8.17 A), and was the first to be de-
Fig.8.15. Typical outcrop of gravel-bed river dominated by element GB. Thin lenses and wedges of SB represent the fill
of minor abandoned channels. Peel Sound Formation (Devonian), Somerset Island, Arctic Canada. Scale is 1.5 m
scribed in this way (Williams and Rust 1969, p. Level 2: level active during flood stages, with few
650-652). The floor of the main channel is here at active channels at other times, sparse veg-
least 3 m below the level of the floodplain. Figure etation cover.
8.17B illustrates another example, the Rangitata Level 3: little continuous water movement, low-
River of South Island. Miall (1977) proposed the energy flow during flood stage, moderate
name "Donjek type" as a general name for this type vegetation cover.
of river and its deposits. In rivers undergoing degra- Level 4: dry islands with dense vegetation.
dation, the topographic levels appear as terraces
The lowest level is that of the active channel, and
within the river valley, but similar topographic
is similar in all respects to the gravel-bed rivers of
differentiation occurs in some rivers on open basin
Sect. 8.2.2. Higher levels are active only during flood
plains, for example, in parts of the Scott and Yana
stages, and characteristically accumulate deposits of
outwash "fans" of southern Alaska (Boothroyd and
element SB. A floodplain mayor may not form a
Ashley 1975; Fig. 3), and in some sandur plains
significant part of the system, depending on valley
(Church and Gilbert 1975, p. 62), where the differen-
width and channel stability. Lateral migration of
tiation is constructional. The lithofacies succession
channels, as for example by distributary shifting on
that occurs in this type of river is comparable to
alluvial fans, is accompanied by deposition on the
those in other deep, gravel-bed rivers as discussed in
insides of meander bends and the aggradation of
the sections on those rivers (Sects. 8.2.4, 8.2.5).
abandoned channel reaches, resulting in the super-
Four levels may be recognized in the Donjek
imposition of deposits formed at successively higher
River:
topographic levels, and the generation of fining-
Levell: level of main channels and principal sedi- upward successions (Williams and Rust 1969; Rust
ment dispersal route. Little or no vegeta- 1972). These will be at least as thick as the depth of
tion, bars exposed only during low stage. the channel, but may be greater if they are developed
Facies Models 211
Fig. 8.16. Outcrop of Molasse Rouge, Provence, France. lenses are interbedded with thick sandstone-siltstone-
Stratigraphic top is to the right. Lenses of conglomerate mudstone deposits representing rapid floodplain sedi-
represent channel fills formed in gravel-bed river. These mentation
by distributary shifting on a rapidly subsiding fan. sits may show upward fining, as in the stratigraphic
As shown by Reinfelds and N anson (1993), the upper section measured at this and an adjacent cliff face
topographic levels of this type of river represent (Fig. 8.20). Another example is illustrated in Fig.
what is normally classified as the floodplain of the 8.21. Conglomerate occurs in sheets 1.5-4 m thick,
river. They may have a fair preservation potential, with erosive bases showing up to 2 m of relief (ele-
and accumulate significant, if thin, deposits of finement GB). Tops are esentially flat, but are cut by the
grained sediment (Sect. 10.3.2). scours at the base of the succeeding trough cross-
Figure 8.18 is an architectural model of this type stratified sandstone (element SB). S.A. Smith (1990,
of river. Element GB is predominant, but large-scale p. 212) interpreted the sandstone-conglomerate
gravel bars forming macroform elements LA or DA contact as a fourth-order bounding surface, indicat-
may also be present. An increasing predominance of ing successive stages of deposition within a single
the element LA indicates a gradation into one of the channel.
higher sinuosity fluvial styles described in the subse-
quent sections. Several types of cycle may be present,
cycles meters to tens of meters thick representing 8.2.4 Gravel-Bed, Wandering River
distributary migration and fan evolution (Sect.
10.2), channel-fill cycles up to a few meters thick The definition of this class of river follows from work
(Fig. 8.8C), and those a few decimeters thick repre- in the gravel-bed braided rivers of British Columbia,
senting bar progradation or flood events. Origins of Canada (Church 1983; Desloges and Church 1987;
cycles are discussed further in Chap. 10. Figure 8.19 Brierley and Hickin 1991), but comparable rivers
illustrates a conglomerate-sandstone succession in- have now been studied elsewhere (e.g., Ferguson and
terpreted to have formed in such a river. The depo- Werrity 1983), and it is generally realized that this
Fig.8.17. A Vertical air photo of part of the Donjek River, deep, gravel-bed braided river. Flow is toward the viewer.
Yukon Territory, showing the four topographic levels At least three topographic levels can be defined, based on
(1-4) defined by Williams and Rust (1969). Part of photo the density of vegetation. Note alluvial fans entering the
AlS728-89, National Airphoto Library, Canada. Widest river from both sides. (Photograph courtesy of Lloyd
reach of the river is 3 km across. B Oblique aerial view of Homer, Institute of Geological and Nuclear Sciences Lim-
the Rangitata River, South Island, New Zealand, a typical ited, Lower Hutt, New Zealand)
type of river occupies an intermediate class between sand lithofacies (St, Sp, Sr, Sh) forming the upper
low-sinuosity, multiple channel rivers - the classic part of bar deposits. Fining-upward successions are
braided river - and high-sinuosity, single channel typical (Fig. 8.8D). At the architectural-element
rivers - the classic meandering river (Fig. 8.22). level, the style is dominated by lateral-accretion de-
Sinuosity is intermediate on average (in the range posits (element LA). Slough deposits are floored by
1.2-1.5), although nearly straight and highly sinuous gravel and may be filled with massive silt and clay or
stretches may occur, and the braiding parameter is by cross-bedded gravel or sandstone.
also intermediate (in the range 1-3). In places, these The similarities between the three classes of ma-
rivers have a single channel, in others two or more. jor gravel-bed rivers (those illustrated in Fig. 8.22),
Large, bank-attached point-bars or side bars are in terms of lithofacies and vertical profile, indicate
common. A model for this class of river is shown in the need for careful architectural work to distinguish
Fig. 8.23. them. Orientation data, as obtained from paleocur-
The lithofacies assemblage is similar to that of rent analysis of cross-bedding and clast imbrication,
deep, gravel-bed-braided (Sect. 8.2.3), and gravel- are critical. An example of such a study was reported
meandering (Sect. 8.2.5) rivers. Channel-fill deposits by Billi et al. (1987). They showed that gravel bars
are dominated by lithofacies Gh, Gp and Gt, with accreted downstream and developed a model of al-
Facies Models 213
Fig.8.17B
Fig.8.18. Model of deep, gravel-bed braided river with well-defined topographic levels. Model 3 of Miall (1985)
ternate-bar growth. Where paleocurrent data can be 8.2.5 Gravel-Bed, Meandering River
obtained from successive bar deposits, it may be
possible to determine sinuosity from the change in As shown by Jackson (1978), high-sinuosity rivers
orientation of the bars, using the sinuosity equation range from steep, torrential, gravel-bed rivers, to
of Miall (1976). sluggish, suspended-load streams. Gravel-bed me-
Fig. 8.19. Example of a conglomerate-

sandstone sucession formed in a ma-
jor braided system with topographic
levels, probably an alluvial-fan com-
plex. Peel Sound Formation (Devo-
nian), Prince of Wales Island, Arctic
andering rivers have not received much attention by bars and side-bar complexes. These commonly show
sedimentologists, but increasing use of architectural a downstream decrease in grain size, with gravel
methods to study conglomerate deposits is showing sheets, lobes or dunes at the head, and sand dunes at
that they are relatively common in the stratigraphic the tail. Lateral accretion deposits (element LA) are
record (Ori 1979; Arche 1983; Ramos and Sopena characteristic (Fig. 6.16). Information about the
1983). One of the first studies of this environment floodplain of this class of rivers is sparse, and is is not
was by Bluck (1971), who examined the modern known whether crevassing and crevasse splays are
Endrick River, Scotland. common. Arche (1983) recorded the presence of
There is typically one main, active channel, with fine-grained deposits in abandoned channels, and
scattered bars and islands, and occasional subsidiary thin overbank silt-mud successions, in his study of
channels (Figs. 8.22, 8.24). New channels are com- Quaternary river terraces in Spain. Forbes (1983)
monly initiated as chute channels, as modeled in the showed the modern floodplain of the Babbage River,
flume experiments by Ashmore (1991; see Fig. 6.13). Yukon, to be composed entirely of fine-grained sedi-
Sedimentation occurs on large, flat-topped point ments.
Facies Models 215
B SILTSTONE
'9-:-:':-:-:
An architectural model for this class of river is
shown in Fig. 8.25. A characteristic vertical profile is
IA 0 illustrated in Fig. 8.8E, and examples of ancient de-
a
FINE - COARSE SS.
posits are illustrated in Figs. 6.37 and 8.26. Note the
CONGLOMERATE presence of well-defined lateral-accretion deposits
in these exposures. Other examples are illustrated ~n
~ TROUGH X-STRAT.
ft. Figs. 6.16 and 6.35A.
. . ..
...
.........
.. ... ~
..
~ .; ;
. . . 0. ,0.
o •
8.2.6 Gravel-Sand Meandering River
50
Rivers of this type, in the coastal plain of Texas and
.. -0- .. Louisiana (e.g., Brazos and Amite rivers), were
'9'.·.·.· .
. . .'. b,oQ.;
amongst the first to be studied by sedimentologists
',6," in the 1950s and 1960s (Fig. 2.16; Bernard et al. 1962;
Bernard and Major 1963; McGowen and Garner
1970), and were used in the construction of the clas-
sic point-bar model (Fig. 2.20). Other modern rivers
of this type include stretches of the Wabash River,
Indiana-Illinois (Jackson 1976a,b,d), part of the up-
o
per South Platte River, Colorado (Crowley 1983; Fig.
8.27), the upper Congaree River, South Carolina
(Levey 1978), and the South Saskatchewan River of
east-central Saskatchewan (Campbell and Hendry
1987). They are examples of the so-called coarse-
grained meandering streams. They are bed load
streams, in which the channel and macro form sedi-
ments consist of sand and pebbly sand, or sand with
debris flow a lag of gravel. Abandoned channels and meander
scars are common on the floodplain, and are pre-
served as clay-silt plugs.
debris flow Several excellent studies of hydraulics and sedi-
mentation have been carried out on these rivers.
Jackson (1976d) and Levey (1978) demonstrated the
development of the spiral or helical flow pattern at a
meander bend (Fig. 8.28). This pattern is best devel-
oped in bends of intermediate curvature. During
high-stage conditions, the channel floor and bar sur-
Fig. 8.20. Stratigraphic successions formed in a major faces are covered with numerous bedforms. A
braided alluvial-fan complex. One of the outcrops is illus-
trated in Fig. 8.19. (MiallI970b, 1973)
fathometer trace along one meander wavelength of
Fig. 8.21. Couplets of horizontally bedded conglomerate sets and in some dune troughs (thick black lines). Flow is
(lithofacies Gh) and trough cross-bedded sandstone (St). into the face. Budleigh Salterton Pebble Beds (Lower Trias-
Thin mudstone lenses are present in places above the St sic), southwest England. (S.A. Smith 1990)
A. BRAIDED REACH: Note three to four active channels and ex1ensive

bar platform areas.
B. WANDERING GRAVEL-BED REACH: Note one to two active channels,

sinuous outline, and less bar platform area than up-valley.
C~ MEANDERING REACH: Note single channel, sinuous outline. and

restriction of bar platform areas to the insides of bends.
Fig. B.22A-C. The transition from braided, through wandering, to meandering styles in the Squamish River, British
Columbia, Canada. (Brierley and Hickin 1991)
Facies Models 217
Fig. 8.23. Architectural model for wandering, gravel-bed river
the upper Congaree River showed three types of accretionary bank of the point bar, and occupying
bedforni (Fig. 8.29). Large, straight-crested dunes up the downstream tail of the point bar (Fig. 8.32). The
to 2 m high, termed "transverse bars" by Levey upper point-bar facies consists of interbedded sand,
(1978), occur on point-bar surfaces. In the relatively silt and clay, including layers of cross-laminated and
straight reach between the bends, a population of parallel-laminated sand (lithofacies Sr, Sh), and
smaller, straight-crested dunes ("sandwaves") aver- fme-grained units (FI). These beds have accretionary
aging 85 em in height are the predominant bedform. dips of up to 12°. They are interbedded with the units
Superimposed on these, and in dynamic equili- of St, and typically downlap onto the surface of the
brium, are fields of small, straight to sinuous-crested gravel bar. Their uppermost surface forms the active
dunes ("megaripples") averaging 18 em in height. inner accretionary bank of the point bar. The differ-
Small-scale ripples may be superimposed on these. ence in accretionary dip between these three facies is
Chute channels with large chute "bars" occur in the noteworthy, It has also been observed in the ex-
downstream parts of point bars. Figure 6.38 illus- humed Eocene point bar studied by Nijman and
trates a field of two-dimensional dunes migrating Puigdefabregas(1978}, a model of which is shown in
across the upper surface of one of the point bars in Fig. 8.33. The clay plug deposited after the abandon-
the upper Congaree River. ment of this channel is well preserved, where it rests
A useful study of a river terrace by Campbell and on the cutbank.
Hendry (1987) and a well-exposed Eocene deposit by These examples show the upward fining of the
Nijman and Puigdefabregas (1978) provide many point bar that was noted in some of the classic early
facies and architectural details for the documenta- research (Bernard et al. 1962; Bernard and Major
tion of this type of river. These and other examples 1963). More recent work has shown that this profIle
were used to construct the architectural model is produced only in the median portion of a meander
shown in Fig. 8.30 and the vertical profIle in Fig. 8.8F bend. The upstream part of the bar (bar head) may
(see reference list in Table 8.3). show upward coarsening because of the presence of
In the South Saskatchewan example, the lower gravel lobes that migrate out of the channel and
part of the point bar consists of gravel sheets, typi- across finer bar deposits (e.g., Bluck 1971). Jackson
cally dipping across the channel at accretionary (l976b) termed this the "transitional" zone, because
angles of about 5° or less (Fig. 8.31). Lithofacies Gh is the helical flow pattern is not fully developed at the
predominant. The upstream part of the bar in this entrance to each meander bend. His profIle shows
type of river tends to be coarser than the down- the coarsest part of the section near the middle of the
stream end, which may be composed largely of sand bar profIle (Fig. 6.41). The bar tail may consist en-
(Fig. 8.32). The top of the gravel bar maybe a distinct tirely of one facies, and show no grain-size trends
fourth-order bounding surface, but in places (as in (Jackson 1976b).
the center of the cross section shown in Fig. 8.31) the
gravel interfingers with the overlying facies, which
consists of cross-bedded sand and pebbly sand 8.2.7 Sand-Bed Meandering River
(lithofacies St). This facies develops as lobes that
migrate across the gravel bar platform, filling chute This fluvial style was amongst the first to be exam-
channels and sloughs, migrating up against the inner ined by sedimentologists, and provided us with the
Fig.8.24. Typical gravel-bed meandering river. Babbage River, Yukon, Canada. Part of photo A21924-199, National
Airphoto Library, Canada (see McDonald and Lewis 1973)
Facies Models 219
Fig. 8.25. Architectural model for gravel-bed meandering river. Model 4 of Miall (1985)
~D-.-
• 0 c
\V~E
Fig. 8.26. Example of Quaternary gravel-dominated succession deposited in meandering river. Jarama River Terrace,
central Spain. (Arche 1983)
classic sandy-meandering model of Allen (1963b, rizons are present, and the unit is interbedded with
1970b), as illustrated in Fig. 2.15. It is one of the the deposits of an inland sabkha, suggesting a rela-
most-studied fluvial styles in the ancient record, as tively dry environment, although the presence of
can be seen from the list of selected references in abundant plant and vertebrate remains indicates
Table 8.3, although studies of modern rivers of this that the climate was not truly arid. Accretionary
type are few. The modern Mississippi River is a good bundles within the lateral-accretion deposits com-
example (Fisk 1944, 1947, 1960; Farrell 1987; Jordan monly show cyclic character, indicating that they
and Pryor 1992). Channel and bar sediments are were formed by floods or seasonal high-discharge
dominantly sand, although some intraformational events (Fig. 8.37). Some sandstone sheets are made
conglomerates may be present as channel-floor lag up of thin, overlapping channel-fill bundles or accre-
deposits, where they accumulate as a result of tionary wedges separated by scours, indicating con-
cutbank erosion and caving. The sedimentology of siderable channel instability (Fig. 8.38). Crevassing
this type of river is similar to that of the gravel-sand took place onto well-developed floodplains, preserv-
meandering river, differing only in the details of the ing small abandoned channels filled with alternating
lithofacies of the channel and point-bar deposits. thin mudstone and sandstone beds (Figs. 8.38, 8.39).
Crevasse-channel and crevasse-splay deposits may Another example of a crevasse-splay deposit is
be more common, as suggested by descriptions of illustrated in Fig. 6.21. In this case the environment
this type of river by Farrell (1987), Plint (1983), and was humid, as indicated by the underlying coal.
Munoz et al. (1992)(see Chap. 7 for examples from
the Mississippi River).
Figure 8.34 is an architectural model for this type 8.2.8 Ephemeral, Sand-Bed Meandering River
of river, Fig. 8.8G is a typical vertical profile, and
Figs. 4.9, 6.35B,D, 8.35, and 8.36 are outcrop photo- This fluvial style (Figs. 8.8H, 8.40) differs in detail
graphs of ancient examples. Architectural details of from that of the previous model, based mainly on
a Triassic coastal-plain system are seen in Figs. 8.37 subtle lithofacies features that reflect climatic differ-
and 8.38. Evaporite pseudomorphs and calcrete ho- ences. As discussed in Sect. 12.6, the determination
•
I. , • H4
100m
Fig.8.27. Examples of a coarse-grained meandering Europe and North America, has radically changed from its
stream. Upper South Platte River, near Denver, Colorado. natural state as a result of flood control works, artificial
US Soil Conservation Service photograph (Crowley 1983). channel diversions, and the extraction of water for agricul-
The present style of this river, as with many rivers in tural and other needs
(
Facies Models 221
Fig. 8.28. Velocity patterns around two successive meander

4.2-4.4 1.1 MT. CARMEL
0"1490-1620 M3/SEC
STAGE I~~ ,.,
bends of the Wabash River. The development of the spiral flow
pattern is demonstrated by the divergence in surface and bottom
T W "25-28°C flow vectors. Note the shifting of the position of maximum sur-
face velocity across the channel, and the fact that this does not
NEAR-SURFACE impinge against the outer bank until half way around the Helm
VELOCITY bend. (Jackson 1976d)
\' )~~~~g",
NEAR-BOTTOM
f ~ :g ,
I~ W\
( \9 59 0M
\ ,J6
(i'ilJtll~ HELM IT, '/11
5 (,
~-£!~y:4f}//7
~ 11//r~7
Jfo ""i --
------....::;::
A B
0 ._._._._._._._._._._._._._._._._._._._._._._._._--._._._._._._._._._._._._._._._._._._._.-._._._._._._.- a
·2
• FLOW DIRECTION
0 0
"'~ m
~
::< .4 ·4
I
I
~
~
·6 ·6
·8
POINT BAR 3
• • CROSSOVER •
~--~----~---------~----~--~----__--------________- -__- - - -__- -________-----"
• POIN T BAR 2
• ·8
o 1000 DISTANCE Iml 2000 3000
Fig. 8.29. Fathometer trace through one meander wave- surfaces. Two sets of smaller-scale dunes are in dynamic
length of the upper Congaree River during bankfull disequilibrium in the relatively straight stretch betwen the
charge. Large dunes up to 2 m high occur on point-bar bends. (Levey 1978)
Fig. 8.30. Architectural model for gravel-sand meandering river. Model 5 of Miall (1985)
of climatic signatures from the facies record of sand- these sandstone bodies were 3-5 m deep and up to
bed rivers is a difficult undertaking, especially since 100 m wide. Sections C-G in Fig. 8.42 illustrate com-
most of the discussion is based on interpretations of plex sandstone bodies, in which lateral accretion is
the ancient record, not from studies of modern ex- only one of several mechanisms that appear to have
amples. To what extent, for example, can rivers char- occurred. In places, this has preserved overlapping,
acterized by seasonal discharge, such as those which laterally accreted wedges separated by mud drapes
are interpreted to have deposited the succession and third-order surfaces, as in sections C, F, and G,
illustrated in Figs. 8.37 and 8.38, be distinguished the left side of section D, and the right side of section
from deposits formed in arid environments charac- E. Note the excellent preservation of scroll-bar mor-
terized by ephemeral flow? This section is included phology on the top of the point bar, and the shale-
here to emphasize the need for careful facies analy- filled abandoned channel, in section E.
sis. Floodplain deposits in this fluvial environment
The key indicator of ephemeral flow is that evi- include thin sandstone sheets and mudstones. The
dence of aridity and flashy discharge is abundant. sandstone beds consist of laterally coalesced lenses
The deposits contain numerous internal scour sur- deposited from an anastomosing network of ephem-
faces, and lithofacies Sh and Sl are abundant, indi- eral floodplain channels. As noted by Stear (1983),
cating the common occurrence of high-energy, internal scour and fill structures and siltstone drapes
shallow flow near the transition from subcritical to between accretionary units indicate multiple epi-
supercritical. Mud drapes occur on second- and sodes of erosion and sedimentation. Irregular, un-
third-order surfaces, and these may be associated dulating top surfaces to the sandstone sheets
with desiccation cracks, mud curls, and various indicate erosion by rivulets crossing the floodplain.
types of evaporite nodules and crystals. Wedges and Well-developed levees are lacking in these Beaufort
laminae of eolian sand may also occur along major Group examples, but broad wedges of alternating
bounding surfaces, as in the modern examples de- sandstone and siltstone flank some channel sand-
scribed by Shepherd (1987). These can be distin- stone bodies, and are interpreted as low-relief levee
guished by their distinctive texture, presence of deposits.
eolian ripples, etc.
The architecture of this fluvial style is character-
ized by sandstone ribbons and sheets, the latter 8.2.9 Fine-Grained Meandering River
including lateral-accretion deposits. These sand
bodies may not be very different from those formed Figure 8.43 illustrates a highly sinuous, suspended-
in perennial sand-bed rivers. A brief description of a load stream. D.G. Smith (1987) reviewed this fluvial
modern example in an ephemeral river was given by style and provided several well-documented ex-
Shepard (1987), and ancient examples were de- amples. The overall geometry is similar to that of the
scribed by Lawrence and Williams (1987; upper part sand-bed meandering streams, but differs in detail
of their association 2) and Stear (1983). Figures 8.41 because of the finer-grained sediment load (fine
and 8.42 illustrate Stear's examples, from the Beau- sand, silt, mud). Point-bar accretion surfaces dip
fort Group of the Karoo Basin, South Africa. Stear steeply (up to 25°) and have a simple geometry,
(1983) estimated that the channels which deposited typically planar or with banks and benches (Figs.
'Tl
Il'
('\
Ib'
tI>
FF-21 FF-20 FF-19 FF-18 FF-17 FF-16 FF-I~ FF-14 Sf:

0
341 0-
~
tI>
340
339
338
337
336 ~
>
3~ ~
...
334 '"' '''
333 '"
1>
CD
...
332 <II
w
a:
331
....w
::E
330
329
328
327
326
Fig. 8.31. Cross section through river terrace adjacent to the South Saskatchewan River, east-central Saskatchewan, Canada (dark shading): gravel (lithofacies
Gh); medium shading cross-bedded sand and pebbly sand (St), light shading interbedded sand, silt, and clay (Sr, Sh, FI), (Campbell and Hendry 1987)
tv
tv
VoJ
Fig. 8.32. Reconstruction of meander bend in

the Saskatchewan River, based on data shown in
Fig. 8.31, and other sections. 1 Inner accretionary
bank (lAB); 2 scarp eroded in lAB; 3 sand sheet
with 3-D dunes lapping onto lAB; 4 sand sheet on
gravel platform or filling chute channel; 5 gravel
platform buried by encroaching lAB; 6 slough
(former chute channel) floored by gravel, and or
mud; 7 bar-tail sand; 8 bar-head gravel; 9
microdelta formed by late-stage runoff; 10 flood-
plain; 11 channel. (Campbell and Hendry 1987)
Cutbank Scour Lower 2-D dune Upper Accretionary bank

pool pointbar subfacies pOintbar Sr, FI
Gm St Sp Sr I Chute I
Fig. 8.33. Cross section model of point

/ bar in the Castisent Sandstone, Spain.
(Nijman and Puigdefabregas 1978)
Fig. 8.34. Architectural model for a sand-bed meandering stream. Model 6 of Miall (1985)
6.40,8.44,8.45) indicating downstream flow separa- may form on the point bar, and have been termed
tion and development of incipient scroll bars (Fig. inclined heterolithic stratification (IHS; Fig. 2.38).
2.37; Nanson 1980). Ripple marks are typically the An example of a unit containing IHS is the Athabasca
most abundant flow-regime bedform present (litho- Oil Sands of Alberta, Canada (Mossop and Flach
facies Sr). Gravel lags and cross-bedded medium to 1983; Flach and Mossop 1985), which contains point-
coarse sands may occur at the base of the point bar bar successions up to 25 m thick (Fig. 8.47). Typi-
(typically lithofacies St). Upward fining of grain size cally, these consist of a lower 5-10 m thick
and thinning of bedding may occur (Figs. 8.81, 8.46; succession of trough cross-bedded sandstone, over-
Stewart 1983). Thick, fine-grained floodplain depo- lain by up to 15 m of low-angle, cross-bedded,
sits may be present, including thin sandstone splays heterolithic deposits (IHS), consisting of decimeter-
and coal seams (Figs. 8.44, 8.45). scale couplets of sand and mud dipping at an average
As noted in Sect. 2.4.4.1, this fluvial style is char- angle of 12° (Fig. 8.48). Bioturbation of the upper
acteristic of low-energy estuarine environments, in part of the succession is common.
which the rivers may be subject to tidal influence (D. Other, very well-exposed examples of this fluvial
G. Smith 1987; Thomas et al. 1987; Zaitlin et a1. 1994). style were described by Wood (1989). He demon-
Decimeter- or centimeter-scale sand-mud COUplEts strated variations in lithofacies within the point-bar
Fig. 8.35. Point bar in lower delta plain deposits. Carboniferous, Pocahontas Basin, Kentucky
Fig. 8.36. Close-up of bar seen in Fig. 8.35, showing slump blocks of mudstone at the base of the bar, resulting from bank
caving
LATERAL ACCRETION SURFACES N l00 0 E
::........
:: '.::':.:::.:::':::::::'::::'.:::'::.::'.:::'
..................... ,...........-_.-:. - S h
MINOR SEOUENCES
IN POINT BARS
Fig. 8.37. A lateral-accretion deposit in the Buntsandstein (Triassic) of central Spain, showing details of cyclic accretion-
ary bundles. (Munoz et al. 1992)
NE
NE sw
,< 225· ,,( 270·
'", \.. ,( 210 ,, 205· Ssr
NE sw
Sh
I
Sr o. ~mL
Fm - Fe
o 0 .5m
Fig. 8.38. 1 and 2 Complex sandstone sheets resulting from repeated shifting of small channels. 3 Fill of small abandoned
crevasse channel. Buntsandstein (Triassic), central Spain. (Munoz et al. 1992)
Facies Models 227
Fig. 8.39. Outcrop photograph of the floodplain deposits illustrated in Fig. 8.38, showing small crevasse channel
Fig. 8.40. Architectural model of ephemeral, sand-bed meandering river
Fig. 8.41. Outcrop of ribbon and sheet sandstone bodies evidence of lateral accretion, indicating deposition within
interbedded with floodplain fines, Beaufort Group. Karoo high-sinuosity channels (Fig. 8.42)
Basin, South Africa. Many of the sheets sandstones contain
B!B_ B.Qtt_ ~~~QSTQ~E__ QQQ.'E.S. tv
tv
10 00
o IIIIIIIIIIIIIII ~ ]~;;?'r'
,,"--, . _ . . . 1 _~
~
._ !
o 40 0 60 10m
o
~!:!.EE:r_ S~NQS.I9tiE... _~qQIES
10
~ 7TP --1 i7D25i? C[~ ®

~O - 100 HO 200 250 300m
50
CD
~~
-~~
~~ C0
10 20 3D Um
'TJ
o Sandstone o i:"
• Shale :J.
1 · 6 Storey' sequences ~
Fig.8.42A-G. Ribbon and sheet sandstone bodies in the Beaufort Group, Karoo Basin, South Africa. Note the variable scale. S Scroll bar; P point-bar surface;
'"
~
F shale fill of abandoned channel. (Stear 1983) ~
::l
p...
'TJ
I»
n
Iii'
'"a::
o
p...
~
'"
Facies Models 229
Fig. 8.43. Architectural model of a fine-grained meandering stream. Model 7 of Miall (1985)
deposits from sand-dominated, trough cross-bed- which they branched. These characteristics of anas-
ded units to muddy IHS successions, and suggested tomosed rivers apply in both humid and arid cli-
that these variations reflected varying transport en- mates, in both settings anastomosed rivers are to be
ergies within a meander bend (sand-dominated in found. Floodplains are characterized by muds which
the upper part of the bend, mud dominated down- may contain pond deposits, coals, calcretes, or
stream), much as have been reported from coarse- evaporites, depending on climate (Sect. 12.7). Sandy
grained meander bends. Wood argued that the crevasse-channel and splay deposits are also com-
processes of meander evolution, including the mon. Table 4.4 lists the architectural elements iden-
changes in local sediment load and slope caused by tified in a typical ancient anastomosed fluvial
neck and chute cutoffs, would affect the grain-size deposit, the Cutler Group of New Mexico. Figure 8.8J
and facies assemblage of the point-bar, accounting is a typical vertical profile.
for lateral variations in point-bar style within the Channel bodies are typically straight to sinuous
same macro form. Wood (1989) also reported the ribbon sandstones (e.g., Fig. 6.5). Friend (1983) sug-
presence of ribbon sandstones within a dominantly gested that a width/depth ratio of < 15 is typical of
sheet-sandstone succession, and suggested that ribbon sandstones, but the new data and review
these represented rapid cut-and-fill of channels, fol- provided by Nadon (1994) indicate that a WID ratio
lowed by avulsive abandonment before significant of < 30 is a more appropriate definition. Commonly,
lateral migration could occur. the channel sandstones consist of simple, cross-bed-
ded channel elements, as in most of the channels
described by Eberth and Miall (1991). Figures 8.50
8.2.10 Anastomosed River and 8.51 illustrate typical examples from that study.
Note the presence of minor internal erosion surfaces
The distinctiveness of the anastomosed fluvial style in these sandstone bodies, These are third-order
was established byD.G. Smith and N.D. Smith (1980) surfaces that divide the sandstone bodies into depo-
and D.G. Smith (1983), based on work on small mod- sitional increments. In body B (Fig. 8.51), the incli-
ern rivers in the Rocky Mountains of Canada (Sect. nation of the surfaces suggests local lateral accretion
2.4.4.3). Most of the examples of this type of river and minor lateral channel migration. Kirschbaum
that have been described in the literature have low and McCabe (1992) demonstrated that anastomosed
gradients, and low stream power. Lateral channel channels may fill by lateral or vertical accretion or by
migration is minimal, and the floodplains therefore concentric accretion, in which bedding in the chan-
lack scroll bars and oxbow lakes (Fig. 8.49). Channels nel-fill parallels the channel floor. Nadon (1994)
tend to be isolated - bounded in most places by demonstrated that in his examples the channel sand-
floodplain deposits. As these are normally fine- stones aggraded vertically at the same time as the
grained, channel banks are typically cohesive and levees, so that channel and levee facies are
steep-sided. Channel evolution takes the form of interbedded.
crevassing, and the development of stable crevasse Crevasse channels and splays typically form com-
channels, which may rejoin a main channel down- plex sheet-sandstone bodies cut by numerous inter-
stream and divert the flow from the channel from nal erosion surfaces. Figure 8.52 illustrates two large,
N
OJ)
o
'TI
a-;S.
e.
C/)
'$.
(t)
Fig.8.44. Outcrop of fluvial succession deposited by fine-grained meandering stream. Figure 6.40 is a close-up of part of this outcrop, and Fig. 8.45 provides an interpretation of '"
I»
::l
the entire outcrop. Tertiary, Ellesemere Island, Arctic Canada Q..
'TI
~
;;;.
'"a::
o
Q..
~
'"
'Tl
P>
("\
A :;.
'"
~
op..
~
'"
-::~:::-
---- - -
o tOm
...... ,!eot and
----------====::- ---===== ==::::: ---------- --= == =:::: -::_-::::::::::::- --
hOI'izonlal scole
8 lithOlOOies C
CV~LE 5
.c OFI
OFm [ililI 5
~;iriWf~\'
- - - --- - - - -
Fig. 8.4SA-C. Interpretation oflarge outcrop showing fluvial architecture formed by a fine-grained meandering stream system. Figures 6.40 and 8.44 illustrate parts of this
outcrop. Tertiary, Ellesmere Island, Arctic Canada. (MiaIl1979)
tv
W
LEGEND
Th,n mud loyen Convolute beddl"g
I
(-d'
~
'-4<ldslone I Claystone
Slump, and can'"'''' beds
Siltstone
Sandstone Laleral accretion un.' dip
I I Synceresls crocks ... Cross - beddIng current direction
J Burrows N Anunol Iralls
"--~ Sand lenses ..(:) Dlnosour foot prin's
-
0 Pion' debrrs ErOSion surface
Wood fragmenTs - Fine laminae E (F'9.78)
Pebbles ~! Sivaive shells
C- Cross - bedding (j ~1QmOfP.!!2 burrows
Cross - lamll'lal,Qn 'i!!I Guller scours
Graded sd'i sond laminae M Mottled beds
A B C ?
t
0
---
~-=...
Fig.8.46A-E. Architectural cross section and detailed vertical profiles through the deposit of a fine-grained meandering
river system. Hastings Beds (Cretaceous), SE England. (Stewart 1983)
modern splays, and Fig. 8.53 provides cross sections moderate to poor lateral and vertical intercon-
through two small splays. Note the large difference nectedness, and are separated by flood-basin mud-
in scales between the rivers illustrated in these dia- stones containing interbedded coal seams.
grams. Anastomosed channels of the Magdalena In some large rivers, such as the Brahmaputra, a
River are up to 0.5 km wide and 10 m deep, with form of anastomosis develops, in which normally
splays extending up to 10 km away from the channel. braided channels flow around relatively stable is-
Splay channels are 2-10 m deep. The main channels lands. The islands develop by the process of amal-
of the Saskatchewan River are up to about 5 m deep gamation of braid bars that build up to floodplain
and 100-200 m wide, with splays extending only 1-2 level (Coleman 1969; Bristow 1987; Thorne et al.
km into the floodplain. Where interpretations of an 1993). Switching between channels and the develop-
ancient deposit are made from limited exposure, it ment of new channels occurs by channel aggradation
may be difficult to assess the overall scale of the and overbank flooding.
system. Thus, the main channels of the modern Modern anastomosed channels that had been
Saskatchewan system and the ribbon sandstone bo- identified in the deserts of the Lake Eyre Basin,
dies described by Eberth and Miall (1991; Fig. 8.51) Australia (Rust 1981), are now regarded as a distinct
are comparable in scale to the crevasse channels of fluvial style (Nanson and Croke 1992; Nadon 1994).
the modern Magdalena splay systems. The channels are incised into the floodplain and are
The larger-scale architecture of an inferred anas- not in equilibrium with it, in the sense that channel
tomosed system is illustrated in Fig. 8.54, in which and floodplain deposits aggrade together. Sand in
ribbon sandstones can be seen to comprise less than the channels is derived by erosion of an underlying
50% of the total rock volume. They show generally sand bed, and the floodplains are underlain largely
Facies Models 233
Fig. 8.47. Outcrop of Athabasca Oil Sands, near Fort McMurray, Alberta, Canada, illustrating very large-scale point bars
of the fine-grained-meandering fluvial style. (Photo courtesy of G. Mossop)
by relict mud "braids" composed of sand-sized mud The alternate bar is therefore analagous to the lat-
aggregates (Nanson et al. 1986; Rust and Nanson eral-accretion elements of point bars. However,
1989). there are significant differences to point bars.
Firstly, the direction of accretion is obliquely down-
stream rather than perpendicular to the channel
8.2.11 Low-Sinuosity River, with Alternate Bars margin, as in high-sinuosity rivers. Secondly, the
accretion commonly takes the form of a single, large-
Just as the wandering gravel-bed river (Sect. 8.2.4) is scale, high-angle cross-bedded unit, rather than an
regarded as an intermediate form between braided assemblage of smaller sets advancing down first- to
and meandering patterns, this fluvial style repre- third-order surfaces. Examples oflarge macroforms
sents sand-bed rivers that are intermediate between developed in this way were described by McCabe
braided and meandering rivers. The style is illus- (1977; Figs. 2.27, 6.43). Crowley (1983) recognized a
trated in Figs. 8.8K, 8.55, and 8.56. It contains char- continuum of geometries from the braid bar to the
acteristics of both of the end members of which it is point bar in the Platte River of Colorado and Ne-
an intermediate form, and great care is needed to braska, and designated the intermediate form the
distinguish this fluvial style. South Platte-type macro form (Fig. 8.56). Wizevich
In some respects, the fluvial style is that of a (1992b) distinguished alternate bars as a special type
braided river with only one active channel. Channel of downstream-accretion element, to which he gave
sinuosity is low, and there may be both midchannel the field code DA2 (Fig. 8.57).
and bank-attached bar forms. The main in-channel Other modern and ancient low-sinuosity fluvial
architectural element is the alternate bar. This ele- systems have been described by Okolo (1983), Bridge
ment is a macroform comparable to a large braid bar et al. (1986), and Olsen (1988). Bridge et al. (1986)
but deposited by accretion of a bank-attached bar. described a process whereby bank-attached bars
Fig. 8.48. Close-up of inclined heterolithic stratification in the Athabasca Oil Sands of Alberta, showing the decimeter-
scale interbedding of sand and mud and the abundant bioturbation. (Photo courtesy of G. Mossop)
Fig. 8.49. Architectural model of an anastomosed river. Model 8 of Miall (1985)
evolve into midchannel forms by the development of 8.2.12 Shallow, Perennial, Sand-Bed Braided River
chute channels (Fig. 8.58). The resulting macroform
may evolve in several ways. Figure 8.59 illustrates This is the classic "Platte-type" braided river, the
one model, in which the chute channel fills with model for which was proposed by Miall (1977), based
sediment and the bar subsequently accretes laterally mainly on work by N.D. Smith (1970, 1971, 1972) on
into the larger channel, as flow is concentrated there the Platte River of Colorado and Nebraska. Later
and the incipent bend translates downstream. work on this river by Blodgett and Stanley (1980) and
Little information is available on the nature of Crowley (1983) is also of importance (Sect. 2.4.4.2).
floodplain deposits of this type of river, but it is The William River of northern Saskatchewan is an-
anticipated that they would be similar to those of other example of a modern river of this type (N.D.
meandering rivers. Smith and D.G. Smith 1984). An architectural model
Facies Models 235
-1 shallow channel occupying almost the full width of

A 2 the floodplain. As a result, fine-grained floodplain
deposits constitute a minor part of the lithofacies
assemblage. The Platte River itself no longer shows
these extremes of stage because of human interfer-
ence, including the presence of dams and the
withdrawl of water for agricultural purposes.
The main architectural component is the pres-
ence of extensive sheets of in-channel, lower flow-
178·
regime dunes constituting element SB. The deposits
B are typically dominated by either lithofacies Sp or St,
1
but not both together, because these lithofacies rep-
resent dune types formed under different hydraulic
conditions. Extensive references to examples are
given in Sect. 6.5, and a typical example is illustrated
in Fig. 4.5. Simple macroforms may be present, such
as the cross-bedded or plane-bedded simple bars of
Allen (1983a), and the Platte-type macro forms of
Crowley (1983), but available information on these
27m
deposits indicates that the lithofacies sets are typi-
cally not organized into higher-order lithosomes,
and bounding surfaces of fourth-order are uncom-
mon. The architecture of the deposits tends to be
that of simple, tabular sandstone sheets. Cyclic suc-
cessions of autogenic origin, such as are common in
deeper rivers of braided and meandering type, are
rare in Platte-type river deposits.
8.2.13 Deep, Perennial, Sand-Bed Braided River
In braided rivers larger and deeper than the Platte,

the channel topography shows greater differentia-
tion between channel floor, bar, and bar-top envi-
Fig.8.S0A-D. Outcrop patterns of four typical ribbon ronments, and the corresponding depositional
sandstones in the anastomosed Cutler Group, New facies are similarly more complex and varied. Most
Mexico. Heavy arrows indicate paleocurrent directions,
of the constructional activity takes place within large
small arrows with numerals indicate positions of vertical
sections illustrated in Fig. 8.51. (Eberth and Miall 1991) macroforms that have variously been termed com-
pound bars (MiallI977), sand flats (Cant and Walker
1978), or sand shoals (Allen 1983a). Miall (l978c)
is illustrated in Fig. 8.60, and aerial photographs of suggested that the South Saskatchewan River of
the Platte River are shown in Fig. 8.61. Figure 8.8L southern Saskatchewan provided a good type ex-
reproduces the vertical-profile model of Miall ample of this type of river, based on the work of Cant
(1977). and Walker (1976, 1978). An architectural model is
The fluvial style is characterized by the presence illustrated in Fig. 8.62.
of fields of large, flat-topped three-dimensional The most characteristic depositional feature of
dunes, termed linguoid bars in the older literature this type of river is the presence of macro forms that
(see Ashley 1990), which are active during high stage. contain internal accretionary geometries, quite dif-
Internally, these are constructed of planar cross-bed ferent from the tabular bedding style typical of the
sets. Smaller 3-D dunes building trough cross-bed shallow braided rivers described in the previous sec-
sets may occupy deeper channels. These rivers are tion (Figs. 8.8M, 8.63). The macroforms are bounded
braided only during low discharge stages, when the by fourth-order surfaces (which mayor may not be
tops of the dunes become exposed. At times of high preserved, depending on the amount of scour that
discharge there may be a single, very broad, but occurs between depositional events), and commonly
A
7 6 54 3 2
SI GM
51
~ ~
5c L
Om
10m
6 5
B
50C~
-=l
10m
7 5
2 0 6 4
50cmL -
50cm L - 10m
10m
Fig.8.S1A-D. Vertical profIles of the channel sandstones illustrated in Fig. 8.50. (Eberth and Miall1991)
Fig.8.S2. Maps of two large splays of the modern Magdalena River, Colombia. (Smith 1986)
Facies Models 237
V.E . • 15:1 .._ _ _ _-.,

o s'om
1 I
FACIES
&?WI Sand
1[-=-8 Silt
_ S il ty Clay
• Peat
o som
V.E. - 15:1
""---~'
Fig. 8.53. Cross sections through two abandoned splays in the Cumberland Marshes, Saskatchewan. (N.D. Smith et al.
1989)
B B'
9 10 11 12
C03l',enin8·u~"'ard
~g~tals~~fn lce/
Shelly mud "on~
m3nne
Ribbon bodies .
an asl o mo~c: d chaonc: ls
Tabu 131' s3ndSlone &
mud:~.mn '-"·
nood basin depos ilS
Coal z.Ones . m
ires
Sandstone wil n
con g ~ omer.1 te l :ly.:rs. .
braided 5oUCan1S
Fig. 8.54. Cross sections through an interpreted ancient ribbon sandstones with floodbasin mudstones and coals
anastomosed fluvial deposit, Dakota Sandstone, southern (unit M in the legend at right). Note the relatively low
Utah. The anastomosed interval is that characterized by paleocurrent dispersion. (Kirschbaum and McCabe 1992)
Fig. 8.55. Architectural model of the low-sinuosity river, with alternate bars
consist of successive growth increments separated 10 km in width, with a maximum depth of 45 m. Even
by third-order surfaces. Many of the macro forms the smallest-scale individual channels are up to hun-
show an upward decrease in grain size or the scale of dreds of meters wide. As shown by Bristow (1987,
sedimentary structures (Haszeldine 1983a,b), and 1993b), the internal arrangement of channels and
fine-grained bar-top facies, including thin muds and bars and the architecture of the macro forms of this
carbonaceous deposits, may be preserved (Cant and river are not unlike those of smaller braided rivers,
Walker 1978). Therefore, vertical profiles through although greatly scaled up. A particular challenge in
the macroforms commonly constitute fining-up- the analysis of this type of river is therefore to deter-
ward successions. The macro forms may be classified mine the scale of the river and its constituent chan-
as downstream-accretion elements (DA) or lateral- nels and macroforms from the complex hierarchy of
accretion elements (LA), or may contain both geo- preserved lithosomes. An ancient example of this
metries in different parts of the same bar complex. type of river, comparable in scale to the large mod-
Their sedimentology is described in detail in Sects. ern rivers of the Indian foredeep, was described in
6.6 and 6.7. detail by Willis (1993a,b). Bars in this ancient system
Early descriptions of rivers of South Saskat- are up to 3 km in length.
chewan type, by Cant and Walker (1976, 1978) did
not discuss the three-dimensional geometry of the
macroforms because little deep trenching was car- 8.2.14 High-Energy, Sand-Bed Braided River
ried out as part of that study. Subsequent work in
ancient deposits (references in Table 8.3) has shown As discussed in Sect. 6.9, a distinctive type of scoop-
that accretionary geometries are common in ancient shaped architectural element, termed a hollow (ele-
braided river deposits. However, it needs to be em- ment code HO) is common in certain types of
phasized that this work, and the distinction that it braided systems, and may be rare or absent in others.
suggests between the "shallow" and "deep" sandy The recognition of this element type in the ancient
braided systems described in this and the preceding record is based on the work of Cowan (1991) in the
section, are based entirely on architectural interpre- Westwater Canyon Member of the Morrison Forma-
tations of the ancient record. Data derived from tion' New Mexico. Another characteristic feature of
small-scale trenches in modern sandy braided rivers this unit is the abundance of plane lamination
reveal very similar lithofacies assemblages. There is a (lithofacies Sh) and low-angle cross-bedding
need for architectural studies of modern sandy (lithofacies Sl), which Cowan (1991) interpreted as
braided deposits, perhaps using ground-penetrating the result of common transitional- to upper-flow-
radar, comparable to the studies of modern point regime conditions. The combination of these fea-
bars reported by Gawthope et al. (1993; see Sect. tures suggests a distinctive fluvial style, in which
9.5.5). sedimentation takes place during high-energy, pos-
Sandy braided rivers vary enormously in scale. sibly shallow, discharge events. These events lead to
The reaches of the South Saskatchewan River de- considerable scour of the channel floor, and the
scribed by Cant and Walker (1978) are about 600 m planing off of the tops of macroforms such as DA and
wide, with individual channels ranging from 70-200 LA units, so that these may not even be recognizable
m in width. They average 3 m in depth, with deeper in the resulting deposits. Scour hollows are formed
scour pools. The Brahmaputra River, above its at channel confluences, at tributary junctions, and
confluence with the Ganga, in Bangladesh, averages below macro forms where flow converges. The rapid
100m 100m
Fig. 8.56. Aerial photographs of the South Platte River, Colorado, showing reaches that are organized into low-sinuosity
channels with alternate bars. Photo from the US Soil Conservation Service. (Crowley 1983)
Small scale dunes superposed Fig. 8.57. Block diagram model of an alternate bar in
on stoss side of macroform Subordinate lateral accretion the Pennsylvanian Lee Formation, central Appalachian
by small scale dunes on Basin. The main slip face is oriented in a downstream to
obliquely-downstream direction. (Wizevich 1992b)
I Sm
A B
Fig. 8.58. Evolution of channels and bars in low-sinuosity, following chute cutoff. successive stages in channel posi-
single-channel rivers, based on the Calamus River, Ne- tion are shown by the numerals 1, 2, and 3. (Bridge et al.
braska. A Initiation of chute cutoffs (dashed arrows). CB 1986)
Chute bar; SB scroll bar. B Changes in channel geometry
flow conditions are the cause of the distinctive rivers was well described by Williams (1971). An
lithofacies assemblage. An architectural model of architectural model is given in Fig. 8.64. Terminal
this fluvial style is shown in Fig. 6.44, and a typical fans (Sect. 8.3) may consist partly or largely of the
vertical profile is given in Fig. 8.8N. deposits formed in this fluvial style (Parkash et al.
Although the HO element has been observed in a 1983; Kelly and Olsen 1993).
few other stratigraphic units (Miall 1993; Stephens The sediments are dominated by sheets, lenses,
1994) it does not constitute a major component and wedges of element SB, consisting of a wide range
of the architectural assemblage of these units, and of sandy lithofacies. Flood cycles up to a few meters
it remains for future work to demonstrate whether thick may be preserved (e.g., Miall and Gibling 1978;
the Westwater Canyon Member of the Morrison Munoz et al. 1992; Figs. 6.22, 8.80, 8.65). Fine-
Formation is anomalous. It is suggested that a grada- grained overbank deposits are likely to be rare to
tion into the flashy, ephemeral fluvial style (Sect. absent, as the rivers, when active, tend to overtop
8.2.16) may occur in some cases. This fluvial style their banks and form broad sheets transporting bed
has not yet been documented in any modern river load.
system. A variation on this fluvial style has been identified
in the arid basins of interior Australia, where it now
appears that thick mud beds are not overbank de-
8.2.1 S Distal, Sheetflood, Sand-Bed River posits but were formed as accumulations of sand-
sized, pedogenic mud aggregates (Nanson et al.
This model is characteristic of distal braid plains, 1986; Rust and Nanson 1989). These can be trans-
particularly in arid regions where ephemeral runoff ported as sand grains and may form hydrodynamic
forms a network of shallow, interlacing, possibly sedimentary structures, but the distinctiveness of
poorly defined channels. The sedimentology of such these may be lost as a result of diagenesis. Micro-
Facies Models 241
UPSTREAM
\ !
" ~~~~~_;:;C;;T=I=V:::E=-C_H_A_N_N_E_L ____ 7-
DOWNSTREAM
"- I t \
1m ~
VC VERY COARSE - GRAINED SAND
~
PEAT, WITH SAND AND SILT C COARSE
::::::: VEGETATION-RICH,SMALL-SCALE CROSS-STRATIFIED SAND m MEDIUM
LARGE-SCALE CROSS-STRATIFIED SAND I FINE
"0 UNSORTED SANDS AND GRAVELS vI VERY-FINE
Fig. 8.59. Generalized model of a midchannel macroform channel at left. The main active channel is shown at right.
in a low sinuosity river. Three sections are shown, oriented Late-stage lateral accretion deposits are shown on the
perpendicular to channel trend. At the center, are ob- midchannel flanks of this channel, reflecting downstream
liquely downstream-oriented accretion deposits of a mid- translation of the adjacent meander bend. (Bridge et al.
channel macroform consisting of cross-stratified sand and 1986)
gravel. Fine-grained, vegetation-rich deposits flll a chute
Fig. 8.60. Architectural model for the shallow, perennial, sand-bed braided river. Model 9 of Miall (1985)
scope examination of the mudstones may, however, ter described in this section. Ancient examples
reveal relict "clotted" textures and other pedogenic include the interbedding of the fluvial Kayenta and
features. Ancient examples of this fluvial style have eolian Navajo formations in Arizona (Middleton
been described by Rust and Nanson (1989), Ekes and Blakey 1983; Herries 1993), the Silurian
(1993), and Talbot et al. (1994). Tumblagooda Sandstone in Western Australia
A particularly common association in hot, arid (Trewin 1993a), Triassic rift-fIll deposits in the Irish
regions is that between eolian dunes, interdune Sea area (G. Cowan 1993), and the Permian
pond deposits, and ephemeral-fluvial flash-flood Rotliegende Formation of the North Sea Basin
deposits. Fluvial deposits typically have the charac- (Glennie 1972, 1983; George and Berry 1993).
242
Fluvial Styles and Facies Models
100m
Fig. 8.61. Aerial views of the Platte River, Nebraska. Photos from US Soil Conservation Service. (Crowley 1983)
Facies Models 243
Fig. 8.62. Architectural model for the deep, perennial, sand-bed braided river. Model 10 of Miall (1985)
GENERALISED SECTION
Sb.
-
... .. EroSion sul'fOC'
~ Ac·CftllC:\ 'fOnt -
~:: } tlow dlrKtion
Fig. 8.63. Vertical profile through a macro-

,: TBI, t.
IB form in a deep, perennial, sandy-braided
I'.~~b re> J
1 ' I
g.' ._" ;: .~.:,:.,,,.,,: ,' < ~ - ' _ ' : 0 ,'" •

river, showing architectural reconstructions
of the cross-section profile and plan view of
", T co tb
the bar. (Ramos et al. 1986)
Fig. 8.64. Architectural model of the distal, sheetflood, sand-bed river. Model 11 of Miall (1985)
8.2.16 Flashy, Ephemeral, Sheetflood, Sand-Bed River the creek of that name in Colorado, which had been
described by McKee et al. (1967). The difference
This fluvial style was distinguished from the distal, between this and the previous model is that in this
sheetflood braided model by Miall (1985) based on a model high-velocity, flashy discharge leads to the
recognition of differences in the lithofacies assem- preferential deposition of transitional to upper flow-
blages. Miall (1977, 1978c) had proposed a vertical- regime beds, dominantly plane-laminated sand
profile facies model named the "Bijou Creek-type", (lithofacies Sh) and low-angle cross-bedded sand
based on a comparison with flash-flood deposits in (Sl). Together, successions of these lithofacies con-
Fig.8.65. Typical vertical profiles of the "second stage" of

Upper Buntsandstein (Triassic) sedimentation, showing flood
cycles. (Munoz et al. 1992)
-- S~
~m----- __~__~--___~
Fm
- - - - Sr
- - - - Co++
Fig. 8.66. Architectural model ofthe flashy, ephemeral, sheetflood, sand-bed river. Model 12 of Miall (1985)
Fig.8.67. Simplified lateral profile and interpreted block-diagram facies model for part of the Kayenta Formation
(Jurassic) of the Mesa Creek area, western Colorado. (Bromley 1991a)
Alluvial Fans and Other Fluvial Distributary Systems 245
stitute sandstone sheets of the element LS (Figs. 8.8P, upon a plain or broad valley, or where a tributary stream is
8.66). Cotter and Graham (1991), in their studies of near or at its junction with the main stream, or wherever a
constriction in a valley abruptly ceases or the gradient of
ephemeral-fluvial sandstones of Devonian age in the stream suddenly decreases; it is steepest near the
Ireland, described examples of lithofacies Sl which mouth of the valley where its apex points upstream, and it
they likened to hummocky cross-stratification, and slopes gently and convexly outward with gradually
concluded that the facies represented deposition decresing gradient."
from transitional to upper-plane bedforms, similar Miall (1990, p. 345) suggested that:
to those described by Cowan (1991). Channels are
poorly defined or absent altogether. Flood cycles "the term alluvial fan may be used for any entirely
nonmarine and nonlacustrine fluvial system whose chan-
may be recognizable, showing cut-and-fill relief nel network is distributary rather than contributary. Fans
along their basal bounding surfaces. Overbank fines that prograde directly into a standing body of water are
are rare to absent. Antidunes are common in the termed fan deltas or coastal alluvial fans .... Alluvial fans
modern stream deposits described by Langford and and fan deltas are a sedimentary response to flow expan-
Bracken (1987), but do not appear to have a very high sion at a basin margin."
preservation potential. Evaporite crusts may occur The term "fan" places natural emphasis on the
on the upper surface of sand beds, and disruption of surface geomorphology of this depositional system.
bedding by evaporite crystallization is common The radiating distributaries and the cone-shaped
(Mertz and Hubert 1990). architecture of the deposit are the two features that
The succession of events in a single flood cycle would be the key elements of most geologists' defini-
was described by Bromley (1991a), based on his tion of the term alluvial fan. However, there is con-
studies of the Jurassic Kayenta Formation of western siderable controversy amongst sedimentologists as
Colorado. He mapped a broad channel with gently to the scope of the definition beyond this initial
dipping margins that appears to have focused flow in point. Miall's (1990) definition (which reflects wide-
the initial stages of the flood (Fig. 8.67). At the begin- spread usage) implies that the term has no unique
ning of the flood event, a phase of channel-floor facies sense. As noted by Miall (1992, p. 134), there
erosion and dune sedimentation occurred. During are small, steep, gravel-dominated fans, within
peak discharge, flow extended in a shallow sheet over many of which debris-slow processes are important;
a broad area, beyond the confines of the original, there are sandy systems deposited by braided
poorly defined channel, and sheets of Sh were depo- streams; there are giant modern fans, such as the
sited. During the waning flood stage, incision of Kosi of India, which grades from boulder conglo-
these sheets occurred, and pockets of two- and three- merate near the mountains to fine sand-silt-mud 140
dimensional dunes (lithofacies Sp, St) were depo- km downslope at its distal end; and there are fine-
sited. grained, arid systems, termed "terminal fans" de-
posited by ephemeral flows that percolate and
dissipate into the substrate at their distal fringes.
Geomorphologically, these are all alluvial fans, and
8.3 Alluvial Fans
all fall under the definitions of this term provided by
and Other Fluvial Distributary Systems Bates and Jackson (1987) and Miall (1990, 1992). A
recent review by Stanistreet and McCarthy (1993)
Alluvial fans are a form of fluvial depositional sys- has followed this inclusive approach. They described
tem distinguished on the basis of geomorphic char- a new class of large, low-energy fan based on the
acter rather than by a characteristic fluvial style. As example of the modern Okavango fan of Botswana,
discussed here, the term is in broad, general use and erected a simple classification of alluvial fans,
amongst sedimentologists, but there is considerable including all the types mentioned above. This is
confusion and disagreement as to the range of types described below. They also discussed comparisons
of fluvial depositional systems that should be in- with submarine fans, a class of depositional system
cluded under the heading of alluvial fans. -about which there have also been problems of defini-
An alluvial fan is defined in the Glossary of Geo- tion.
logy (Bates and Jackson 1987, p. 17) as: Contrasting with this approach is that of
McPherson and Blair (1993) and Blair and
"A low, outspread, relatively flat to gently sloping mass of
loose rock material, shaped like an open fan or a segment McPherson (1994). These authors accepted many of
of a cone, deposited by a stream (esp. in a semiarid region) the standard components of the geomorphic defini-
at the place where it issues from a narrow mountain valley tion of fans, noted above, including their conical,
radiating form, their steep depositional slopes, and (Holmes 1965; Gole and Chitale 1966; Wells and Dorr
their association with high-relief source areas. They 1987; Singh et al. 1993) has been cited as a "textbook"
pointed out a significant contrast between fans and example of one style of channel evolution within fans
most other fluvial systems. The latter typically ex- (e.g., Collinson 1986). Also, the term "terminal fan"
hibit concave-up transverse profiles, the channel be- (Mukerji 1976; Friend 1978; Parkash et al. 1983) is
ing the lowest point, whereas fan systems are now widely used (e.g., Kelly and Olsen 1993), but
characterized by convex-up transverse profiles, most terminal fans are not alluvial fans according to
which lead to overbank flow and sheet flooding of the definition of Blair and McPherson (1994).
bed load. However, these authors argue that there is Continuing detailed work on modern and Recent
a distinct break in the range of longitudinal slopes fans suggests that the proposals of Blair and
exhibited by the various types of fluvial distributary McPherson (1994) may need modification. Nemec
system. They recommended limiting the term allu- and Postma (1993) provided a detailed description
vial fan to steep systems, which have a range of slopes of some Quaternary alluvial fans in Crete that do not
between IS and 25 0 (slope: 0.026-0.466). These fans fit their definitions. A suite of fan deposits there has
are characterized by sediment-gravity-flow deposits, preserved depositional dips of up to 13 0 , yet the
especially debris flows, by hyperconcentrated flows deposits consist largely of clast-supported stream-
and sheetfloods. Traction-current deposits are rare flow gravels forming gravel sheets (element GB).
to absent. According to the same authors, fluvial Use of the restricted definition of Blair and
depositional systems having slopes in the range McPherson (1994) will lead to the recognition of fans
0.4-1S (0.007-0.026) are rare, with most meander- in the ancient record on the basis of their facies
ing and braided fluvial systems, including the radial, characteristics alone, the deposits being coarse-
distributary systems that many authors would clas- grained debris-flow and related facies banked
sify as fans, exhibiting slopes less than 0.4 0 • This against an uplifted source area. Such classification
provides a natural break, which they proposed to already is implied by the use of such terms as
emphasize in their classification. They recom- "fanglomerate" for ancient conglomerates, even
mended excluding depositional systems on slopes of where evidence of the geomorphology of the original
< IS « 0.026) from the category of alluvial fans, depositional system is not available. Such a designa-
mainly on the basis of an historical analysis of the tion will likely be suggested without evidence of the
evolution of the term fan. They pointed out that the distributary nature of the fluvial system, a feature
original work of such researchers as Drew (1873) and that is difficult to prove in the rock record without
Surell (1841, 1870), which helped to establish the detailed paleocurrent evidence. Whether the use of
term "fan", dealt with small, steep, coarse-grained the Blair and McPherson (1994) definition in this
fans (Sect. 2.2.2). Heward's (1978a) important review way will prove to be useful and will serve to reduce
of fan depositional cycles also focuses on coarse- controversy and confusion remains to be learned
grained fans, although he did not discriminate or from experience. In the meantime, the more inclu-
subdivide fans on the basis of sediment transport sive classification of Stanistreet and McCarthy
mechanisms. Blair and McPherson (1994) suggested (1993) is provided below.
that the inclusion of fan-shaped glacial outwash dis- In this classification (Fig. 8.68), it is recognized
persal systems under the heading of the term alluvial that various fluvial processes may occur in different
fan serves no useful purpose. proportions, depending on climate, the nature of
Following this approach, fine-grained, fan- the source terrane, etc. This led Stanistreet and
shaped deposits, such as those of the Kosi River, McCarthy (1993) to propose a triangular classifica-
terminal fans, and those not associated with moun- tion that illustrates the varying importance of the
tainous uplifts, would not be included under the three main processes, sediment-gravity flows, braid-
heading of alluvial fans. They would be termed ing, and meandering (Fig. 8.69). The first class of fan
distributary fluvial systems or braid deltas. While in their classification is the debris-flow-dominated
this proposal is clear, and may be applied with some fan, the only type accepted by Blair and McPherson
consistency, at least to modern environments where (1994) as a "true" fan. These fans are small, typically
slopes can be measured, it leaves many depositional less than 10 km in radial length (Fig. 8.70), and steep
systems and deposits that have traditionally been (slope> 0.1). They are particularly common in arid
termed fans out of the classification. This may be of regions, where sediment-gravity flow processes are
little importance in the case of sandy systems, such . more common (but are not exclusively found in this
as the Kosi fan, even though the pattern of channel climatic setting, as detailed in Sect. 12.5). Blair and
wandering and avulsion documented on this fan McPherson (1994) recognized two varieties. The first
type are those constructed mainly of sediment gra- Stanistreet and McCarthy (1993) termed the
vity-flow deposits, together with colluvial slides, losimean fan, an abbreviation of "low sinuosity-
bedrock slides, rock avalanches, and rockfalls. meandering". The latter type is exemplified by the
Sandy interbeds are not common. Secondary pro- Okavango fan of Botswana, the largest subaerial fan
cesses, including overbank sheet flow, wind erosion, system yet described, at 1SO km in axial length.
weathering, bioturbation, soil development, and Depositional processes in these braided and mean-
case hardening are common. The second variety is dering systems are those of the respective fluvial
built primarily of sheet flow, and hyperconcentrated styles, described in Sect. 8.2. As discussed by
flow deposits. Sandy overbank deposits are com- Stanistreet and McCarthy (1993), such "fluvial" fans
mon, and a distal sand fringe may also be present. as the Kosi and the Okavango, are difficult to recog-
The other two types are the braided fluvial fan nize in the ancient record, particularly in the subsur-
(e.g., tributary fans seen in Fig. 8.17B), and what face where the absence of paleocurrent data would
DEBRIS FLOW
INNER FAN:
Chonnelised proximal ----------/.~-A
debris flows
~ID FAN :
Distal debris flows
and sheet f:ows
OUTER FAN :
Sheet flaws and
shallow braided stream FAN SURFACE STEEP
(SLOPE -0.1 -0.01)
LITTLE VEGETATION
BRAIDED FLUVIAL FAN

INNER FAN:
Longitudinal grovel bars
~ID FAN :
Longitudinal grovel bars
and transverse sand bars
OUTER FAN :
Transverse bars ------::r'S""---~
and dunes
FAN SURFACE NOT STEEP

(SLOPE - 0.001-0.0003)
SOME VEGETATION ALONG
CHANNELS
LOW SINUOSITY/MEANDERING FLUVIAL FAN

INNER FAN:
Active meander belts run
through permanent swamps
with vegetated levees
~ID FAN:
Low Sinuausity/ Anastomosing
river with vegetated levees
OUTER FAN:
FAN SURFACE SHALLOW

(SLOPE -0.0003)
HIGHLY VEGETATED
Fig. 8.68. A classification of alluvial fans. (Stanistreet and McCarthy 1993)

DEBRIS flOW
PROCESSES Fig. 8.69. A triangular classification of
rAN TYPES
alluvial fans, showing the subdivision
into the three main depositional styles.
[ill DEBRIS FlOW DOMINATED
o BRAID DOIoiINATEO
(Stanistreet and McCarthy 1993)
S LOSIMEAN OOIoiINATED
• RECENT FANS
.. ANCIENT FANS
~I~~~~~~~~~~~~~~~~~~~~~~(O~VAN~ FAN
lOW SINUOSITY I
t.4EANDERING RIVER
(lOSIt.4EAN)
DEBRIS FLOW FANS LOW SINUOUSITY/MEANDERING

HANAUPAH ,AN (LOSIMEAN)
/),~
a ':,Qo1.
l
(DEATH VALLEY) FLUVIAL FANS
1° 0 :o~ooo\
I , "oq,f 01
,
\.,t !o 0 0 1
/ ,'0/7"'0l~
i-- "
TROLLHEII.4
,AN
(DEATH VALLEY)
,j<
...
""~
,AN IL...V
(S.E.ALASKA) \
OKAVANGO
,AN
(N.W.BOTSWANA)
... _ _ ..,/1
/-
.,
N
, . . ......
... cz;/ Ie
!
20 lO 40 tIC
, ! ! ,
'm
Fig. 8.70. A comparison of the sizes of modern alluvial fan systems. (Stanistreet and McCarthy 1993)
make the radial dispersal pattern almost impossible Some ancient fluvial deposits are marked not only
to prove. Two examples of sandy ~ystems from the by a downstream decrease in grain size, but by a
ancient record were described by Nichols (1987) and decrease in interpreted channel scale, and a distal
Hirst (1991). A gravelly system was described by transition into fine-grained flood basin or lacustrine
Miall (1970a). In the first two cases, paleocurrent deposits. Friend (1978) was the first to discuss the
data were used to reconstruct radial dispersal pat- sedimentological implications of these characteris-
terns. In the gravel example, clast counts were inter- tics. He suggested the terminal fans of India as
preted in terms of radial dispersal within a system of analogs for these ancient systems (Mukerji 1976),
laterally coalescing alluvial fans. and a modest literature has now developed in which
Facies associations :
@=
Fig. 8.71. A facies model for terminal fans. 1 Feeder zone, possibly with more than one channel; 2- 4 correspond to
proximal, medial, and distal fan zones; 5 floodbasin. (Kelly and Olsen 1993)
his embryonic ideas have been applied to sedi- of some Paleocene conglomerates in the Rocky
mentological interpretations (Parkash et al. 1983; Mountains. These sediments are characterized by
Abdullatif 1989). A useful summary, with additional stream-flow facies deposited mainly under lower
examples, was provided by Kelly and Olsen (1993), flow-regime, traction-current conditions, and
who developed the general sedimentological model would be classified as the deposits of distributary
illustrated in Fig. 8.71. They concluded that indi- fluvial systems or braid deltas by Blair and
vidual fans are unlikely to exceed 100 km in radius. McPherson (1994). The classification of DeCelles et
Fluvial styles within terminal fans are characterized al. (1991) is described and illustrated in Sect. 4.5 (see
by broad, shallow channels in the proximal and Fig. 4.3), and is discussed in Sect. 10.2.
feeder zones (typically the shallow, perennial type of Other syntheses of alluvial fan sedimentology are
Sect. 8.2.12), grading distally into sheet-flood styles given in the books edited by Nilsen (1985) and
with poorly defined channels (the distal, sheet flood, Rachocki and Church (1990). Alluvial fans and braid
and ephemeral styles of Sects. 8.2.15 and 8.2.16). deltas are very sensitive indicators of basin-margin
DeCelles et al. (1991) pointed out the hierarchical tectonism (Heward 1978a), a point discussed at
nature of alluvial fan deposits, and applied Miall's length in Chaps. 10 and 11.
(1988a) bounding-surface hierarchy to an analysis
Chapter 9
The Stratigraphic Architecture of Fluvial Depositional Systems
9.1 Introduction The construction of a basin-scale stratigraphy is a

question of correlation and mapping, and it is the
practical details of this procedure that are addresse~
In the preceding chapters we have seen how to de-
later in this chapter. General discussions of strati-
scribe and interpret the various lithofacies compo-
graphic correlation have been provided in several
nents of a fluvial depositional system (Chap. 5) and
recent textbooks (e.g., Miall1990, Chap. 3). Conven-
how these components may combine to form chan-
tional stratigraphic methods are concerned with cor-
nel fills and floodplain blankets (Chaps. 6, 7). Studies
relation at the formation or member level and are
of lithofacies and channel geometry then permit the
not the main interest here. The focus of this chapter
recognition of fluvial style, of which 16 typical ex-
is on methods whereby the stacking of individual
amples are described (Chap. 8). These earlier chap-
large-scale elements of a fluvial succession, such as
ters describe units comprising groups 1-7 of Miall's
the meander belts and floodplains and alluvial fan
(MiallI991b) hierarchical classification (Table 3.1).
lobes can be reconstructed using techniques of out-
The next step is to reconstruct fluvial systems on a
crop and subsurface mapping. On the larger s:ale,
basin scale, that is, at the level of groups 8-11. How
fluvial units combine into sequences of vanous
many separate fluvial systems were there in a given
types. Unless specific depositional elements can be
basin? Which way did they flow, and how did they
traced, and their scale, geometry, and lateral rela-
interact? What is the nature of vertical and lateral
tions determined, it is not possible to address the
transitions into other nonmarine facies and into
larger questions regarding the action of autogenic
marine deposits?
and allogenic sedimentary controls. Several of the
Table 9.1 provides a listing of examples oflarge-
examples of the various scales of stratigraphic unit
scale fluvial units and depositional processes orga-
discussed in Sects. 9.2-9.4 are used to illustrate map-
nized in accordance with the hierarchical
ping methods in Sect. 9.5.
classification of Chap. 3. Many terms, such as
cyclothem and megasequence, have been assigned a
variety of different meanings. Also, terms such as
first-, second-order, etc., have been employed in 9.2 Channel Belts
many conflicting ways for processes and products,
with the numbering arranged in order from both
The subject of this section is the deposits of groups 7
large- to small-scale features and long- to short-term
and 8 (Table 9.1), corresponding to depositional
processes, and the reverse. The result has been that
units formed over time periods of 10L lOS years.
the use of "order" classifications has become very
Over this time scale channels and channel belts are
confused. It is hoped that the listing in Table 9.1 of
formed, and alluvial fans may undergo a prograda-
examples of processes and depositional products
tional-trenching cycle. Deposits of group 8 compare
according to the time scale they represent (recur-
in duration to the high-frequency (fifth-order)
rence interval) and the use of the nongenetic
Milankovitch cycles of Vail et al. (1977) and Miall
"group" classification will help to clarify terminol-
(1990).
ogy and provide insights into processes. Many more
Examples from studies of the Siwalik Group of
examples could be cited in this table. A representa-
Pakistan are illustrated in Figs. 2.32 and 9.1 (see also
tive selection of ancient examples has been given, in
Fig. 7.13). The Siwalik sections were constructed
which constraints on timing and rates of sedimenta-
from the measurement and physical correlation of
tion are available to enable the units to be classified
numerous outcrops in areas of excellent exposure.
into the appropriate group.
The isochron and time scales were derived from
252 The Stratigraphic Architecture of Fluvial Depositional Systems
Table 9.1. Terminology and examples, large-scale fluvial stratigraphic architecture

Group' Time scale Terminology and examples References
years
6 102 Architectural elements: Miall (1985)
-10 3 e.g., crevasse splay, macro form.
Fan channels DeCelles et al. (1991)
Simple pedofacies sequences Kraus (1987)
7 103 Channels.
-104 Autogenic channel-belt models Allen (1974b), Bridge and Leeder
(1979), Bridge and Mackey (1993)
Sand-belt architecture models Galloway (1981), Hirst (1991),
Cowan (1991)
Avulsion events (buff system, Siwaliks) Behrensmeyer and Tauxe (1982),
Behrensmeyer (1987)
Fan progradation-trenching cycle DeCelles et al. (1991)
(fifth-order lithosome)
Fan sequences. Heward (1978a)
Compound pedofacies sequences Kraus (1987)
Typical coal seams Nemec (1988)
Units I to VIII, Old Red Sandstone Allen and Williams (1982)
8 104 Channel belt, alluvial fan.
-105 Long-term geomorphic thresholds Schumm (1977)
(third-order geomorphic cycles) Schumm (1977)
Fifth-order sequence: Milankovitch cycle: Vail et al. (1977), Miall (1990)
Return period of
autogenic channel-fill Siwalik cycles Johnson et al. (1985),
Channel-belt migration (buff/blue-grey syst.) Behrensmeyer and Tauxe (1982)
Fan megasequences. Heward (1978a)
Major coarsening-upward cycles. Gloppen and Steel (1981)
Sixth-order lithosomes DeCelles et al. (1991)
Pedofacies megasequences Kraus (1987)
Devonian cycles, E. Greenland Olsen (1990)
Minor cycles Heckel (1986)
Incised valley-fills, Louisiana shelf Suter et al. (1987)
Channel-belt alloformations, Louisiana Autin (1992)
9 105 Major depositional system, fan tract.
_10 6 Long-term geomorphic thresholds Schumm (1977)
(third-order geomorphic cycles) Schumm (1977)
Fourth-order sequence: Milankovitch cycle: Vail et al. (1977), Miall (1990)
Depositional-system axis Galloway (1981)
Seventh-order lithosomes DeCelles et al. (1991)
Pedofacies megasequences Kraus (1987)
Catenas of tectonic origin Atkinson (1986)
Cyclothems Klein and Willard (1989)
Minor cycles Holdsworth and Collinson (1988)
Major cycles Heckel (1986)
Castlegate Sandstone sequences, Utah Van Wagoner et al. (1990, 1991),
Miall (1993)
Medium-order cycles, Tr-Jur, North Sea Nystuen et al. (1989)
10 106 Basin-fill complexes
-10 7 Second-order geomorphic cycles Schumm (1977)
Tectonic cyclothems Blair and Bilodeau (1988)
Third-order stratigraphic sequences Vail et al. (1977)
Mesothems Ramsbottom (1979)
Major cycles Holdsworth and Collinson (1988)
Ancient examples:
Third-order Brazeau-Paskapoo cycles McClean and Jerzykiewicz (1978)
Tertiary formations, Gulf Coast Galloway (1981)
Karoo basin cycles, S. Africa Visser and Dukas (1979),
Turner (1983)
Channel Belts 253
Table 9.1 (Contd.)

Group' Time scale Terminology and examples References
year
c-u cycles of L. and M. Siwaliks Johnson et al. (1985)
Formations in U. Cret clastic wedge of Utah Molenaar and Rice (1988)
High-order cycles/megasequences of Nystuen et al. (1989)
Triassic-Jurassic, North Sea Steel and Ryseth (1990)
Megacycles, Cutler Gp. New Mexico Eberth and Miall (1991)
Red Molasse (Olig), Digne, France Crumeyrolle et al. (1991)
Sequences of Straight Cliffs Fm, Utah Shanley and McCabe (1993)
Chinle Sandstone members, Utah Dubiel (1991)
Cafiizar Sandstone sheets, Spain Lopez-Gomez and Arche (1993)
11 107 Basin-fill complexes
_10 8 First-order geomorphic cycles Schumm (1977)
Second-order stratigraphic sequences Vail et al. (1977)
Sloss-type sequences Sloss (1963)
• As defined by Miall (1991b): see Table 4.2; c-u=coarsening-upward
.: :
: '
~ "_ " I .,..--J--'

.
--~-..:::7
"
_--7
~-----..-..-~
'-'"
'.:,
\ ... ________
~
CH .--- - --/
'- /-
"-.../
Fig. 9.1. Interbedding of contrasting fluvial styles. The posited by a trunk-braided fluvial system. Model 8 repre-
depositis of fluvial model 11 of Miall (1985) are the "blue- sents the "buff" deposits laid down by an anastomosing
gray" deposits of Behrensmeyer and Tauxe and were de- tributary system
magnetostratigraphic studies. The deposits consist Behrensmeyer and Tauxe (1982) demonstrated
of major channel sandstone bodies 10-30 m thick that these deposits represented two quite different
and extending laterally for up to tens of kilo- fluvial systems that interfingered with each other
meters. They include both ribbons and sheets over a time period of about 8 x 104 years (Fig. 9.2). A
(Behrensmeyer 1987). Ribbon sandstones are par- sandy braided fluvial system formed sheet-like
ticularly well documented in section 4 of Fig. 2.32 channel bodies. Behrensmeyer and Tauxe (1982) re-
Overbank deposits include clay, marl, pedogenic ferred to this as the "blue-grey system" (Fig. 9.1).
carbonate, sand, and intraformational conglome- Tributaries, consisting of narrower, possibly anasto-
rate, representing levee, crevasse-channel, crevasse- mosing channels, formed mainly ribbon sand bod-
splay, and minor-channel environments. Several ies, constituting the "buff system" of Behrensmeyer
major abandoned channels more than 1 km wide are and Tauxe (1982). The interaction of these two flu-
indicated in Fig. 2.32 by cut-and-fill erosion surfaces vial systems, and the autogenic processes that con-
with 10 m or more of relief, followed predominantly trolled them, are discussed in Chap. 10.
by fine-grained deposits.
o. practical examples based on studies of the ancient

record.
Galloway (1981) described Cenozoic fluvial sys-
tems of the Gulf Coast of Texas from a combination
of outcrop and subsurface data. The distribution of
sand bodies in the subsurface is shown schematically
in Fig. 9.3. Interpretations of one of these, the
o
O BLUE GRAY
SYSTEM
Gueydan system, are shown in Figs. 9.4 and 9.5. This
---- 50km
~ BUFF
system has the characteristics of the "low-sinuosity
I2...J SYSTEM river with alternate bars," the style described in Sect.
b 8.2.11. According to Galloway (1981), channel sand
bodies are locally stacked vertically along "deposi-
tional axes" (as described in Sect. 9.3). Channels and
associated splay deposits form sand belts up to
several kilometers wide. Several other fluvial styles
occur in this Cenozoic succession and are described
by Galloway (1981).
Continuous lateral proftling of large outcrops in
the Ebro Basin, Spain, produced detailed maps of
sand-body distribution of an Oligocene-Miocene
fluvial system by Hirst (1991). He classified the sand
bodies into ribbons, sheets, and amalgamated com-
plexes, following the definitions of Friend et al.
(1979) and Friend (1983). Figure 9.6 shows examples
of the various sandstone types, and Fig. 9.7 is a small
example of part of one of his lateral proftles.
The pedofacies concepts of Bown and Kraus
(1987) and Kraus (1987) may be useful for gaining
Fig.9.2. Reconstruction of the two interacting fluvial sys-
tems that constructed the Siwalik sediments in a part of
information on channel architecture in areas of
Pakistan. The outcrops illustrated in Fig. 9.1 are shown by limited exposure. As shown in Fig. 7.24, pedogenic
the heavy dashed line near the right-hand edge of each maturity is related to distance from a main channel,
block diagram. (Behrensmeyer and Tauxe 1982) which governs the exposure time of the paleosol, and
its dilution with clastic detritus. Kraus (1987) sug-
gested that vertical successions of pedofacies record
lateral movements of major channels. Such ordered
Several detailed studies of the fluvial stratigraphy vertical successions she termed "compound pedo-
of coal-bearing units have been reported by R.M. facies sequences." For example, Fig. 9.8 shows a
Flores and his colleagues (e.g., Flores 1981, 1983b, stratigraphic section in which paleosols are inter-
1984; Flores and Hanley 1984; Warwick and Flores bedded with levee and crevasse-splay deposits. The
1987). An example is illustrated in Fig. 7.2. In each paleosols show an upward increase in maturity from
case, coals provide major marker beds, permitting stages 1 to 3 (0-19.5 m interval of section), but the
lateral correlation of stratigraphic sections for dis- last paleosol then returns to stage 2. Kraus's (1987)
tances of up to tens of kilometers. As discussed in interpretation of this succession is shown in Fig.
Chap. 13, coals have been used in some studies for 2.31. The section begins with the channel and flank-
the definition of stratigraphic sequences. ing levee deposits formed when the channel was in
Theoretical sand-belt models have been con- its first position and ends with the channel in posi-
structed by many workers for the purpose of explor- tion VI, having returned to approximately its origi-
ing the lateral spacing and vertical stacking pattern nal position on the floodplain. Guccione (1993)
of channel belts. A basic classification was offered by suggested that, in much the same way, the grain size
Friend (1983) and is illustrated in Fig. 2.35. These of overbank clastic deposits could be used as an
models, the most recent of which employ computer indicator of distance from the source channel. As
simulation procedures, are discussed in Chap. 10. In noted in Sect. 7.4.2, however, other factors, including
this section we illustrate and briefly discuss some substrate type and differential subsidence rates, may
Channel Belts 255
N sw NE
!'-
;::7~ I
~
,
I
~. H
•I I
+ I
I
I
I t
I I
., 1-. j I I
• ~
DATUM IT e~TOM CA TllHOUI. A

@ CHITA AXIS
t ."
sw
TOP OAt<",' .. U
....,j
® -
i N
GUEYDAN AXIS .........
S ; ~ • I a
I
~i~+-"'rn !
r_ +,
I I H TOP CAUMOlJlll.
\TOP CATAMOUr.... . iii
® ®
!
GEORGE WES T AXIS MOULTON STREAM PLA IN
Send
Chonnel lo ll
... -
t
I
.. Mulldllferol sond bod ies
MulliSlory sand bodi es

Concenfrafions of channel
l ill ond ossocioled sploy
foci es
0i--_""_.,..:;2O
o KPoI 30
':J: '00
Mud / coy
l
v Ex. ~ 200 I 500
Fig. 9.3. Representative stdke-parallel cross sections of shown in diagram A, is illustrated in Figs. 9.4 and 9.5. The
m:l]cr ccastal-plilin fluvial deposits, Cenozoic of the Gulf locations of these cross sections are shown in Fig. 9.14.
Coast, Texas. The fluvial style of the Gueydan system, (Galloway 1981)
Fig. 9.4. Schemat~c

••
BAR ACCRETION
FLOODPLAIN AGGRADATION
reconstruction of the Gueydan fluvial style, Texas. (Galloway 1981)

1 - - - - -- - - - - - - ---=-:--=---10" r.., Ot ".,,"'0'-- - - - - - -- - - - -- -----1

I~" - - - - -03 fee Ot ..,.""'"'------1
_~I~~~I_ _ _ _ _..j----Chomel F . I I - - - -- - - - l
B o
EXF\.ANATIQN E
CJ Mtd ltoNn 10 toGIM tond ./u.!. MI.Id doltl, p;ebbIn, pellett, C)"Id dro04t ~ Mldil.m- cn:lIlIftOl~" If'Ou9h crot' - llralif-c.otion
c::::J
DSond. __
F.... .a~ 10 lilt ~.....",... PtDbI... •• 'lIoP)ritt - fbpptl Crou·1ominaI1On
to.. OuH,td iron nod",1n • • Aant dltlrit . . . CI,mbn; ,we tfOS!l-1omftl1101"1
0--", <1Oy>- lITlT_ - "ca,bonate- nodu"" ! 1J \' Root ttOctt
Fig. 9.5. Schematic facies architecture of the Gueydan B proximal, mudstone-filled crevasse channel, C mixed-fIll
fluvial system, Texas, illustrating the lateral relationships crevasse channel, D sand-silt-filled channel, E main fluvial
and scales of the various channel and overbank elements. channel fill. (Galloway 1981)
Measured sections illustrate A distal crevasse-splay,
RIBBON SANDSTONE wrr<15 also affect soil character, and so it is necessary to

.---W-- examine and interpret paleosols within an overall
\ ~
A Wing Laterally stable flow basin context. Marriott and Wright (1993) provided
an example of such a stUdy with their work on the
SHEET SANDSTONES wrr>15 well-known Old Red Sandstone (Devonian) paleo-
sols of South Wales. They showed how variations in
B
sedimentation rate, erosional episodes, and overall
changes in the style of floodplain development could
c
be deduced from the details of paleosol profiles.
Multistory/Multilateral Progressive burial of a soil profile without disrup-
D === .!: ... · .~=-h.?- .... : ....... __ . nr l ' l'·
tion of pedogenesis produces a cumulate paleosol,
Poorly channelized flow short-lived interruptions of pedogenesis may lead to
overlapping soils, termed composite paleosols, and
Unconfined flow so on. Spectacular outcrops of channel and over-
bank architecture in Miocene deposits in Pakistan
enabled Willis and Behrensmeyer (1994) to offer a
AMALGAMATED COMPLEX more thorough discussion of the possible styles of
F ~ meander-belt development. These deposits, to-
z:'~··"~;::.':.liJjt~ gether with a description of the nature of avulsive
processes and their control on meander-belt archi-
Fig. 9.6. The range of sand-body geometries in the Huesca
tecture, are discussed in Sect. 10.3.1.
fluvial system (Oligocene-Miocene), Ebro Basin, Spain. Conglomeratic alluvial fan deposits commonly
"Wings" are levee deposits. (Hirst 1991) display coarsening- or fining-upward successions
Channel Belts 257
o;. ~~~~~~~~
Fig. 9.7. A small example of the profiles constructed from extends for more than 1 km. Sand bodies are numbered in
outcrop data by Hirst (1991). The drawing is at true scale order of deposition within the entire profile
(without vertical exaggeration), and the complete profIle
r.y . . . . SpllY
2 Pale-osol
Crevl ••• Channel
p.leolo l
. . . . Spl.y
. . . . Splay
1,,,·.... ,,,..
,
'------'
I I ~, ..
...... Fig. 9.8. Stratigraphic section of the Willwood
allla.
Formation (Eocene), Wyoming, showing a
ft!Iucl.t compound pedofacies sequence. (Kraus 1987)
/t'oit4
I
HJORIESIEGVA I NEI FA
_ FLOODBASIN DEPOSITS
1Ill[]]J F,UV I. ,
Fig.9.9. Cross section of the Hjortestegvatnet fan (Devonian), Norway, showing coarsening- and fining-upward on at
least two scales. Vertical graphs with dots show maximum particle size in centimeters. (Gloppen and Steel 1981)
Fig. 9.10. Drawing and interpretation of an outcrop of the in Fig. 4.2. Stippled areas are poorly exposed siltstones.
Beartooth Conglomerate (Paleocene), Wyoming. Numbers (DeCelles et al. 1991)
are bounding surfaces, the classification of which is given
(cycles), and these have been much studied for the 9.10. Sequences of 15- to 65-m thickness showing
information they yield regarding autogenic and either coarsening- or fining-upward trends are com-
allogenic processes (Heward 1978a; Gloppen and mon in these examples. They are bounded by fifth-
Steel 1981; DeCelles et al. 1991). Examples of profiles order surfaces in the classification by DeCelles et al.
from two fan complexes are shown in Figs. 9.9 and (1991; Fig. 4.3), and may extend laterally for hun-
Channel Belts 259
dreds of meters. The examples illustrated in Fig. 9.9 difficult to map in practice, because fans coalesce
are upward coarsening in type. Those in Fig. 9.10 are laterally, with complex interfingering relationships
upward fining. They are bounded at the base by ero- developing as individual distributaries on adjacent
sion surfaces with up to several meters of relief. The fans meander and shift by avulsion. Compositional
top of each sequence may be marked by a thick silt- differences between adjacent fans may develop be-
stone unit several meters thick, which may contain cause of different source-area geologies, and this
pedogenic features. may permit individual fans to be mapped on the
Thicker successions, termed megasequences by basis of sandstone petrography or clast composi-
Heward (1978a), are also common in alluvial fan tion, a mapping technique that has been in use for
deposits. Figure 9.9 shows that the entire succession many years (e.g., MiallI970a).
of the Hjortestegvatnet fan in Norway constitutes a Coastal fluvial systems are strongly influenced
megasequence 250 m thick, which coarsens upwards by base-level change, a topic that forms part of the
through the first 200 m of the section and then fines basis for the study of sequence stratigraphy. This
upward. Heward (1978a) illustrated several examples subject is considered in detail in Sect. 1l.2.2 and
of fan profiles showing sequences and megasequen- Chap. 13, but a few descriptive points are noted
ces of several types and thicknesses (Fig. 9.11). Indi- here for the sake of completeness. Among the most
vidual fans constitute "group 8" deposits in the detailed studies of coastal stratigraphy are those
hierarchical classification of this book. In the undertaken on the Gulf Coast of Texas and Louisi-
classification of DeCelles et al. (1991) entire fan bod- ana. Late Cenozoic deposits contain the evidence of
ies are bounded by sixth-order surfaces (Fig. 4.3) but, numerous glacioeustatic changes in sea level, and
as DeCelles et al. (1991) noted, such surfaces may be fluvial systems contain much evidence of this. For
o.
&iBI
I
dost wpported conglomerate b. c.
t d :"." M
I
matrix supported conglomerate
VERTICAL SEQUENCE OF COARSE GRAINED SEDIMENT TYPES
r
, 11
~
cs Cross-bedded sandstone
PS Pebbly sandstone
SC Stral'of,ed conglomerate
~
~~
~.
~K
UC
SC
PS
Unstratjfied conglomerate
Stratified conglomerate
Pebbly sandstone
~ -5 UC Unstratified conglomerate CS Cross -bedded sandstone
(5
200
P5
(5
P5
(5
0
" ' b...........'" 0
0
fining and 0
thinning
upward I 1SO
~
megoseqvences
fining and coarsening
thinning
upward
megosequence
II and thickening
upward
mega sequence
I
700
o f
"
o{
65.
i:
m
IJ
5llcm II
4Dcm
em em
ove.mo.x,
dostslze
Fig. 9.11. Examples of fan sequences and megasequences in the Carboniferous alluvial-fan succession of northern Spain.
(Heward 1978a)
LOU s A ,.. A
"
i' I'
~ """'''I
,
lOt IO-.W.I
o Fluvial d.poslta fm~ SheH ph... delta
Erg] Sh.lf ph... d.... • Shelf ....,.tn delta

(thlckne.. In . ,
Fig. 9.12. Depositional systems on part of the continental paleogeography developed was one of many fifth-order
shelf of Louisiana that developed during the Pleistocene at glacio eustatic (Milankovitch) cycles that affected conti-
a time oflow sea level. Note the fluvial systems occupying nental margins during the late Cenozoic. (Suter et al. 1987)
incised valleys. The sea-level cycle during which this
example, during low stands of sea level, fluvial sys- and rate of base-level rise). An example of the depo-
tems become incised into the continental shelf and sitional systems that result is illustrated in Fig. 9.12.
extend their courses to the shelf margin. During the Commonly, the scale of such incised valleys is little
subsequent sea-level rise, these broad incised valleys different from that of channel systems generated by
are then filled with fluvial or estuarine deposits (de- autogenic processes, and correct interpretations re-
pending on the balance between sediment supply quire very careful mapping.
Depositional Systems 261
ml a m b
450
I
1100
l !t
'r'
1000
G
400 900
+
t
F
800
t
E
MEGA-CYCLE A
350
+
t
.D
j
;
C
300
I
B
400
250 300 ~
o 5 10 m
Fig. 9.13. Variations in sand-body thickness in a Devonian fluvial section in East Greenland. CR. Olsen 1990)
Another possible example of fifth-order cyclicity 9.3 Depositional Systems

preserved in nonmarine deposits is that given by H.
Olsen (1990). He described a Devonian fluvial sec-
tion in East Greenland containing prominent fining- Group 9 deposits are those representing hundreds
upward cycles that show systematic variations in of thousands of years of deposition. The genera-
thickness. Variations in channel sand-body thick- tion and abandonment of entire depositional sys-
ness are shown in Fig. 9.13. The average thickness of tems or alluvial fan tracts may be accomplished dur-
the fluvial fining-upward cycles is 20 m, with a range ing such time intervals. Long-term (fourth-order)
of 13-30 m. Channel sandstones vary from 2 to 13 m Milankovitch cycles are also of this time duration.
in thickness. Bundles of 3-6 cycles can be defined, as Deposits of this scale are typically regionally exten-
can thicker trends that trace out megacycles. Mega- sive and are "mappable", in the sense that they can
cycles range from 79 to 135 m in thickness, with an be defined on conventional geological maps on
average of 101 m. H. Olsen (1990) suggested that scales of 1:50000 or larger. Many such units have
these cyclical changes in thickness are the result of been therefore assigned lithostratigraphic formation
discharge variations brought about by orbital forc- or member names in the older literature. They are
ing, a point discussed further in Sect. 12.12. bounded by seventh-order surfaces (Tables 4.2,9.1).
;;'fMJ'i ~ 150 It (45 m) sand

,~:;"',~l!f,. Based on - 850 wells
50mi
~----~I
80km
Fig.9.14A,B. Simplified total-net-sand isolith maps of correspond to channel-belt axes. These are separated from
two Cenozoic fluvial systems on the Gulf Coast of Texas. A each other by sand-poor zones corresponding to interfluve
Catahoula Formation, B Oakville Formation. Areas con- areas, crossed by small tributary streams and also occu-
taining more than 75 m of total sand are indicated by pied by floodplain lakes. (Galloway 1981)
stippling and define broad, sand-rich belts or aprons that
--.c:
--'"'"
o ..
.!!!
-'" :a"O;~
0
-0
...0
'" ......
...'"0:> ..
--.0; .~ C
sw tl :x:""0
Ol
-;;;3: CD oIV._
.- c: 3: 0 .s::. ";:;
'" NE
Ee ~ E t O~Q) "0 '" ~
Ol, ::J
till :.:;en 0:::; :.J en (!)~O U 3:ct:
t
0]0
I
: Generally uniform sand/mud ratio of system San Mal'cos

arch Generalized
1Upward decrease in sand/mud ratio of system
I
c m ft vertical
scale
300 1000
Fig.9.15. Generalized strike-parallel cross section through the Oligocene-Miocene succession of Texas, showing the
principal depositional elements and fluvial "axes". Note the long-term vertical persistence ofthese axes. (Galloway 1981)
Galloway (1981) showed from detailed subsurface Kraus (1987, 1992) noted that over vertical inter-
work that large-scale depositional systems may oc- vals in the order of hundreds of meters of section,
cupy constant positions within a basin margin for average pedofacies maturity may undergo long-term
extended periods oftime (Figs. 9.14, 9.15). His fluvial changes. For example, in the northern Bighorn Basin
"axes" within the Cenozoic Gulf Coast fluvial succes- of Wyoming, compound pedofacies sequences near
sion correspond to group 9 deposits. the base of the Willwood Formation are character-
Depositional Systems 263
ized by stages 1- 3, whereas at the top they are typi- of differences in drainage conditions and fluvial to-
cally stages 3, 4, and rarely 5. She termed such long- pography related to differential subsidence within a
term changes "pedofacies megasequences" and foreland basin. This is an example of a paleosol
attributed them to changes in deposition rates. In catena controlled by tectonism. Examples from the
the Willwood example it seems likely that the basin Recent were described from the Gangetic plains of
subsidence rate decreased during accumulation of India by Srivastava et al. (1994), who attributed local
this formation, permitting individual paleosols to changes in fluvial style and paleosol character to
mature for longer periods before being covered by neotectonic movements of the basin.
overbank flooding. Atkinson (1986) described varia- The classic cyclothems of North American geo-
tions in paleosol character displayed laterally on a logy are fourth-order stratigraphic sequences and
regional scale and interpreted the variations in terms correspond to group 9 of the classification in this
Shallow
Nonmarine
Marine
Contin- Inter- For

ental Mar- mediate Off-
Erotlon edimln- Swamp Off-
totion ginal shorl shore
Shale (marine)
AIQal limestone (conlains
nlarlhore and brackish
water Inver lebratn)
Limestone (contain. for
off-shore invertebrates,
~i~~ esplcially fusulinids)
~ L imestone, Impure to .haly
'h;:;;~"'-~ (contains intermediate
..., off-shore Invertebrat,,)
Shale, marine (contains

near-shore invertebrates)
Coal
Underclay
Nonmarine shale, commonly
landy
Nonmarine sandstonl
Oisconform lty _ _ __
Limestone (contains lor
~ ~~~~
- - - -
off-.hore invertebrates,
especiolly fusulinids)
Limestone, sholy (contains
intermediate off-.hore
invertebrates)
Shale, marine (contains

nearshore invertebrates)
Coal
--,
Nonmarine sha le, sandy
Nonmarin. landstone
J-'-,rj-'-'.;..;.;.....- - Oilconformity _
L imestone (contains
'I--'T---L_ fUlulinid.)
Fig. 9.16. Two typical Illinois-type Carboniferous cyclothems showing interpretation in terms of transgression and
regression. (Crowell 1978, based on Moore 1964)
book (Table 9.1). They may contain a significant Another example of this type of cyclic sedimenta-
fluvial component, representing incised valley fills tion on a lOS-year scale is that of the Dunvegan delta
deposited following the period of maximum regres- complex of Alberta (Bhattacharya 1991; Fig. 9.17).
sion. Two typical Illinois-type cyclothems are illus- The Dunvegan Formation consists of seven subdivi-
trated in Fig. 9.16, and their origins are discussed sions, classified as allomembers, each of which is
further in Chap. 13. Similar regressive fluvial-deltaic estimated to represent about 200 ka. Fluvial channel
sandstones, termed "grits" in the older literature, fills occupy valleys incised into the top of several of
form regressive units in comparable British the allomembers.
cyclothems (Ramsbottom 1979; Holdsworth and Examples of fourth-order cycles that include
Collinson 1988). Note the confusion in terminology more extensive fluvial deposits are those that consti-
for these cycles: In Heckel's {l986} work on the tute the Castlegate Sandstone and Upper Desert
Pennsylvanian rocks of the Midcontinent of the Member of the Blackhawk Formation of Utah (Van
United States these cycles constitute the larger scale Wagoner et ai. 1990, 1991; Miall1993; Yoshida et aI.,
of cyclicity, and they are termed "major cycles". in press; Fig. 9.18). Each of the fourth-order se-
The study by Holdsworth and Collinson (l988) of quences represents several hundred thousand years
similar cyclicity in rocks of broadly comparable age of geologic time. The sequence corresponding to the
in Britain refers to cycles of similar lOS-year duration lower Castlegate Sandstone is an extensive braided-
as "minor cycles", compared with the major cycles, fluvial sandstone sheet resting on a regional erosion
or "mesothems" of Ramsbottom {l979}. surface. It can be traced down dip along the Book
NORTHWEST SOUTHEAS1
"'-
'00£--
-
- -- ---- ----
CRiEifil_
ARIC'Eit-_
.....
-------- ----
_ _ _ _ _ _ _ DH\I!
IC 1-----------
Lotor SI",on.lI. IIlltton.
o
c
--
z
<
'"
CD E
---------- _. o
------ --
~
z ....
F --~
..,. ..-""""
G ......
..... _ ./
/" ..... ~--
.,," r---:--__-...,
_ ____ _ _ _" ALBERTA
...- -- --- -
-----."......... -
.- -- i!
1
1/
,.
U)
:a:: _- ."
/ ~A80UT 200km----'
.
--.-'..-
~ FSU-- " " ALLOMEMBER BOUNDING DISCONTINUITIES
EDMONTON
"./ - (MojOI' Morini Flooding Surface'
'" ,/
.
(I) .......-.-.. EROSION SURFACES
./
CII ,/
c: ,/ ~ LAPOUT
C.~CA.Y
:. FIS~_./
~
SCALES
Fig.9.17. Schematic regional dip-oriented cross section separate offlapping shingles can be mapped, as shown by
of the Dunvegan Formation, west-central Alberta. Heavy the numbers. Root symbol, nonmarine facies; light stipple,
lines are regional flooding surfaces that subdivide the for- marine sandstone; heavy stipple, channel fills; blank, ma-
mation into seven al1omembers. Within each allomember rine shale. (Bhattacharya 1991)
Basin-fill Complexes 265
ILEGBND I
~ Type I Sequence Boundary (3n1 order)
B Type I Sequence Boundary (4th order)
IHSTI Highstand systems tract
ITST 1 Transgressive systems tract
ILSTI Lowstand systems tract
En vironment
PC
LOCATION
MAP
vertical
scale 20
.
m
1liOMPSON
r
grain size
c1 ay
I~
30
~
yf
§ open marine CANYON

IlIlllIIID shallow marine
[':7.:1 estuarine
E::;~<l deltaic dislriburary channel
(:- : -~ coastal plain I lagoonal
1-",--1 tidally - influenced fluvial
I : ::: ~ braided fluvial (sand - shale)
E:::3
Fig. 9.18. Sequence stratigraphy of the Castiegate Sand- years. This passes down dip into a series of fourth-order
stone and equivalent deposits, Book Cliffs, Utah. In prox- sequences consisting of fluvial and estuarine sandstones
imal areas (left-hand sections) the succession consists of interbedded with shales. (Yoshida et aI., in press)
a single third-order sequence spanning about 5 million
Cliffs for some 140 km and passes eastward into ond-order stratigraphic sequences, in the classifica-
deltaic and marine deposits. It is assumed to extend tion of Vail et al. (1977) and Miall (1990). Regional
laterally along strike (north-south) for a consider- tectonism and eustatic sea-level variations are the
able distance. As discussed in Chap. 13, the sedi- major generating causes of these sequences, as dis-
ments are interpreted as having been deposited cussed in Chap. 11.
during or immediately following a low stand in re- Ramsbottom (1979) argued that cyclothems in
gional base level. the British Upper Carboniferous succession could be
Some other important examples of nonmarine grouped into larger-scale packages, which he termed
fourth-order cycles are those which form part of mesothems, corresponding to group 10 deposits in
complex cyclic sequences in the North Sea Basin the present terminology. Part of the evidence for this
("medium-order" cycles of Nystuen et al. 1989); form of cyclicity is that regressions are more exten-
these are discussed in the next section. sive at the close of mesothems than at the termina-
tion of individual component cyclothems, leading to
thicker and more areally extensive fluvial-deltaic
successions, and the more widespread establishment
9.4 Basin-fill Complexes
of fully nonmarine conditions, with deposition of
major coals. The major sandstones that terminate
Groups 10 and 11 correspond to sedimentary succes- each cycle are called "grits" in the British strati-
sions having durations of a few million to a few graphic literature. A chronostratigraphic section
hundred million years. This includes third- and sec- through Namurian mesothems in part of northern
South tv
0\
0\
Alston
Meso· I Stages
Block Stainmore Colsterdale Wharfedale Centre of Basin thems Zones
'Secon d gr it ;::lli::;:;:::r~~:r:,:::.:;:.;:~:;:.:::.:::.:::.::::::_:;:;~;~;:~~;~~:~~;:~:;:_i:::::i:: Lai~?~~?~ sst ~;:;:;:;:;:;:;:;:;:;:::: ~n"gh rock.:::::::::::::::::::::!:':::::::::::::::::::::::::::::::::::::::::::::::: N 11 /G
la, b Yeadon ian
' ~~?na.~r;lt.;.t::;}:;::~.: ;.: :.:,l:.: ;.: : ,:;:r:::,: :,: :,: 1,: :i.:::r;:t::::}};:::;}=.~.~:~.~.~~;,}:~,1 !, : ,~~~J,: : : :.: :,:FI!~~~;~~i :~:i:~:i~~:~:;i;:: ~i:~:: 1~:;;:1:i:)~~:~)~:[~li~l:i: ::i: j: ;:;: j:i: : : : f:~: ... N10 I R2c
r Marsdenian
N9 I 2b
r R 2a
First !lrit:':::;'·:·:·;·:·:::·:·:·:·:·:·:·: :::::~:~::::::~ I bnmhamgri t'.;:;:;:;:;:::;:.?'!~.:.::::~':;~~;~·;;;;~·~· ·~;·i~·f.Ti;i;'...'...:
·m.·.·.·.·.·.·.·.·.·.·.....· •• ·:teldingham edge grit:::;:::::
. ---_ . N8 I R1c
:~ ~ ~~..~~~~ ~:~.1~~~~.~:~:~:.~:.:.~~:-~:~i~:.~:~~~~i:;~_~~i:~;:;~;.:;~:;~;:;:;~~:;:.i:~.:;:.;:;.:;:.;{:,~:.::~.~.~~.<~.:.~.~.~_~.~_ ._~.~i.~~~~:~:i~~:;~i:;.:~~;:;.::~.::~~::;.~.:~r.:~~:.:I: :.): :.): :. ).i':i

-. N7 I R lKinderscoutian
1b
.i_ •
::~:;:;:;7::~:;~:=;:;:::~:;:;.:~:.;:;.~::.::;.:::.;:;.:~:.;:;:~:~ :;~~;:~i~i~:~.:~~:~:~;~:~::~:.:~:.::=.:.::.::~.~:~.:~:.:~:.:~:_:.~:~._~ ~.~ ~.~:~.~ ~~ }::~(i~;.:~;.~:~.:::.~~.:::.~:~.~~:.~::.:.~~~.::.:1e.B.~~~.~.? ~ R~. ~.~ I_~ '.:~:.i::j~.:.~:~:.:,:i~.:.~;:: :1..
N6 ~a
2c
N5 I § Alportian
18, b
~
rt>
,..............::::::::::::::::::;:;=;=;=;:;=;=;=:=:=:=:=:=:=;=;=:::::=~Ur. foilifoot grit:::::::::;:::::;:~::;~:;::::::;.:arock.s t!an k. grit:::
...... •............ _ ;. ;. ;
. .._._._ ............................. .............................................. ............. ;.;. ;. ;. ;.;.;.;.;.;.;.;.;.;.;.;.;.;.;.;.;.:.:.:.:.:. !4'
N41 H 1•• b Chokier ian ...
I»
,..
~.
I grn; I»
"0
N3 H 2c ::r
(;.
~ Arnsbergian >
...
.:.:.:.::;.~:;.~~;.~:~.~~;~::~~~;~:;~~.~:~.~:;'~:!.~:;.;::~:: ;:: ~:::;~: ~~:~;~::};~::~:;.~~;.~:;.~t:E~;~1!;:J:~:~I.!:!::!;t:~;:.:!:JLI·:·;'~:~~::!~:::J~r~j~ht~~~~;;;~;~;;;~f!J;~:t~:~~.~~~~·j:~~.~:.:~~.~;';:::::·:::·::::.::~.~::t:.::~:r.:.~ , N2 E 2b [.rt>
,..
"
~
:~ :r~~mmm~~~:~~~~~:~:;;;:~:!ft~;l;~t~!~~!}'~7~~.~.~;~.~:~~!1~l~I~~,l~l,\~J~~.~~~.l~~!.~~!.~ !.E};~~:~~~~~j~~~~~j[~~jlj!~1!t!1:1:I: N1 I E2a ...rt>
E 1a ) Pendleian o
.....
::::::::::::::::::::::~:~:~.:. >rj
.............
:.:.:.:.:.:.:.~
lliill]
....... Sandstone ill Other lithologies [
e-
Fig. 9.19. Chronostratigraphic section through Namurian mesothems in part of northern Britain showing the repeated occurrence of regressive o
sandstones ("grits") of fluvial deltaic origin. (Ramsbottom 1979)
.g
o
~.
o·
::s
e-
~
~
'"
England is shown in Fig. 9.19. The debate regarding trates the extent of six such cycles in the Triassic
the reliability of the mesothem concept (Holdsworth Molteno Formation of the Karoo Basin in South
and Collinson 1988; Leeder 1988) illustrates the Africa. One of these cycles extends for more than 400
difficulty in reconstructing complex litho- and bio- km down depositional dip into the basin interior. A
stratigraphic patterns in tectonically active areas comparable subsurface example, from a different
and in unraveling the various causes of cyclicity basin, is illustrated in Fig. 9.21. This diagram shows
(Chaps. 11, l3). Leeder (1988) noted that new that the "high-order" cycles (those corresponding to
biostratigraphic data and revised sedimentologi- group 10 of this book) form the largest of three scales
cal interpretations that emphasize autogenic del- of cyclicity, including allogenic "medium-order"
taic progradation cast doubt on the reality of cycles and fully autogenic "low order" cycles. Steel
Ramsbottom's mesothems. and Ryseth (1990) extended this type of analysis
Large-scale, regionally extensive, sand-domi- throughout much of the northern North Sea basin,
nated, fining- and coarsening-upward cycles, typi- on the basis of seismic evidence (Fig. 9.22) and addi-
cally 100 m or more in thickness, are common in tional subsurface well correlation (Fig. 9.23). They
nonmarine deposits. References to several examples were able to relate the origins of the sequences to
are given in Table 9.1. Commonly, such cycles are the tectonic history of the area, as discussed further
areally very extensive. For example, Fig. 9.20 illus- in Chap. l3. Biostratigraphic evidence available to
,
\ 0 100
6 I 111m
5 Sase of Molteno oulcrop
\
,
4 {~}
LESOTHO
2 ,
1::-:
~Flal bed<I.... . Shale wilh ~

~hatei
~ rar.coal
~ cross-beds
Planar D Massive 10
irregularty
bedded
[,] ~ Ripple
...... :..-: IalTinltion
'-
~o Pebbles,
0 a ~
0 CObbles and '" CQ Plant rnalerial
boUders
Fig. 9.20. Six cycles in the Molteno Formation (Triassic) of the Karoo Basin in South Africa. (Turner 1983)
r
::; :rJ
m
Fig.9.21. Example of a subsurface wireline-Iog
r CJ)
m LOG SEQUENCES GR. OEN INTERVAL RES record through Triassic nonmarine deposits in the
0 :rJ gfcmJ TRANSIT
....
CJ)
< 1.95 295 TIME Ohms Snorre field area of the northern North Sea, show-
:rJ 0
jj
CJ)
HIGH MEDIUM LOw API N'\?R us fI mol m ing three categories of nonmarine sequence. The
ORDER ORDER ORDER _1~~~"."
high-order sequences correspond to third-order
0 cycles of Vail et al. (1977) and are allogenic in
Gl C
t origin. (Nysteun et al. 1989)

-c Z
.c;
Z
....
CJ)
U.
,-
."
»
....
~
,=
;::." L_ §
r
. c-
o L. LS 10
:rJ
0
~
A
,-
.'~
"-i- ~
'.
'- -;,.
~~J
~~ l' 2.
~1
r
C
Z BC
0
m
."
0
:rJ
;::
»
....
6
Z
C
-c
-c
m
:rJ
;::
m 0
;::
(IJ
m
:IJ
LS 9 c
u
8
LS 8
Nystuen et al. (1989) indicated that their high-order Nested sequences on two scales occur in the
cycles have durations of 4-12 million years. Steel and Castlegate Sandstone of Utah (Fig. 9.18), as noted
Ryseth (1990) suggested a duration of 10-15 million above. The main Castlegate sequence in proximal
years for their megasequences. areas had an estimated duration of about 5 million
3114 - 3 ProteCttOn of w ei
HORDA PLATFORM 31/6 - I along strike HORDA F AUL T ZONE
1;)::1
I»
en
!- - Sub- basin (e) "1 - - & - - -Sub- baSin (b) _I " Sub- basin (a) -I S·
§
()
o
.g
~

Fig. 9.22. Seismic cross section through the Permian-Lower Jurassic succession in the Horda Platform area of the northern North Sea Basin, showing the subdivision
of the Triassic succession into megasequences PRJ, PR2, and PR3. (Steel and Ryseth 1990)
tv
0\
\C)
IV
30/8-15 31/4-3 '-I
TRIASSIC SAND AACHTECTUAE, o
@ 3112-4
© HOADA PLATFORM
@
! STATIUOfIO
FM.
..---r ~-=:?c:::=.
...,:---- ~
--=""...".. _-::7 ~-
~~l~~~':;"" <='
-=7
t~
>-l
::r-
Ib
en
.......
a
ciQ'
...
i;:;.
_ ,-"I4Oj;'DI"l'lII',IIII.U'NC~lIn'f1IOMaArlC)ll"
>
...
_ _ _ _ WlOAMOUINCt: ~AJIIIT I' .... -t.. . . . .'
____ LtJI"'OEIMKMJII.'~~TiC)III
[.
Ib
n
~ 1M ... \, SAHOS'OtC' ...2
Ib
~~I ....... ".... T.tONl: MID..-uDS'OMI
o
...,
••
:;j !!i! i:! ];; :!!
$.
~
t1
Ib
Fig. 9.23. Correlation of nonmarine Triassic megasequences on the Horda Platform, North Sea basin. (Steel and Ryseth 1990) "0
o
fI)
a:o
::s
~
~
~
3
fI)
Fig. 9.24. Schematic cross sections perpendicular to active

•. ::.:. .:. . :. :->. . :.:: B
~
r.:::::
margin of A pull-apart, B rift, and C foreland basins, showing
.' . . . ' ~ .: : . .
tectonic cyclothems. Large-dot pattern, alluvial-fan and
braidplain deposits; small-dot pattern, fluvial, lacustrine,
VI'eo
and marine deposits. (Blair and Bilodeau 1988)
o. 102C30
4 SA 6 8 88 9 II 12 14 16 168
4-00001-......+ -+--1f-+-+ -t- H 65 000 years
G 5!l ZOO y ... r ·
36 100 years
E Z7 000 years
16800 years
IZ ZOO years
9500 years
A a years
IA 3 5 58 7 8A 8e 10 T1A I3 . 15 16A
10 ............. coars~ s~4 u .. nct:' B- marker tuff
(bas~ "f)
Fig.9.25. Schematic summary of an alluvial suite contain- scale is proportional to time. Thickness of the succession
ing up to seven marker tuff beds, showing the use of these varies between about 13 and 27 m. Old Red Sandstone
beds to provide a detailed subdivision of the succession. (Lower Devonian), southwest Wales. (Allen and Williams
Horizontal scale is arbitrary, but includes vertical profIles 1982)
mapped over a lateral distance of about 35 km. Vertical
years. These sequences are thought to reflect the in- tectonic origin are common in the geological record,
dependent operation of tectonic mechanisms on two in all types of tectonic setting (Fig. 9.24). They pro-
time scales (Yoshida et aI., in press; see Chap. l3). posed the term "tectonic cyclothem" for these suc-
Blair and Bilodeau (1988) reviewed a wide range cessions. Their tabulation shows cycles of 120 m to
of literature on sedimentation and tectonics and more than 2 km in thickness and representing 2.5-15
pointed out that large-scale cyclic successions of million years of sedimentation.
.IA 2 ,', 3', 4, 5,

•I
,, .."
"
"' v I'
"" .
"
,
· ,I
"I v ,
", :\~ v
.,.,'
coarse
~ sequence 5
,
.'.
"
II IIA ,.,
:I! :;:~12 _____ ----- ,.
bV
,'. . .
v::'"
·
'I
"
'",~
j
"""
Fig. 9.26. Details of unit V of the succession shown in Fig. corresponding to tuff F. Circles with arrows show variable
9.25 (section continues from proflle 10 in upper right to paleocurrent patterns. Dark spots in upper part of each
profIle 11 in lower left). Lower datum is tuff E, which is proflle indicate nodular calcrete development. (Allen and
truncated by a major channel to the left of profIle 12. Left Williams 1982)
(west) of profIle 10 the section is hung on an upper datum
9.S Methods of Correlation and Mapping 9.5.1 The Use of Marker Beds
Fluvial deposits are difficult to map in detail because Channel-fIll deposits are not laterally extensive rela-
of the frequency oflateral facies changes and the lack tive to many other types of sedimentary deposit. This
of distinctiveness of individual beds in successions limits their usefulness for mapping purposes. Super-
consisting of repeated similar channel and overbank imposed channel units may be lithologically very
units. similar, so that, in the absence of complete exposure
Basin mapping methods in general are discussed or some other data, lateral correlation of partial
in Miall (1990, Chap. 5), and most of this detail is not outcrops of channels is not likely to be very reliable.
repeated here. The focus in this section is on specific Detailed architectural examination of laterally ex-
techniques that have been found to be useful in tensive outcrops, including "walking out" of indi-
fluvial settings. The discussion is organized to flow vidual beds or tracing them on photomosaics, may
from the more generalized and widely applicable provide enough data for detailed mapping (e.g., Fig.
methods (Sects. 9.5.1-9.5.4) to the more specialized 9.7), but such situations are rare. However, a few
methods that are suitable for use only in certain field other types of fluvial facies may be extensive enough
situations (Sects. 9.5.5-9.5.9). Both outcrop-based to be useful for regional mapping purposes. Notable
and subsurface methods are discussed. among these are coal seams and paleosols. In addi-
Mapping may be based on the tracing of tion, nonmarine environments are the ideal location
unconformity-bounded units (sequences) in the for the preservation of tephras, and these may ex-
surface or subsurface, as illustrated in Figs. 9.18, tend for tens or hundreds of kilometers, providing
9.21, and 9.23. The recognition of such sequences is ideal marker horizons,
an interpretive procedure and is discussed in Chap. Allen and Williams (1982) documented a succes-
13. sion of Lower Devonian beds in southwest Wales
Methods of Correlation and Mapping 273
consisting of channel sandstones and interbedded tion profiles from the Carboniferous South Wales
floodplain units with calcretes and including eight Basin (from Woodland and Evans 1964).
airfall tuffs. They were able to map the tuffs over a As discussed in detail in Chap. 13, base-level
lateral distance of more than 30 km, and this enabled changes may lead to the generation of widespread
them to subdivide the succession into eight units units reflecting regional increases or decreases in
ranging from about 2 to 13 m in thickness. Each unit accommodation space. Coal seams and paleosols are
represents an average of about 8000 years of sedi- commonly deposited in response to such base-level
mentation, so that very detailed architectural recon- changes. Estuarine and lacustrine flooding during
structions are possible from these rocks. Figure 9.25 base-level rise can generate distinctive tidal sand
shows the architectural summary, and the details of bodies and widespread shales, the study of which can
one of the depositional units is given in Fig. 9.26. In aid in the definition and interpretation of non-
this interval a major channel up to 10 m deep incises marine sequences.
the alluvial plain, and is only partially filled with
typical channel-fill sand lithofacies. This led Allen
and Williams (1982) to suggest that the channel in 9.5.2 Wireline Logs
fact represents an incised valley formed during an
interval of base-level lowering, and that the channels The recognition of characteristic log "shapes" and
that subsequently filled it with sediment were on a distinctive vertical-profile character may be a useful
smaller scale. Note the mature pedogenic calcretes tool for correlation purposes and an aid for inter-
that blanketed the area prior to the deposition of tuff preting fluvial style. The technique is almost as old as
F. modern sedimentology itself (e.g., Nanz 1954), as
Individual coals and paleosols (discussed in noted in Chap. 2. The technique depends on the
Sects. 7.4.1 and 7.4.2) may extend for tens of kilo me- distinctiveness of the vertical profile through a sand-
ters across a floodplain and are therefore among the stone body and on our ability to interpret this profile
most laterally extensive and mappable units within in terms of depositional processes and environment.
fluvial systems. For example, Fig. 7.2 illustrates a The well-known bell-shaped gamma ray or sponta-
coal-bearing alluvial succession in Wyoming and neous-potential log response yielded by a typical
Montana; Fig. 9.26 illustrates a thick and are ally fluvial fining-upward cycle is a classic example.
extensive calcrete unit. Nemec (1988) argued that However, such interpretations are simplistic and
because coal seams are laterally persistent and can may be quite incorrect. Thus, gamma ray logs record
be correlated over wide areas, they can be used to the presence of natural background radioactivity,
examine the details of differential subsidence within which typically is highest in clay minerals because of
a basin. He developed a form of graphic correlation the concentration in these minerals of naturally ra-
between coal seams to be used for this purpose and dioactive isotopes of potassium and thorium. The
demonstrated the technique using detailed correla- log response is therefore interpreted as an indicator
,..------,~
GAMMA RAY GAMMA RAY GAMMA RAY
API API API
150 150
w
10m U
Z
w
20m ::J
o
w
30m (f)
--.J
W
40m Z
Z
«
I
u
SHALE ~""'""N-~
CONGLOMERATE
Fig. 9.27. Channel sandstone succession as seen in,three adjacent wells, and corresponding gamma-ray logs. Note the
interval of high gamma-ray values in the center well, reflecting the presence of a shale-clast conglomerate. (Rider 1990)
of" clayeyness", which is normally inversely propor- about 20 m are unlikely to be the product of within-
tional to grain size. However, the presence of clay- channel (autogenic) processes and are more pro-
rich clasts in an otherwise clean, coarse sandstone bably related to allogenic causes, such as sequence
may distort the reading (Fig. 9.27), and feldspar-rich development or tectonic pulses.
sandstones (e.g., "granite wash") also yield high Wireline logs may reveal the presence of region-
gamma-ray readings because of their high potas- ally mappable marker beds, distinguished by a rec-
sium content. Other problems with interpretation ognizable type of log deflection or because they
are addressed by Rider (l990). subdivide the vertical profile into intervals of dis-
A quite different type of problem is that, as distinctive log character. An example of such a profile
cussed in Chap. 8, fining-upward cycles in fluvial containing marker beds is shown in Fig. 9.28. Such
deposits are not amenable to unique interpretations. markers may serve to divide a thick nonmarine suc-
Channel fill and abandonment, point bar growth, cession into thinner, more readily mappable hori-
and certain tectonic processes can all yield fining- zons or "operational units". These markers typically
upward cycles and corresponding bell-shaped log reflect allogenic controls, such as changes in base
patterns. In the first instance, attention should be level (stratigraphic sequences) or regional changes
paid to the scale of the cycle. Cycles thicker than in paleoslope brought about by tectonic tilting. Their
G .N
BALDHU
OPERA 1'IONAL lNT
LJ'PER SLLGO
OPERATlONAllNT
I
..·.·...
·
LOWffi B\-l.GO
OPERA TIONAllNT
..··::·....
..·.:
:.:.:
.·-.:
'
...
..:: .
.' .
'
..~..·:." .
YoClMlARRA Fig. 9.28. Operational subdivision of a nonmarine

OPERATIONAl LNT succession using wireline log character. Note the
presence of various "marker" beds. G, Gamma ray
log; N, neutron log. Narrabeen Group (Permian),
Sydney Basin, Australia. (Hamilton and Galloway
1989)
origins are discussed in Chap. 11. Hanneman et al. the mixed marine-nonmarine Latrobe Group of the
(1994) discussed the subsurface mapping of paleo- Gippsland Basin is now based largely on the se-
sols and their use for defining sequence boundaries. quence-stratigraphic model of Rahmanian et al.
Calcic paleosols within clastic fluvial successions are (1990). Melvin (1993) discussed the influence oftec-
characterized by their high density and high interval tonism on basin development in the Endicott field
velocities, which makes them readily identifiable on area of Alaska.
density and sonic logs and in seismic records (Sect. Ryseth (1989) reported a subsurface study of the
9.5.4). Middle Jurassic Ness Formation in the northern
The use of sequence-stratigraphic concepts can North Sea Basin. He was able to subdivide this lOO-
considerably aid in regional mapping, even within m-thick succession into its main architectural com-
entirely nonmarine successions. For example, the ponents and to correlate them between four
Statfjord and underlying units in the northern North exploration wells over a distance of about 10 km
Sea can be subdivided into shale-rich and sand- using a combination of core and wireline-log infor-
stone-rich intervals, the development of which re- mation. Coals and paleosols defined the tops and
flects the balance between changing accommodation bases of a series of autogenic cycles which could be
space and sediment supply, as described in detail in traced between wells.
Chap. 13. Mapping of these variations and correla- Visher et al. (1971) used log character and shape
tion of major internal erosion surfaces interpreted to to discriminate various fluvial and deltaic environ-
be sequence boundaries forms the basis for field- ments in the Bartlesville Sandstone of Oklahoma.
wide and regional correlation (Steel and Ryseth Wightman et al. (1987) reported on a similar type of
1990; MacDonald and Halland 1993). Exploration in exercise applied to the Mannville Group of the
A B C
7·34·55·14W4 7·36·55·14W4 15·12·55·13W4
1700 '
1750 '
1800 '
1850'
D E
1OD·29·47·1W4 Depositional
6B·15·53·2W4
Environments
~Shallow
r<:J
1400 ' L.;.::J Marine Bay
1750 '
Distributary
.... .... Mouth Bar
flItUSplay
@i:bl Channe l
: , \ : ~.-:'.'
1500'
Fig. 9.29. Typical log character of sand-
1850'
stones in the Upper Manville Sandstone,
Lloydminster area, Alberta, showing in-
terpreted. environments. (Wightman et al.
GR 1987)
Fig. 9.30. Log facies types for the Permian Narra-

LOG FACIES TYPES FOR THE NARRABEEN GROUP
been Group, Sydney Basin, Australia. (Hamilton
and Galloway 1989)
(gamma log)
Discrete Stacked/amalgamated
chemeI-'.
~y
'~ard c:nan--fi•
.sandstone bocies
3uc
-<
·fIr"'
l,k)ward coat'$eI"Ing
,-
6
~Ione
~ ~iP'ethl1
lining-1.oward
cycles ..
Poaly defined
sandslone/m..dslone
Lloydminster area, Alberta (Fig. 9.29). Hamilton and ter (see Fig. 9.28). A sand distribution map of
Galloway (1989) discriminated various fluvial styles the same unit is given in Fig. 9.32. The two maps
in a Permian fluvial unit in Australia using log shape provide a good insight into the paleogeography of
(Fig. 9.30). Golin and Smyth (1986) used the tech- the unit.
nique for a Jurassic unit in Surat basin, and Jirik
(1990) used the same technique to recognize fluvial
subenvironments in a study of the Frio Sandstone in 9.5.3 Lithofacies Mapping
Texas. Davies et al. (1991,1993) suggested that verti-
cal-profile character could be used to discriminate Conventional methods of lithofacies mapping are
meandering from straight (including braided) chan- described in detail in standard textbooks (e.g., Miall
nel styles in a study of the Travis Peak Formation 1990, Sect. 5.3). Such techniques as mapping of net
(Lower Cretaceous) of east Texas, although this type sand content, or sand/shale ratio, or net pay are
of approach may not always be reliable, as argued by standard procedures for the delineation of reser-
Miall (1980). voirs and for use in subsurface trend prediction.
The most effective use of log shape as a mapping However, these techniques are oflimited usefulness
tool is in conjunction with other lithofacies mapping for exploration in fluvial environments, because of
techniques, such as those discussed in the next sec- the extreme facies variation in these rocks, the thin-
tion. Thus Fig. 9.31 shows the distribution of log ness of the sandstone bodies, and their lack of direc-
"facies" within an informal subsurface stratigraphic tional predictability. In most exploration situations
unit that had been defined on the basis oflog charac- well spacing is considerably greater than the average
EXPLANATION
====:> Ouor 11.0$.e SOUrce
crrr:rp:. Mixed sourCe
~ Lilhlt $Ol,Irce
~ Pebbly OIJt'Wll'lClSh onCl bro~ed (nonne.1 tOties

~ Cooru-9fo ineo meonde,~lt compl el
~ FloodplOln focies
~ Loc.l,nMinoe tocles
Fig. 9.31. Distribution of characteristic log facies types in data. The combination of data types provides an excellent
the Wombarra Operational Unit, Sydney Basin. Arrows paleogeographic map. (Hamilton and Galloway 1989)
indicate sediment sources interpreted from petrographic
dimensions of the typical sandstone body (except, quite possible for the same data to lead to different
perhaps, for the length of major paleovalleys). Nar- mapped contour patterns, reflecting the different
row channels and small point bars may therefore be paleogeographic models of the authors. For ex-
missed, except where well spacing is unusually ample, compare the "fan" model for the Westwater
dense, as in some enhanced recovery projects. It is Canyon Member of the Morrison Formation of New
t
Scale
0 10 20 30 40 50
km
Fig. 9.32. Lithofacies map of the W ombarra Operational Unit, Sydney Basin, showing sandstone thickness expressed as
a percent of total unit thickness. (Hamilton and Galloway 1989)
Mexico (Galloway 1980) with the channel model of reservoir engineering purposes, is normally carried
Saucier (1976). out by subdividing the producing formation into
Mapping of units defined by the sixth- and "sheets" or "flow units" or "petrophysical zones"
higher-order surfaces according to this book can be that extend across the entire field (e.g., Wadman et
based on conventional methods of subsurface al. 1979; Jin et al. 1985; Yin an et al. 1987; Lawton et al.
wireline log correlation, provided well spacing is 1987; Struijk and Green 1991; Melvin 1993). Prin-
adequate (many examples of this are described and ciples of sequence stratigraphy, incorporating ideas
illustrated in the sections on petroleum geology). In regarding eustatic and tectonic control of fluvial
most mature basins, with a well spacing of one to a depositional systems, may be introduced at this
dozen wells per township (wells 1.5-5 km apart), stage to facilitate stratigraphic interpretation (e.g.,
such correlation should be possible down to the Melvin 1993). Production engineering and computer
"sequence" or submember level, although local modeling for development purposes are normally
problems may arise as a result of internal erosion based on this type of stratigraphic subdivision of the
and the resulting amalgamation of sandstone sheet reservoir units (Stanley et al. 1990; Melvin 1993;
units. An example is illustrated in Fig. 9.23. Com- Martin 1993).
puter modeling of lithologic units at this scale, for
Fig. 9.33. Pattern of ribbon sand-

50 km stone bodies in the Upper Mannville
Sandstone, as proposed by Putnam
24 20 20W3M
(1983). Insets show some ofthe details
15 10 ,5
i ' 64 of data-point distribution
J ,
Mapping of ribbon sandstone bodies in the sub- alternative interpretation. They showed from de-
surface can cause particularly difficult problems, tailed core studies of lithofacies, trace fossils, and
because the width of the sandstone bodies is usually vertical sequences that many of the sand bodies are
less than well spacing, except in densly drilled fields. of deltaic origin, including mouth-bar and splay
In addition, it may be difficult on wire line logs to sandstones, and others are marine sheet sands (Fig.
discriminate between channel sandstones and the 9.29). The anastomosed fluvial model was decisively
various overbank sandy facies, including levees and rejected in favor of a more elaborate deltaic model (a
splays, which may occupy a much wider belt flanking deltaic setting does not, by itself, exclude the possi-
the channel than the channel itself. An example of bility of an anastomosed fluvial style. As discussed in
the problems in interpretation that can arise is the Sect. 13.3.3, this fluvial style is commonly developed
pattern of anastomosed ribbon sandstone bodies during times of rising base level).
that were reported in the subsurface Mannville For- Figures 9.34 and 9.35 show differences in strati-
mation of Alberta and Saskatchewan by Putnam graphic interpretation that illustrate the difficulties
(Smith and Putnam 1980; Putnam 1982a,b, 1983; Fig. in defining narrow channel sand bodies. The inter-
9.33). His analyses were challenged by Wightman et pretation by Putnam (1982b) shows the sandstone as
al. (1981) on several grounds. They pointed out that a continuous, ribbon-like body (Fig. 9.34). However,
Putnam used an arbitrary classification of Wightman et al. (1987) demonstrated, with pressure
petrophysical log character to define channel sand data, that the sandstone in well 7-33 is not in fluid
bodies, that some of the bodies correlated between connection with that at 7-34 and 7-35. There are,
adjacent wells are of different thicknesses, and that therefore, at least two separate sandstone bodies
they are at slightly different stratigraphic levels, sug- (Fig. 9.35). Note also the interpreted presence of a
gesting that the channel patterns may be an amal- lower channel sandstone in well 7-34 in Fig. 9.35, a
gamation of different channels that were of different unit that is incorporated into the main sand body in
ages, and not all active at one time. The channels in Putnam's intepretation.
Putnam's reconstructions are also much larger than Mapping and reservoir modeling of ribbon sand-
any described from modern anastomosed systems. stone bodies is best attempted when a large volume
In a later paper Wightman et al. (1987) presented of subsurface data is available from a well-developed
detailed sedimentological evidence in support of an field (well spacing of 32-54 ha; 8-4 wells per one-
NW SE
10-2 7-1-56-15W4M 7-32 7-33 7-34 7-35 7-36-55-14W4M
8 i)-1.4k.m -¢-. 4 km----... ¢--1.6km-i)-1.6km-i)-1.6km-i)-1.1k~-i) 1 8
~ GAS WELL
£L 1km
SP
C
R
Spontan.oul Pot.ntiat
Ruil/ivity
Conduct ivity
. . CHANNEL SANDSTONE
Fig.934. Stratigraphic interpretation of sandstone bodies in the Upper Mannville, near Uoydminster. (Putnam 1982b)
7-32 7-33 7-34 7-35 7.J6.5S-14W4
* - 1.6km-*-sp 1 .6km-s"·';':)
I l:Er;t-.·_1.6km~I.7km-*
Sp
Datum W P sp SP E
Viking - ---.H----""'P'-t--------t-t----+-If------..:'+--
1600'
1650 ' 1650'
1650'
Top 1700 '

Man nvi lIe---..J...+--::=~;;~'::
Group 1750'
Channel
Sandstone
1850 '
Top
Middle _-..I.4---?-'t'--
Mannville
Fig. 9.35. Stratigraphic interpretation along same line of cross section as Fig. 9.34, showing alternative interpretation.
(Wightman et al. 1987)
mile-square section). Yin an et al. (1987) noted the He demonstrated that only where well spacing was
difficulties in producing from "straight" and "con- equal to or less than the average valley or channel
fined channel" (anastomosed?) sandstone bodies width could a reliable map be constructed.
because of the discontinuous lateral geometry and Conventional lithofacies mapping of fluvial
the need for a particularly dense well spacing. stratigraphic units, for example, sand/shale ratios,
Robinson (1981) discussed the problem of mapping net sand isopachs, etc. (e.g., Tyler and Ethridge 1983;
paleovalleys from a statistical point of view (prob- Hamilton and Galloway 1989), may not be a very
ability of intersection depending on well spacing). effective technique for reservoir development pur-
s
r:. S::::
Bosin Member 51
'"
ET-
0
0-
Cr>
0
....,
n
d":1 ,,- - ----=-:> 0
-===--:..""=-"'"
......~
~
o·
0
": : : : ~?'.I . W~~h : ~~~~~r: : :::: 0
"'
0-
s:::
m
r
"'
'""E.
o .m 0
<)Q
W E
B'
II m B
OTO 5 - 3 SR8 LOCATION OF CROSS SECTIONS
r Iii
o ~m
20 I I
KXl
200 10
'0
o 2m
~i~
J '..J
'0
I ,• IV • E9 n o r
o 3.m
1~ ~'E!...
: OolOI n Co
~ CroVOSh splay D Floodplain dtpoSlts
I
[ill] Po.nt bor ~ Abondontd chonnt' lill
I.....
n Low · sinuosity ~f.
LJ channe l depOSItA .... • Do1o't Cleek
Fig. 9.36. Mapping of sand bodies in the subsurface. Strike- (A -A') and dip- (B-B') oriented cross sections through the Salt Wash Member of the Morrison
Formation (Jurassic), westernmost Colorado. (Tyler and Ethridge 1983)
tv
00
Contour Inlerval 10m experience with the Westwater Canyon Member

(Miall 1988a), macro form bar units, bounded by
fourth-order surfaces, and the subunits within them
that are defined by third-order surfaces would re-
quire well spacing of 32 ha (8 wells/section) or less if
their geometry is to be reliably reconstructed from
wireline log and core data (Fig. 9.38, bottom). In the
. '" Westwater Canyon Member, even with a well spac-
ing of 4 ha (64 wells/section) many of the smaller
individual fifth-order units (e.g., Mia1l1988a, Fig. 4)
would not be correia table. In larger river systems the
macroforms are correspondingly more extensive,
and a wider well spacing might be adequate. For
example, a point bar 6 km in diameter described by
Busch (1974) was mapped using a 16-ha (16 wells/
section) spacing (Fig. 9.39). In their study of the
Gypsy Sandstone of Oklahoma, Doyle and Sweet
(1995) determined that channel deposits bounded
by fifth -order surfaces were the smallest scale of unit
I' that could be reliably correlated with a 100-m well
1;2 spacing. Other examples of subsurface channel and
!II i' " ,',' 0'
~~_/ )/./ . point-bar mapping are described in Chap. 15.

The necessary data for detailed lithofacies map-
u _"/'
o >'~ ./ .'
IOll.m
A. ...-----,----6.....~, ping should become available from pilot studies car-
ried out at the commencement of enhanced recovery
Fig. 9.37. Isopach map of the upper Salt Wash Member, projects. Engineering grid-block modeling at this
Colorado. Note linear trends of isopach "thicks" which scale may be a useful intermediate step between
indicate the general position of channel belts. (Tyler and models of the large flow units and those constructed
Ethridge 1983)
from core-test data.
Mapping of a point bar provides information on
poses. The construction of such maps requires the the scale of the river, such as its meander wavelength.
definition of a mappable stratigraphic unit, but in This may become a useful prospecting tool, as scat-
fluvial deposits there are few mappable markers, and tered well data may be used to predict meander
so such units tend to be much thicker than the chan- position by the sketching in of possible meander
nel belts, with the result that the maps show the positions to accommodate the available data (e.g.,
average position of several or many channels with sandstone penetration corresponding to bar, shale
different position and trend. Figure 9.32 is a "sand" possibly interpreted as abandoned channel fill). This
map of one of the Permian operational units in the technique has, to the writer's knowledge (consulting
subsurface of the Sydney Basin, Australia. Figures contacts), been used as an aid in the exploration of
9.36 and 9.37 provide cross sections and an isopach Tuscaloosa prospects in the Gulf Coast area and
map of the Salt Wash Member of the Morrison For- Permian sands in Texas and may be in widespread
mation in Colorado. These diagrams illustrate this use in other mature basins. As discussed in Sect.
point. The isopach maps provide only the most gen- 1004.1, use of surface information to model the scale
eralized information regarding channel position and of channel sandstones may also be a helpful ap-
could not be relied upon as a subsurface mapping proach.
tool for locating actual channel bodies. In the case of One of the most detailed published examples of
the Salt Wash Member, channel belts are 2-10 km subsurface mapping that has been carried out in
wide (Fig. 9.37), whereas individual channels are fluvial sandstones is illustrated in Fig. 9040. Here
only a few hundred meters wide (Tyler and Ethridge gamma-ray logs from very closely spaced wells have
1983). been used to construct detailed ("ross sections across
However, where well density is sufficient, detailed two sandstone bodies. They show that minor inter-
lithostratigraphic subdivision and channel mapping nal erosion surfaces can be detected and mapped in
may be a very effective exploration tool. Based on the subsurface, given adequate data. These may be
Fig. 9.38. Depositional systems and geo-

GEOGRAPHICAL SCALE DEPOSITIONAL ELEMENT SUBSURFACE morphic elements compared at three
MAPPING scales. Top: The San Juan Basin (Colo-
METHODS rado-New Mexico) and a speculative
braidplain system (Westwater Canyon
Member), shown side by side at the same
Son scale; middle: sheet and ribbon sand-
Juan correlation stones in the Westwater Canyon Mem-
01 ber (right) shown at the same scale as a
Basin wire line I-square-mile section with two different
logs densities of well spacing (left); bottom:
enlargement of a section, shown at the
same scale as a small macroform. lsd,
Legal subdivision (Canada). (Miall
1988a)
L..-J
·5 0 km
section (1 mile2 )
-
- -
8 we lls / section
64 wells / section
••••••••
• • • • • • • •
1 km
__ 64 wells / sec tion - 7 _

(4/lsd)
~--=:;)
macro/arm area : 40 ho
correlation
01
core ?
~~
--:>
~ Isd (1320 It / 402 m ocross }------~)I
100 m
fifth-order surfaces, recording successive positions features are readily mapped using seismic methods.
of a channel as it migrated laterally, or they may be For example, Fig. 9.41 illustrates the essential seis-
third-order surfaces, indicating erosional breaks in mic-facies characteristics of a fluvial-deltaic clastic
the filling of the channel by lateral or downstream wedge prograding from a fault-bounded basin mar-
accretion. gin. The mountains (a) may be flanked by an allu-
vial-fan zone (b), which would likely yield reflections
of variable amplitude and poor continuity, because
9.5.4 Seismic Methods M the poorly stratified and variable lithofacies of
most alluvial-fan deposits. Braidplain deposits,
Many of the major structural features typically asso- downstream from the fan (c), may yield good to
ciated with fluvial deposits are readily detectable on discontinuous reflections, depending on the degree
reflection-seismic records. For example, as shown in of channelization and preservation of floodplain de-
Chap. 11, thick clastic wedges are likely to be associ- posits. Shale beds in sandy braided stream deposits
ated with rift faults or fold-thrust belts, and regional are commonly are ally extensive (e.g., Morrison For-
unconformities may be crossed by incised paleo- mation: MiallI988a), especially where they have de-
valleys filled with fluvial deposits. These large-scale veloped in response to regional tectonic adjustments
Fig. 9.39. Isopach of producing sandstone, Little Creek

field. Mississippi. (Busch 1974)
fIElD
17 .. "
ISONoCH IHUIYAl · 20 It
~ C-C-'-[~~g-:~~
ttfFff(fif€~\E<r:~ t ( ~-%
Fig. 9.40. Two gamma-ray log cross sections through the subdivide the sandstone into four discrete lenses. (Walton
producing sandstone in the Bronson-Xenia field. Kansas. et al. 1986)
Note presence of internal, dipping bounding surfaces that
or base-level changes, and may be thick enough to regional unconformity. For example, there may be
yield good reflections. As in the alluvial-fan zone, widespread transgressive deposits formed on a
channels marked by lithologic contrasts between the "maximum flooding surface" (f). Such facies, and
channel fill and the channel banks may give rise to the unconformity with which they are associated, are
seismic scatter and to the development of short, commonly readily traceable on seismic records
confused reflections, or to diffractions. Close to the (Chap. 11). An example is illustrated below.
coastline (spot d in Fig. 9.41) seismic facies may Meadows and Beach (1993) demonstrated how
change as more floodplain deposits are incorporated seismic data could be used to distinguish between
into the succession, giving rise to a greater and more fluvial-dominated and eolian-dominated sandstone
are ally extensive acoustic contrast between sand- successions in a Triassic rift fill, and they docu-
stone bodies and overbank materials. Reflections mented the distribution of these broad facies varia-
may therefore be more continuous and of higher tions relative to active faults in the basin. Brown and
amplitude. The fluvial wedge may pass laterally into Fisher (1977) described the typical seismic-facies
a delta, distinguished by clinoform relections (e). characteristics of alluvial and delta-plain facies in
Where the clastic wedge is part of a sequence con- Brazilian offshore basins. Reflections are typically
trolled by base-level change or tectonism it may be "horizontal, parallel, rarely divergent, layered to lo-
sandwiched between other facies, as marked by a cally reflection-free; locally, erosional channels may
Fig. 9.41. Simplified seismic facies
Wf::-~~~_~_ -Z-~
characteristics of a fluvial-deltaic
clastic wedge prograding from a fault-
bounded basin margin (Anstey 1980).
- -,- - - -- See text for explanation
0)
O~r-~F~W ------_--__
-V-IA-L----_---------- --- ' 0.5 r=-------------==-----:===.----,
---=-= - -- -=ALiiNlAL ~
-= --:=--=,\ - - OELTA=!'LAIN--
1,0""""-".... . ___ -- - .....
=
_ "'";5:oELTA:- RRIER BAR
"-=--.~
---
I KM. QlP SECTION
A
STRIKE SECTION
B
l.o-:l~:: ~~~~ ~~~

- =- -- ---
------..
I KIA DIP SECTION
D
3 .5
DIP SECTION DIP SECTICN

C E
----
I KM DIP SECTION
F
Fig. 9.42. Deltaic and associated seismic facies patterns, redrawn from original seismic sections through Brazilian
offshore basins. (Brown and Fisher 1977)
be inferred. In strike sections, the reflections are variable (high in continuous reflections and poor in
weak, parallel-layered to subtle-mounded, chaotic- chaotic zones), and spacing is very regular in zones
to-drape patterns. Continuity of reflections ranges of high-continuity reflections but irregular in the
from excellent to fair in dip sections ... , but conti- remainder of the unit. The reflections collectively
nuity is poor to fair in strike sections ... ; amplitude is define a tabular external geometry ... " (Fig. 9.42).
WAVELET
seismic reflection survey of a fluvial-lacustrine de-
ZERO PHASE posit (Ray 1982). A synthetic seismogram is shown in
10·15·45·50 HZ
REFLECTION NORMAL POLARITY Fig. 9.43 and Fig. 9.44 is a seismic cross section. Five
COEFFICIENTS unconformities have been recognized in this succes-
sion, at 810,1000,1295,1360, and 1670 ms (depths on
, "I ''''''~ SONIC VELOCITY o the right-hand side of the section, Fig. 9.44). These
KIt/sec
r
unconformities separate distinct stratigraphic se-
~ quences. A particularly instructive example of the
r~ 1.,.........--=-
j
...... lateral changes in seismic facies discussed by Anstey
(1980) is shown by the sequence between the 1000-
t 1
and 1295-ms unconformities (sequence C). Note the
F ~ east-dipping clinoform structure of this sequence.
~ miND ntH
r21RiVER Note also the change in reflection character from
~ FM. .S
........
It1"
low-amplitude, low-frequency, low-continuity re-
~3~ .-.- .....

IIIII
flectors in the upper part of the section, particularly
in the west, to a region of high-continuity, high-
~ ~; ~ ~ amplitude, high-frequency reflectors in the middle
Yo!
of the unit (the clinoform structure is particularly
.;:E
j: ~E4~__~ "" t
clear here), to a zone oflow- to moderate-continuity,
.> :z:
1.0 ~
< =:5
Yo!
1.0
low-frequency, low-amplitude reflectors in the
lower part of the sequence. These changes are inter-
Q preted as lateral facies changes from discontinuous
o
:~
I-
6"1
j - ...... ...
.....
- fluvial sandstones and shales in the west, to
interbedded sandstones and shales of the lake-mar-
I....·
~~!'~ ..
gin delta clinoforms, to uniform lacustrine shales in
7
ra-
...l-
.........
the east.
The ideas in the preceding paragraphs, including
t -............... 1.S
8
~~
-.
those of Anstey (1980), Brown and Fisher (1977), and
Ray (1982), relate to the recognition of large-scale
1.S SEa. A
1-
.uu fluvial depositional systems within other types of
--
9 ~~~~~
stratigraphic units. Much more difficult is the map-
ping of individual sandstone bodies within a fluvial
1O~LANCE system. As pointed out by Neidell and Beard (1985),
~ careful attention to processing and presentation of
seismic data can lead to considerable refinements in
Fig. 9.43. Synthetic seismogram for a well through the our ability to map subtle stratigraphic objectives. A
Fort Union Formation and related deposits, Powder River knowledge of the structure of the propagating wave-
Basin, Wyoming. (Ray 1982, reprinted by permission)
form and the use of color to display amplitude varia-
tions are two vital aspects of this work. Fulthorpe
Paleosols may be characterized by distinctive (1991) demonstrated that seismic resolution of
petrophysical properties, and may therefore be stratigraphic features is increased in areas of high
mappable using wireline logs (Sect. 9.5.2) and seis- sediment supply, because the deposits are thicker
mic methods. Paleosols are commonly distinctive and therefore more readily resolvable. Also, deposits
stratigraphic markers (Sect. 9.2) and in some cases that develop by lateral progradation are much more
are used to define sequence boundaries (Sect. readily subdivisible into their component units than
13.3.1). They can therefore be of considerable strati- those which accumulate by vertical aggradation, be-
graphic importance. Hanneman et al. (1994) cause of the greater scale of depositional units in the
mapped calcic paleosols in some Miocene beds in horizontal dimension. Cartwright et al. (1993)
Wyoming using seismic methods and wireline logs pointed out that the improved resolving power avail-
and demonstrated that their petrophysical visibility able from modern seismic acquisition and process-
was caused by their high densities and interval ing techniques, relative to those available when
velocities. seismic stratigraphy was first introduced in the
The Paleocene Fort Union Formation of the Pow- 1970s, has led to an increase in the demonstrable
der River Basin, Wyoming, provides an example of a complexity of stratigraphic units and to a need for
WEST SEISMIC LINE A EAST
~
'"ET-
0
0.-
en
0
.....,
FU ()
0
....
....
1.1l-
['"
o·
. l FU ::s
I»
::s
0.-
~
l AN
~
"Cl
5'
C1Q
2.0
Fig. 9.44. A typical seismic line through the Fort Union Formation, Powder River Basin, Wyoming. The Lower Fort Union (LFU) shows a gradation from low-continuity, low-
amplitude, low-frequency fluvial strata at left, through high-frequency, east-dipping clinoform shoreline deposits, to lacustrine deposits at lower right of unit. (Ray 1982,
reprinted by permission)
IV
00
"
w ..
4
.
1
.
2 t E
.l20-~-
--,.- - . - \-_~"",,~'~
'-------:#:-::C:C"""":--:-
'NG----:AV[C:--ftAG(-- .120
...
1\ I. • .' .~. ..~
(E!SEISMIC
r\L OF~vr. ISOCHPtOM ~~
' ./ ~~.,..:.....--.", i=~
!
.110- -'
'. ,110
-l.OO ---+-- -- - - - -- -- -- - - - - - + - - - - - -··>000:::
-3•• 0----+-- - - -- - - - - - - -- - - - - - 4 - - - - - - -·»00
! i em
A ·,••O -~-------------------------- -16oo
Fig_ 9.45. A Cross section made from seismic records across a paleovalley filled with Red Fork sandstone, South Ceres
field, Oklahoma. B Reflection picks which determined the channel. (Lyons and Dobrin 1972, reprinted by permission)
considerably more sophistication in the interpreta- thin to be picked up on most seismic data. Dobrin
tion of seismic data. This is of particular importance (1977) reported the difficulty in mapping the three
in the study of stratigraphic sequences, as discussed major productive fluvial sandstone lenses in the
in Chap. l3. Candeias field, Rec6ncavo Basin, Brazil, using good-
Fluvial deposits are commonly characterized by quality seismic data. The sandstone bodies them-
lenticular sandstone bodies, including paleovalleys, selves were not visible on seismic sections, and the
channels, and bars, but these may be difficult to structure was confused by a velocity pull-up gener-
detect on reflection -seismic records, because of their ated by an adjacent, massive sandstone unit.
small size and the poor acoustic contrasts between Sandstone bodies and channels can be identified
the sandstone body and its host strata, such as a and mapped using seismic data, but the best results
channel fill and the sediments into which it is cut. tend to be achieved only when a considerable
The general comments of Tipper (1989) regarding amount of local information is available regarding
the nature of facies "mosaics" apply in particular to the stratigraphy, the seismic character of the units in
fluvial settings. Many fluvial sandstone units are too question, and the likely associations, orientations,
. "c "
-=--[
,
and configurations of the bodies to be mapped. In
other words, seismic data cannot readily be used as a
prospecting tool for fluvial sandstone bodies in fron-
fA
~
I
!I tier areas, but they may prove invaluable for extend-
ing fields or finding infill fields in well-known areas.
=~ Success depends on using the information obtained
L
from existing fields to carry out very refined seismic
I
I modeling and interpretation.

t::=-
fA
--so I Paleovalleys, where a sandstone is cut into a lime-
~ stone, may yield good reflections, as maya sand-
fA
p-
-J stone that is porous and gas filled. These are ideal
situations (Anstey 1980). In some cases the sand-
stone of the valley fill may resist compaction more
than the host strata, in which case a slight structural
II anomaly may appear in the records. Figure 9.45 illus-
I 8 ~
trates an early success story in the mapping of a
'0
I valley fill in which the shale became compacted
around the sandstone, leading to a slight drape
structure.
Fig.9.46. Velocity logs and synthetic and recorded wave-
forms for five types of stratigraphic section; Morrow Sand- The Morrow Sandstone (Pennsylvanian) of Kan-
stone, Oklahoma. (Dobrin 1977, reproduced from work by sas and Oklahoma has been extensively studied us-
Waters and Rice 1975, reprinted by permission) ing seismic methods (Clement 1977). This is a
transgressive marine unit that fills paleovalleys
eroded by fluvial processes during a preceding re-
gression. The sandstone bodies are therefore similar
to fluvial paleovalley fills and other types of channel,
and the details of the techniques by which they have
been mapped are instructive.
Dobrin (1977) reported on a study of the channel-
fill sandstones in Oklahoma using seismic modeling,
followed by the application of pattern-recognition
techniques in an attempt to map distinctive types of
wavelets in the subsurface. Synthetic and actual
mapped data are compared in Fig. 9.46, and Fig. 9.47
is a map showing the interpreted distribution of
several types of channel in part of Oklahoma.
A modern and much more sophisticated study
was reported by Clark (1987). Figure 9.48 is an ideal-
ized geological and velocity model for a Morrow
channel in an area of southern Kansas, and Fig. 9.49
is the synthetic seismic response generated using
this information. The channel is constructed to be 18
m deep on the left, thinning to 6 m on the right. It is
• s, CNnrwt filled with shale, except for 3 m of sandstone at the
ms'''''-
~s, ~ TvPM "'~oNl.
~ top.
$4 T"IM III'rilCloft!o,,...IK
The channel is at 200 ms in Fig. 9.49. It is an
- 01"""
S. ~""' IC L~
obvious anomaly, caused by the slow velocity of the
Fig.9.47. Mapped distribution of ptpes of channel body, channel fill in contrast to that of the overlying and
based on statistical analysis of waveforms recorded along underlying rocks. Note also the sagging of the over-
seismic lines. (Dobrin 1977, reproduced from work by lying Marmaton anomaly at 155 ms, as a result of
Waters and Rice, reprinted by permission) differential compaction, with the channel fill com-
pacting more than the host strata. The apparent
faulting in the underlying Arbuckle reflector, at 260
ms, may represent only apparent breaks caused by
0' Fig. 9.48. Idealized geology and velocity model

0' '
1.0,00' 2P,00' .
aQOO' 4,0.00'
.
5,0.00'
0' for a Morrow channel, southern Kansas. (Clark
17,000' ISEC 1987, reprinted by permission)
1,000' , ~,OOO'
MARMATON....., C 11,600 ISEC
MORROW 14,900'/SEC 16,600'/SEC
g;1,wremRE~- - =-=-::-- -----==:::---
--~.
2,000' 11,800' I SEC 18,400' ISEC 2,000'

-- ---_______ aau
ARBUCKLE ----rc11~O-o~is~c
LIMESTONE
3,000' I1J SANDSTONE 19,600' ISEC 00'
IHi!I SHA LE
.1 i
j
:~
I
BANDPASS
14-65 HZ
.3 PHASE: 0
.4 4
SHOT POINT INTERVAL: 165' Fig. 9.49. Synthetic seismic cross section generated from the
TIME SAMPLE RATE: 2 MSEC
data shown in Fig. 9.48. (Clark 1987, reprinted by permission)
150 155 160 165 170 175 180

, I . I . I ... , I . ' . , I ! I ! ! ! • I ...
SOUTH LADD 2-8 CANNON
SEeS 30S 25W
LADD 1-9 HARRIS
SWSESE
GAM ..... • DENSITY
CHEROKEE
MORROW
1.0_. ._1.0
U U
Fig. 9.50. Unmigrated seismic line
across a Morrow channel, southern
Kansas. Note three wells projected
onto the line. The channel is the
strong positive reflector immediately
above the Mississipian limestone
(stippled) at 1 s, between the 160 and
173 shot points. (Clark 1987, re-
printed by permission)
ray deflection from the edge of the Morrow channel. similar to those described above for the Morrow
All these anomalies may be used to locate actual channels.
channels. An actual sesimic line through this area is Many of the stratigraphic traps discussed later in
reproduced as Fig. 9.50. Note the "sagging" of the this book, especially the CB-type fields, are charac-
Marmaton, the "faults" in the Arbuckle, and the terized by structural drape over a laterally limited
strong positive reflection from the channel itself. sandstone ribbon or lens (Chaps. 14, 15). Potentially,
Other examples of channel and paleovalley map- these small structural anomalies can be mapped us-
ping include those by Wood and Hopkins (1989), ing refined seismic techniques, once the acoustic
who illustrated a seismic section across a Mannville properties of the rocks in question have been tho-
paleovalley fill in the Little Bow area of Alberta. roughly investigated by regional seismic mapping
Here, the differential compaction of the valley fill and the use of well logs for velocity control.
relative to the host strata provided the key for the Brown (1985, 1991) and Weber (1993) discussed
identification and mapping of the channels. the applications of three-dimensional seismic data
Sonnenberg (1987) mapped paleovalleys in the "D" to the mapping of stratigraphic objectives. Channels
sandstone of the Denver Basin using techniques that yield very subtle reflections in vertical cross
Fig. 9.51. Horizontal 3-D seismic section showing meandering channels in Miocene-Pliocene fluvial-deltaic deposits,
Gulf of Thailand. The original illustration was in color. (Brown 1991, reprinted by permission)
Fig. 9.52. Horizontal slice section, part of a 3-D seismic are superimposed. The original illustration was in color.
section showing bifurcating deltaic distributaries, Ceno- (Brown 1991, reprinted by permission)
zoic, Gulf of Mexico. Structural contours, in milliseconds,
section may be quite spectacular when viewed in 9.5.5 Ground-Penetrating Radar

horizontal sections. Figures 9.51 and 9.52 illustrate
two examples of the mapping of channels. The use of Ground-penetrating radar (GPR) is a relatively new
3-D seismic method holds great promise for the de- technique that has found increasing application for
tailed mapping of channels and bar deposits in PV- the delineation of the shallow subsurface in a wide
and CB-type fields, and even for the better definition range of geological applications (Davis and Annan
of reservoir heterogeneity in SH-type fields (termi- 1986, 1989; Moorman et al. 1991). The technique
nology is defined in Chap. 14). Increasing refine- involves the transmission of high -frequency (10-100
ment in acquisition and processing techniques will MHz) electromagnetic pulses and the recording of
lead to the definition of ever-smaller features, al- reflected signals, which are processed much as are
though limitations in resolution are imposed by the seismic signals. Reflections occur at the interface of
nature of the seismic wave, which has a spherical beds with contrasting electrical properties, and the
wave front and generates reflections over a zone depth of penetration depends on the attenuation of
termed the "Fresnel zone" that may be several to the signal. Greatest penetration and lowest re-
many meters across. A. Nur (personal communica- flectivity are yielded by unconsolidated sands and
tion, 1993) noted that in some circumstances, such gravels and dry sandstones. Attenuation increases
as in the shallow tar-sand studies being carried out in with saturation, and with decreasing grain size,
northern Alberta, 3-D seismic resolution may be able reaching the highest values in the case of wet
to resolve features as small as 4 m across. At depths muds. Characteristic reflection strengths and con-
of 2 km objects 20 m wide should be resolvable, figurations permit the erection of "radar facies" and
which should permit the mapping of individual mi- "radar sequences" in much the same way that seis-
nor channels and bars. mic facies have been delineated (Huggenberger
1993; Gawthorpe et al. 1993; Alexander et al. 1994). are shown in Figs. 9.55 and 9.56), and with the assis-
Core or outcrop data are desirable to "ground-truth" tance of biostratigraphic data or information from
the radar facies. radiometric dating of interbedded tuffs, the local
Several applications have demonstrated the abi- record may be correlated with the global time scale
lity of the tool to map architectural details in fluvial (Fig. 9.57). The procedure is summarized in Miall
deposits down to depths of as much as 30 m, with (1990, Sect. 3.7.5).
resolution as high as 10 cm, although where the Not all fluvial deposits lend themselves to paleo-
section consists of saturated fine-grained deposits, magnetic study. The quality of the results decreases
as in the case of modern floodplains, penetration sharply with increasing age, because of diagenetic
may be only a few meters. Both modern (Moorman complications, and, because the magnetic signature
et al. 1991; Huggenberger 1993; Gawthorpe et al. is best obtained from fine-grained units, reliable
1993) and ancient (Stephens 1994) fluvial deposits results depend on the presence of such facies at least
have been examined. every few meters through the succession. The work
There are two main potential applications of the on the Siwalik Group, which is the Late Cenozoic fill
technique. of the Himalayan foredeep basin in Pakistan
(Behrensmeyer and Tauxe 1982; Behrensmeyer 1987;
1. For the the evaluation of sand-body architecture
Johnson et al. 1985; Burbank and Raynolds 1988;
as an aid to the study of reservoir heterogeneity. A
Burbank et al. 1986; Mulder and Burbank 1993) is of
particularly powerful approach is to examine an
the highest quality and has been widely quoted be-
ancient unit in outcrop, using two-dimensional
cause of the insights the work has provided on strati-
lateral profiling techniques (Chap. 4), and to run a
graphic architecture (this chapter) and tectonic
GPR survey over the top of the outcrop, in order
controls (Chap. 11). The data permit the subdivision
to calibrate the radar data and to extend the archi-
and regional correlation of fluvial successions down
tectural analysis back into the third dimension.
to the group 8 or 7 level, as discussed in Sect. 9.2.
Pratt and Miall (1993) reported a preliminary
Under ideal conditions, this is comparable to the
study of a Silurian carbonate shoal deposit car-
subdivisions that can be erected based on inter-
ried out this way. Gawthorpe et al. (1993) recon-
bedded tuffs, coals, and paleosols, as described in
structed lateral-accretion bedding in a modern
Sect. 9.5.1.
point-bar deposit. Stephens (1994) examined an
ancient fluvial system and was able to document a
history of complex braid-bar development.
2. The evaluation of the flow of groundwater and 9.5.7 Paleocurrent Analysis
toxic wastes through shallow aquifers, including
ancient sediments and modern surficial deposits Methods of outcrop paleocurrent measurement and
(Huggenberger 1993; Knoll et al. 1994). documentation are discussed in Sect. 4.4, in which it
is shown how orientation data relating to cross-bed-
A single example of a radar line and its interpreta-
ding and macro form accretion directions may be
tion are illustrated in Figs. 9.53 and 9.54. This survey
used to reconstruct channel and bar configurations
was carried out by Stephens (1994) over the top of a
(Fig. 4.1). Elsewhere, the concept of the hierarchy of
large outcrop of the Kayenta Formation, a Lower
depositional units is introduced, and it is shown how
Jurassic ephemeral fluvial system in southwest Colo-
the directional properties vary at each level of the
rado. Most of the reflections are interpreted to repre-
hierarchy (Fig. 2.10). Reconstruction of macro forms
sent bounding surfaces, probably of third-order
using paleocurrent and facies data is discussed in
rank, that record the lateral to oblique accretion of
Chap. 6 (e.g., see Fig. 6.24c).
DA units in a broad sandy-braided river channel.
Paleocurrent analysis may be used as a supple-
mentary mapping tool to investigate the following
kinds of information:
9.5.6 Magnetostratigraphy
1. Changes in channel and bar orientation and di-
The record of magnetic reversals preserved in sedi- rectional variability through a stratigraphic unit,
mentary rocks has, in recent years, become a power- as one of several indicators of vertical or lateral
ful tool for intrabasin and global correlation. changes in fluvial style
Carefully sampled stratigraphic sections can lead to 2. Reconstruction of tributary or distributary pat-
the erection of a local chronostratigraphy (examples terns, e.g., determination of radial alluvial-fan
Fig. 9.53A,B. A radar line (A) and its inter-

Position (m)
pretation (B), run across an outcrop of
! ~ the Kayenta Formation (Lower Jurassic),
Dolores River, southwest Colorado.
o 5070 (Stephens 1994)
~
1m;,
50!:0
~0
!5030
50221
!5010
5003
"ggQJ
g
"QS0 c:
0
~<1l
W
"goo
"0e0
"~0
4Q30
4~
4;10
4912J12l
11!lZl0 4892l
drainage and location of entry point of drainage of the elements, together with readings of the dip and
into a basin strike of accretion surfaces. Element 2 contains an
3. Reconstruction of local and regional paleoflow internal bounding surface, Ei, which separates incre-
patterns ments showing opposite accretion directions (in the
4. Vertical changes in flow direction through a left-hand half of the outcrop). It is therefore inter-
stratigraphic section as indicators of interacting preted as a fourth-order surface. Accretion surfaces
fluvial systems (e.g., trunk and tributary drain- at the left-hand end of the outcrop in element 2A dip
ages), or vertical changes in orientation of a sys- NNW and NNE (Fig. 9.58A), while the mean cross-
tem, in response to paleogeographic changes in a bed azimuth from the same beds is to the NNE
basin (011°). These data indicate that element 2A is a DA
unit, accreting obliquely downstream at about 70° to
An example of the use of detailed paleocurrent the outcrop mean. This suggests that the flow within
information to reconstruct channel and bar orienta- and adjacent to the bar was oriented at a relatively
tion is illustrated in Figs. 9.58 and 9.59, taken from high angle to the mean valley orientation, as indi-
the author's study of the Castlegate Sandstone in cated in Fig. 9.58C. Another example of this type of
Utah (Miall1993, 1994). At a location named Tusher analysis is given in Fig. 6.32.
Canyon-B, the outcrop is subdivisible into a series of A repetition of this type of analysis through a
elements separated by fifth-order bounding sur- vertical section can provide insights into how a river
faces. The exposure extends around a bend above a system evolved through time. Figure 9.59 illustrates
dry creek bed. Figure 9.58A and B illustrate the three successive bar deposits superimposed on each
paleocurrent data for the entire outcrop and for one other at the Tusher Canyon locality. There is no way
Fig.9.53B
Una 1 Una 2 Una 3 Une4
'~::
SURFACE
CLIFF --- - - -
Q ~-
~ - ---..>--_._'---- ------ "
~~~~ -~
FACE
.::===-,_._--_.-..-----_. ._------_._-_.-_. .-.. . -

-_..._ ------
I
V.E.:1.4
• Surface
____ } •These lines (from diagram above) represent s urfaoes correlated

with those in other radar sections
. Represents surfaces th at could not be correlated
PROFILE 2
"=iiA:i,.sWr".4f=
Pos~ion 01
-<
3·0 block diagram
Point bar or mld-d\annel bar
10mL
, 10m
DA • Downstream accreting macrofonn

237
o~
Orientation cliff-faoe
of
Fig. 9.54. Block-diagram reconstruction of a fluvial system, based on the correlation of seven radar lines run through the
Kayenta Formation, SW Colorado. Line 5 (the front of the block diagram) is illustrated in Fig. 9.53B. (Stephens 1994)
3 5 7 8 10 12 16 17 18 19
CH HL OK MKM GK I!W RK TWO KAK
DR 5
-- --
KUL ON 5
o
m.
----
OR4
10
20
50
ON 4
DR 3
iii Ii i iii i i
o 2 4 6 8 10 kn>.
Fig. 9.55. Magnetic sampling of a fluvial succession, the This is the field data base from which the fluvial recon-
Miocene-Pliocene Siwalik Group of Pakistan. Black sample struction shown in Fig. 9.1 was derived. (Behrensmeyer
points, normal polarity; open points, reversed polarity. and Tauxe 1982)
to determine whether the three bars followed each bounding surfaces (as noted above) is insurmount-
other closely in time, or whether they were separated able. Another problem is that flood-stage deposits
by long or short intervals of time in which other bars may be preferentially preserved, and these are likely
were formed and destroyed. However, this type of to preserve lower directional variance than other
analysis does provide some insights into the vari- deposits in the system because of the tendency
ability of river behavior at a given site and may for sinuosity to decrease at high discharge (Bridge
provide information on subtle vertical changes in 1993).
fluvial style brought about by allogenic mechanisms, The use of paleocurrent data to reconstruct the
such as the response of the river to a base-level details of a fluvial depositional system is a well-
change or a regional tectonic tilt. These processes are established practice (e.g., see Miall 1990, Sect. 5.9).
discussed at length in Chap. II. Integrated basin-analysis methods couple paleocur-
In an earlier analysis of vertical changes in rent data with data on facies and grain-size trends,
paleocurrent patterns it was suggested that the verti- and perhaps information on detrital sediment com-
cal changes in mean paleocurrent orientation could position and possible sediment sources, the combi-
be used to calculate river sinuosity (Miall 1976). It nation providing a powerful mapping technique. A
was proposed that the angular change in mean orien- few selected examples of the application of this inte-
tation, as measured by a moving average technique, grated methodology to the study of alluvial basins
could be converted to sinuosity using an empirical include Friend and Moody-Stuart (1972), Friend et
equation adapted from the work of Langbein and al. (1976), and Miall (1979a,b). Lawton (1985,
Leopold (1966). A slightly different approach was 1986a,b) used paleocurrent and petrographic data to
taken by Le Roux (1992, 1994). These techniques explore the history of tilting of the regional
now seems simplistic, because paleocurrent direc- paleoslope and basement uplift in the Rocky Moun-
tions record the orientation of bedform and bar tain Foreland Basin of Utah during the Late Creta-
deposits, not the channel itself, and may not corre- ceous and Tertiary (Sect. 11.3.3). Miall and Gibling
spond closely to channel orientation at anyone loca- (1978), Nichols (1987), and Hirst (1991) recon-
tion. Also, the problem of section missing at major structed radial paleocurrent patterns along a basin
VGP LATITUDE
-fd -YJ' +y! +6l:t
[)t{)K PATHAN FM.
1
z
Q
No t;
w
NAGRI FM. V>
en
«
::.:
a::
j
~
a:l
~
VGP LA TIT UOE

-00· .3d' +?If. 00·
N,
EOCENE LMS.
Fig. 9.56. Magnetic sampling data and indicated pole po- samples. Open circles are "class-B" data, in which one of
sitions for two stratigraphic sections through the Miocene- the three samples shows statistical divergence from the
Pliocene Siwalik Group, Pakistan. Closed circles indicate other two. A series of magnetic chrons (N) is identified at
"class-A" data, consisting of results supported by three the right. (Johnson et al. 1985)
margin and deduced from this pattern the presence of 63° between the lower and upper part of the sec-
of radiating distributaries in sandy alluvial-fan dis- tion. Note also the upward decrease in directional
persal systems. variance and the increase in set thickness through
Vertical changes in paleocurrent directions this profile. These data are interpreted as the result
through a stratigraphic section, coupled with of the rejuvenation of a tributary stream flowing into
changes in detrital composition, may be a simple the basin from the southeast following uplift along
means to reveal tectonic movements in a basin (e.g., the basin-margin fault. The northwestward-flowing
Miall1979a). In some cases, however, the changes in stream may have been deeper and more vigorous,
bedform character and orientation may be very accounting for the larger bedforms and the lower
subtle. Figure 9.60 is a plot of bedform and paleochannel sinuosity. Information on bounding sur-
current data obtained from a vertical profile through faces and architectural elements is not available for
a sandy braided stream deposit near a faulted basin this locality.
margin. Note the change in mean current direction
I OOSERI/ED cality for studying fluvial systems. Very few pub-

I POLARITY
lished examples of the successful application of
I these tools are available, which either may be a re-
?tf< FliHAN
\
\
\
\
flection of the industry confidentiality surround-
\
ing an invaluable technique, or may (more likely,
+
\
\
N' \
\\ in this author's experience) indicate that successful
I
\ \ applications are sparse. Descriptions of the tech-
\ 8 \ POLARITY nique and processing routines are given by
\ \ TIME SCALE
\
, \
\
Schlumberger (1970) and Vincent et al. (1977, 1979)
NAGRI FM. \ \ and have been summarized by Miall (1990, Sects.
j
7.88m.,.
VOLCANIC CH~\ 5.4.3, 5.9.6).
TUFF
6l:l.At:J.6 9 " 8.56
19.5~.5ooy1 Figure 9.61 illustrates the principal of the map-
N, ping of drape over bars or within channels. The only
10.00
10 example known to this writer of an application of
this technique to a practical case study is that of
t
11.20
II
------- 11.79
Muwais and Smith (1990). They found that in the
CHINJI FM. large, tidally influenced fluvial channel fills of the
12.1~
Athabasca Oil Sands of Alberta, the main surfaces
1 14
15
13.46 recorded by the dip meter were those defined by the
fill of vertically accreting channels and the lateral-
accretion surfaces oflarge point bars. Cross-bedding
1
15.12
is rare in these deposits, except near the base of the
CHRQN channels, and was not normally picked up by the
KAMLIALlt.lIRREE 16
FM. dipmeter. Figure 9.62 illustrates the interpretive
j 17.56
principles, and Fig. 9.63 is an example of a dipmeter
log interpreted in terms of three vertically aggraded
channels. The interpretation in this case is con-
firmed by outcrop studies of large surface mines
Fig.9.57. Correlation of the magnetic chrons of Fig. 9.56 nearby.
with a standard magnetic time scale. (Johnson et al. 1985)
Potentially, third- and fourth-order surfaces,
such as those mapped in Fig. 9.40, should be recog-
9.S.8 The Dipmeter nizable in cores and dip meter logs (especially using
the Schlumberger Formation MicroScanner) by
Logging companies, such as Schlumberger, promote their gentle depositional dip and their association
the use of the dip meter and related tools, such as the with shale drapes or lag deposits, but in practice their
Formation MicroScanner and the Fullbore Forma- identification is very difficult (author's consulting
tion MicroImager, as tools for the detection and experience). However, there is considerable poten-
mapping of sedimentary dips in the subsurface tial here for the detailed study of sand-body anatomy
(Serra 1989). There are three main potential applica- in field development situations, where abundant
tions in the study of fluvial sandstones: core and log data should be obtainable. Figure 9.64
illustrates a typical point bar and the type of dip-
1. The mapping of dipping fourth- and fifth-order
meter pattern that might be expected to result from
surfaces corresponding to bar-top and channel-
a well drilled through it. Figure 9.65 illustrates an
floor surfaces, and the drape associated with
example from the Athabasca Oil Sands, where the
them, providing information on the shape and
accretionary dips are not confused by the presence
orientation of these features
of dipping cross-bed surfaces.
2. The mapping of internal, second- and third-order
The potential for mapping cross-bedding (Figs.
erosion surfaces, that would facilitate the map-
9.66, 9.67) would seem to be high and, indeed, this
ping of macro forms, such as point bars
technique has been successfully applied to the study
3. The mapping of cross-bed orientations for the
of large-scale eolian cross-bedding, in which the
paleocurrent information they yield
structure of the dunes is simple and their orientation
These three applications are arranged in order of relatively consistent (see Miall 1990, p. 327 for dis-
decreasing scale and utility, in terms of their practi- cussion). However, in fluvial deposits there are many
®-,~~
:,,: ':',:,-,-" ,:,:,-,
E ~ - ~ E
E/C~40" 070~C
D ~_1.QP m ,
Thsher Canyon - B __ lora. X!:J ~_,.-::_:", ",regional
element 2 ~'fI,. - "_' , _3·0 bedform r>,\i" ' paleocurrent
V;;.. .:·~~.. .
_ OUo _ ',, ' -:' mean
°
~
n -7 , " -'
090,....,.
Iwu>d " , ,"-
of """"'P
~-- ~03 ~
(;\A"- ~;
~ 340°
,
5~ 070° © fIP\
~ lUsher ~anyon - B
all readlOgS
076° n=56
trace of
1 - 3 order accretion surfaces
on top of element 2A-DA
Fig. 9.58. Example of the interpretation of an architec- outcrop is also shown. Note the slight divergence between
tural element using orientation data from cross-bedding the outcrop mean and the regional mean, which may re-
and accretion surfaces. A, Orientation ofTusher Canyon-B flect the dominance of channels and macroforms oriented
section, Castlegate Sandstone, Utah, showing measured oblique to the main channel trend at the Tusher Canyon
accretionary dip directions and interpreted accretion outcrop. Note also the large divergence between the ele-
surfaces on the top of element 2A-DA, and current rose ment-2 paleocurrent mean and the outcrop mean, reflect-
diagram obtained from measurement of cross-bed orien- ing the locally high divergence of accretion directions from
tations in element 2. B, Perspective view of element 2A-DA. the mean channel trend. Accretion in element 2A is in-
C, Interpretation of braided river that deposited the terpreted as DA rather than as LA because of the within-
Tusher Canyon section. Note the position of the rusher element similarity between indicated paleoflow and
Canyon-B section. The river is oriented according to the accretionary dip directions, (Miall 1993)
regional paleocurrent mean, The rose diagram for this
within the range of modern imaging tools). In addi-

tion, there are many types of surface in cross-bedded
units that can produce confusion. For example,
G trough cross-beds have curved dips, and many
cross-bed sets are deformed by slumping or over-
turning, or contain reactivation surfaces. Cameron
et al. (1993) provided a useful discussion of the prob-
E"
lems and the filtering techniques that can be used to
sharpen the results.
Williams and Soek (1993) tackled the problem of
dip complexity by progressively filtering out low
dips and testing the variability of the indicated ori-
entations at each stage. Their data were collected
t:<' flow in main __ direction of from outcrop measurements in order to simulate
channel local flow subsurface dipmeter records, the purpose being to
facilitate interpretation by selecting a stratigraphic
Fig. 9.59. Superimposition (stacking) of macroforms in a
multistory sand body, showing three elements at Tusher unit already well known from surface studies. One of
Canyon. Each panel shows the two adjacent outcrops, their stratigraphic sections with dip records is shown
rusher Canyon-A and -B, straightened out to facilitate in Fig. 9.68, and Fig. 9.69 shows the results of filter-
viewing. Each element is shown in the process of forma- ing. Note that as the range of dip magnitudes is
tion. The reconstruction of the low$St element, 2A -DA, is
progressively narrowed in favor of readings ap-
shown in Fig. 9.58
proaching 35° (typically the highest angle of repose
for loose sand), the percentage of readings falling
complications. The scale of the cross-bedding, at a within the known channel orientation (north to
few tens of centimeters or less, approaches the limit northeast) increases. A similar result was obtained
of resolution of the dip meter tool (although it is well by Cameron et al. (1993) in their outcrop simulation
CAPE VESEY
HAMILTON
30 \
w.m
~
Li gni re
Pebble beds
25
20
t
0 - 339
n: 26
~"
L=;96.0
S2 ~342
o
l
(
/
}t
Other sands
~
10
.-J
5
~
1=
..
L - 88_6
~
(9')
£
M
'!l 8
It')
~
.. ,
0
"
....0 ~ 0 S2=870
N
Azimuth Movin,i averaee Movina avera8e MDving average Foreset dip
uimulh vtll'ianc~ set th ickness (cm)
Fig. 9.60. Paleocurrent data and bedform scale measured Canada. Moving averages are based on a moving set of ten
over a vertical profile through a sandy braided stream readings. (Mia111976)
deposit, the Isachsen Formation, Banks Island, Arctic
® ®
O· 10· 20· 30·
Fig. 9.61. Dipmeter patterns associated with bars and of the tails, with north being toward the top.
channels. Dip magnitudes are indicated by the position of (Schlumberger dipmeter manual 1986; see also Mia1l1990
the head of the "tadpoles", and azimuth by the orientation for general interpretive principles)
i - - -- - - - -----ABANDONED MEAtilER CHAPKL DEPOSIT -----------+.DEP~O~SIT~~IPOIfT~~BARI~~t:;------j

I =~
lei Ib) 10)
Fig. 9.62. Schematic illustration showing how vertical in dip (profile a) may indicate lateral accretion, whereas
profiles from dipmeter logs may be interpreted in terms of upward decrease in dip (profile c) may indicate vertical
the large-scale features of a channel fill. Upward increase aggradation. (Muwais and Smith 1990)
DEPTH S DIP ANGLE ANO DIRECTION DEPTH S DIP ANGLE AN D DIRECTION

(M)
t
N
(M I
f
N
i()
I
20 030 40 50 0 10
I
20 30 40 50
0 5
5
:. 0
0 5
~~
5- 20
1-
20 25
I-
~
2 3
5~1i4
~~
Fig. 9.63. Dipmeter log interpreted as the record of three Fig. 9.65. Dipmeter log interpreted as the record of three
stacked, vertically accreted channels, in the intervals stacked lateral-accretion deposits, in the intervals 35-25,
26-14,14-8, and 8-4 m. (Muwais and Smith 1990) 25-17, and 17-8 m. (Muwais and Smith 1990)
a\ "" a'
~wpattem
sUrface]
flow pattern
bottom
...I.- dip of lateral-accretion

surfaces
orientation of smaller
scale cross-bedding
hypothetical corehole A
CoreA
dip of lateral-accretion
...
surfaces
orientation of smaller
scale cross-bedding
Fig. 9.64. Typical point bar and the dip-

ping surfaces associated with it. Note the
distinctive low dip and orientation of the
L.£..~'----!.-'-.L ____________ •
third-order surfaces (epsilon cross-bed-
approximate
ding), A speculative dipmeter log is also
scale 1. planar cross-bedding 2. trough cross-bed edge 3. trough axis 4. climbing ripples shown
Fig. 9.66. Detection of cross-bedding us-

ing the dipmeter and its dependency on
.... the sensitivity and sample interval of the
tool. (Schlumberger dipmeter manual
1986)
....
....
.t
D· 15· 3D·
O· 15· 3D·
Short Interval
Long Interval (attitude of unit boundaries +
(attitude of unit boundaries) cross-beds)
Deep perennial Synthetic Geodip Log Shallow perennial Synthetic Geodip Log
braided river o 10 20304050 70 90 braided river 0 10 20304050 70 90
I. low angle cross-bedding 2. trough cross-bed edge 3. trough axis 4. planar cross-bedding
5. reactivation surface 6. set bounding surface
Fig. 9.67. Two types of braided river profile (see Chap. 8 for a discussion of these) and the dip meter patterns that might
be expected from them
of a dip meter survey run through deltaic distribu- of the cross-bed sets within the overall hierarchy of
tary channels. The conclusion of both studies is that directional indicators (Fig. 2.10), they were able to
actual subsurface dipmeter studies are likely to pro- make predictions about channel trend from these
vide useful paleocurrent data, even when obtained readings, predictions that revealed the presence of a
from single exploration holes. previously unsuspected meander bend. Several new
Dueck and Paauwe (1994) provided an excellent producing oil wells were drilled using these data.
case study of the gradual development of small-scale
- but nonetheless productive - exploration pros-
pects in channel sandstones based mainly on the 9.5.9 Surveillance Geology
careful use of paleocurrent data obtained from bore-
hole imaging techniques. They were able to plot The term surveillance geology is used to describe the
cross-bed foreset surfaces as little as 5 cm apart work done by production geologists to monitor the
within a borehole succession, allowing for the de- pressure and fluid composition of producing wells.
monstration of variable dip angle from base to top of Pressure data can be used to determine the connect-
individual sets. By carefully considering the "rank" edness of specific sandstone bodies, and fluid com-
GfWNSIZE
pre-etrudural rotaIion poat·IlNCIu,aI roIaIton Fig. 9.68. Example of sedimentary log and
20406080 020406080 dip profIle through stacked channel sands,
Carboniferous, western Ireland. Note that
16
,
~ rotation of readings was carried out to re-
, move regional structural tilt. (Williams and
15 ,' Soek 1993)
\
~j
14 , 4
13 ,
/t/
,~
12 ,
•
",
11 ....L
--"- I
-4-
--"- tI"' Channel palaeoflow
,
10
--"- and sandbody
--"- elongation dlractlon
,,
,
I~
....L
9
en
l'! \
j
8
\ III wHhln
Dip magnHude
range 17 ·35
[i
•
7 :
t
6
~
,
,
,
~
•
~
•
I'
position, especially water saturation, can be used to graphic mapping of the reservoir and therefore pro-
track the movement of the oil front through a reser- vides data which can be used to refine the deposi-
voir body as it is swept by injected water or steam. tional model. Figure 9.70 illustrates a pressure-depth
Thakur (1991) and Martin (1993) provided reviews plot constructed for the Mannville sandstones of
of this specialized subject. Lorenz et al. (1991) de- part of Alberta.
scribed a useful case study. The pattern of the flow of oil and water depends
Pressure-depth plots can be used to test reservoir on the porosity-permeability architecture of the res-
body connectedness (e.g., Lorenz et al. 1991). Al- ervoir, and careful attention to the details of the fluid
though this information is of importance in its own movement through surveillance methods can lead to
right, as an aid to proper planning of a production continual refinements in the architectural model of
program, the information is of use in the litho strati- the reservoir. Figure 9.7l illustrates patterns ofwa-
Fig_ 9.69. Filtering of dip readings from

Killard distributary channel
the Carboniferous of western Ireland to
- Stack 1 progressively narrow the dip range in

----0-
Stack 2 favor of steep dips expected to represent
cross-bedding. Each curve shows data
-
--0--- Stack 3
---¢-- All stacks from a different stack of fluvial channels.
Stack 2+3 Note that, with one exception, the per-
centage of readings within the expected
range (channel orientation measured
from outcrop) increases, as dips are li-
mited to high readings. (Williams and
Soek 1993)
0-90 5-35 9-35 13-35 17-35 21 -35

Dip magnHude (in degrees)
GAS GIIADIENT
_ • "0 t o.D. 1
I
.. .."
....
••
~
~
~
..
~ .110
o
~ . 1o,
:. •
...• • '0
~
.......
z
Q • "
o
.• uoo o+---~----~--~----~----~
'''0 JS OO )150 . 1 00 . "0
__ ~
•• 0 0
PAESSUII[ ( IIPA I Fig.9.71A-F. Schematic representation of waterflood
patterns in various sand-shale configurations. Permeabil-
Fig_ 9.70. Use of pressure-depth plot to test lithostrati- ity log (K) is given for each section, showing increase in
graphic correlation. The points fall into two groups, indi- permeability to the right. An injection well is shown by the
cating that they represent two channel-fIll sandstone circle with the arrow, and a production well is shown by the
bodies isolated from each other by fine-grained units. black circle. Water is indicated by the dotted areas, with
Mannville Sandstone, Alberta. (Putnam and Oliver 1980) arrows indicating direction of flow. Irregular shales are
shown by heavy black lines. A Slumping of injected water
front as a result of upward decrease in permeability. B
Gravity drainage of oil around shale beds (open arrows). C
terflood that can be predicted from various configu- High sweep efficiency reflecting presence of shales to
rations of an interbedded sandstone-shale unit. The guide the flow. D Viscous fingering of waterfront in homo-
presence of high-permeability zones in a reservoir geneous strata. E,F Viscous fingering, plus channeling
along a high-permeability zone. (Hopkins et al. 1991, re-
may lead to the development of "fingers" or "chan- printed by permission)
nels" in the flood front (Fig. 9.7ID,E,F) and to early
breakthrough of water to a production well. "Slump-
ing" of the water front may also occur where there is stone of Alberta. The reservoir is an estuarine sand
an upward decrease in permeability (Fig. 9.7IA,B,C). body within an incised valley-fill succession. Figure
This can lead to isolation of pockets of oil, which may 9.72A shows a longitudinal cross section through the
be partly recovered by gravity drainage down to the field, and Fig. 9.72B is an interpretation of flow pat-
main flood pathway (Fig. 9.71B). Hopkins et al. terns through the field. Slumping and channeling are
(1991) used monthly fluid production data to test the interpreted to be occurring in this field, as indicated
applicability of these models to a water-injection by water-production patterns in the northwest end
production situation in the Upper Mannville Sand- of the field.
U POOL
NW
...
KEY WEll SE
01-24
•
11 -24 102110-24 07-24
•
14 - 24
•
04-25 13- 24
o
• • fZJ •
50
DETRITAL
100
A
...1--_ __ _ _ __ _ _ _ 1.8 km - - - - - - -- - ---1.... MISSISSIPPIAN
SE
11 - 24 102/10-24 07-24 01-24

13-24
•
14-24
• P • • •
15
00 10
iii
I-
W
~ 5
o ~;:;:J
... . > 24% porosity 20- 24% perosity 111 20%
$~ ..
< porosity
..~.-_ _- - - - - -_ - - -__ 1.8km __---_-----_--~

~
B
Fig. 9.72A,B. U Pool, The Little Bow field, Alberta. A stone; small dots, sandstone-shale; horizontal dashed
Stratigraphic cross section. Porosity values higher than lines, shale. B Porosity distribution and inferred fluid flow
21 % are shaded black on porosity logs. Large dots, sand- pattern. (Hopkins et al. 1991, reprinted by permission)
Martin (I 993) described various engineering using bright-spot techniques to illustrate the pres-
techniques that are used to evaluate production ence of gas, then the movement of the fluid can be
in braided fluvial reservoirs. Pressure tests can be charted.
used to assess the size and orientation of permeable Karlsen and Larter (1989) demonstrated that the
flow pathways. Pulse tests are used to determine petroleum in a given field is not homogeneous, but
connectedness between wells. One well is produced, varies in terms of gas/oil ratio and oil type. These
and the degree of pressure drop at an adjacent well variations occur because the diffusion of petroleum
is monitored. Lorenz et al. (1991) described com- into and through a reservoir is a slow process, re-
parable procedures as part of their detailed case quiring millions of years for complete homogeniza-
study of the Mesa Verde Group of Colorado. They tion. Detailed study of the petroleum itself may
also used injected tracers to test reservoir heteroge- therefore reveal reservoir heterogeneities that are
neity. impeding the homogenization process.
Robertson (1991) described the use of3-D seismic Aquifer studies and problems of toxic waste dis-
reflection data to monitor fluid flow through a reser- persal through alluvial deposits may also constitute
voir. Using identical data acquisition and processing a form of surveillance geology, in which studies of
routines, repeated surveys are run over intervals of fluid movement throw light on facies architecture
several months. If fluid composition can be estab- models and thereby provide data for improving the
lished by appropriate processing, for example, by flow models. An example is given by Blair et al.
(1991), who examined the flow of pentachlorphenol of stratigraphic units and a new method of defining
in groundwaters around a chemical plant. and mapping them. This is the method of
Information on reservoir architecture and fluid allostratigraphy. "An allostratigraphic unit is a map-
movement patterns is used in production simulation pable stratiform body of sedimentary rock that is
models, as discussed in Chaps. 14 and 15. Field defined and identified on the basis of its bounding
checking of these models against measured data on discontinuities." (NACSN 1983, p. 865) A hierarchy
pressures and fluid contents is a process called his- of units, including the allogroup, alloformation, and
tory matching. Examples are described in Sneider et allomember, was proposed, and rules were estab-
al. (1991) and Martin (1993). lished for defining and naming these various types of
units. Allostratigraphic methods enable the erection
of a sequence framework that avoids the cumber-
some and inappropriate rules of lithostratigraphy,
9.6 Stratigraphic Nomenclature
whereby lateral changes in facies within a unit of
comparable age require a change in name. An ex-
The methods of lithostratigraphy have served sedi- ample is illustrated in Fig. 9.73, in which it can be
mentologists and basin analysts well in the establish- seen that by using an allostratigraphic approach to
ment of the broad stratigraphic and paleogeographic the subdivision of a fluvial-lacustrine assemblage
subdivisions of a basin fill. These are standard meth- the natural subdivision of the succession into four
ods and are described in several textbooks (e.g., sequences, bounded by breaks in sedimentation, can
Miall 1990, Sect. 3.4). However, the advent of se- readily be formalized in a stratigraphic nomencla-
quence stratigraphy and a renewed emphasis on the ture.
importance of breaks in sedimentation have created It will be apparent that the ideas behind the
the opportunity for a much more refined approach bounding-surface hierarchy used throughout this
to stratigraphic documentation. The current North book may readily be adapted to the methods of
American Code of Stratigraphic Nomenclature allostratigraphy. In fact, the three systems of strati-
(North American Commission on Stratigraphic No- graphic subdivision, allostratigraphy, the bounding-
menclature 1983) includes proposals for a new class surface hierarchy, and sequence stratigraphy, are in
EXPLANATION
4 Allollr lIiQr aphic un~
~BuriedlOiI
- - - DilCOnformitY
Fig. 9.73. Example of the allostratigraphic classification Using lithostratigraphic methods, each gravel, sand, and
of a fluvial-lacustrine assemblage in a graben. Numbers clay unit would typically be given a separate name.
1-4 correspond to alloformations which are defined by (NACSN 1983, reprinted by permission)
breaks in sedimentation and cut across facies boundaries.
Stratigraphic Nomenclature 307
reality closely related, differing mainly in emphasis. group 9 depositional systems of Tables 3.2 and 9.1
Allostratigraphy has been proposed as a general and are bounded by sixth- or seventh-order surfaces
system of descriptive, nongenetic, regional strati- in Miall's (1988a) scheme. The subdivisions of each
graphic subdivision based on the recognition of of the alloformations into allomembers, indicated by
breaks in sedimentation. The bounding-surface hi- the clay, sand, and gravel areas in Fig. 9.73, represent
erarchy proposed by Miall (1988a,b) and used the group 8 subdivisions of the depositional system
throughout this book focused initially on smaller- into alluvial fans, channel belts, floodplains, etc.,
scale features, such as the individual components of bounded by fifth-order surfaces. The scheme of
depositional systems but has now been expanded to DeCelles et al. (1991), illustrated in Fig. 4.3, should
include larger-scale features. For example, the num- be compared with this. There are some differences,
bered alloformations of Fig. 9.73 correspond to but the same principles of hierarchical subdivision
30· 30' 00'"
',.' "
:j .:.:.:.
- j ::: : : ;':-':
30° 32' 30'
30· 30'
, ' :. :. , .. ' :.' . ,' ' ..'

.. ' ':: .. ,' .. ,"...... , .
. ... . --: ...... .. ",'
...... -.
,,': .. '
-. -: : ...
.. , ...... . Fig. 9.74. Distribution of alloformations
in the Recent sediments of the Amite River
valley, Louisiana. (Autin 1992)
91 ° 00' 00" 91° 57' 30"
A Fig. 9.75. Conceptual block diagram of the relationships

between the various alloformations shown in map view in
Fig. 9.74. The oldest unit is the Prairie Complex (PR).
Successively younger alloformations, WAT, DS, and
MAG, are incised into each other, with steep erosional
lateral contacts and diffuse horizontal margins on the
floodplain, where fine-grained units are in contact across
boundaries that have been disturbed by pedogenesis. A,
Modern Amite River; B, point bar topography of DS
alloformation; C, DS ridge and swales topography; D,
WAT abandoned channels; E, WAT floodplain flats.
(Autin 1992)
c::::J SANDY fACIES
~ PEORIA LOESS
~ PlAIRIE COMrLEX
---, ~~~~~~MATION
BOUNDARY
apply. Groups 1-7, bounded by surfaces of first- to surfaces for more than 60 km and suggested that they
fifth-order rank, represent the smaller-scale strati- are probably, in fact, basin-wide in extent. They
graphic features, not shown in Fig. 9.73. Sequence subdivided the formation into six sandstone sheets.
stratigraphy focuses on the larger-scale features of Lopez-Gomez and Arche (1993) suggested that the
this subdivision and interprets them in a genetic seven "main bounding surfaces" are regional ero-
context, as discussed in Chap. 11. sion surfaces that developed as a result of reorgani-
Several groups of workers are now explicitly em- zation of fluvial depositional systems following
ploying an allostratigraphic methodology. Walker pulses of basin-margin faulting and gentle regional
and James (1992) recommended the general use of tectonic tilting. The sandstone sheets could there-
allostratigraphic methods in stratigraphic and facies fore be defined as alloformations.
studies. A few examples of the application of the No rules for relating the various systems of no-
technique to fluvial and deltaic deposits are now menclature to each other have yet evolved. However,
available. For example, Bhattacharya (1991) subdi- the examples described in this section indicate that
vided the deltaic Dunvegan Formation ofwest-cen- alloformations are defined by surfaces that would be
tral Alberta into a series of allomembers using classified as sixth-order or higher in Miall's (1988a)
mappable flooding surfaces (Fig. 9.17). Autin (1992) scheme, and that allomembers are bounded by sur-
subdivided the postglacial terraces and associated faces of fifth- or sixth order. The examples quoted
sediments in a Holocene fluvial floodplain succes- here reveal a range of time scales for alloformations
sion in Louisiana into alloformations (Figs. 9.74, and allomembers, reflecting the wide variation in the
9.75). Each represents a separate channel-belt com- rates of change imposed on alluvial systems by
plex that formed in response to changes in climate allogenic controls. For this reason, it is not possible
and sea level. They are incised into each other and to definitively assign allostratigraphic units to spe-
partly overlap, and are therefore comparable to suc- cific groups within the hierarchy of this book (Tables
cessive, superimposed incised valley fills of the type 3.2, 9.1). The six sandstone sheets comprising the
described in Chap. 13. Cafiizar Sandstone described by Lopez-Gomez and
In other cases, data may be available for allo- Arche (1993) span 10 million years, indicating an
stratigraphic subdivision, but the terminology has average duration for each of 1.7 million years (group
not yet been employed. For example, the detailed 9). It is suggested here that they correspond to
subdivision of the Old Red Sandstone using tephras alloformations. The schematic alloformations of
by Allen and Williams (1982) led to the erection of a the NACSN (1983) proposal (Fig. 9.73) are equivalent
detailed stratigraphy based on regional channel-belt to group 9 deposits. The eight units of Allen and
erosion surfaces (Figs. 9.25, 9.26). The eight units Williams (1982), termed allomembers above, span
defined by Allen and Williams (1982) are in effect an average of8000 years (group 7). The Recent flood-
allomembers. Lopez-Gomez and Arche (1993) map- plain alloformations of Autin (1992) have time spans
ped seven sixth-order surfaces in a 170-m-thick Tri- of a few thousand years. However, it is questionable
assic sandy-braided system in Spain. They traced the whether Autin's (1992) units would be assigned
Stratigraphic Nomenclature 309
alloformation rank if they comprised part of an an- abled the complex oflaterally interfingering units to
cient nonmarine succession. A comparison with the be related to each other and provided the basis for a
eight units of Allen and Williams (1982) is suggested simpler system of named units, but uncertainties
here, with a change in rank to allomembers. The remained regarding the definition of boundaries,
entire postglacial valley-fill complex would then because in places the units are conformable, and no
rank as an alloformation. discernible lithologic changes could be discovered to
A proposal for a correlation between allostrati- enable placement of the boundaries to be completed.
graphic units, the bounding-surface hierarchy, and To then compare this allostratigraphic framework
sequence concepts is presented in Chap. 13. with one constructed using sequence concepts is
A recent study of a stratigraphic succession in perhaps inappropriate, because allostratigraphy is
Wyoming was carried out for the specific purpose of intended to be purely descriptive, whereas sequence
comparing the use of lithostratigraphic, allostrati- stratigraphy is genetic and can lead to very specific
graphic, and sequence nomenclature for the sub- predictions regarding stratigraphic composition
division and interpretation of an interfingering and geometry (Chap. 13). Sequence concepts can be
marine-nonmarine clastic succcession. The study used to construct a complete stratigraphic frame-
was by Martinsen et al. (1993), who discussed the work as long as some of the basic elements of the
stratigraphy of the Hanna Basin, an Upper Creta- sequence model can be observed, although in this
ceous foreland basin. A system of lithostratigraphic case difficulties arose because the rapid subsidence
nomenclature was available from earlier work and and the particular tectonic style of the Hanna Basin
provided a complete framework of named units for made comparisons with standard sequence models
descriptive purposes. Allostratigraphic concepts en- difficult.
Chapter 10
Fluvial Depositional Systems and Autogenic Sedimentary Controls
10.1 Introduction be included under this heading is discussed in Sect.

8.3. This section is concerned primarily with what
Blair and McPherson (1994) term braid deltas.
Beerbower (1964) was the first to clearly distinguish
There is a very large literature on alluvial fans and
between within-basin and extrabasinal sedimentary
other distributary systems, dealing with such data as
controls. The first, which he termed auto cyclicity,
depositional processes, downstream changes in
refers to the redistribution of sediment within a
slope, sediment grain size, sorting, facies, facies as-
depositional system as a result of processes inherent
sociations, and clast shape (Denny 1967; Bull 1972,
to the system, such as channel migration and bar
1977; Heward 1978a; Nilsen 1985; Rachocki and
development. As discussed in Sect. 2.3.6.2, during
Church 1990; Fraser and DeCelles 1992; Blair and
the 1950s and 1960s it was realized that fining-up-
McPherson 1994). A significant body of field data is
ward cycles in fluvial systems could be explained by
presented in these publications, together with much
such processes. The term autogenic processes is now
useful theoretical discussion. A summary of this
preferred, because they are not necessarily strictly
would require a book of its own, and the discussion
cyclic in nature. Much of the sedimentological re-
below is limited to a presentation of the more impor-
search in the 1960s and 1970s was concerned with the
tant concepts which control large-scale fan architec-
recognition and description of autogenic processes,
ture and vertical profile. The main theoretical
as discussed in Chap. 2. The purpose of the present
framework for interpretation has been built from
chapter is to review the present state of our under-
studies of modern and ancient gravelly braided sys-
standing of the subject. Extrabasinal, or allogenic
tems, supplemented by experimental work on small
(formerly allocyclic), processes, are discussed in
laboratory models. However, it can be shown that
Chap. 11.
the same principles apply to all distributary systems,
The most important autogenic processes are
such as terminal "fans" and the giant rivers of the
those which lead to changes in the position of chan-
Himalayan foredeep.
nels. Channels may change in position and shape by
Basin-margin gravel rivers are highly susceptible
gradual lateral movement, or they may change posi-
to allogenic influences, especially tectonism. Much
tion entirely to another part of the floodplain by the
research has been published on the sequences and
process termed avulsion.
megasequences in the resulting deposits generated
by pulses of basin-margin faulting (e.g., Steel et al.
1977; Steel and Aasheim 1978; Heward 1978a; Blair
10.2 The Evolution and Bilodeau 1988). These processes act over a phy-
sical scale and time scale corresponding to sediment
of Distributary Fluvial Systems
groups 8 and above of this book (Table 9.1) and are
discussed in Chap. 11. However, autogenic processes
Distributary systems include alluvial fans and deltas. are of equal importance. They concern primarily
They are characterized by networks of channels radi- depositional processes and products of group 7 and
ating from a single trunk channel. There may be one are the subject of this section.
or more active channels, together with several or There are two main autogenic processes in fluvial
many dry channels that are occupied only during distributary systems, the avulsive shifting of dis-
floods or as a result of the autogenic processes de- tributary channels and a cycle of trenching and
scribed in this section. The current controversy re- backfilling. The experimental work of Schumm et al.
garding the definition of the term "alluvial fan" and (1987), and the observations of DeCelles et al. (1991)
the range of distributary fluvial systems that should and Fraser and DeCelles (1992) suggest that these
312 Fluvial Depositional Systems and Autogenic Sedimentary Controls
processes are intimately related. As shown in Fig. 4.3, indicated that a natural, autogenic cycle of entrench-
conglomeratic distributary systems cycle through ment occurs, and DeCelles et al. (1991) confirmed
two main depositional styles, a cycle of entrench- the existence of this cycle in their architectural
ment and a cycle of trench backfilling and dispersion mapping of a Paleocene gravel complex in Wyoming
of flow. Avulsive shifting of distributaries may be and Montana. Figure 10.1 is the schematic and
associated with the establishment of new deposi- generalized model of the entrenchment-backfilling-
tionallobes during the entrenchment phase, or with dispersion cycle developed from experimental work
the more minor shifting of distributaries that can on small laboratory models (from which Fig. 4.3 was
occur during the dispersion phase. Events upstream developed by DeCelles et al. 1991), and Fig. 10.2 is an
in the drainage basin may also affect the develop- architectural model of a distributary complex that
ment of the deposits, as the tributary network grows develops by this process, based on observations of
and sediment is stored and periodically flushed from ancient deposits.
the hinterland out into the basin (Fraser and The cycle begins with an interval of rapid aggra-
DeCelles 1992). In detail, over periods of 10L 1OS dation on the upper part of the river system as a
years, the resulting stratigraphy is controlled by the result of the lateral flow expansion and abruptly
action of various geomorphic thresholds, as de- decreased gradient that occurs at the mouth of the
scribed by Schumm (1977, 1979). feeder canyon. "Eventually the local gradient on the
Schumm et al. (1987; Table 9.1) listed numerous upper fan becomes steep enough that a trench forms
mechanisms that lead to fan-head entrenchment, in order to re-establish grade between the floor of the
including climate change and reduction in source feeder canyon and local base level ... Sediment then
area-basin relief as a result of normal erosion. How- begins to bypass the upper fan, through the trench,
ever, their experimental work on small, artifical fans and accumulate in the form of a depositional lobe on
the lower part of the fan, below the intersection point
... This part of the cycle is referred to as entrench-
ment" (DeCelles et al. 1991, p. 583; Fig. 1O.1A).
Trenching may also be initiated by a major flood
event (Denny 1967). A new drainage system is there-
by established, a new sediment lobe forms, and the
A B
trench begins to backfill with sediment (Fig. 10.lB).
The trench is eventually obliterated, and sediment is
then spread out in a fairly uniform manner across
the fan surface - the stage termed "dispersion" by
Schumm et al. (1987) (Fig. 10.lC). Oversteepening of
gradients on the upper part of the system replicates
the first stage, and a new cycle is initiated, with
c D
trenching occurring in a different part of the dis-
tributary complex showing a lower topographic pro-
file (Fig. lO.lD-F).
Other processes may also lead to diversion of
the main distributary. Denny (1967) and Schumm
(1979) described stream capture by secondary
streams occupying topograpically lower profiles,
E
such as streams developed on the basin floor that cut
Fig. 10.1. Generalized model of autogenic fan develop- back toward the apex of the distributary system by
ment, based on experimental work on small, artificial fans. headward erosion.
A Apex incision results from source-area erosion and leads
to deposition near the toe; B the channel backfills and the DeCelles et al. (1991) recognized a category of
locus of sedimentation (the intersection point of channel fifth-order litho somes and two types of fifth-order
slope with the fan surface) moves up to the midfan region; bounding surface that they interpreted as the pro-
C channel is obliterated by backfilling, and flow is dis- duct of the cycle of entrenchment and backfilling
persed from near the apex in all directions; D over- (Figs. 4.3, 10.2). Fifth-order surfaces, in this model,
steepening of the apex occurs again, and incision along a
steep flank of the fan leads to renewed entrenchment; E, F may be erosional or accretionary. Erosion surfaces
depositional backfilling commences in a new part of the are produced by the cutting and lateral migration of
fan. (Schumm et al. 1987) entrenched channels. These may include trench
The Evolution of Distributary Fluvial Systems 313
TRENCH INCISION Fig. 10.2. Architectural cross section
~ J,
FAN (6th-ORDER 0/
LlTHOS~ME)
through the upper part of a fan, trans-
verse to fan axis, showing the develop-
ment of a fifth-order lithosome (see
also Figs. 4.3 and 10.1). 1 Trenching; 2
lateral migration and filling of trench; 3
backfilling of trench; 4 dispersion of
,,S,'"os=F
flow; 5 abandonment of active part of
fan, with development of pedogenic
modifications and initiation of re-
newed trenching elsewhere; 6 cross sec-
tion through resulting fifth-order
3 lithosome, showing architectural fea-
tures and types of bounding surface.
(DeCelles et al. 1991)
WIDESPREAD BRAIDED-STREAM DEPOSITS~
4 ~ --.e=::=s =J
NEW
TRENCH INCISION _SILTSTONE DRAPE
5 l 5th-ORDER LlTHOSOME
< - - - - ; - ' - - - - - - 100. to 1,000. 01 metre. -------'-;-__>
~~fL~AT~'O~'~-""~~jj~i~~~~~~~
<:TSTON£
10, orJ,m.~tr:
1" / ~~~~~~
•• Sil OiltAPr--.., ~ ISPERSION
TRENCHING
Jro-OAoER SUlllFAces
cutbanks several meters deep, although such sur- traction currents tend to migrate laterally, whereas
faces may be difficult to detect within continuous on fans with significant sediment-gravity-flow depo-
conglomerate successions (see Fig. 6.17 for possible sition the trenches formed by traction-current inci-
examples). Accretionary surfaces are developed by sion are simply plugged by flow deposits during the
lobe construction during the backfilling stage. These backfilling stage (Schumm et al. 1987).
surfaces are hundreds to thousands of meters in Much attention has been paid to the nature of the
lateral extent in the conglomeratic deposits studied vertical profile through fan deposits. In a thorough
by DeCelles et al. (1991). discussion of this topic, Heward (1978a) stated:
Fifth-order litho somes formed in each cycle may
be separated from the next overlying lithosome by "Despite the apparent individual characteristics of mod-
ern fan flood events, vertical sequences of progressively
thin siltstone intervals showing evidence of pedo- changing grain size, bed thickness and depositional pro-
genesis. These likely form during the interval of en- cesses characterise many ancient alluvial fan deposits ...
trenchment that initiates a new cycle. Wind erosion Such sequences are generally interpreted to represent the
and the formation of desert varnish may also charac- gradual progradation, avulsion, abandonment, or lateral
terize these temporarily abandoned areas of the fan migration of depositional elements; rather than fraction-
ation of sediment within a depositional element, or pro-
surface (Denny 1967), a process particularly impor-
gressive changes in the sediment supplied" (Heward
tant on debris-flow dominated fans (Blair and 1978a, p. 676).
McPherson 1994).
Some differences in architecture may occur The term "sequence" was used by Heward (1978a) to
between systems dominated by traction-current refer to autogenic fan successions. They are typically
processes and those with significant sediment-gra- tens of meters in thickness and are classified as
vity-flow (debris-flow, mud flow) sedimentation group 7 deposits (Table 9.1). Examples are illus-
(the "true" fans of Blair and McPherson 1994). trated in Figs. 10.3 and lOA. Variations in upward
Trenches on fans that are formed predominantly by fining and coarsening character reflect the details of
20
5
OJ
m
"'0
-<
:0
:0 m
oGlu> z
Ou> 10
:0 0 I 0
»
o 'T1
r=
~ r
6
z
..,
om
SPC SCgl S CB
Heward. 1978b Gloppen & Sleei. 1981 This Paper.
Sec lion 2.
198·218 m
Fig. 10.3. Vertical profiles through three ancient conglo- profile, from Gloppen and Steel (1981), is from Devonian
meratic fan deposits, showing different types of autogenic basin in Norway. The right-hand profile is from the
succession. The left-hand example is a Carboniferous suc- Beartooth Conglomerate (Paleocene) of Wyoming-Mon-
cession in Spain, from Heward (1978b), and is interpreted tana. (DeCelles et al. 1991)
to have formed from a single debris-flow event. The middle
_Cool
Lt:GENO
~ U
inmotono
__
...:_'_11,,,
JIl _ _ '_WIg
'-' 1n>UgII ."...•• -w.a . ... -

-
..IJ. -
..,prigM _ In ~h pooiU""
-' - .",
Q" ~""'c'o" , _'"
COl ....."..... ls -_",bed_ing
~
...
-#' ......
CHmtMng r'ppIe _ lion
_ ripple _ "",
o.
d
~
Ii
I § 1-§
t ~
il
0
"&
~
a.
0 cloy,", IGnd ~
Fig. 10.4. Examples of distal, sandstone-dominated, sheetflood successions in ancient fan deposits, Carboniferous,
Spain, showing variations in upward coarsening and fining character. (Heward 1978)
channel and lobe initiation (Fig. 10.5). Upward fin- in their field area of sharp-based trench-fill succes-
ing above a basal erosion surface indicates erosional sions. Progradational units are thought to be more
trench or distributary initiation by avulsion, or em- likely to predominate in distal parts of fans, espe-
placement by lateral migration, followed by gradual cially where the fans interfinger with basinal depos-
filling and abandonment. Upward coarsening indi- its, such as lacustrine facies or the deposits of a trunk
cates progradation. DeCelles et al. (1991) found that river, as in the rightmost profile of Fig. 10.4.
their fifth-order litho somes typically did not show The concepts developed in the preceding para-
upward coarsening or fining (e.g., right-hand pro- graphs can be applied, with some modifications, to
file, Fig. 10.3), and attributed this to a predominance all nonmarine distributary systems, including termi-
The Evolution of Distributary Fluvial Systems 315
rt:-'\
Fan
surface Intersection
--_
Depositional
lobe ~
~
Fanhead
trench profile
longitudinal section
Plan view of alluvial fan (after Hooke J967)
eX location of vertical sequence examples
, gradual progradation of active fan segment followed by rapid abandonment
i avul.ive initiation of active fan segment fallowed by gradual abandonment
sequence scale
m-~sm
t gradual progradation and abandonment of active fan seQment
Fig. 10.5. Fan profile and model of possible vertical successions resulting from channel and lobe switching. (Heward
1978)
grade. Switching is gradual, because channel banks

o 25 are stabilized by vegetation and resist erosion; but
-=-=-
km avulsive switching of channels is nevertheless an
important part of the evolutionary process
(McCarthy et al. 1992). It is not known what vertical
successions develop in this river. The Oligocene-
Miocene Huesca fluvial system of Spain (Figs. 9.6,
9.7) was compared to the Okavango fan by Hirst
INDIA
(1991), and avulsive mechanisms that operated there
were also attributed to channel aggradation, perhaps
with the additional trigger of tectonic tilting at the
fan apex.
In the case of the Kosi River of India, many stud-
ies have reported the avulsive migration of the main
distributary channel across this sediment surface
(Holmes 1965; Gole and Chitale 1966; Schumm 1977;
Gohain and Parkash 1990; Wells and Dorr 1987). The
main channel has migrated westward a distance of
Fig. 10.6. Courses of the Kosi River, India, between 1731
110 km between 1736 and 1964 (Fig. 10.6). The cause
and 1963. (Schumm 1977, after Gole and Chitale 1966) of this migration was examined thoroughly by Wells
and Dorr (1987), who were able to demonstrate that
nal fans and the large sandy systems discussed in avulsive events were not correlated in time with
Sect. 8.3. In particular, the process of channel or lobe major floods or with tilting episodes that occurred
aggradation followed by avulsion is common to all during earthquakes. They confirmed the conclu-
distributary systems, although the details differ. sions of earlier workers that the channel shifting is
In the case of the large, sandy Okavango system of an entirely autogenic process. Gole and Chitale
Botswana anastomosis takes place as channels ag- (1966) demonstrated that channels shift to occupy
N
66"0'
500
t
260 500
250
200
250
200
175 ~~-- ____- -____ 175
150
Fig. 10.7. The Kosi fan in 1936. Solid contour lines show with the shaded areas indicating deposition. (Schwnm
configuration of the fan at that time. Dashed lines are 1977. after Gole and Chitale 1966)
contours re-surveyed following an avulsion event in 1957.
I. Recent shift: river course i. stab I.
"".11 .Iroam ,..,.,nUy occupied enannol
2. A<;Igradation: course is stable; some drainage into adjacent channels
3. River course is unstable: preferred drainage into lower channel
Fen probably comprises laterally stacked 15 km by 2 m channel belts

Fig. 10.8. Model for the avulsive shifting of the
main channel of the Kosi River. (Wells and Dorr
1987)
Avulsion in Fluvial Systems and Its Effect on Alluvial Stratigraphy 317
low-lying areas on the depositional surface (Fig. dynamic equilibrium, normal autogenic processes
10.7). The triggering process seems to be instability will lead to the gradual or sudden and episodic shift-
caused by channel aggradation, with the main chan- ing (avulsion) of the channel, with consequences for
nel moving to occupy low-lying areas traversed by the ongoing development of the alluvial stratigraphy
old, former channels, or tributaries of the Kosi sys- (Schumm 1977; Richards et al. 1993).
tem (Fig. 10.8). Wells and Dorr (1987) noted several There are two kinds of avulsion, that which occurs
other examples of similar lateral shifting of distribu- within the active channel system, and that which
taries on modern distributary systems, but it is un- displaces flow into a new channel position. The sim-
known whether the very regular one-way (westward) plest example of the first case is where the natural
sweeping of the Kosi distributaries is typical. migration of a river meander results in one reach of
Bluck (1979,1980) proposed a model for the regu- the meander bend cutting into another reach, result-
lar, autogenic, pendulum-like sweeping of an allu- ing in a neck cutoff. This process has been known
vial fan by its major distributary, based on some and understood for a very long time. For example, it
observations of modern fans and Devonian fan de- was described and illustrated in Lyell's (1830) great
posits. Vis eras and Fernandez (1994) described a textbook (Fig. 2.2). Chute cutoff occurs where a flood
Pliocene succession in Spain which they suggest fits channel develops across the top of a point bar (Fig.
this model. The mechanism for the model, suggested 8.58). The complex of channels, cutoffs, crevasse
by Bluck (1979, p. 201), involves the preferential splays and levees created by this process may be-
building of bars on one side of the distributary, with come slightly elevated from the main floodplain, and
bank erosion and channel widening occurring pre- is termed an alluvial ridge.
ferentially on the opposite bank. There is some The avulsive shifting of braided channels is an
stratigraphic evidence for this, as reported by Bluck integral part of the braiding process. Several auto-
(1980) and Viseras and Fernandez (1994), in the genic mechanisms occur. They are discussed in Sect.
form of bar accretion sets dipping preferentially in 6.3, where the emphasis is on the development of
one direction at a given stratigraphic level. However, gravel bars and bedforms. Avulsion is also the main
the cause of this preferential lateral migration of the process whereby channels in anastomosing systems
channel remains unclear. As discussed in Sects. 6.6 evolve, as described in Sects. 7.2 and 8.2.10.
and 8.1, bar growth and cutbank erosion are related Maps of meander migration in sinuous rivers may
to meander development both in braided and mean- be produced by observing the relict topography of
dering systems, and this, in turn, is related to the abandoned channels and of the scroll bars on point-
turbulent properties of flowing water. Therefore, it bar surfaces. Such maps began appearing in text-
seems unlikely that lateral bar accretion and cutbank books in the early part of this century, as exemplified
erosion would take place on the same sides of a river by those in Chamberlin and Salisbury (1909; Figs. 2.7
for more than one meander length. It seems more and 2.8 of this book). Maps of this type, or schematic
likely that a process of channel aggradation and versions of them, are popular as generalized illustra-
avulsive reoccupation oflow-Iying areas is the main tions of sedimentary processes (e.g., see the cover of
cause of lateral channel movement in fans, as de- Walker and James 1992). They are easily made from
monstrated (above) for large fans such as the Kosi. A aerial photographs, in which patterns of ancient
possible tectonic mechanism may also be sought for meanders are commonly quite spectacular. Current
the lateral migration of channels and the preferential work on meander evolution and migration and its
preservation of accretion sets dipping in one direc- stratigraphic results is discussed in Sect. 6.7. Cutoffs
tion, and just such a mechanism is described in Sect. in modern rivers were reviewed by Lewis and Lewin
11.2.1.1. (1983). Within-channel avulsion in gravel-bed
braided rivers was modeled by Ashmore (1991) and
Leddy et al. (1993) and is discussed in Sect. 6.3.
10.3 Avulsion in Fluvial Systems The second, more stratigraphically significant,
form of avulsive channel shifting is where the entire
and Its Effect on Alluvial Stratigraphy
river changes position, abandoning its original
course and flowing in a different part of the valley.
Fluvial style is governed mainly by a range of This process is most important in the case of mean-
extrabasinal controls, as discussed in Sect. 8.1. How- dering rivers, because these normally occupy only a
ever, given a normally graded fluvial system flowing small part of their total valley width, so that when a
in a steadily subsiding basin under conditions of shift occurs it has a profound effect on the resulting
stratigraphic architecture. However, avulsion is im- as a general term for highly sinuous rivers because of
portant in the case of other fluvial styles as well, as its characteristic winding pattern. Anderson (1961)
discussed later in this section. documented shifts in the Rufiji River in Tanzania.
On coastal plains avulsion may lead to the aban- Schumm (1977) illustrated old and new courses of
donment of a delta lobe and to the initiation of a new the Murrumbidgee River, Australia, using aerial
locus of deposition laterally displaced by tens of photographs, and discussed the climatic changes
kilometers. Maintenance of a new channel in such that accompanied this avulsive shifting. Holmes
cases is commonly facilitated by the fact that the (1968) described the shifting of the modern Indus
diversion results in the occupation of a shorter, River.
steeper route to the sea. This avulsion process is By far the best-known history of channel avulsion
analogous in some ways to that which occurs on is that of the Mississippi River, which was investi-
alluvial fans. gated in detail by Fisk (1952). Numerous changes in
the course of this river have been documented, and
many of these triggered the growth of new delta lobes
10.3.1 The Development of Meander Belts along the Gulf Coast (Fig. 10.9). Fourteen separate
delta lobes are known to have developed in the past
Much work on avulsion in meandering systems has 7500 years, since the last deglaciation (Frazier 1967).
been carried out by geomorphologists and civil engi- All of these required a major channel avulsion to
neers, because of the importance of the process from occur. Other avulsive events which occurred far up
the point of view of navigation and flood control. the Mississippi valley, such as at Cairo and Thebes
The review by Schumm (1977, pp. 297-305) is par- Gap (Fisk 1944), are probably not associated with
ticularly useful in this regard. deltaic diversion. According to Fisk (1952), the Mis-
There are two types of avulsive process in mean- sissippi has followed 20 distinct courses during the
dering rivers: firstly, high discharge, or aggradation past 2000 years, meaning a change in course some-
of the channel by a large sediment load, may lead to where in the lower alluvial valley on average every
the channel overtopping its banks, causing cre- 100 years. Not all workers agree with this reconstruc-
vassing and the diversion of flow into the floodplain. tion. For example, Saucier (1974) recognized only
Under certain conditions, this can evolve into long- five main meander belts (one shift every 400 years).
term channel diversion. Secondly, avulsion can Leeder (1978) concluded that in major rivers such as
occur when normal, gradual, meander migration the Mississippi "realistic mean avulsion periodicity"
results in the channel encountering a pre-existing, is in the order of one to two events per 1000 years. In
older channel at a lower elevation, resulting in the the hierarchical classification of this book this would
reoccupation of that channel. Such earlier channels assign meander-belt avulsion to group 6 or 7 (Table
may have developed as a result of the first of these 9.1).
two avulsive process. The case of the Atchafalya River, a distributary
Documented cases of channel avulsion in modern branching off the Mississippi (Fig. 10.9), may be
rivers are fairly limited. In some river valleys the taken as a type example of channel avulsion in a
existence of older channel belts has been recon- meandering system (details from Fisk 1952; sum-
structed from archeological or historical data in mary in Schumm 1977). The Atchafalaya is an old,
areas of early human civilization; for example, former course of the Mississippi, known from his-
Holmes (1965) referred to the lateral shifting of the torical records to have been an active distributary in
Euphrates River in Mesopotamia (Iraq) by a distance the early 1500s (Shlemon 1975). It was intersected by
of some 16 km, around the position of the ancient the modern river as a result of channel migration in
city of Ur. MacKay (1945; quoted by Schumm 1977) the late 1940s. Its course from this point to the Gulf is
documented the rise and fall of ancient cities, 225 km long. The modern channel is much longer, at
including those in Mesopotamia, whose prosperity 515 km, partly because of the lengthening that has
depended on the proximity to a navigable river. occurred at the mouth by delta construction. By 1951
Russell (1954) noted the existence of "old" and the Atchafalaya was carrying 21 % of the flow of the
present courses of the Meander River in Turkey, and Mississippi, and by 1956 it was the third largest of all
summarized historical evidence relating to the rivers flowing into the sea from the United States.
change in position of the meander belt. According to Diversion of the flow was gradual initially, because
Russell (1954) this river was mentioned in the work of the originally small size of the channel, but natural
of Herodotus and Strabo, and its name became used widening and deepening occurred, facilitated by the
steeper course along the Atchafalaya relative to the floor is aggraded by bedload sedimentation is there
main Mississippi, and it was estimated that once the likely to be a significant elevation difference be-
flow reached 40% of the total, complete diversion tween the channel and its floodplain, providing a
would be accomplished rapidly. Large new deltaic force to drive flow diversion. However, this is an
deposits developed very rapidly in the Gulf of Mexico important cause of diversion in some braided river
at the mouth of the river (Shlemon 1975). In 1963, systems, as discussed in the next section.
engineering control works were completed which What are the stratigraphic effects of major
maintain the diversion at no more than 30% of Mis- avulsive diversion in a meandering system? Most
sissippi flow, and in this way New Orleans, one of the obviously, they include abandonment of the origi-
great port cities of the United States, has been at least nal meander belt and initiation of a new one. In the
temporarily saved from the fate that befell Ur and case oflarge rivers, such as the Mississippi, this can
other ancient cities when their navigable waterways result in the displacement of the locus of sand-body
were abandoned. sedimentation by several to many tens of kilome-
The development of crevasses in rivers such as the ters. The steeper course of the new channel may
Mississippi is not thought to lead to avulsion in the result in straightening or scouring of the main chan-
normal course of events (Fisk 1952; Schumm 1977). nel above the point of diversion, although the geo-
Crevasse splays may form (Sect. 7.2.3), but unless logical results of this, in the form of unusually deep
there is a gradient away from the main channel there incised channels and changes in facies architecture,
is no force to drive the diversion further, and the may be destroyed by continued evolution of the
crevasse channel will simply become plugged with river system.
sediment and abandoned. Only where the channel
33°
32"
o 50 100
1 . . ; . '_ _---!''--_ _....l'
miles
31°
2 9°
92" 91° 90" 89° 92° 91° 90° 89°
Stage 3 - - - - - Stage 10
- - - - - - Stage 4 - - - - - - Stage 15
~ Miss. R. ~Miss.R.
....................... Atchafalaya R . ...................... Atchafalayo R.
Fig. 10.9. Former courses of the Mississippi River, according to Fisk (1944), as redrawn by Schumm (1977)
• COAL ANDCARBONACEOUS SHALE I:?=-~l SILTSTONE ANDSHALE

rn SANDS TONE ~ LIMESTONE
Fig. 10.10. Lithogenetic model for the autogenic evolution of a meander-belt complex, showing three stages in the
development of the stratigraphic succession illustrated in Fig. 7.2. (Flores 1981)
An ancient example of channel avulsion inter- avulsion. The deposits have been interpreted as fol-
preted to be by autogenic processes was described lows:
from the Himalayan foredeep by Behrensmeyer and "The upper cross-section (A) portrays the buildup of a
Tauxe (1982) and is illustrated in Fig. 9.1. These major channel complex prior to abandonment. The broad
authors suggested that the pattern of channel bodies lateral extent of this channel system reflects high sinuosity
in the "buff lithofacies" may represent avulsion or meandering of the stream. As the channels fUled, ad-
along two channel belts spaced 10-15 km apart, joining peat swamps spread into submergent abandoned
channel ridges, forming thick, laterally extensive peat
comparable to the small drainage systems occupying coals. The abandoned channel system, which was main-
the Himalayan front at the present day. Alternatively tained as a high topographic area, influenced the
the observed architecture could represent the activ- autocydic shift of a new channel into the adjoining topo-
ity of a single river. In the first case, given the de- graphically low area to the right (northeast). The middle
tailed magnetostratigraphic control available to date cross-section (B) shows that the area formerly occupied
by the oldest channel complex was transformed into a
these rocks, it can be shown that avulsion would have floodplain that probably supported poorly drained
had to occur on one or another of the rivers at a backswamps. The backswamps were relatively stable and
maximum spacing of once about every 6000 years. If long-lived as evident by deposition of thick peat coals.
only one river was present it would have had to Occasional intrusions by crevasse-splay and overbank de-
undergo an avulsive shift at a minimum rate of once tritus account for the splitting of the peat coals and testify
to the last activity in [the] adjoining channel before being
every 2000 years. This is therefore a "group 7" pro- inundated by peat swamp. The detritus probably came
cess. from a major channel that was formed at this time to the
Another excellent example of the alluvial archi- right (northeast) of the oldest channel complex. The small
tecture of a meander-belt complex is illustrated in channel sandstones associated with the crevasse-splay
Fig. 7.2. The interpretation of this composite cross sandstones are deposits of active feeder channels. In the
floodbasin to the right of the major channel, a very thick
section is illustrated in Fig. 10.10 and may serve as a peat accumulated, indicating development in a more
good example of the kinds of stratigraphic complex- stable and long-lived backswamp. The absence of thick
ity generated by autogenic crevassing and channel crevasse-splay deposits within the very thick, laterally ex-
----"77-.
",ro.
;; "', I Hypothesis This condition may have been in response to differential
compaction of earlier formed interchannel and channel
sediments, allowing subsidence of the floodbasin, which
ponded some open water." (Flores 1981)
A contrast with the humid meandering system

Plan VIeW
described above is the arid system documented by
Stear (1983). These Permian deposits of the Karoo
I Hypothesis 2, Basin, South Africa, exemplify an «ephemeral sand-
bed meandering river" fluvial style (Sect. 8.2.8; Figs.
8.40-8.42). Isolated sandstone sheets and ribbons
were deposited by channels that were free to mean-
der and avulse on a broad floodplain. These deposits
are comparable in facies to those of terminal fans,
but Stear (1983) presented no evidence of down-
I Hypothes is 3, stream fining and thinning, or of radial paleocurrent

patterns.
Willis and Behrensmeyer (1994) examined auto-
genic processes in a meander belt from a different
perspective, that of the overbank deposits. Excellent
outcrops of the Miocene Chinji Formation (Siwalik
ew Group) in Pakistan (Fig. 7.13) provided details on
the large-scale architecture of these deposits, and
I Hypothes is 4, provided the basis for a discussion of the possible
mechanisms for meander-belt evolution, with par-
ticular reference to the development of the flood-
plain. They proposed five hypotheses (Fig. 10.11),
most of which have been applied in different settings
to specific fluvial units by other workers. The first
hypothesis predicts that the entire floodplain ag-
grades episodically, producing uniform, basin-wide
I Hypothes is 5 , floodplain units. There do not appear to be any
documented examples of this case in the fluvial lit-
Pa leosol oounaea seQuence
erature. The second hypothesis relates the thickness
<II~
~
%.
, ~~
,
• •
....
~
-,'
.
and grain size of floodplain units and paleosol type
to proximity to an «alluvial ridge". This hypothesis
L...-~'--u.....::"""'-=< CrolS-net lonal view
has commonly been invoked and is the basis for the
«pedofacies" concept described in Sect. 7.4.2. Hy-
Fig. 10.11. Hypotheses to explain the development of
floodplain deposits in a meander belt. This diagram was pothesis 3 is based on alternation of degradation and
drawn as a basis for a discussion of the Miocene Chinji aggradation, as in the cycles of channel incision and
Formation, Pakistan, but has a general applicability. Ruled filling that occur during changes in base level (Sect.
areas on maps indicate floodplain areas of above-average 11.2.2, Chap. 13). Hypothesis 4 proposes that over-
elevation, dashed line indicates cross section location. See bank sedimentation is controlled by deposition in
discussion in text. (Willis and Behrensmeyer 1994)
topographically low areas of the floodplain, which
shift with time as a result of channel-belt aggrada-
tion and differential compaction and subsidence.
tensive peat coal suggests that the northeast side of the
channel system probably was not breached by flood wa- This hypothesis has been invoked for the modern
ters, thus allowing uninterrupted peat deposition in an Mississippi and Kosi rivers. The final hypothesis is
isolated, broad floodplain (e.g. marginal to abandoned similar to no. 4 but links floodplain aggradation in
meander channels propagated by avulsion). The lower low-lying areas to the avulsive shifting of major
cross-section (C) shows abandonment of the channels at channels. The specific flooplain units studied by
the right (northeast) accompanied by autocyclic shifting of
a major channel to the left (southwest). Correspondingly, Willis and Behrensmeyer (1994), which stimulated
the central floodbasin remained an interchannel low; how- this discussion (see Fig. 7.13), appear to have evolved
ever, it became a well-drained interfluviallake or pond. in the manner described by hypotheses 4 or 5.
10.3.2 Avulsion in Braided Fluvial Systems The spectacular and catastrophic diversions of the
Yellow River in China are of this type (Holmes 1965;
Braided rivers tend to occupy most of their alluvial Li and Finlayson 1993).
valley, leaving little room for floodplains, and there- The Eocene Escanilla Formation of the Spanish
fore only minor areas for the channel to avulse into. Pyrenees appears to represent an exception to the
However, avulsive shifting is a significant sedimen- generalization that in braided systems channels oc-
tary control, and large- and small-scale effects on cupy most of the alluvial valley (Dreyer et al. 1993;
stratigraphic architecture can result, as discussed in Bentham et al. 1993; see Sect. 8.2). Channel-fill archi-
this section. tecture indicates a braided fluvial style. Fine-grained
By far the most well-known examples of the large- deposits total more than 40% of the succession, by
scale avulsive shifting of braided rivers are those of volume. Bentham et al. (1993) attributed this fluvial
the Kosi River in India, and the Yellow River in style to rapid subsidence and frequent avulsion
China. The Kosi River is referred to elsewhere in this within an unconfined basin. There are indications
book as an alluvial fan (e.g., Sect. 8.3), and the avul- that the depositional system was distributary in
sion processes there are described in Sect. 10.2. How- nature, and there may therefore have been a ten-
ever, it may serve as the type example of a particular dency for the depositional surface to have been con-
avulsion mechanism, that which is triggered by vex-up, which would have facilitated avulsion, as in
aggradational elevation of the channel on the flood- the various alluvial fans described earlier in this
plain surface. This process seems to be particularly chapter.
important in braided systems, which are character- The alluvial architecture of the Old Red Sand-
ized by large bedloads and by rapid aggradation of stone (Lower Devonian) as exposed in coastal sec-
channels and t.ars. Other types of distributive sys- tions in southwest Wales, is discussed and illustrated
tems, such as terminal fans, also show this type of in Sect. 9.5.1 (Figs. 9.25, 9.26) and serves as a good
autogenic avulsion process (Sect. 10.2). Richards et example of tightly documented regional autogenic
al. (1993) noted that other fans in the Indo-Gangetic control of a largely braided fluvial system. Although
plains also show avulsive behavior similar to that of base-level change may be one contributing factor in
the Kosi. Several other examples of avulsive channel the development of this alluvial suite, autogenic
and bar changes in modern rivers are discussed in switching of the main channel belts on average about
Best and Bristow (1993). every 8000 years was also suggested by Allen and
Where the river is in the form of a distributive Williams (1982) as a major sedimentary control. The
system, as in the case of "inland deltas" such as depositional environment is interpreted as that of a
the Kosi River or terminal fans in arid regions, the coastal-plain tidally influenced fluvial and mudflat
channel and adjacent floodplains are elevated with complex, in which the channels were oflow sinuosity
respect to the regional surface as a result of aggrada- and may have been braided. Earlier work on these
tion. Processes such as those which occur in the Kosi beds by Allen (l974b), based on a much less exten-
River system are therefore likely to be characteristic. sive data base of vertical profiles, had led to specula-
West East
,,
'00 mL.I_=::...-_
Campbell"s ·channel system·
Fig. 10.12. Cross section of the Morrison Formation, connaissance correlation of channels within the Weswater
northern New Mexico, redrawn from the work of Campbell Canyon Member, and the top of the diagram does not
(1976). Units prefixed with ''J'' are members of the correspond to the top of the Dakota Formation. rectangle
Morrison Formation: Jmr Recapture Member; Jmw West- at left shows the area re-examined in detail by Cowan
water Canyon Member; Jmb Brushy Basin Member; Kd (1991)
Dakota Formation. The diagram was constructed by re-
tions by him regarding alluvial architecture and pro- to downstream accretion. As far as can be deter-
posals for the nature of the architectural response to mined, both the stacking of macroforms and, at the
various combinations of autogenic and allogenic next level, the stacking of sandstone sheets represent
controls. The eight models that he illustrated in that autogenic migration and switching (avulsion) of
paper formed the conceptual basis for all the later major and minor channels on the braidplain.
work in this area (Sect. lOA), including the modern Through the limited vertical thickness of this out-
work on the architecture of fluvial deposits in a crop there are no discernible systematic changes in
sequence-stratigraphic context (Chap. 13). channel or bar style, or in orientation, that would
A well-known example of an early study of point to some external control on sedimentation.
alluvial architecture is that by Campbell (1976), Autogenic stacking of macroforms was explored
who described the predominantly sheet-sandstone in more detail in another outcrop of this unit, at
bodies of the Westwater Canyon Member of the Tusher Canyon, near Green River, Utah, where de-
Morrison Formation in northern New Mexico. Some tailed paleocurrent analysis of cross-bedding and
authors have interpreted this unit as the deposit of a accretion-surface orientations permitted the recon-
giant sandy alluvial fan (e.g., Galloway 1980). struction of three of the bar units in a vertical succes-
Campbell's (1976) reconstruction is illustrated in sion. This example is illustrated in Sect. 9.5.7 (Fig.
Fig. 10.12. The apparent westward migration of the 9.59).
channels with time caught the attention of many The example of the Castlegate Sandstone illus-
sedimentologists, and the deposit has been com- trates the existence of autogenic processes on two
pared to the Kosi fan ofIndia (Miall1981a; Collinson scales, the intrachannel switching of bars and their
1986), which shows a similar westward migration resulting macroform deposits, and the avulsive
(Fig. 10.6). However, a recent detailed remapping of switching of the channels themselves, typically as a
this area by Cowan (1991) demolished this recon- result of aggradation. Campbell (1976) used a com-
struction and proposed a quite different fluvial style parable subdivision of sandstone bodies in his archi-
for the beds (Fig. 10.13). Cowan (1991) was able to tectural reconstruction of the Westwater Canyon
show that Campbell's (1976) mapping of individual Member in New Mexico, the larger-scale features of
sandstone sheets was based on outcrop color differ- which have now been shown to be in error, as dis-
entiation. He was able to demonstrate that the differ- cussed above. His reference to sandstone sheets as
ent colors of the sandstone bodies were formed by the deposits of "channel systems" composed of
diagenetic leaching, which was only partly related to stacked and amalgamated channel deposits has,
depositional architecture. The curved channel mar- however, been confirmed by Godin (1991), although
gins that make Campbell's reconstruction so appeal- in modified, reinterpreted form. Wizevich (1993)
ing are seen to be mainly inferred. Cowan's (1991) performed a similar type of cycle analysis on the Lee
new map of the extensive cliff outcrops shows that Formation (Pennsylvanian) of Kentucky.
the Westwater Canyon Member in fact consists of An example of Godin's (1991) detailed vertical
laterally extensive sandstone sheets that commonly proflles through the Westwater Canyon Member is
have indistinct, feathered margins and contain con- illustrated in Fig. 10.15. Small-scale fining-upward
siderable internal complexity, resulting from the cycles 1-6 m in thickness are amalgamated into four
migration of bar complexes in a braided fluvial envi- major cycles over the total 41.9 m thickness of the
ronment. Internal channel scour was particularly entire section. This section constitutes a complete
important in this river system. As discussed in Sect. profile through a mappable middle submember of
8.2.14, these deposits are now regarded as an ex- the Westwater Canyon Member. Godin (1991) de-
ample of a distinctive fluvial style, that of the "high- fined three types of autogenic cycle, based on analy-
energy, sand-bed braided river." sis of this and other proflles in the Gallup area of New
On a smaller, intrachannel scale, the stacking of Mexico (Fig. 10.16). The small-scale cycles, bounded
channels and macroforms in braided systems is lar- above and below by surfaces of third order or higher
gely the product of autogenic avulsion of minor and rank, he interpreted as the fining-upward succes-
major channels triggered by channel and bar aggra- sions formed by macroform accretion or scour fill.
dation. For example, Fig. 10.14 illustrates stacked Channel-scale cycles range from 1 to 10 m in thick-
sandstone sheets in the type section of the Castlegate ness and are composed of stacked macroform cycles.
Sandstone (Upper Cretaceous), near Price, Utah. They are bounded by surfaces of fifth order or higher
Most sheets are composed oflaterally and vertically rank. These cycles are not distinguished in Fig. 10.15.
amalgamated macroforms that were built by lateral Sheet-scale cycles are 4-16 m in thickness (heavy
VJ
6 7 8 • overbank N
, Mid get Mesa 9 depos~s
...
Mess Burre ,
wMe-COlored
, , Kd
D sandstone
~ buH-coIored
sandstone
Jmw ~ ~ • red-colored
sandstone
Jmr = 50m
15 krn 20 krn
\/
I
6 7 8 9
r
" "
Fig. 10.13. Comparison of part of Campbell's (1976) reconstruction ofthe area between his sections 6 and 9, below (see Fig. 10.12 for location), and Cowan's "r:I
detailed cliff map of the same exposures, above (Cowan 1991). Legend applies to Cowan's reinterpretation, Westwater Canyon sandstone bodies were not [
differentiated by color in Campbell's reconstruction, Kd= Dakota Sandstone, Jmw= Westwater Canyon Member, Jmr= Recapture Member E-
O
.g
oen
::;.'
0'
::I
E-
~rI>
S
en
P>
::3
0-
~
...o>-
~;:;'
Vl
rI>
0-
S'
rI>
::3
...
P>
-<
(J
o
::I
::;
8.-
en
=-
CASTLEGATE -A Q,-_ , - --,2P m ~ paleocurrent

_100· \ nor1ll
Fig. 10.14. Architectural analysis of part of the type posed mainly of laterally to downstream-accreted macro-
section of the Castiegate Sandstone (Upper Cretaceous), forms. A-I Major (fifth-order) bounding surfaces. (Miall
near Price, Utah, showing subdivision of the unit into a 1993)
series of stacked sandstone sheets, each of which is com-
arrows in Fig. 10.15) and may be hundreds of meters Some stretches of the Waimakariri River valley
to several kilometers in lateral extent. Each type of are unoccupied by active channels for significant
cycle may be terminated at the top by fine-grained periods of time, for two main reasons. Mid-channel
deposits representing bar-top deposits, channel- floodplain areas may represent large bar complexes
abandonment deposits, and floodplain deposits that have accreted to the point that channel diver-
formed following channel-belt avulsion events. sion around them conserves more energy than does
However, these fine-grained deposits commonly are their elimination by erosion (e.g., area A in Fig.
absent because of scour accompanying the initiation 10.17). In other cases, floodplains develop in pro-
of the overlying deposit. For example, the top of the tectedareas (e.g., area E in Fig. 10.17), although long-
first sheet-scale cycle in Fig. 10.15 is indicated not by term migration of the channel complex may lead to
fine-grained deposits but by a prominent intraclast periodic flushing out of any of these floodplain areas
lag breccia at the base of the next cycle. Amalgam- (Fig. 10.18). Aggradation of channels may lead to
ated, multistory sand bodies showing little internal reoccupation of former channels on a floodplain
cyclicity are commonly the result, and distinguish- surface, and this may contribute to the erosion of a
ing the various types of cycle from vertical profile long-term floodplain area.
data alone may be impossible. The stratigraphy of floodplains in gravel-bed
Godin (1991) pointed out that because of the pres- braided rivers consists of amalgamated units of
ence of several superimposed scales of cyclicity channel-bed gravel deposits interbedded with thin
within the same sand body, the use of simple one- units of floodplain fines, such as lithofacies FI or Fm.
dimensional Markov-chain techniques for the analy- The fine-grained units are a few decimeters thick,
sis of cyclicity is not to be recommended. This tech- ranging up to a few meters thick where they fill
nique was popular in the 1970s (e.g., Miall 1973; abandoned channels. The lateral extent of these fine
McDonnell 1978), but the advent of two- and three- units is unlikely to be greater than a few hundred
dimensional architectural methods of analysis has meters.
revealed its inadequacy as a method of investigating
the complexity of depositional processes in fluvial
deposits. 10.3.3 Avulsion in Anastomosed Fluvial Systems
A useful study of floodplain development in
gravel-bed braided rivers was reported by Reinfelds The cause of anastomosis is generally agreed to be
and Nanson (1993). They mapped modern flood- the low gradient of the river. Smith and Smith (1980)
plains in the upper reaches of the braided cited the backwater effect caused by the damming of
Waimakariri River in New Zealand and showed that a river by temporary obstructions, such as ice or log
significant areas within the river valley are occupied jams, or by more permanent damming, for example,
by floodplain deposits (Fig. 10.17). These floodplain by the progradation of a tributary alluvial fan across
areas correspond in origin to the upper topographic a confined valley. Immediately above such impedi-
levels of the Donjek River, Yukon, the type example ments the slope is reduced and crevassing and avul-
of a gravel-bed braided river (Sect. 8.2.3). sion are frequent events. A rise in base level, or the
sr·sa \
sr.w~
MEASURED VERTICAL SECTIONS
SEDIMENTARY STRUCTURES
FACIES ShlSI· low

angle to horizonlal
~ FACIES SI. low
E::::::::;:::l angle 1I'Oughs
~~I r.w'
~
~ FACIES SI/St • low to r-:l FACIES 51-1 · large
~ normal angle x·strata I.2:::::iJ trough x· strata
sr-r
D FACIES St·$ · small

trough x·strata
I;;• ':1 lithic
0 '
FACIES Se · hetera·
scours
~ FACIES Se • intraclastL-:l FACIES FmlFI · OVer.

~ scour tacil~s ~ bank tines; fine sand
to mud
~ FACIES Sp • p/atlar f7::\l Bounding surface rank sa·r
~ cross· stratifocatlon ~ (trom Miall. 1988)
IfiiSVI No structure visible·

~ maSSive appearance
0 Coset boundary
sa·r
t;;I Undulose contact sr·sa

r-sr
"
SORTING
vw very well
w well
moderately well
LITHOLOGY
moderately
fines· mudstone
poorly
fines· siltstone
v very poorly
sandstone
pebbly sandstone
extraclasts
ROUNDING
intraclasts
~
r well rounded sa· ,
UJ
r rounded sr-sa ..J
«
U
sr subrounded r-$r
sa subangular r·sr
'"
:::i! UJ
Ill: ..J
«
a angular r-sr
f2 u
m ..
u,.
Si>
0
Ill:
U
«
'"
~
UJ
:::i! :r
WHITE ROCK MESA EAST
'"
Fig. 10.15. Detailed vertical profile through the West- showing the stacking of fining-upward cycles on at least
water Canyon Member of the Morrison Formation (Upper three scales. (Godin 1991)
Jurassic) at White Rock Mesa, near Gallup, New Mexico,
tectonic rejuvenation of a structural high crossed by the Rhine-Meuse system of the Netherlands, the
the river, will also have the same effect of reducing avulsion frequency has been about 1 ka during the
slope (D.G. Smith 1983, 1986; Tornqvist et aL 1993). past 10 ka, and that the number of active distribu-
The last two controls are, of course, allogenic, not taries was greatest during the period 6-4 ka B.P., at a
autogenic. Tornqvist (1994) demonstrated that in time when the postglacial sea-level rise was at its
Quantitative Studies of Alluvial Architecture 327
Fig. 10.16. Model of cyclicity in the Westwater Canyon Mem-

-"""II~-~-.:r-;r-..,""'"- 5th order surface
ber of the Morrison Formation, based on vertical profIles,
such as the one illustrated in Fig. 10.15. (Godin 1991)
SHEET .
,
S"Nf)STONE 3rd or 5th order surfaces
..-.,;;;;:::;;;::;;;:;:;:t:l:t 3rd order surface

5th order sur face
MACROFORM ASseM81AGE
A·B Macroform scale fining upward cycle

C-O Channel/scour scale fining upward scale
E-F Sheet sca.l . fining upward cycle
•
N
I
[J
.. F\.OCXlI'I.N<
~
NOn-o:ACUS I VAI,UY
so;
..........,
0.5
Kllom _tru lSJ u>:JO<)NSI:JoPf'
0
Fig. 10.17. Distribution of channels and floodplain areas location (A) and, more commonly, in areas protected by
within part of the Waimakariri River, New Zealand. Exten- bedrock, bends in the valley, or projecting alluvial-fan
sive areas of floodplain occur in unprotected midchannel deposits (E). (Reinfelds and Nanson 1993)
most rapid and the anastomosing fluvial style was 7.2.2 and 7.2.3. An example of the type of stratigra-
dominant. phy generated by avulsion in an anastomosed sys-
The Cumberland Marshes of east-central Saskat- tem is illustrated in Fig. 8.54.
chewan have served as an excellent field location for
the study of avulsion in anastomosed fluvial systems
(Smith et al. 1989). This area, nearly 100 km across, 10.4 Quantitative Studies
consists of a complex of shallow lakes and anasto-
of Alluvial Architecture
mosed channels (Fig. 10.19). The Old Channel of the
Saskatchewan River was dammed by ice in 1873,
causing a diversion to the north, where a "New Chan- Information regarding the architecture of alluvial
nel" developed. Most of the other channels in this deposits is vital in the petroleum industry for the
area are now largely inactive, with the exception of purpose of estimating reservoir volumes and petro-
Carrot River, in the south. The New Channel has leum productivity. There are three questions to be
continued to evolve by processes of autogenic cre- asked:
vasse formation and avulsion, as described in Sects.
1. What are the lateral dimensions of the sand body?

2. Does it contain internal heterogeneities, such as
individual point bars, shale drapes, and aban-
doned channels, and if so what is their scale and
disposition?
3. What is the connectedness of the sand bodies?
In this section we examine the various empirical
studies and statistical and computer-based simula-
tion models that have been developed to address
these questions. We focus here on alluvial suites
built by autogenic processes, those discussed in the
first part of this chapter. The effects of tectonic and
other allogenic modifications are discussed in Chap.
11.
10.4.1 The Dimensions of Fluvial Sand Bodies
Channel dimensions are set by external (allogenic)

o ~~~~~~~;~EETATED controls, including discharge and overall sediment
t:J YOUNG FLOODPLA IN
supply, and by internal (autogenic) mechanisms that
~ MATURE FLOODPLAIN govern the geometry of the channel, including its
c;J ALLUVIAL FAN width/depth ratio and sinuosity (Sect. 8.1). Channel-
o VAL LE Y SIDE scale turbulence patterns and local bed and bank
materials (within-basin autogenic controls) are
Fig. 10.18. Model for the development of floodplains in largely responsible for determining channel form at
braided rivers, (Reinfelds and Nanson 1993)
.: IM -
."10
.....2,0 k m
' ~'?,--..,...
°
0"'1._.:..
' 'T
A .bmi
Fig. 10.19. The Cumberland Marshes, Saskatchewan, showing the complex of anastomosed and meandering channels
formed by autogenic avulsion, (Smith et al. 1989)
B.
A.
1m.11 K."
x-str.t.
upper ph •••
plene bedI
Fig. 10.29. Relationship of fining-upward cycle to depth of channel and thickness of point bar. (Leeder 1973)
.. .,-.......,.
• •
. 0
0 . >0'
•\.-\.
0
· 0 0
..
0
.~ l!
:. ~ ~·oo 0 0
• o 0 0 0
8
0
0.. ~o
•• • 0 0
0 0 0
"0. 0 0 ~
o
•• • o. ~oo
reP 0 000
o o
h • bankfull dept., ) 1m
w • bankfun width
• • ct\inr\el S!l'lUMlty) I · 7
0- ct\3r'Wlet s'~!Jily0' r:i
W metres
Fig. 10.21. Plot of 104 measurements of bankfull depth h against width w, for modern rivers. (Compiled from various
sources by Leeder 1973)
a given location. Reconstruction of the local channel Mannville Sandstone in the Lloydminster area,
form is the subject of this section. Alberta, in Sect. 9.5.3 for an example of the difficul-
Vertical profile data from outcrops and drill holes ties encountered in such work).
provide information on sand-body thickness and Much work has been carried out to investigate the
internal lithofacies composition but yield no direct internal relationships between the various dimen-
information regarding the lateral dimensions of the sions of fluvial channels, meander bends, and their
sand body. Where the outcrops are large and con- major deposits. Measurements have been collected
tinuous it may be possible to map entire bodies, as and tabulated for such data as channel width and
described in an example below. However, this is depth, meander radius, point-bar dip angle, and the
rarely possible from subsurface data. Well spacing relationships between these data and sediment grain
generally is inadequate for accurate sand-body map- size and discharge. Two useful papers are those by
ping, except in cases of tight well spacing during Leeder (1973) and Ethridge and Schumm (1978).
enhanced recovery projects (see discussion of They showed that for typical meandering rivers
h
metres
log w • I-54 log h + 0-83

r • 0-91
Sd = 0-35
sample of 57 rtvers where:-
h ~ bankfull depth> 1m
w =bankfull width
Sinuos l ty.,.I·7
10 10 10 10
W metres
Fig. 10.22. Plot of 57 measurements ofbankfull depth h against width w, for modern meandering rivers (sinuosity> 1.7).
(Compiled from various sources by Leeder 1973)
some useful statistical relationships can be devel- data reflects the fact that cycles may be incompletely
oped for predictive purposes. The thickness of a preserved, and that cycle thickness will not neces-
fining-upward cycle in the deposits of such rivers is sarily be the same in every location around a mean-
approximately equal to the bankfull depth of the der bend. It should be noted that no comparable
river (Fig. 10.20). The width of a point bar, as mea- statistical relationship has been demonstrated for
sured by the horizontal extent of a single accretion modern low-sinuosity « 1.7) rivers, which are
surface from top of bar to base of channel, is ap- mainly those in the braided category.
proximately two thirds of the former channel If point-bar dip angle J3 can be measured in the
bankfull width (Leeder 1973). Unfortunately, accre- field or obtained by identification of dipping sec-
tion surfaces are rarely fully exposed in outcrop and ond- or third-order surfaces in a core or dipmeter
are extremely difficult to reconstruct from subsur- log, another method of estimating channel width is
face data. As noted in Sect. 9.5.8, attempts may be available (Leeder 1973). Width and depth are sys-
made to use dipmeter data for this purpose, but tematically related to this angle, as shown in Fig.
results to date have not been promising. 10.23. The relationship is:
Leeder (1973) compiled data to demonstrate the
relationship between channel depth and width (Fig. w = 1.5 h/tan J3
10.21) and showed that the width/depth ratio is Point bars are discussed in detail in Sect. 6.7. Several
greater for channels of low sinuosity, as expected. studies of fluvial point bars in the subsurface are
For meandering rivers (those of sinuosity> 1.7), reported in Chaps. 14 and 15 and reveal the regular
Leeder (1973) was able to demonstrate a reasonably geometry of these bars. See, for example, the discus-
close statistical relationship between width and sions of the Fall River Sandstone in Wyoming, the
depth (Fig. 10.22). The relationship is: Tuscaloosa Sandstone in Mississippi, and the Chero-
log w = 1.54 log h + 0.83 kee Group in Missouri.
or w = 6.8h 1.54 Leopold et al. (1964) demonstrated that channel
dimensions are related to the scale of meander
r = 0.91, Sd = 0.35 log units bends, through the equation:
where w=bankfull width and h=bankfull depth. The Lm = 10.9w1.01
value of this equation is that, using the 95% confi-
dence limits, upper values of channel width can be where Lm=meander wavelength. Using this equa-
estimated from cycle thickness. The scatter of the tion some estimate of the radius of a point bar can be
• - palae~('hannel dimenSionS from

epsilon cross shatlftcatlon
1
10
a ~:::::!
..::::',",. : .:
1 2
10 10
Wmetres
Fig. 10.23. Relationship between bankfull width and depth and point-bar dip angle, p. Note decrease of dip angle with
increasing river size. (Leeder 1973)
obtained. This information may of considerable po- of the meanders (the distance between two succes-
tential usefulness in attempting to determine the sive meander cutbanks measured perpendicular to
best location for step-out wells, when prospecting in the channel trend). This relationship would not ap-
point-bar sandstones. The user should be aware, ply where sedimentation rates are slow and the river
however, of the built-in potential error in these esti- has an extended period of time to comb back and
mates. Precise predictions cannot be made, and the forth across its own deposits. Lorenz et al. (1985)
technique is best suited for determining maximum developed the following relationship between chan-
possible dimensions of the various parameters of nel width and meander-belt amplitude (Wm), using
interest. Thus, there may not be a simple relation- data from Leopold and Wolman (1960) and Carlston
ship between sand-body width and channel dimen- (1965):
sions. As shown in Fig. 6.3, channels can aggrade
Wm = 7.44 wl. OI
both vertically and laterally, leading to sand bodies
having very different width/depth ratios than the Assuming average conditions, this equation can be
channel which deposited them. The sand body clas- used to estimate the expected width of a sand body.
sifications dependent on geometry, such as that pro- This should be a useful technique for subsurface
posed by Friend (1983), should therefore be used petroleum exploration, where information is needed
with caution. An example of the variability in sand on potential reservoir dimensions.
body geometries is illustrated in Fig. 6.4. Seventeen Collinson (1978) assembled data to explore the
units bounded by fifth-order surfaces can be de- direct relationship between channel depth and me-
fined. These range in thickness from 5 to 43 m, with ander-belt width. He was not able to be certain of
a mean of 16.3 m. The full width of many of these eliminating some nonmeandering sheet sandstones
units is not exposed. They range from a minimum of from his data set, and the resulting equation shows
160 m to an exposed maximum of860 m. The average some scatter:
exposed width is 379 m. Even if the true mean width
Wm = 64.6h1. 54
is double this, the width/depth ratio is 46, which is
low for a sandy-braided system. Lorenz et al. (1985) reported a test of his method,
Lorenz et al. (1985) suggested that, given average where both outcrop and core information were
sedimentation rates and channel residence times, available to permit a cross-check between the vari-
the width of a typical meandering-river sand body ous measurements and estimates. They measured
would be approximately equivalent to the amplitude cycle thickness in cores (Fig. 10.24), and plotted a
'" { \ CORE
.I LITHOLOGY
-;;. \LOG
CORE GAMMA-RAY LOG DOWNHOLE

GAMMA-RA Y LOG
!'" 11011/3.05 m
5,300
DOWNHOLE
CORE GAMMA-RAY LOG GAMMA-RAY LOG
4,850
INDIVIDUAL POINT- 110 1I/3.05m
BAR DEPOSITS
1
10 Davies et al. (1993) used the estimation methods

'"c:0 described here to predict the scales of sand bodies
; deposited by high- and low-sinuosity rivers in a sub-
>
~
 surface example in Texas, but no independent data


'"0
.0
from outcrop architectural studies or closely spaced
5
wells were available to test the estimates, and it is
'0 risky to base predictions of fluvial style and sand
~
.0 body geomtry on a few examples of vertical profiles
E
:I
c: in wireline logs or core, as in this study. Thus, sheet
0 sandstones are commonly the product of braided
0 5 10 15
preserved point-bar thicknesses noted in core (It)
fluvial systems, as might be expe<;ted from the large
width/depth ratio of such rivers, but can also be
deposited by high-sinuosity rivers. For example, the
Fig. 10.25. Histogram of interpreted point-bar thick-
nesses in cores through the Mesa Verde Group, Colorado. width of the braided Brahmaputra River valley
(Lorenz et al. 1985, reprinted by permission) reaches 14 km, with an active channel up to 3 km
wide (Coleman 1969). However, Bristow (1987,
1993a) showed that lateral accretion of sand flats is
histogram of the results (Fig. 10.25). Rare, complete one of the dominant modes of bar construction in
fining-upward cycles (Fig. 10.24) provided a maxi- this river. Sheets consisting mainly of lateral accre-
mum thickness for the cycles, whereas many of the tion deposits (element LA), with occasional incised
cycles had been truncated by internal erosion. The minor channels (element CH) are locally common
thickest cycle, at 4.3 m, was assumed to represent and indicate an abundance of sinuous channel
the high end of a normal distribution, with the lower reaches. This is but one example to demonstrate that
end of the thickness distribution, at 2.7 m, suggested the presence of lateral-accretion deposits is not a
by the preservation of nontruncated cycles of that reliable indicator of a meandering style and does not
thickness. Using an estimated "typical" range of automatically permit the application of the equa-
cycle thicknesses of 3-4 m, Lorenz et al. (1985) then tions discussed above. Stear (1983) described ex-
used the equations given above to estimate channel amples of sandstone sheets with LA deposits more
width and meander-belt amplitude (=reservoir than 250 m wide in the Beaufort Group of South
sandstone width). They arrived at a range of350-520 Africa (Figs. 8.41, 8.42) that were deposited in an
m for predicted meander-belt width. ephemeral meandering system. In the Dakota Sand-
Nearby outcrops of the same stratigraphic unit stone of New Mexico and Arizona, similar sheets
enabled Lorenz et al. (1985) to test their estimates. exceed 1 km in width. R.M.H. Smith (1987) described
Using air photographs, they measured the widths of an exhumed sandstone sheet at least 6 km across,
meander-belt sandstone lenses, making allowances consisting of the preserved point bar and associated
for the angle of the exposure and the orientation of facies formed within four successive meander bends
the bodies relative to the outcrop face. This proce- of a major high-sinuosity stream in the Permian of
dure yielded a range in width values of 60-2130 m, South Africa. The width of the meander belt in this
with a mean of316 m, a standard deviation of275 m, example is approximately 3 km.
and a median of250 m. In some cases point bars with Fielding and Crane (1987) assembled data from
lateral accretion surfaces were exposed. Lorenz et al. numerous studies, including those described above,
(1985) measured the width of these, multiplied these and plotted a graph in which five distinct styles of
values by 1.5, and used the resulting estimated chan- sand body geometry could be evaluated (Fig. 10.26).
nel widths to determine meander-belt widths. The Their case 1A represents streams with the lowest
resulting estimated range of values is 430-650 m. width/depth ratio and is limited to incised, straight,
These three methods of determining reservoir width nonmeandering rivers. Case 1B represents the upper
yielded reasonably consistent results, and Lorenz et bounding line for meandering channel deposits.
al. (1985) demonstrated their applicability to a few Case 2A is the best-fit line for all the data and repre-
other cases described in the literature. sents a geometric mean. This relationship could be
Fig. 10.24. Determination of fining-upward cycle thickness from core and wireline log data. Mesa Verde Group
(Cretaceous), Colorado. (Lorenz et ai. 1985, reprinted by permission)
I
• LOW SINUOSITY
o MEANDERING CHANNEL
• BRAIDED MORPHOLOGY
" ANASTOMOSED
" CHANNEL MORPHOLOGY
NOT KNOWN
100
,,--
!
iA.!=
10
lA-- "X
--
"•
W
Q
....I
W
Z
Z
lB •
...
C
:E:
C) X"
• 2A
• • 2B
!! I , " ,I , , "'" I " " ,I

10 100 1000 10000 100000
CHANNEL BELT WIDTH (m)
Fig. 10.26. Plot of channel depth, h, versus sand body width, Wm, for various types of modern and ancient fluvial
systems, using 45 published sources. (Fielding and Crane 1987)
used where a variety of channel types are suspected Examples of the type of data collected by Dreyer
to coexist, or where the formative channel type is not (1993b) are illustrated in Fig. 10.27. These data were
known. Case 2B is Collinson's (1978) relationship for measured in an ephemeral system, in which sheet
typical, "fully developed" meandering streams. Case and ribbon sandstones were deposited on an arid
3 is the lower relationship for typical, "fully devel- fluvial plain and frontal splay sheets were formed at
oped" meandering streams. Case 3 is the lower the mouths of incised channels. Other quantitative
bounding line for all the data and represents sand studies of sand body dimensions were reported by
bodies deposited by laterally unrestricted rivers, Cuevas Gozalo and Martinius (1993).
such as braided rivers.
In recent years a great deal more field work has
been carried out on the architecture of exposed flu- 10.4.2 Estimating Probabilities of Sand Body
vial sandstone bodies in order to improve the data Penetration and Interconnectedness in the Subsurface
base for subsurface modeling and prediction. AN or-
wegian program, led by T. Dreyer, was initiated fol- Figures 10.28 and 10.29 illustrate the problem to be
lowing the discovery of large petroleum reserves in addressed by exploration geologists attempting to
fluvial reservoirs in the Norwegian sector of the determine the location of sand bodies in the subsur-
North Sea (Dreyer 1990, 1993a,b; Dreyer et al. 1993). face. The success rate of exploration wells depends
Dreyer and his colleagues examined several fluvial on the width and spacing of the sand bodies. The
and deltaic units using the kinds of architectural degree of interconnectedness of the sand bodies will
techniques outlined in Chap. 4. They reconstructed affect their petroleum productivity and determine
fluvial styles, examined the control of climate and the strategies to be used for field development. The
tectonism, and compiled data on sandbody geom- first important parameter, that of the scale of the
etry at several scales of the architectural hierarchy. sand bodies, has been addressed in the previous
Another Norwegian group examined the exposed section. Here, we review briefly the work that has
fluvial crevasse-splay deposits of the well-known been done to explore the distribution of sand bodies
Ravenscar Group of Yorkshire (Mj0s et al. 1993). within a basin-fill.
20 7~~--~---r--~--~--~~~
A
- 6
"E 16 .§.
'"
:;'l12
'"'"OJ
~ ~ 4
<J .<'!
£ 8
x
::;;;'" 4
o.5.j....t.-!J'-----'----'----'----'----'-----'-4
0 o 40 80 120 160 200 240 280
1 1000 10000 Width (m)
6~------~--------~~----~
o
o• 0~5-------2~5~------~5~0------~75
o ~ ~ ~ M1001~1~1~1M~0
Width (m) Width (m)
9.5 ~--..__----~---__r~--_
"E 8 :[ 8
:;''"l 6 :;''"l 6
c c
.x .x
£
<J <J
£4 x
4
x
i 2 i 2
20 40 50 75 o ~O--10J..0-:'2.L.00-:'3....00:-"':4..1.00--5..1.00-5...1.0-0-7.J.0-0 -8...1.0-0 -9"'00

Width (m) Width (m)
Fig. 10.27 A-F. Plots of thickness versus width for various frontal-splay sheetflood sand bodies; D single-story rib-
types of sand body in the ephemeral Esplugafreda fluvial bon bodies; E multistory ribbon bodies; F multilateral
system (Paleocene), northern Spain. A All coarse-member channel bodies. (Dreyer 1993b)
sand bodies; B channel-to-sheet transitional bodies; C
bodies within a fluvial basin fill, depending on such

factors as subsidence and sedimentation rates and
the nature of the channel-shifting patterns, specifi-
cally, steady meandering versus sudden switches in
position (avulsion). Given input data from prelimi-
nary studies within a basin, it may be possible to
employ these models to predict sand body distribu-
tion and well-penetration probabilities.
Fielding and Crane (1987) developed a statistical
model for prediction of well-penetration success
rates. They proposed that first the channel sand bod-
ies be identified in each well log by analysis of
lithofacies characteristics and wireline-log response.
The use of wireline logs is discussed in Sect. 9.5.2. An
Fig. 10.28. The importance of well spacing in effective analysis of fluvial style must also be carried out so
exploration for lenticular sand bodies. Where only wells A that the appropriate case from Fig. 10.26 can be
and B have been drilled a sand body of intermediate depth selected. The thickness of each identified channel
will not be discovered. (Fielding and Crane 1987)
sand body is then multiplied by its predicted width,
using the case selected from Fig. 10.26. In this way
Leeder (1978), Allen (1978), and Bridge and the cross-sectional area of all channel sand bodies in
Leeder (1979) developed numerical models which the well can be calculated and totaled. Given that the
explored the size and spacing of meander-belt sand wells already drilled in an area have penetrated a
50 IOOkm
J
I
I
. ' .
.
• • • •• ........ .- . . . - ·1:1. • • • •• •
·c · ..
F • ~ .
• ••
.
• • I~
• • •
-
Fig. 10.29. A-E Results of successively denser sample and 7 fail to find any sandbodies because of the random
spacing in the exploration of a fluvial section 100 km spacing of these units, but such failure means little in
across. Black squares indicate location of channel bodies. terms of the overall distribution of the sandstone bodies.
Stippled areas represent areas of outcrop or well sections. (Leeder 1978)
The location of all channel bodies is given in F. Wells 4, 5
.
wellp well p+1
I~.--------------~:~i ----I
wellp
I
I~ W J
Wn •
I
Sn
•+ r
I
tn Sn ____ I
I
•r ~D.1
Wn·1 I I
~ ____.__________~LI
tn·1 Sn·1 I
t
•
Wn.2 [J
+r
t - 1_ _ _
Sn_.2_.......
tn·2 Sn.2 •••••••••••••••••••••••••••• 1
I:::::::::::::··
1···························1
•........................... ,
I ......... . ~---------:::l
I~~ --~
t2 ~ --=----=--=-
IW1 S2 Sl - - - - - --=- --=--:.
- -:::.
t1 1::::1
I Sl
I
Cross-sectional area of all sandstones present (b)
Cross-sectional area of sandstones intersected (a)
L
n
= i i=1
tjWj
=
i=1
tjW
Success rate = alb
Fig. 10.30. Statistical description of sandstone beds. In present can be predicted based on well spacing. The well
average well p there are n sand bodies having thicknesses tn "success rate" is the proportion of the total cross-sectional
and widths W n; the latter values are estimated from Fig. area that is penetrated by the development wells. (Fielding
10.26. In well p+ 1 the cross-sectional area of all sandstones and Crane 1987)
en parameters are very difficult, if not impossible to

a> 50
=c0 40 make from reflection-seismic or wireline data, and
.c even core data provide very limited information on
'0
<:
30
three-dimensional sand body architecture. Bridge
'en"
Qi
20 and Mackey (1993b) also pointed out the inherent
.c
Qi
variability in fluvial systems and the consequent sta-
<:
<: tistical error in the approaches summarized in this
10
'"
J::
(,)
and the preceding sections. Most of the data sets
'0
a> from which the various equations have been derived
'0 represent limited suites of measurements collected
a>
<:
5
<:
0 4 in quite specific fluvial settings. The methods may
l:! 3 therefore be used to suggest limits, or ranges, for
a>
~ sand body dimensions and relationships but should
'0 2
not be relied upon to provide precise estimates.
a;
.c
E
:J
<: 1
0.0 0.2 0.4 0.6 0.8 1.0 10.4.3 Alluvial Stratigraphy Models
sands fraction _
The processes by which fluvial channel and flood-
Fig. 10.31. Sand body interconnectedness as a function of
the proportion of sand in the section. (Fielding and Crane plain deposits are stacked to form stratigraphic units
1987) have been the subject of several specialized studies
over the past two decades. As noted by Friend (1983),
Allen's (1965) speculative diagrams (Fig. 2.17) were
representative succession, then any hypothetical very influential, because they represented the first
well drilled through the interval of interest will pen- attempt to interpret the details of alluvial stratigra-
etrate a set of sandstone beds equivalent to those phy in terms of autogeniC and allogenic sedimentary
found in the test well. A "success rate" can be de- controls. Allen (1974b) returned to this subject in an
fined, which is the proportion of the total channel- evaluation of the architecture of the Old Red Sand-
belt sandstone volume in a given interval that will be stone in South Wales, in which he drew attention to
penetrated by planned development wells a distance the significance of pedogenic carbonate units as in-
W apart (Fig. 10.30). dicators of periods of nonsedimentation in inter-
Theoretical simulation studies (Fig. 10.31) sug- fluve areas and as marker beds that could be used for
gest that channel bodies are effectively in isolation correlation within suites of laterally impersistent
if they comprise less than about 50% of the succes- channel and floodplain deposits. Based on qualita-
sion (Allen 1978; Fig. 10.31), although intercon- tive, deductive reasoning, Allen (1974b) developed
nectedness may be higher locally where sand body five distinct models, with additional variations
distribution is not entirely random but is con- bringing the total to eight. These represent essen-
strained by local tectonic factors, or by such factors tially abstract models in which an interpretation of
as bank stability (e.g., presence of clusters of resis- the Old Red Sandstone was an almost incidential
tant shale plugs filling abandoned channels, which byproduct of the analysis. Two of the models are
resist lateral channel migration). reproduced here because of their historical interest
Estimates of sand-body interconnectedness are (Fig. 10.32). The first represents a purely autogenic
one of the important outputs of the computer simu- model, and one which has now been produced by
lation models discussed in the next section. computer simulation, as discussed below. The sec-
As noted by Bridge and Mackey (1993b), great ond represents, in effect, the first model of fluvial
care must be employed in attempts to use outcrop sequence stratigraphy, as discussed in Chap. 13.
data as a basis for statistical estimates or modeling of Attempts to quantify these models were made by
a subsurface fluvial unit. The exercise will be mis- Allen (1978) and Leeder (1978). Both authors make
leading if the fluvial systems in outcrop and the reference to the need to quantify sand body geom-
subsurface represent different fluvial styles, have etry and architecture as an aid to the study of hydro-
different channel scales, and were deposited in dif- carbon reservoirs.
ferent tectonic and climatic regimes. Unfortunately, The generation of alluvial stratigraphy models by
many of the observational details that are used to computer simulation was pioneered by Bridge and
interpret these autogenic and allogenic controlling Leeder (1979). This is a much-quoted paper that was
,.' U ....lIIIf. IOII

.·.. l ' ."
LEGEND
o Floodplain d,posils (mud,'on,.) CJ Chonn,1 sandslon" ~ P,doOln"iC ca.bonol, un it.
A E Major eros ion surfaci 26 Number of av ul sion
:'<~;{Wi'~li:i~~':;~i;~~;
\JSlli
......,.,.,..,
/:.::~.";:;~:';jjjjj;l
'""'-
/:\'-
, U:· ........
LEGEND
o Floodploln depol'" (mudlfon..l b:i\:,:;','J Channel, ,0n"1one, ~ PHo9'"'C CorbOlld Ullltl
B E Malo' I . ollon Iurfoee ~~7 Mu"I-lfore, lolld bod)'
Fig.lO.32A,8. Two of Allen's (1974b) qualitative models combination of autogenic channel avulsion and a cycle of
of alluvial stratigraphy. A Model 3, developed by autogenic dissection and aggradation, in which the base level
avulsion of the channel belt. B Model 58, developed by a changes over a ve rtical interval greater than channel depth
built on an analysis of the mechanics of channel-belt paleoslope. The model is quantitative, in that magni-
and floodplain construction and a systematic docu- tudes, rates, and scales are stated, and many of
mentation of the rates and scales of various the parameters are designed as variable input for
autogenic sedimentary processes. The main purpose successive computer runs to permit an exploration
of the model is to build cross sections of alluvial of the complex interaction between dependent and
stratigraphy in an orientation perpendicular to independent variables. Channel-belt sand bodies are
D
~
P>
g.
P>
~.
en
~
0.-
n; '
en
o
.....,
>-
~
:5.
e.
..,>-
n
...e:
(1)
n
~
..,
(1)
Fig. 10.33. Example of a simulated cross section of alluvial stratigraphy constructed using the numerical model of Bridge and Leeder (1979).
The cross section is lO km wide. Channel-belt aggradation rate is 20 m/ka. Channel width is 600 m and depth is 3 m
VJ
VJ
\C)
>I>-
'"
o
Ck.\::~Y·:'::·:·>·:~;<>I;·~:::·<::~:·(:~·.::..::::.:·::·.~~·::;~:·.::··::;.·[kk:···;·::.·.;·/Pb~:·:··:·'···· ·:·I:::::.:'.:-:·)+·;·:.·:.:··}'
'TI
:::.:.:};:;.:.;..:::: ::: .~~~.~...-:::'::::". ~ ,:: ,::' :::;'J [

e-
O
Fig. 10.34. Another model run from the Bridge and Leeder (1979) experiments, In this case channel units are set at 2 km wide and 7 m deep .g
oCI>
::;.'
0'
::l
e-
~E!
CI>
~
p..
~
~
n'
en
(1)
~
a
~
~
~
e..
CI>
assumed to be constructed by any appropriate that reduces the aggradation rate across the width of
fluvial style, following the hypothesis that the two the model cross section in proportion to the distance
major styles, meandering and braided [the "mobile from the channel belt. This reflects the fact that
channels" of Friend (1983)], both construct com- floodplain sediments are derived primarily from
plexes of bars and minor channels within relatively overbank flooding. Compaction of the accumulated
stable meander belts. This is certainly the case for deposits is accomplished, layer by layer, using preset
meandering systems, in which lateral migration is compaction factors. This process generates relief
constrained by the alluvial topography of levees on the floodplain surface because floodplain fines
and the resistance met by channels when they erode compact less than channel sands, and the least com-
laterally into the fine-grained deposits of abandoned paction occurs where channel sands are stacked ver-
channels and the floodplain. The assumption of a tically. The model therefore builds in a feedback
confined channel belt may not be so appropriate for effect, in that the site of the next lowest position for
braided systems. channel occupancy will be determined by the out-
The alluvial stratigraphy models discussed here come of the preceeding sedimentation-compaction
are autogenic, in that the external, allogenic vari- history. Two examples of Bridge and Leeder's (1979)
ables that control the long-term behavior of the river model are illustrated here (Figs. 10.33, 10.34). The
system are set at constant background values. Tec- most realistic of these is the first, in which sand body
tonism and sea-level change are not assumed to distribution appears to be relatively random after
affect the system, so that average accumulation rates the initial ten avulsion events. The second model
and regional paleoslope remain constant through- shows a much too regular stacking of the channel
out a model run. Climate change and its effect on bodies, an effect that occurs when channel width and
discharge, fluvial style, and sediment supply are also depth are set too large relative to aggradation rate.
not considered. Bridge and Leeder (1979) were The effects of the stacking and compaction of nu-
among the first to consider the quantitative effects of merous preceding events are downplayed in this
tectonism on fluvial aggradation, by building in the model, and successive channel positions are deter-
ability in the computer model to tilt the depositional mined primarily by the position of the two or three
surface in a direction perpendicular to paleoslope. immediately preceding channels. The result is a dis-
This aspect of the model is noted in Sect. 11.2.1. tribution of channel bodies not unlike that of bricks
The most important controlling factor in auto- in a wall. Nevertheless, the Bridge and Leeder (1979)
genic models of alluvial stratigraphy is the determi- model has been much used, particularly by petro-
nation of the rate and style of avulsion. As noted leum geologists, because of the insights it yields into
earlier in this chapter, little is known about the de- channel stacking patterns and interconnectedness,
tailed avulsion history of most rivers, and so some factors of considerable importance in the under-
generalizing assumptions have to be made. Leeder standing of reservoir predictability and fluid migra-
(1978) and Bridge and Leeder (1979) assumed in tion behavior. The model has recently been
their models that avulsion is not dependent on the expanded and updated by Bridge and Mackey
existing position of the channel within the model, (l993a), but the improvements are at a level of re-
but occurs sufficiently far upstream from the loca- finement that does not need to be addressed here.
tion of the model that the channel is free to move to The application of these model concepts to the
the current lowest position on the floodplain. Avul- interpretation of real stratigraphic data can be a
sion frequency was determined by reference to the complex problem. Willis (1993b,c) provided ex-
limited information on stratigraphic and historical tremely detailed architectural descriptions of the
data (Sect. 10.3), and the recurrence interval was Siwalik Group in a segment of the Himalayan
assumed to follow a distribution function similar to foreland basin in Pakistan, including quantitative
that of earthquakes, which in any location of tectonic studies of a 2-km-thick section keyed to the magne-
stress become more likely to occur as time passes. tostratigraphic time scale. Variations in sedimenta-
Mean avulsion periods ranging from III to 1780 tion rates, sand body thickness and sand body
years were used by Bridge and Leeder (1979). density cannot readily be interpreted using simple
Given an avulsion event, the channel is moved to autogenic models because of lack of knowledge of
the lowest point on the floodplain and allowed to the number of rivers in the basin and a lack of
accumulate a channel sand body at a preset accumu- detailed information on local tectonism.
lation rate, ranging from 5 to 40 m/ka. Floodplain Quite different approachs to modeling have
accretion is accomplished using a simple equation been taken by other researchers. Several modeling
body interconnectedness (e.g., Hirst et al. 1993; sev-

eral papers in Buller et al. 1990). The nature of the
a depositional process is not the primary concern in
these studies, and many of the models are simply
Wing. designed to simulate sand body geometries using a
(OY.rbanka)
stochastic approach, rather than building a stratig-
raphy layer by layer using a mainly deterministic
model based on actual depositional processes, which
is the approach used by Bridge and Leeder (1979).
The model of Hirst et al. (1993) generates three-
b \ dimensional hexahedra corresponding to channel
CIIann.. Belt
reaches and links these in chains by stochastic pro-
cesses to form channel belts (Fig. 10.35). The sizes
and shapes of the channel "boxes", the sinuosities of
the simulated reaches, the presence or absence of
channel bifurcations and oflevee and crevasse-splay
wedges are all determined by sampling from distri-
butions that are supplied as input. Channel belts are
Fig. 10.35. Simulated channel belts, built by a stochastic positioned in a simulated three-dimensional space
computer model. Each belt is composed of rows of
"boxes", with the size and shape of the boxes and the
in order to agree with sand body sizes, dimensions,
sinuosity implied by the nature of the contacts between and positions, as determined from well penetrations.
them all determined by stochastic sampling from built-in Channel belts are added until the observed sand-
sampling distributions. Crevasses and levees may be speci- stone/nonsand-stone ratio (the net/gross ratio) is
fied. (Hirst et al. 1993) achieved. The availability of a reliable data base for
the various parameters of the sand body architecture
is clearly essential for the running of this program.
programs have been developed by petroleum com- Some of the statistical relationships referred to in the
panies to generate realistic sand body architectures preceding sections can be used, or the modeler may
for partially known petroleum reservoir intervals, in wish to generate a unique data base built on available
order to aid in the determination of essential reser- know-ledge of the specific reservoir interval under
voir data, such as total sandstone volume and sand investigation.
Chapter 11
Tectonic Control of Fluvial Sedimentation
11.1 Introduction 11.2 Tectonic Control of Alluvial Stratigraphy
The major external, or allogenic, controls on fluvial It has been realized since ancient times that the
sedimentation are tectonics and climate. This chap- deposition of coarse-grained sediment in rivers is
ter considers tectonic controls; climatic controls are related to the presence of elevated source areas. As
discussed in Chap. 12. Base-level changes, driven noted in Sect. 2.2.1, a clear link between the grain size
either by tectonism or eustasy, are of great impor- of gravels and proximity to a mountainous source
tance in controlling fluvial style and sedimentation area was suggested by Playfair (1802). It has been a
rates in the lower reaches of rivers, through the standard interpretation, at least since the work of
control base level exerts on regional slope and sedi- Barrell (1917), that beds of coarse clastic deposits
mentary accommodation space. The role of eustasy close to a basin margin indicate tectonic rejuvena-
in controlling stratigraphic processes remains con- tion of that margin.
troversial and is described in Chap. 13, where the The links between tectonic movement and
main theoretical discussion of sequence stratigraphy coarse-clastic sedimentation have recently been re-
is presented. In this chapter changes in base level are examined by Jordan et al. (1988), Heller et al. (1988),
considered as relative changes, and their effects on and Blair and Bilodeau (1988). These researchers
stratigraphic processes in the fluvial environment have challenged the conventional idea that wedges of
are considered without entering into the tectonism- coarse sediment are syntectonic in origin, present-
versus-eustasy debate. ing several lines of argument to suggest that in many
There is a vast literature on the tectonic control of cases the most widespread coarse units form after
sedimentation (Sect. 11.2). In this chapter it is con- tectonism has ceased. The actual link between sedi-
sidered under the following subheadings. In Sect. mentation and tectonics is not questioned; it is not
11.2.1, the local effects of intrabasin fault and fold doubted that rejuvenated relief is required to gener-
movement are considered, including the effects of ate and transport coarse debris in large quantities.
rejuvenating fluvial systems and the diversion of What is questioned is the detailed timing of the
drainages. Relative changes in base level are dis- various events, relating to the speed of basinal and
cussed in Sect. 11.2.2. Regional, basin-scale effects depositional response to uplift and subsidence. This
are discussed in Sect. 11.3, including the differing topic is examined in Sect. 11.3. Here the discussion is
responses of depositional systems to the major types limited to a presentation of the immediate evidence
of tectonism. Particular attention is paid to the styles for tectonic influence in nonmarine basins - the
of nonmarine sedimentation that occur in exten- effects of changes in slope on grain size and dispersal
sional basins and foreland basins. Recent quantita- patterns. There are two main kinds of effects to
tive models for sediment dispersal are evaluated. be described: the effects of fault pulses at basin
Nonmarine sedimentation is then discussed within margins, and the growth of faults and folds within
the context of plate tectonics (Sect. 11.4), and, fi- basins. Many data on these topics from the 1960s and
nally, some generalizations regarding the tectonic 1970s were summarized by Miall (1978d) and
control of regional facies and dispersal patterns are Heward (1978a), and numerous case studies have
given (Sect. 11.5). been published since these papers appeared. A re-
344 Tectonic Control of Fluvial Sedimentation
search program initiated by M.R. Leeder in the 1980s (Fig. ILl). Tectonic slopes are produced by a combi-
set out to investigate the quantitative relationships nation of footwall uplift and hanging-wall subsi-
between geomorphic, sedimentary, and tectonic dence (Leeder and Gawthorpe 1987; Prosser 1993).
processes, and some of this work is referred to in this The footwall slope normally is steep, and typically is
chapter (see also his summary article: Leeder 1993). blanketed by coarse debris in the form of alluvial
It should not be forgotten that sediments are si- fans. The source area for these fans is small, because
multaneously affected by both autogenic and allo- it is limited by the size of the area comprising the
genic processes, and the allogenic processes footwall scarp, unless repeated uplift occurs over a
described here are overprinted on the autogenic very long time period and vigorous downcutting
variations in stratigraphy, such as those generated results in the tapping of source areas behind the
by geomorphic thresholds and lateral channel footwall scarp. The hanging-wall slope is gentler but
migration and avulsion, as described in Chap. 10. may tap a much larger sediment source area, result-
Recent work also suggests that changes in fluvial ing in the formation of larger depositional systems
facies, channel density, channel style, and other fea- depositing finer-grained sediment than those blan-
tures that have been attributed to tectonic mecha- keting the footwall scarp. Subsidence of the basin is
nisms in earlier studies may in fact be driven by typically asymmetric, with rotation occurring about
climate change. For example, in Sect. 12.12.2 it is a fulcrum point (Fig. U.l). In the case of basins
shown how changes from aggradation to degrada- bounded and underlain by listric faults, rotation
tional regimes in Texas river terraces were caused by against the fault results in the fulcrum point migrat-
climate changes associated with the late Cenozoic ing up the hanging-wall slope, generating strati-
glaciation and not by tectonism or base-level change. graphic onlap onto the basement.
G.A. Smith (1994) documented a complex record of The stratigraphy of the clastic units derived from
changes in fluvial style in a late Cenozoic extensional the marginal fault scarps depends on the balance
basin in Arizona at a time of tectonic quiescence, and between uplift and erosion rates on the basin mar-
was able to suggest climatic causes because of the gins and on that between subsidence and sediment-
availability of a proxy climatic record in the form of accumulation rates within the basin. Typically, the
oxygen and carbon isotopic signatures. marginal deposits form lenticular clastic wedges,
with the lateral extent of the wedge and vertical
changes in grain size through it recording changes in
11.2.1 The Effects of Syndepositional Fault the balance between these processes (Miall 1978d;
and Fold Movements Heward 1978a). These are large-scale, long-term
controls on sedimentation, relative to the autogenic
11.2.1.1 The Effects of Basin-Margin Faulting processes of avulsion and fan-trench evolution
described in Sect. 10.2. They represent processes
Extensional basins are commonly bounded by nor- occurring over time spans of 10L 106 years, corre-
mal faults. These may be steep faults that penetrate sponding to groups 8 and 9 of Table 9.1. Two styles of
to at least mid -crustal levels, or listric faults that may fan evolution may occur, fan segmentation and fan-
flatten out onto decollements at some shallower head trenching (Bull 1964; Hooke 1967). Segmented
level. Graben basins are bounded on both sides by fans are those in which the radial profIle shows two
faults, whereas half-graben basins are underlain by or more straight-line segments steepening toward
tilt blocks and have a bounding fault on one side the apex (Fig. 11.2A). Typically, each segment repre-
-ve dlaplacement vectors
HANGING WALL DIP SLOPE +ve displacement vectors
FOOTWALL SCARP
Fig. 11.1. Nomenclature for tectonic slope as-
HANGING WALL FOOTWALL sociated with a simple tilt-block, half-graben
basin configuration. (Leeder and Gawthorpe
1987)
Tectonic Control of Alluvial Stratigraphy 345
the fan deposits is the result. Between periods of

tectonic rejuvenation the rate of source-area erosion
may exceed the rate of isostatic rebound. Progressive
down cutting results in channel incision, leaving ear-
lier fan segments as terraces and depositing a new
cone of sediment as an offlapping wedge, with a
gentler depositional slope, at the distal end of the fan
(Fig. 11.2B). Significant fan-head trenches may be
formed in this way, although, as noted in Sect. 10.2,
they can also be formed by autogenic processes.
Progradation and retrogradation of the fan de-
posits, with corresponding vertical changes in grain
Fig. 11.2. Two alluvial fans in cross section, showing re- size, are a characteristic feature of basin-margin
sponse to changes in the balance between tectonism and alluvial-fan deposits and have been described by
sediment transfer. Numbers 1-4 are successive fan pro-
fIles, A-D are successive wedges offan sediment. a Uplift numerous researchers (see references in Miall
(to the right) more rapid than basin subsidence and sedi- 1978d; Heward 1978a; Leeder and Gawthorpe 1987;
mentation, causing fan segmentation, upward coarsening Blair and Bilodeau 1988; Prosser 1993). Some ex-
of the debris, and basinward progradation; b fan-head amples of typical stratigraphic configurations are
trenching, caused by relative decrease in the rate of tec- illustrated in Fig. 11.3 (see also Fig. 9.11). Heward
tonic rejuvenation. Older fan segments are left as terrace
remnants (unornamented areas of diagram), and fan sedi- (1978a) termed the resulting clastic wedges "mega-
ments become finer-grained with time. (MiallI978d) sequences", to distinguish them from the "se-
quences" of autogenic origin (the term sequence has
now, of course, assumed a different meaning).
Prosser (1993) recommended the erection of "tec-
tonic system tracts," based on the three-dimensional
architecture and composition of tectonically con-
trolled stratigraphic units. As described in Sect. 9.2,
allogenic cycles are commonly as much as 250 m
thick. Clast size and bed thickness commonly in-
crease upward through most or all of the section,
attesting to the increasing competence and capacity
of the fan streams.
The details of the tectonic geomorphology of
modern faulted tilt blocks and sedimentary basins
were considered in detail by Leeder et al. (1991) and
Leeder and Jackson (1993), based on the study of
tectonically active extensional basins in Greece and
the Basin and Range area of the United States. The
~ proximal molosse F::::~ olher foc ies
evolution of fault systems, the evolving geomorphol-
ogy of the uplifted fault scarps and drainage basins,
k}):ll dislOI molasse ~~:~'d local bosemenl
and the interaction between adjacent basins were
Fig. 11.3. Examples of alluvial-fan wedges banked against
considered in detail, and it was pointed out that
basin margins. a Simple faulted margin with continuous these modern systems can be used as analogues for
variations in width and grain-size of clastic wedge, reflect- the interpretation of economically important areas
ing both autogenic and allogenic processes; b,c retrograda- of rift-basin tectonics, such as the structural traps in
tion of fan deposits caused by backstepping of basin Mesozoic fluvial deposits in the northern North Sea
margin on successive faults. (MiallI978d)
Basin (Leeder and Gawthorpe 1987; Prosser 1993; see
also Chap. 15).
The basins that develop along strike-slip faults
sents the sedimentary response to a pulse of uplift in have faulted margins similar to those of rift basins.
the source area, resulting in steepening of the depo- However, the gradual lateral displacement of the
sitional slope and coarsening of the transported de- marginal clastic belts relative to their feeder streams
bris. Upward coarsening and lateral progradation of introduces some characteristic features. Nonmarine
strike-slip basins in California were discussed by Pyrenees (Anad6n et al. 1986; Puigdefabregas et al.
Crowell (1974a), and Crowell and Link (1982), and 1986; Burbank et al. 1992), the Andes (Jordan et al.
many examples of their structure and stratigraphy 1988), and the Himalayas (Burbank et al. 1986;
were given by Ballance and Reading (1980) and Burbank and Raynolds 1988), in the Canadian Arctic
Biddle and Christie-Blick (1985). A general discus- (MiaIl1991d), the modern Persian Gulf (Baltzer and
sion of this class of basin was provided by Miall Purser 1990), the Po basin, Italy (Ori 1993), Papua-
(1990, Sect. 9.3.3). A recent study by Mastalerz and New Guinea (Crook 1989), and in many other loca-
Wojewoda (1993) focused on the details of the effects tions (other references in Blair and Bilodeau 1988;
of strike-slip displacement on a modern alluvial fan Miall 1995b). However, In contrast to extensional
prograding across the basin-bounding strike-slip basins, where the downthrown block passively sub-
fault. The deposits contain small syndepositional sides, perhaps with a rotational component of move-
grabens and shear zones and show a complex pattern ment, in compressional basins the down thrown
of laterally offlapping depositional lobes, formed as block may undergo significant folding and uplift.
older fan deposits are structurally displaced along One of the most important effects of this local
the basin margin (Fig. 11.4). tectonism is the development in many foreland
Syntectonic to post-tectonic fanglomerate wed- basins of a distinctive class of structures termed
ges also characterize compressional basins bounded syndepositional or intraformational unconformities
by thrust faults. Well-described examples occur (Fig. 11.5). These were first described in detail using
worldwide, for example in the Rocky Mountain examples from the Ebro basin, Spain (Riba 1976;
foreland basin (Lawton 1986a,b; DeCelles et al. 1987, Anad6n et al. 1986; Puigdefabregas et al. 1986) but
1991), in the foreland basins flanking the Spanish have also been observed in many other foreland
basins, including the Indo-Gangetic basin (Johnson
and Vondra 1972), the Venetian Alps (Massari et al.
1993), and the Rocky Mountain foreland basin
(Spieker 1946; DeCelles et al. 1991). In present day
basins Recent alluvium may be tilted and incised by
modern streams. Fanglomerate deposited against a
rising (growing) fold or fault is uplifted, deformed,
and eroded by continued movement of the structure.
Younger deposits onlap the unconformity toward
the basin margin. These syndepositional structures
tend to be laterally restricted. The unconformities
described by Anad6n et al. (1986) typically die into
the basin within 500 m and extend along strike for as
012345 km little as 5-10 km.
The deepest part of a fault-controlled sedimen-
tary basin is typically located close to the bounding
J
-N-
fault. In the case of extensional basins it will be
immediately adjacent to the fault (Fig. ILl), whereas
T in the case of compressional basins the trough may
be located a short distance out from the bounding
_ 2 fault, corresponding to the axis of the syndeposi-
tional fold generated in the footwall. The position of
=3
this axial basin and its transverse and longitudinal
slopes will vary depending on the progress of defor-
~5 mation on the basin-margin structures, and this on-
going deformation has an important effect on the
D evolution of river systems flowing through the basin.
Commonly, the axial trough is occupied by a river
Fig. 1l.4A-D. Evolution of an alluvial fan during left- flowing parallel to the basin margin. This will
lateral strike-slip displacement along the basin-margin
bounding fault. 1 Main basin-margin fault; 2 sediment
migrate laterally or undergo avulsive switches in
dispersal; 3 graben; 4 shear zones; 5 listric faults. position in response to pulses of fault movement.
(Mastalerz and Wojewoda 1993) Alexander and Leeder (1987) and Alexander et al.
TYPES OF SYNTECTONIC
TECTONIC SETTING AND PROCESSES LOCAL RESPONSE
INTRAFORMATIONAL UNCONFORMITIES
SOUTHERN PYRENEAN FORMATION OF ASSEMBLED COMPOSITE
~~
THRUST BELT A LATE FORELAND PROGRESSIVE
BASIN MARGIN UNCONFORMITIES
lete phases of nappe
ALONG THE
em.placement
;r~. (Floor thrust emergence)
NAPPE FRONT
n~
_- -:.r-', -- 0 CATALAN ~OASTAL RANGE ISOLATED COMPOSITE
0 STRIKE-SLIP SYSTEM
FORMATION
PROGRESSIVE
UNCONFORMITY
;/ / Uplifting of push-up
I W/f structures OF NEW
(i)
BASIN
CD (i) 'If
/ CATALAN COASTAL RANGE
SHIFTING COMPOSITE
STRIKE-SLIP SYSTEM
OF THE MARGINS
PROGRESSIVE
BASIN UNCONFORMITIES
Uplifting of complex
MARGIN
push-up structures
IV
~
LINKING ZONE BETWEEN SYNTECTONIC SUPRA TENUOUS
IBERIAN AND C.C. RANGES RESTRICTION OF SYNCLINES
CD_ ---CD DEPOCENTERS.
LATER SPREADING
AND RELATED
COMPOSITE
~
Synsedimentary folding OF SEDIMENTARY PROGRESSIVE
and thrusting AREAS. UNCONFORMITIES
Fig. 11.5. A classification of syndepositionaJ (intraformational) unconformities in the Ebro Basin, Spain, showing the
range of tectonic processes and resulting structural styles. (Anad6n et aI. 1986)
(1994) documented this effect in the case of the Leeder (1979) model. Atkinson (1986) attributed this
Mississippi River, where it crosses the New Madrid contrast to the fact that the simulation studies model
earthquake zone, and in the case of the South Fork the evolution of single rivers, whereas in real basins
Madison River in Montana, which has been affected (such as the one he described), more than one river
by tilting of a half graben in a direction tranverse to may be present. High channel densities in the Span-
the river (Fig. 11.6). They pointed out that the tec- ish example are attributed to the activity of several
tonic slope generated by individual earthquake epi- distributaries of a fan system. Other local complica-
sodes may be comparable in magnitude to the tions are described by Alexander et al. (1994).
longitudinal slope of the river, so that significant In addition to the shifts in meander pattern illus-
immediate effects on drainage may be expected. trated in Fig. 11.6, Alexander and Leeder (1987) and
Todd and Went (1991) and Olsen and Larsen (1993) Alexander et al. (1994) pointed to other significant
described possible ancient examples of the lateral effects on the alluvial architecture that could be ex-
migration of rivers in response to tectonic tilting. pected to be caused by tilting. For example, soils are
The effect has also been modeled by Bridge and likely to be more mature on the updip flank of the
Leeder (1979). As discussed in Sect. 10.4.3, their tilted basin, because such areas receive less clastic
computer model of avulsion selects the lowest posi- influx to dilute pedogenic processes and change the
tion on the flooplain for each successive channel position of the water table.
position. In the case of a lateral tectonic tilt the It was noted by Alexander and Leeder (1987) and
model automatically tends to shift channels down- Leeder and Alexander (1987) that changes in river
slope, much as has been actually observed by courses in general and meander patterns in parti-
Alexander and Leeder (1987). However, in real ba- cular could be predicted from a knowledge of the
sins the alluvial architecture may not always follow tectonic slopes developed by fault movement, and
this pattern. Atkinson (1986) documented a case in they developed this idea to explain the changes
the Tremp-Graus Basin, Spain, where channel densi- in meander pattern of the Mississippi River in the
ties are highest in areas of most rapid subsidence, the New Madrid area in response to the earthquakes of
opposite of the effect predicted by the Bridge and 1811-1812. Idealized river flow lines are drawn
bluff bounding floodplain

c.-
'7{/ meander scars
: ....
............. ..-:
,
-.--
...... -.-.
. ~ . - . -.: -.- . -.-.-.
lkm
Fig. 11.6. The meander belt of the South Fork Madison Pleistocene meander belt (Pinedale belt) on the updip
River in SW Montana, showing the asymmetry of meander flank of the modern meander belt, and the preferential
patterns and relict floodplain deposits. The river has preservation of meander scars that are convex to the
gradually migrated northeast in response to a tectonic southwest, as the river migrated laterally away from them
tilting in that direction, and the river now flows against the and eroded the meanders facing down structural dip.
basin-margin fault. Note the presence of an abandoned (Alexander and Leeder 1987)
perpendicular to the contours of the tectonic slope, a Zagros fold-thrust belt, which forms the border to
process that can be simulated by computer (Weston the Mesopotamian foreland basin, and Rasanen et al.
and Alexander 1993). Alexander (1986) combined (1987) discussed the postglacial adjustment of the
information on syntectonic fault movement and Amazon system in the Andean foreland basin of
paleochannel trends derived from outcrop paleo- Brazil to tectonic movements. Changes in the magni-
current data to project the distribution of sand bod- tude of the basement slope lead to changes in river
ies in the subsurface behind cliff exposures of a style (Schumm 1993; Wescott 1993), and changes in
Middle Jurassic sand body on the Yorkshire coast, the orientation of the slope cause changes in flow
England. direction, including reversals in flow, lateral shifts
in meander belts, and river capture. Numerous
neotectonic observations have been made of these
11.2.1.2 The Effects of Faulting and Folding Within Basins features. In particular, the growth of folds within
basins, such as the anticlines that develop over blind
Rivers are sensitive to external forces, such as tec- thrust faults, can be mapped on the basis of changes
tonic subsidence, uplift, and tilting. In tectonically in the drainage system at the surface. For example, in
active areas the drainage patterns may be directly Iraq, ancient drainage canals have been deformed by
related to the patterns of subsidence and uplift gen- the growth of anticlines beneath them (Lees 1955). In
erated by growing structures. Baltzer and Purser central Asia, growing anticlines are bringing about
(1990) showed that this is the case adjacent to the diversions in drainage patterns (Trifonov 1978). Nu-
merous changes in river patterns in the Hungarian local sedimentation rate that affected soil develop-
Plains were attributed to tectonic tilting by Mike ment were attributed to neotectonic tilting by
(1975). Several studies have examined the changes in Srivastava et al. (1994).
fluvial morphology in response to modern tectonic Blind structures may eventually penetrate the
movements, including changes in longitudinal pro- land surface and evolve into topographic barriers
fIle, terrace patterns, channel sinuosity, meander and sediment sources. Such patterns of development
shape and orientation, bar character, and lateral are particularly common in foreland basins (Morley
channel migration (Seeber and Gornitz 1983; 1986), where they may accompany the development
Burnett and Schumm 1983; Ouchi 1985; Leeder and of satellite or piggyback basins (Sect. 11.4.6).
Alexander 1987; Riisiinen et al. 1987; Marple and Ouchi (1985) reported on a series of experiments
Talwani 1993). ModifIcations in fluvial style in the to simulate the effects of anticlinal and synclinal
plains of the Ganges River in India and changes in folding across a river's longitudinal proflle and
Slope deformation River adjustment

c c
a. .2 0 c
~om./+l
Uplift
Pattern '\
,~~~ fIA
~~/d
bor- bra ided
~I
teorroces bar- braided
Or
meanderi ng -talweg
braided
c
o
c
b. a 0
~ c; c
Subsidence
profile/ill
Pattern j.. 1 ~\
~~ ~:;~1
alternate
bars
be r -
braided
I,:
transverse
bars
~~ Fig. 11.7. Adjustment of a braided river to (a)
anticlinal uplift and (b) synclinal subsidence
across it. (Based on experimental work by Ouchi
1985)
a. Uplift il' b. Subsidence f

Slope deformation Slope deformation
Channel pallern Channel pallern
and ad'ustm.mt and adjustment
f
'----0---..- ~
.i'V"
~ ret~
--------
~ ~Ie
f
-cutaff-
.
-cutoff-
", ....
~ ~ ,~
~ : f..,. f 7'.~ Fig. 11.8. Adjustment of a mixed-load me-
:sinuous - or sinuous- or
iSland- anderingriver to (a) anticlinal uplift and (b)
islond- braided
synclinal subsidence across it. Four time
~ braided
~
f\.V' ~
steps are shown, in order from top to bot-
tom. (Based on experimental work by Ouchi
t,
• 1985)
7'" 30' E
KOHAT PLATEAU
~.
33" 00 N
o 10
,
km
KOHAT PLATEAU
~:~~i~,s' •
Shakardarra 0
tilted
block
KALABAGH '
L -_ _ _~ FAU LT .-.. hrnge zone
--=-? ~ to...,:S::O=U:,:th:..-_--
Fig. 11.9. Map and block diagram of part of the Himalayan along the Kalabagh Fault is the main cause of this move-
frontal fault system, Kohat Plateau, Pakistan, showing the ment. Terrace rem ants in the northwest corner ofthis area
eastward migration of the Indus River in response to tilt- represent former Indus gravels now elevated to as much as
ing and uplift of the alluvial plain. Transpressional uplift 630 m above their former position. (McDougall 1989)
described some actual examples of these changes this work are summarized in Figs. 11.7 and 1l.8. In
based on detailed mapping at various locations in summary, changes in slope imposed on the longitu-
the United States, where cratonic warping at rates of dinal profile of a river are compensated initially by
up to 1 em/year has been recorded. Burnett and changes in sinuosity, so that the along-channel slope
Schumm (1983) described additional examples from remains constant. Thus, an increase in slope leads to
the United States. Some of the major conclusions of an increase in sinuosity, and vice versa. There are
limits to the ability of rivers to change in this way. At Earlier changes in the course of the Indus River in
high sinuosities the incidence of chute and neck the Potwar plateau area of Pakistan were recon-
cutoffs increases (reducing the sinuosity again), and structed by Mulder and Burbank (1993), based on an
bank erosion and incision ensue. Reduction in slope extensive synthesis of facies, petrographic, and
may lead to meandering rivers becoming flooded paleocurrent data within a stratigraphic framework
and developing anastomosed patterns, as was pre- that has been rigorously dated using magneto-
dicted by Smith and Smith (1980). At synclinal and stratigraphy. Several studies have shown that the
anticlinal axes, respectively, aggradation and degra- Indus River flowed eastward into the Ganges system
dation occur. These ideas are of considerable impor- prior to about 5 Ma (Fig. 11.10). Its deposits are
tance in connection with the effects of base-level characterized by a distinctive white sandstone in the
change on rivers and are discussed further in Sect. form of channel bodies bearing a characteristic vari-
11.2.2. ety of hornblende in the heavy-mineral suite. These
Recent applications of these ideas include beds are interbedded with a variegated assemblage
the work of Marple and Talwani (1993), who of brown sandstones showing much greater litho-
employed all these indicators in a study of neo- logic variablity and generally smaller-scale cross-
tectonic deformation of the coastal plain of South bed structures and thinner channel units. The brown
Carolina. Upstream from a region of uplift, rivers units are interpreted as the deposits of smaller
are anastomosed and swampy. In addition, some streams draining uplifts of the Himalayas and
rivers flow along one side of their floodplains, sug- the Salt Ranges to the south of the Potwar Plateau.
gesting lateral migration due to tectonic tilting. The structural emergence of the Salt Ranges can
Mather (1993) documented the uplift of a small be determined on the basis of the appearance of
fluvial basin in Spain, which was undergoing ac- brown sandstone units draining the emerging up-
tive alluvial-fan sedimentation in the Pliocene and lift and accompanying shifts in paleocurrent pat-
now is characterized by incised streams, with the terns and, later, by the appearance of a distinctive
courses of the fan feeder channels truncated by river granite clast type known only from Salt Range
capture. sources. Initially, the uplift of the Salt Ranges shifted
These observations based on neotectonic and ex- the Indus River system northward, and continued
perimental studies and observations of the Recent uplift then deflected it westward to near its present
may be applied to analyses of the more ancient course (Fig. 11.10). Minor drainages of the Jhelum
record, where examples abound of the effects of system captured the runoff from the emerging struc-
syndepositional, within-basin tectonic deformation. tures.
Some of the most spectacular examples occur in the This picture of tectonic control of fluvial paleo-
Himalayan foreland basin of India and Pakistan, geography, and the sedimentological evidence from
where fluvial deposits have been accumulating con- which it can be reconstructed, is typical of many
tinuously since the mid-Cenozoic, and major trunk basins marginal to fold-thrust belts, particularly
rivers such as the Indus, Ganges, and Brahmaputra "piggyback" or "satellite" basins, as described in
have occupied the axis of the basin for millions of Sect. 11.3.
years. These· rivers show many changes in channel The development of coal is sensitive to tectonic
patterns (e.g., Holmes 1968; Wells and Dorr 1987), setting. Given the appropriate climate for peat for-
at least some of which are related to fault move- mation, the thickness and quality of a peat unit
ment and block tilting within the sedimentary basin depend on subsidence rate and clastic influx.
(Morgan and McIntire 1959; MCDougall 1989; Titheridge (1993) described the effects of faulting
Srivastava et al. 1994). For example, the Indus River and consequent local basinal tilting on the distri-
has been diverted eastward a lateral distance of 10 bution and thickness of coals. He showed that the
km along the western side of the Potwar Plateau by slightly increased subsidence rate on the down-
uplift on a major transpressional fault during the thrown block of a half-graben leads to an increase in
past 100-500 ka. The former course ofthe river can seam thickness and also to a greater frequency of
be traced across uplifted fault blocks, where terrace seam splitting. The difference between true low-ash
remants are now as much as 630 m above their coal and carbonaceous shale depends on the clastic
former elevation. Lateral migration of the Indus laid influx into the coal swamp. McCabe (1984) demon-
down a sheet of gravel which is now being drained by strated that low-ash coal may reflect the develop-
small consequent streams flowing across the newly ment of raised peat islands, which deflect drainage
tilted block (Fig. 11.9). channels and their contained clastic influx away
A in Utah and Arizona may be a case in point. Peterson

suggested that very gentle structures mapped in a
Jurassic fluvial system were active during sedimen-
tation and controlled sediment thicknesses, paleo-
current patterns, river sinuosities, and the location
of interfluviallacustrine deposits. Heller et al. (1993)
suggested that these structures are part of a suite of
North similar structures in the Western Interior Basin de-
veloped by intraplate stress (see below). Most of the
supposed stratigraphic and facies relationships are
not obvious to this writer, except that one major
"basin", the Henry Basin (which is actually crossed
by several of the minor anticlines), is clearly located
at a confluence between two major fluvial systems
and shows a thicker section and higher channel sinu-
osities (based on paleocurrent variability) than adja-
cent areas.
11.2.2 Base-Level Changes
There are several alternative definitions of base

level. Schumm (1993, p. 279) stated: "Base level is
the imaginary horizontal level or surface to which
sub-aerial erosion proceeds. It is sea leveL" In their
o 50 review of nonmarine sedimentation in relation
Km to sea-level change, Shanley and McCabe (1994)
pointed out that sea level is not always the critical
Fig.ll.IOA-C. Paleogeographic reconstruction of the surface in determining the level to which rivers tend
Potwar Plateau area, Pakistan, for A 6.0 Ma; B 5.2 Ma; and
C 4.8 Ma. Dashed line marks position of basement normal to erode. In inland basins, lake level, or the level of a
fault used as a reference frame. Dotted area shows position tectonic barrier, may be the practical base level. In
of upthrust Salt Ranges. (Mulder and Burbank 1993) eolian basins the surface of deflation, related to the
level of the groundwater table, may be the local base
level for fluvial processes. Shanley and McCabe
from the peat swamp. Wise et al. (1991) and Belt (1994) proposed the adoption of the concept of the
(1993) argued that, on a local scale, basinal tectonics stratigraphic base level as encompassing all these
may also be a factor, with the growth of anticlinal varied controls on fluvial sedimentation.
structures over blind faults acting to dam or divert In coastal plains, base level may change as a result
clastic drainage. They attempted to explain rapid of tectonic movements of the basin or as a result of
(0.5 million years time scale) vertical changes be- eustatic changes of sea level. Changes in sea level and
tween clastic and coal-bearing intervals in proximal their effects on stratigraphy have, of course, come to
parts of the Appalachian foreland basin by this form the basis for the subject of sequence stratigra-
mechanism. Belt (1993) suggested that differential phy and are therefore of paramount importance to
uplift and subsidence during the evolution of the sedimentary geologists (Chap. 13). However, the na-
Laramide basins in the western United States could ture of the fluvial response to base level change is a
explain drainage-pattern changes in the Williston subject that has been bedeviled by controversy and
Basin and the development of coal swamps there on misunderstanding since the nineteenth century (see
a similarly rapid time scale. Schumm 1993for a brief review). Much that has been
Because the effects of tectonism on fluvial style written recently about the effects of base-level
and paleogeography can be very subtle, it may be change is the work of sequence stratigraphers, or of
possible to read into the geological record more than geomorphologists explaining their science in re-
can be conclusively demonstrated. Peterson's (1984) sponse to developments in sequence stratigraphy.
study of tectonic control of cratonic sedimentation This section therefore involves a consideration of
some of the older sequence literature, in order that a (1988). For this reason, it is important that the earlier
clearer understanding of the present state of under- models be described briefly and the problems with
standing can be reached. them mentioned. As noted below, H. Posamentier
Over geological lengths of time, rivers develop a and his co-workers do not now use the part of the
dynamic equilibrium in the form of a graded longi- earlier work that is now described.
tudinal profile (some of the historical steps in the The basis for the model of Posamentier et al.
emergence of this idea are summarized in Sect. (1988) and Posamentier and Vail (1988) is illustrated
2.3.3). Butcher (1990, p. 376) stated: in Fig. 11.11, which is a variation on a diagram pub-
"An equilibrium profile, or graded stream, is one in which lished by Lane (1955), as shown in Fig. ll.l2. The
slope, velocity, depth, width, roughness, pattern, and shift in point of grade from 0 to F is thought to occur
channel morphology delicately and mutually adjust to during the time of maximum rate of sea-level fall,
provide the power and efficiency necessary to tranport the when the equilibrium point in the basin (the point
load supplied from the drainage basin without aggrada- where vertical movements due to eustasy and
tion or degradation of the channels (Leopold and Bull
1979). Rivers that are out of equilibrium will aggrade or tectonism are in balance) moves seaward across
incise in an attempt to achieve a graded profile which ... is the continental shelf. Two observations of modern
a function of tectonic, isostatic, climatic variation ... and rivers which Posamentier et al. (1988) described in
base-level controls." support of this model included the effect of the lat-
Isolating the effects of base-level changes is diffi- eral migration of a modern river meander and the
cult, because in many naturally occurring cases filling of a reservoir. In the case of the meander
changes in base level are linked to other effects. For migration, the junction of the river with a tributary
example, the major base-level changes that affected shifted across the river valley as the meander shifted,
the world's coasts during the late Cenozoic are and this created accommodation space within the
glacio eustatic in origin, and the effects on the rivers tributary, to which the tributary responded by ag-
and their deposits of actual changes in base level are grading. This effect was noticed because the aggra-
difficult to separate from changes in discharge and dation resulted in deposition taking place over
sediment load that occurred at the same time railroad tracks that run along the banks of the main
(Butcher 1990; Schumm 1993; Leckie 1994). Experi- river valley (Posamentier et al. 1988, Fig. 12, p. 133).
ments by Begin et al. (1980) and Begin (1981) in In the second case, a river aggraded in its lower
modeling base-level changes in a flume have been course as it built a coastal deltaic prism across a
useful in clarifying geomorphic concepts. Recent reservoir (their Fig. 13, p. 134).
work by Blum (1994) and Blum and Price (1994) has However, there are three problems with the
demonstrated that climate change may be much model of Po samentier et al. (1988):
more important than base-level change in affecting 1. No account was taken of the rate of change. How
the balance between aggradation and degradation, fast is the shift of the point of grade from 0 to F,
because of the effects of climate change on discharge relative to the ability of the stream to adjust? The
and sediment yield. This topic is dicussed at greater family of curves AF, BF, CF, and DF is only one set
length in Sect. 12.12.2. of possibilities and represents a situation where
Posamentier et al. (1988, p. 135) stated: the river is responding only to changes at the
"The fundamental principle which governs sedimentation mouth by a backfilling process, ignoring up-
in fluvial environments (assuming constant sediment sup- stream influences. Another possibility is the cre-
ply) is that a stream will aggrade if its equilibrium profile
shifts basinward or upward and incise if its equilibrium
profile shifts downward ... subaerial accommodation
occurs by basinward shifts of the stream equilibrium pro-
file."
This statement is quite correct, but there has been
debate regarding the causes and controls on the shift o
in fluvial profiles (MiallI991c; Schumm 1993). The Fig. 11.11. Fluvial aggradation due to shift of stream equi-
book in which the statement appears (Society of librium profile from 0 (original point of grade), with ini-
Economic Paleontologists and Mineralogists Special tial curve AO, to F (final point of grade). According to
Posamentier et al. (1988) the river aggrades in response to
Publication 42) is still regarded as an authorita- this shift, developing in turn profiles BF, CF, and DF (let-
tive source, containing two landmark papers. by ters added to original illustration of Posamentier et al.
Posamentier et al. (1988) and Posamentier and Vail 1988, Fig. 15). (Diagram from Miall1991c)
in base level (the initial filling of the reservoir in

a)
their Fig. 13 is not at issue here).
Baselevel2 In another diagram, Posamentier and Vail (1988,

Baselevel 1
Fig. 8) showed fluvial coastal-plain progradation
and coastal onlap occurring during the late-high-
b) stand phase of the eustatic sea -level cycle, during the
time of maximum rate of fall (times T4-T6). They
Baselevel 1
showed the equilibrium point migrating seaward,
the bayline and fluvial profile shifting seaward, and
Baselevel2 the coastal plain expanding landward as a result of
onlap. Progradation is shown by the conventional
c) rising zig-zag facies contact in the deposited sedi-
ments. Landward of the equilibrium point is a zone
of relative sea-level fall (this is what the equilibrium
point means, as explained in the definition of this
term by Posamentier et al. 1988, p. 116). These dia-
Baselevel 1
Baselevel 2 grams are not consistent with some of the text in the
two companion papers in this volume, which discuss
Fig. 11.12. Movement of stream profile in response to
changes in base level. Note the increase in sedimentary the tendency for fluvial incision during late-
accommodation space (upward movement of equilibrium highstand, relative to sea-level fall. Figures 8 and 15
profile) generated by a horizontal shift in the mouth of the of Posamentier and Vail (1988) clearly indicate sig-
stream in diagram (c). (Schumm 1993, after Lane 1955) nificant late-highstand fluvial deposition, coastal
progradation, and coastal onlap during a time of
relative sea-level fall (as do several figures in Jervey
ation of a series of curves parallel to and above the 1988), yet Posamentier and James (1993, p. 15) and
original curve AO, as the point of grade shifts Weimer and Posamentier (1993, p. 10) do not accept
seaward slowly enough for the river to grade itself that this is what had been described in the earlier
throughout. Much would depend on the rate of studies. Careful examination of their Fig. 8 indicates
sediment supply. that the late-highstand fluvial aggradation and
2. It is not made clear how far upstream these pro- coastal onlap indicated in this diagram have been
files are intended to represent. The shift of the created in two ways. First, deposition has been al-
upper part of the profile from B to D ignores lowed beneath the "tectonic hinge", which cannot
upstream controls. Projection of these profiles occur if the hinge is the stable point about which the
upward takes them back to the stream sources. continental margin undergoes rotational subsid-
The diagram seems to suggest that the response of ence. Second, most of the accommodation has been
a base-level fall is for the mountainous source created by rotational subsidence at a rate compa-
area to rise upward and/or move forward (sea- rable to the eustatic fall in sea level. The equilibrium
ward) whereas, of course, it may be undergoing profile is not shifted horizontally, but is extended
active degradation. There is, in any case, no rea- obliquely downward in a seaward direction.
son why a change in base level should be linked to In retrospect, it is difficult to ascertain the basis
changes in the source area. Given slow tectonism for these models, such as that reproduced in
(as on the passive margin assumed for these mod- Fig. 11.11. Many of the architectural ideas of
els by Posamentier and his co-workers), the upper Posamentier et al. (1988) are based on the computer
parts of the later curves (C, D) should drop below simulations ofJervey (1988), which have led to sche-
B, not climb above it, leading to an overall de- matic models of continental-margin configuration
crease in slope. that are difficult to interpret because they lack scales,
3. The shift from ° to F is drawn as a horizontal and therefore cannot show realistic slopes. Rates of
shift, whereas the whole point of the discussion is subsidence and sea-level change are shown as having
vertical changes in base level. This error follows the same magnitude, which is commonly not the
from the use of the two particular examples by case. The late-highstand fluvial onlap model seems
Posamentier et al. (1988), quoted above. In nei- to have developed from an attempt by Vail and his
ther of these examples was there a vertical change co-workers to reinterpret their "sawtooth" coastal-
TERRIGENruS INR.UX - Sf}, IMt

Fig. 11.13. Shifting of facies boundaries
during eustatic rise and fall. Note the regres-
sion during sea-level fall, and the seaward
I extension of the coastal plain. (Vail et al.
I 1977, reprinted by permission)
I
Iz
IS
t .....
INITIAl
DEPOSITIONAl
I~ R8.ATM SURFACE
I ~ RISE OF
... SEA I.fVB. ......
'a I ..........
lll!j
8 NO OffiRENTW. SUBSIDENCE
......................... .....
I IIID[ ......
: oL-.....l ..................
10 MI OOASTAI. ENCfl(JA()HENT ......
~---------------------------
UNCONFORMITY TYPel ruSTATIC SEA LEVEl

SUBAERIAl - SUBMARINE !TYPE 11 sea-level fall at the same time, in the same strati-
HIGH LOW
t'§§iiiiiJ -~-a1d~i iDElI F~IA~T: ~j~Lf~OGE

graphic units. Onlap is shown at the left in each
panel, where the stratigraphic units terminate
W
against a steeply dipping basement surface (this is
SUBAERIAl /TYPE 2l not a faulted contact). At the same time facies belts
are shown migrating seaward, culminating in
., ~ LESS THAN RATE OF SUBSIOENCl

AT SHELF EDGE
the development of an unconformity, indicating re-
gression and relative sea-level fall. It seems possible
that the observations of small-scale systems by
Posamentier et al. (1988) referred to above suggested
the possibility for this late-highstand regression, and
RAPID RISE
FOLLOWED BY STIlSTAHO the early model of Vail et al. (1977) (Fig. 11.13) was
ANO 5I.DW FAU.
ready at hand. In Posamentier and Vail (1988), Fig. 8
~ AU.UVlAl PLAIN 0 MEAllSHORE shows a small segment of a "sawtooth" coastal onlap
GCOASTr.t. PlAlN EO] fINE- _ED MlRIiE
curve in which late-highstand alluvial onlap is
Fig. 11.14. Definition of type 1 and type 2 unconformities. clearly indicated as the reason for the sharp tip of the
(Vail et ai. 1984, reprinted by permission) sawtooth.
As discussed by Miall (1991c), a drop in sea level
would take the shoreline out and down across the
onlap curves in order to arrive at symmetrical curves continental shelf. The slope of this additional stream
for eustatic sea-level change. The original sawtooth course may be comparable to that of the lower
shape aroused considerable controversy (e.g., Pit- reaches of the original curve BO, but this would
man 1978), and their early interpretive diagrams depend on local, actual continental margin configu-
contain the seed of a possible reinterpretation that is ration (Fig. ILlS). A drop in base level would there-
more fully explored in the papers by Vail and Todd fore not necessarily shift the graded curve, but might
(1981) and Vail et al. (1984). Figure 11.13 is repro- simply extend it. It should not be forgotten that a
duced from Vail et al. (1977) and shows their early newly exposed shelf might add new drainage area to
ideas regarding regression of the shoreline during a river, thus adding its discharge and tending to
sea-level fall. Vail and Todd (1981, pp. 221-223, Fig. lower the slope.
2) incorporated these ideas into their methods for Butcher (1990) summarized some long-known
deriving sea-level change from measurements of ideas regarding the response of river profiles to base-
coastal onlap and showed the same change from level change, and reminded geologists of the concept
transgression to regression of the coastal zone to- of the nickpoint (Fig. 11.16). In most cases, a fall in
wards the end of a cycle of eustatic rise and fall. sea level leads to incision and to the headward mi-
Figure 11.14 is the definition diagram for type-l and gration of a nickpoint. The steepened lower reaches
type-2 unconformities (from Vail et al. 1984) and of the river may change in fluvial style as the river
shows what this writer regards as an impossible situ- cuts down to its new equilibrium position, and rem-
ation: the stratigraphic effects of sea-level rise and nants of the earlier course of the river are left as
1750m Fig. 11.15. Examples of continental margin

1500
1250
profIles, showing the shapes of slopes that
1000 would be added to river profIles if sea level fell,
750 exposing additional areas of the continental
500
Z50
shelf. ProfIles of Babbage and Firth rivers,
coost Yukon, from McDonald and Lewis (1973); pro-
0
0 ZO 40 60 80 100 IZO 140 160 180 ZOO 220 -250 fIles of other rivers from Leopold et al. (1964).
km -500 ProfIles of shelves from Times Atlas and other
70m sources. (MiallI991c)
60
50 SOm
40 40
Lower Indus
30 30
20 20
Murray
10 10
cooS! SN level coost sea level
a a
200 400 600 800 -10 200 400 600 800 -10
km -20 km -20
-30 -30
-40 -40
1000m 1000m
Ria Grande
500 500
NI~e
sea level coost sea level
a a
1000 1500 500 1000 1500 2000 2500
km -300 km -300
Current Profile
Pitman and Golovchenko (1988) discussed a hypo-
Nickpoint Location at Present thetical case which might, in fact, correspond to sea-
Equilibrium Profile (Tl) level change in the way described in the Exxon
1 models:
,:I:---;~r--~- _ Sea Lsvel (n)
"Let us consider a coastal region in which a large graded
'21--=~i!!!Oi;~~ __ Sea Level (T2)
fluvial system enters the ocean at the coastline and in
which the littoral zone has been transgressing slowly for a
long period of time. In this case it is likely that the graded
Fig. 11.16. The nickpoint concept. The drop in sea level slope of the shelf will be less than the gradient of the lower
from T1 to T2 causes steepening of the lower slope of this course of the fluvial system that is being progressively
river. Equilibrium is restored by the development of a transgressed. In that case, if there is a regression exposing
steeper profIle of the lower reaches of the river, which joins the shelf, subaerial deposition will take place. Because the
the original profile at a break in slope termed the exposed shelf is at a gentler slope than the lower course of
nickpoint. This nickpoint migrates upstream as a result of the fluvial system, the rivers crossing the exposed shelf
erosion of the steepened parts of the profile, and may must slow down and hence will deposit a part of their load.
eventually affect the entire course of the river. (Butcher These will be entirely nonmarine coastal deposits. This
1990) coastal onlap will continue during regression."
Their illustration of this process is shown in Fig.
terraces on the valley margins. The response of 11.17. As Butcher (1990) pointed out, the scenario
the river to imposed changes in slope in terms of envisaged by Pitman and Golovchenko (1988) is less
changes in fluvial style is complex, as discussed be- likely to occur than one where the lower course of the
low. Rivers that have undergone such changes re- river has a lower slope than the continental shelf.
cently in response to glacio eustatic changes in sea This is especially likely to be the case where rivers
level reveal complex cut-and-fill stratigraphies re- have had an extended time period to build a coastal
flecting the integrated response to changes in base plain - the very situation most likely to be of interest
level and climate (Sect. 12.12.2). to geologists because of the high preservation poten-
Few workers have considered the actual slopes of tial of such accumulations on slowly subsiding conti-
continental margins and rivers. Pitman (1986) and nental margins. In order to test these points, Miall
Table 11.1. Slopes of rivers and shelves. (Miall1991c)

Location Distance Slope
averaged
(km)"
SEA LEVEL
Rivers:
- -=-=~==~-=-~-~=-=;::~ ~ Mississippi 650 0.000014
,:, , Indus 500 0.00008
.... ...
0.0001
~.
"' .' Nile 1000

Guadalupe (Texas) 0.00025
Rio Grande 500 0.0004
b Babbage (Yukon) 100 0.0008
Firth (Yukon) 100 0.0036
Shelves:
Guadalupe (Texas) 8 0.00023
Off Mississippi R 60 0.0003
South Carolina 75 0.00067
N. Slope, Alaska 65 0.0009
Murray 190 0.0011
N. Slope, Yukon 115 0.0017
Rio Grande 80 0.0026
Nile 20 0.005
Indus 40 0.005
Mississippi delta 20 0.0095
SEA LEVEL
l! 53
"Distance from coast to nearest available contour on shelf,
or length of lowermost, flattest portion of river profIle,
54 measured from mouth.
e ment of accommodation space is where a relatively

small, steep river flows out into a coastal plain
flanked by a broad shelf. For example, if the three
Q COASTAL SEDIMENTS Yukon rivers (Babbage, Firth, Malcolm) were to be
t-=-g MARINE SEDI MENTS extended by sea-level fall across a shelf similar to
that of the Gulf of Mexico between the Mississippi
Fig. 11.17. Model of a fall in sea level from time 53 Ma (a), and the Guadalupe rivers, the newly extended lower
to 49 Ma (e). If the slope of the coastal plain is steeper than courses of these rivers would be at a slightly lower
the slope of the shelf, regression and onlap will occur.
(Pitman and Golovchenko 1988) slope, and a minor amount of aggradation might
take place. In their present relationship with the
Yukon continental shelf, the Malcolm may aggrade
(199Ic) compiled a table of actual slopes of some during a sea-level fall, the Babbage may undergo
modern rivers and shelves (Table 11.1). The results incision, and the Firth would likely not undergo
of this review are illustrated in Fig. 11.15. Contrary to significant change in grade. Such small rivers are of
the assumption of Pitman and Golovchenko (1988), little geological significance. Wood et a1. (1993) car-
many modern shelves have steeper slopes than the ried out laboratory experiments to study the influ-
lower courses of the rivers flowing into them. This is ence of varying shelf angle on fluvial and deltaic
particularly the case where the river has built a sub- deposition during rises and falls of base level and
stantial delta out across the shelf (compare values for produced results very similar to those described
Mississippi and Nile) and also occurs where a sub- here.
marine canyon heads near the river mouth (Indus). The numerical and graphical models of
A fall in sea level would therefore result in steepen- Nummedal et a1. (1993) suggested that during deltaic
ing and a change in the fluvial style of the river. progradation the longitudinal profile of the river will
Incision of the lower course of the river would likely be shifted upward as the river mouth moves seaward,
follow. Leopold et a1. (1964, p. 476) predicted that, in thereby creating accommodation space. They sug-
general, glacio eustatic fall in sea level would lead to gested that for this reason the trajectory of the river
degradation at the mouth of the river. The only situ- mouth has only an indirect relationship to the conti-
ation where a fall in sea level may lead to the develop- nental shelf. But under a condition of falling sea level
a delta will prograde seaward only a limited distance response referred to at the beginning of this section
before autogenic switching leads to the development (and was also rendered more complex by climate
of a shorter, steeper route to the sea. And however far change, as discussed in Sect. 12.12.2).
the river mouth moves horizontally in response to The shift in the entire profile suggested by
progradation, it must also move vertically in re- Posamentier et al. (1988) has not been observed in
sponse to base-level change. Whatever the slope of any of the rivers described by Leopold et al. (1964),
the continental shelf, a fall in sea level forces the river Leopold and Bull (1979), and Schumm (1993). Even
mouth to drop. The slope of the shelf is therefore if such a shift could occur, it would create little
relevant to questions of accommodation space, as accommodation space. Given a slope of 0.001, typical
discussed above. of moderate to large rivers such as the Nile (Table
There is therefore no a priori reason why a fall in 11.1), simple trigonometry indicates that moving the
sea level should be accompanied by creation of ac- mouth of the river horizontally a distance of 100 km
commodation space and consequent fluvial aggra- would create only 10 m of accommodation space,
dation and coastal onlap. The extension of a thin which is entirely inadequate to explain the large
fluvial sheet seaward over shoreline deposits seems thicknesses of fluvial sediments present in some
likely, but thick aggradation and coastal onlap does stratigraphic sequences (e.g., see Galloway 1989b).
not seem probable. For example, note that in the The Yellow River of China illustrates this point (Qian
thick clastic-wedge cycles described by Embry 1990, as quoted by Schumm 1993). The river ag-
(1990a, 1991) there is no wedge of onlapping fluvial graded significantly between the late seventeenth
strata between the subaerial unconformity (se- century and 1855 as a result of the growth of its delta.
quence boundary) and the overlying transgressive During this period the lower reach of the river was
deposits. This supports my objections to the extended nearly 200 km by deltaic progradation, and
Posamentier and Vail (1988) model. Additional sup- the channel aggraded its bed to 10m above the level
port has been provided by the numerical models of of the North China Plain. However, this effect oc-
Rivenaes (1992). curred only because of the confinement of the chan-
With regard to horizontal shifts in the mouth of a nel between artificial levees, which were breached by
river (for example, as it "tracks" a prograding delta: a major flood in 1855. Without the levees the in-
Nummedal et al. 1993), the condition shown in crease in accommodation space would have resulted
Fig. 11.11, Leopold et al. (1964, pp. 260-266) and in channel and delta-lobe switching, probably with-
Leopold and Bull (1979) presented data showing out significant aggradation.
that a rise in base level or horizontal extension Recent work by Leckie (1994) on the rivers of the
of a profile, as in the filling of and progradation Canterbury Plains, New Zealand, demonstrated that
across a reservoir, has a very minor effect on transgression is not necessarily accompanied by the
the overall river profile. Aggradation takes place immediate creation of accommodation space. Be-
for a few hundred meters above the reservoir. cause of very high wave energies along the east coast
Above this the longitudinal profile is characterized of South Island, New Zealand, much of the coastal
by a transitional reach, which curves upstream plain is undergoing erosional shoreface retreat, and
into the remainder of the river, whose slope and the rivers are incising their valleys near their mouths
longitudinal profile remain unaffected. Schumm as they grade down to the retreating shoreline. The
(1993) suggested that the rivers observed by process is analogous to the development of the
Leopold et al. (1964) and Leopold and Bull (1979) ravinement surface that develops during transgres-
were too small and the observation time too short sion of a shallow shelf.
for these to be useful examples from a geologi- Another assertion by Posamentier et al. (1988, p.
cal point of view. The major rivers draining into 135) that should be commented on is the prediction
the Gulf of Mexico show significant incision relat- of an "overall upward-fining trend" in grain size in
ing to glacio eustatic falls in base level; for example, intermediate positions of the stream profile (e.g., a
the Mississippi valley is incised for 370 km up- short distance downstream from A in Fig. 11.11).
stream from its mouth, as far as Baton Rouge This remark is likely to lead to many simplistic de-
(Schumm 1993), with aggradation having taken ductions made from observed grain-size trends in
place during the subsequent rise. However, the well logs. The argument is that in the new accommo-
precise nature of the relationship between base- dation space slopes are initially lower, leading to a
level change and changes in equilibrium profile dumping of coarse sediment, followed by remobi-
in these rivers is confused by the nature of complex lization of this sediment as slopes increase again.
The basic hydraulic flaw in this argument is that the ment supply the major river on the Gulf Coast, the
deposited sediment is assumed to become finer as Mississippi, has prograded significantly.
slopes (and thus velocity and shear stress) increase A recent text on sequence stratigraphy by H.W.
(from slope AF to DF). Of course, the exact opposite Posamentier and G.P. Allen (manuscript in prepara-
would happen. An increase in slope results in in- tion) contains many improvements on the 1988 syn-
creased competency and increased grain size. Sedi- thesis. The following lengthy quote provides the
ments deposited on the fluvial plain represent current state of ideas by the acknowledged leading
temporary sediment storage, with geological preser- experts in the field of sequence stratigraphy:
vation typically occurring because of abandonment "During the early lowstand phase, relative sea level is
of a deposit before it can be remobilized (e.g., lateral gradually falling and the highstand shelf or coastal plain is
accretion on a point bar, channel avulsion). There- subjected to subaerial exposure and fluvial incision. Sedi-
fore, if the pattern of Fig. 11.11 were actually to ments pass through this area with no net deposition. Con-
occur, an upward-coarsening profile would result. sequently, in the absence of tectonic tilting, within the
incising river channels there is no net accumulation of
In conclusion, the model of Posamentier et al. fluvial sediment during this phase. Locally, however, flu-
(1988) and Posamentier and Vail (1988), suggesting vial deposits can be preserved in the form of stranded
that significant fluvial sedimentation, coastal-plain fluvial terraces .... When relative sea level stops falling,
progradation, and coastal onlap occur during times fluvial incision ceases. At this time, examination of fluvial
of rapidly falling sea level (the late-highstand phase, systems would reveal a minimum fluvial deposit concen-
trated along the axes of incised valleys .... After relative sea-
when eustatic fall outpaces subsidence, and an actual
level fall ends, and stillstand and slow rise ensues, rivers
relative fall occurs: times T4-T6 in Posamentier and stabilize and then begin to aggrade as coastal plains en-
Vail 1988, Fig. 8), seems incorrect. It seems much large by normal (in contrast with "forced") shoreline re-
more likely that the main coastal-plain fluvial accu- gression. The total amount of fluvial aggradation during
mulations occur either during relative rises in sea this late lowstand phase results from the combination of
two mechanisms: 1) tectonically-induced aggradation by a
level or during periods of tectonism and source-area
landward tilting of the fluvial profiles, and 2) aggradation
uplift (as described earlier in this chapter). These provoked by the seaward and upward migration of the
conclusions were supported by the numerical mod- fluvial equilibrium profile during normal shoreline re-
eling experiments of Rivenaes (1992), but they need gression under the influence of slow relative sea-level rise
to be evaluated in each field case, in light of local or stillstand." (H.W. Posamentier and G.P. Allen, manu-
script in preparation, p. 51)
variations in subsidence, sediment supply, and ma-
rine energy, as demonstrated by the work of Leckie The landward tilt referred to in this excerpt applies
(1994). The rise in sea level following the last ice age in the case of foreland basins in which subsidence is
is a good example of a transgressive situation. Most more rapid in the proximal part of the basin, but
river valleys were initially drowned and then became would not apply in the case of extensional continen-
filled with estuarine and fluvial facies. Suter et al. tal margins. "Early lowstand" corresponds to the late
(1987) showed that this happened several times on highstand of Posamentier and Vail (1988).
the Louisiana shelf during the late Quaternary sea- Experimental work and geomorphological obser-
level changes (Chap. 13). Yet because of high sedi- vations on the effects of slope changes on rivers have
------Baselevel 1
o ---------10----- H-Baselevel 2
a) DIST ANeE DOWNSTREAM
160 :tOOkm
b) DIST "NeE OOWN VALLEY
100 200km
P -1.5 P-2.0
Fig. 11.18. Effect of base-level fall on channel lengths and patterns. See text for discussion. (Schumm 1993)
Fig. 11.19. Effect of base-level rise on channel length

and patterns. See text for discussion. (Schumm 1993)
\.. __ .a) DII--.!~NCE DOWN8~!d'_

I 110
L It) DIITAHCE D9WH VALLEY
o -~ - - - - ~ --;;0 ----200....
~===9-l
A H F
P -1.5 P-1.0
changes in sinuosity. Figures 11.18-11.20 demon-

strate his arguments. It is important to distinguish
between valley slope and channel slope. The latter is
less, by an amount directly proportional to sinuosity
(see Sect. 8.1). In Figs. 11.18 and 11.19, valley slope is
shown by AB and channel slope by AC. Consider the
~;~, C
effect of an arbitary base-level fall at a point - for
example, by a change in dam height in a sediment-
filled reservoir (a drop from B to D in Fig. 11.18). The
channel profile can lengthen from C to H, as the
A
stream incises downward to the new profile FD. An
increase in channel length over the same valley dis-
I I tance without a change in channel slope or valley
B C
length means an increase in sinuosity. A rise in base
level results in aggradation over the area FBD and a
reduction in sinuosity (Fig. 11.19). These changes
would be gradual. Point F in Figs. 11.18 and 11.19
represents the nickpoint, which retreats upstream.
A B C
Changes in sinuosity are also accompanied by
changes in channel depth and bed roughness, and by
A
changes in the sediment supply. For example, in
~B~
.)
C C ~ 0
experimental studies of base-level lowering, the
channel becomes shallower but wider, and valley
incision feeds additional sediment into the river,
Fig. 11.20. Summary of effects of changes in slope in the which must be compensated by a steeper gradient.
lower course of rivers, for example, as a result of base-level The arbitrary conditions presented in Figs. 11.18
lowering across continental shelves of different inclina- and 11.19 are given greater geological relevance in
tions. See text for discussion. (Schumm 1993) Fig. 11.20, where the effects of base-level lowering
exposing continental shelves of different slope are
summarized. In examples (a) and (b) the newly ex-
long been available, as described in Sects. 8.1 and posed shelf has a gentler slope than the stream chan-
11.2.1.2 (e.g., Ouchi 1985; see Figs. 11.7, 11.8), and nel. In (b) the change in slope can be accommodated
need to be fully integrated into the study of sequence by a reduction in sinuosity without aggradation, but
stratigraphy. The publication of Posamentier et al. in (a) aggradation must also take place to compen-
(1988) stimulated a revival of interest in this area of sate for a much gentler slope (area DBC). In (c) the
research, and Schumm (1993) and Wescott (1993) river extends its course without changes in fluvial
have now clarified this subject. Using geometrical style. In (d) the river can accommodate to a steeper
constructions, Schumm (1993) illustrated a well- course by an increase in sinuosity, and in (e) the
known geomorphic principle, i.e., that rivers are ca- natural limit on the ability of a river to increase its
pable of making significant adjustment to changes in sinuosity because of chute and neck cutoffs leads to
slope without incision or aggradation, simply by incision and to the erosion of area DBC.
T3
c~~~~
O~2
Fig. 11.21. Maps of flume experiments showing A equilib- of aggradation downstream. Eventually, a return to dy-
rium, B aggrading, and C degrading channels; gravel-bed namic equilibrium would reduce the size of the aggraded
rivers at left and sand-bed rivers at right. In degrading braid bars in this downstream area. (Germano ski and
sand-bed channels incision leads to terrace development Schumm 1993)
upstream (numbered in order from TJ to T4) and to a wave
Experimental studies suggest that the responses terns. Section 12.12.2 deals with the response of riv-
described above may apply only to sinuous rivers ers to changes in sediment supply resulting from
(Schumm 1993). In the case of braided rivers, base- climate change.
level lowering causes channel instability and inci- The changes in sinuosity illustrated in Fig. 11.20
sion. Incision itself has the apparently contradictory carry implications for other changes in fluvial re-
effect of increasing valley deposition in the lower sponse, as described by Schumm (1993) and Wescott
reaches because of the increased local sediment (1993). Reduction in sinuosity near the river mouth
supply. Deepening and widening is eventually com- (Fig. 11.20a,b) is comparable to a mountain stream
pensated by significant valley-floor aggradation. entering an inland basin. The formation of a fan
Germanoski and Schumm (1993) reported on a se- delta may be the result. An increase in sinuosity (Fig.
ries of experiments designed specifically to test the 11.20d,e) may be accompanied by development of
response of gravel-bed and sand-bed braided rivers other characteristics of high-sinuosity streams, such
to changes in sediment supply. Increases in sedi- as levee development, with crevassing, cutoffs, and
ment supply cause aggradation, resulting in an in- avulsion. Newly exposed areas of continental shelf
crease in the number of braid bars, whereas a that emerge as a result of base-level fall develop their
reduction in the sediment supply leads to channel own drainage networks that interact with the previ-
incision and the coalescence of braid bars into fewer, ously existing streams. They will be smaller in drain-
larger bars (Fig. 11.21). Changes in sediment supply age area and consequently develop smaller channels.
and the changes in braided style resulting from them A variety of fluvial styles may therefore be present
could be expected to occur as a result of changes in along continental margins. Other implications for
base level. Thus, incision resulting from a reduction sequence stratigraphy are discussed in Chap. 13.
in accommodation space increases the downstream Numerous studies have been carried out recently
sediment supply as a result of increased bank ero- on the effects of tectonism and changing base level
sion and leads to downstream aggradation. Aggra- on alluvial fans and fan deltas (e.g., Muto 1987, 1988;
dation that occurs to fill increased accommodation Gawthorpe and Colella 1990; Bardaji et al. 1990;
space resulting from a rise in base level reduces the Fernandez et al. 1993). The facies and architecture of
incidence of bank erosion, leading to a reduction in fan deltas constitute a specialized topic that can now
sediment supply. The experiments of Germanoski be illustrated by numerous case studies (Nemec and
and Schumm (1993) suggest that the change in the Steel 1988; Colella and Prior 1990), but the details of
dynamic equilibrium in favor of aggradation would this are considered beyond the scope of this book.
therefore be balanced to some extent by the ten- Muto (1987, 1988) discussed the implications of low-
dency for individual channels to incise and rework ering base level across a slope that is greater or less
the bed materials. As they noted, however, their than the. slope of the fan and suggested that the fan
experiments were not scaled to actual full-size rivers, would entrench or aggrade, respectively, based on
and it would be premature to translate their results arguments similar to those of Schumm (1993), as
directly to interpretations oflarge-scale natural sys- shown in Fig. 11.20. A complete cycle of base-level
AGGRADATION
•
.............................................
AGGRADAT ION
•
op
OP : ooen fon
fv: fan valley dc: dissection channel
cf: confined fan df : dissection fill B
Fig. 11.22. A Terminology for alluvial fan and fan-delta morphology; B phases in the development of fan morphology in
response to changes in base level. See text for explanation. (Muto 1988)
change generates a three-stage change in overall fan changes in plate-tectonic configuration. For ex-
architecture comparable to the auto cyclic entrench- ample, a river has occupied the course of the Missis-
ment-backfilling-dispersal cycle described in Sect. sippi River since at least the Late Paleozoic, in the
10.2 (Fig. 11.22). Entrenchment causes the develop- position of an ancient Precambrian rift (Potter
ment of confined fan valleys and dissection of chan- 1978). The Amazon River may also follow the course
nels. These may fill during a rise of base level, leading of ancient rifts, amplified and deepened during the
eventually to the spreading out of fan deposits across Mesozoic opening of the Atlantic Ocean. Other riv-
the entire fan surface. Muto (1988) showed how the ers draining into the Gulf of Mexico were in similar
superimposition of base-level changes on basins of positions throughout most of the Tertiary, as shown
varying subsidence rates would change the overall by the position of modern river courses immediately
architecture of the succession, a topic discussed in above thick Tertiary fluvial-deltaic lobes entering
Chap. 13. In a study of interfingering fan-delta and the basin from the same position on the continental
lacustrine deposits, Fernandez et al. (1993) de- margin (Fisher and McGowen 1967; Galloway 1981;
scribed a type of cycle similar to that of Muto, in Fig. 11.23).
which the entrenchment phase is represented by Some of these large rivers cross major plate-tec-
ribbon-shaped gravel bodies filling narrow, incised tonic boundaries, with the result that in their lower
channels, and the open-fan phase by sheet-shaped courses their size and discharge characteristics may
gravel bodies and by interfingering with fine- bear no relationship to the local tectonic and cli-
grained lacustrine deposits. matic control, while their sediment load may reflect
the mineralogy of far distant sources. Good ex-
amples are the Nile and the Amazon. The headwaters
11.3 Tectonic Control of Basin Style of the Nile are in the tropical grasslands and rain
and Basin-Scale Fluvial Patterns forests of central Africa, and discharge is strongly
seasonal where the river flows through the arid
11.3.1 Big Rivers deserts of Egypt. The Amazon headwaters are lo-
cated in the Alpine snowfields of the Andean moun-
Most of the very large rivers of the world have a long tains, contrasting with the rain-forest basin along its
geological history and follow very long established lower course. The rivers draining the Alpine-Zagros-
courses through structural depressions, even though Himalayan foldbelt show similar contrasts between
their courses may be modified or even reversed by source area and basinal climate.
>-l
ro
()
MT. PLEASANT
I FLUVIAL SYSTEM
8'
i:l
o·
()
o
a....
s.o
...,
tl:I
~
en
s·
(fl
-3:.
ro
~
i:l
p..
tl:I
~
en
9'
(fl
()
e:..
ro
'Tj
Soulh Texas [
Shelf System e:..
"'C
- ~
ro
• HOUSTON :::1
en
LOCAT ION MAP

N Modem
Sobine River
Areas of maximum fluvio l sand (bottom Slratum) deposit ion

~
ANGELIN A I NECHES
DELTA DE LTAS
TEXAS \)~ -1 /
f~ SCALE
~'i/" o
,
25
,
50
!
Miles
ROCKDALE DELTA SYSTEM

WILCOX GROUP (Eocene) TEXAS
Fig. 11.23. The longevity of fluvial depositional systems flanking the Gulf of Mexico. Simplified isopachs outline delta systems of the Eocene Wilcox
Group. Note their coincidence in space with the course of modern rivers. (Fisher and McGowen 1967)
'-"
0\
'-"
Most major rivers enter the sea at divergent plate Rhone, Rhine, and upper Nile (where it flows
margins (Mississipi, Rio Grande, Mackenzie, through part of the East African Rift System), or it
Murray, Amazon, Yellow, Plate, Congo) or where the may be the axis of foreland and forearc basins, such
axes of foreland basins terminate at the margins of as those occupied by the Mackenzie, Po, Tigris,
continental plates (Po, Tigris-Euphrates, Ganges, Euphrates, Indus, Ganges, lower Brahmaputra, and
Indus). Foldbelts that are marginal to continents, Irrawaddy. Others flow along the structural grain of
such as the Appalachians and Cordilleran belts of foldbelts until near their mouths, such as the
North America, have continental divides close to the Mekong. A few major rivers, such as the Columbia,
continental margin, so that the rivers draining their are transverse systems; that is, they are consequent
seaward flanks are short. A major exception is the streams draining directly from an uplifted source
Columbia River, which crosses the coastal moun- area, or antecedent streams that have maintained
tains of the western United States by a geologically their course as mountain systems rose in their path.
complex route. On the landward flank of such It has been suggested that many of the major trans-
foldbelts are the vast continental interiors, which by verse rivers of the Himalayan region are of the
their very size contribute to the development of antecendent type (Holmes 1965, p. 595). Transverse
major rivers. and axial systems normally have very different chan-
nel scales and styles and may carry very different
sediment loads, resulting in the interfingering of
11.3.2 Axial and Transverse Drainage contrasting facies assemblages with distinct sand-
stone colors, compositions, and sedimentary struc-
A particular characteristic of the world's major riv- tures. Typically, transverse systems are much
ers is that most are axial or longitudinal rivers, that smaller and carry coarser sediment loads. They may,
is, their course parallels regional structural grain for example, be alluvial fans. Sinha and Friend
(Kuenen 1957; Potter 1978; MiallI981a,b). The grain (1994) carried out a detailed study of this point with
may be that of major rift systems, as in the case of the reference to the large rivers of the Indo-Gangetic
Mississippi, Rio Grande, Amazon, Niger, lower plains of northern India. Examples of the contrast-
/ Hasl if
bo rdto r
Fig. 11.24. Paleogeography of the Swiss Molasse basin reversed in direction between these depositional phases.
during deposition of the Lower and Upper Freshwater (Van Houten 1981)
Molasse. Transverse alluvial fans fed axial drainage, which
Tectonic Control of Basin Style and Basin-Scale Fluvial Patterns 365
subdivision of drainage patterns are discussed in

Sect. ll.S.
11.3.3 RegionalTectonic (ontrol

Revealed by Basin Analysis
Various regional paleogeographic indicators can be

referred to in the evaluation of tectonic influences.
This has become important in recent years because
of the debate regarding the tectonic-versus-eustatic
origin for stratigraphic sequences. For example,
Embry (1990a) studied the Mesozoic stratigraphic
record of Sverdrup Basin, in Arctic Canada, a succes-
sion up to 9 km thick which he subdivided into 30
third-order stratigraphic sequences. Based on this
work, Embry (1990a) suggested the following lines of
evidence which may be used to indicate a significant
tectonic control of the basin-fill architecture:
"(1) [Tlhe sediment source area often varies greatly from
one sequence to the next; (2) the sedimentary regime of the
basin commonly changed drastically and abruptly across a
sequence boundary; (3) faults terminate at sequence
boundaries; (4) significant changes in subsidence and up-
lift patterns within the basin occurred across sequence
boundaries; and, (5) there were significant differences in
Fig. 11.25A,B. Regional drainage of the Alberta Basin at
two stages in its development: A during the Albian (D.G. the magnitude and the extent of some of the subaerial
Smith, in Mossop and Shetsen 1994); B during the earliest unconformities recognized on the slowly subsiding
Paleocene. (Dawson et al., in Mossop and Shetsen 1994) margins of the Sverdrup Basin and time equivalent ones
recognized by Vail et al. (1977, 1984) in areas of high
subsidence."
Although applied here to the Sverdrup Basin, all of
ing styles of axial and transverse river systems are these criteria have a general applicability. They are
described in several places in this book (e.g., see particularly useful in cases such as the Sverdrup
discussion of Figs. 9.1,11.10). Basin, which does not flank any contemporaneous
Axial drainages are particularly sensitive to the foldbelt that could have served as the sediment
larger scale of plate-tectonic changes in a region. For source. The sediments filling the Sverdrup Basin are
example, as shown in Sect. 11.2.1.2, the Indus River, now thought to have been derived by intraplate up-
which formerly flowed longitudinally southeastward lift of the cratonic interior ofN orth America (Embry
for the full length of the Himalayan foredeep 1990a, 1991).
(Burbank and Beck 1991), was diverted by growth of Lawton's (1986a,b) work on the foreland-basin
the Salt Range and uplift of the piggyback basin clastic wedge of Utah is a good illustration of the use
behind it and now flows southward. Van Houten of basin-analysis techniques to demonstrate tectonic
(1981) demonstrated a reversal in axial drainage control. Lawson used paleocurrent and petrographic
along the Swiss Molasse basin during the Oligocene, evidence to indicate shifting sediment sources and
as a result of variations in rates of uplift and sedi- changes in regional paleoslope during deposition of
ment supply along the Alpine Orogen (Fig. 11.24). A the Mesa Verde Group and associated units in Utah
similar reversal in drainage occurred in the Alberta (Fig. 11.26). He was able to demonstrate the unroof-
basin during the Cretaceous (Fig. 11.25), for similar ing of intrabasin uplifts from which early basin-fill
reasons. sediments were cannibalized, and he also docu-
The relationships between drainage patterns, mented changes in dispersal patterns related to ba-
basin subsidence, and sediment supply are further sin tilting. Cowan (1993) used petrofacies data to
discussed in Sect. 11.3.6. Regional paleogeographic demonstrate shifting sediment sources in the
models that incorporate the transverse/longitudinal Permo-Triassic Sydney Basin, Australia, resulting
Fig. 11.26. Evolution of fluvial dispersal systems in Utah fold-thrust belt. The late Campanian pattern indicates the
from the mid-Campanian (Castlegate Formation, left) to development of basement uplifts within the basin and
the late Campanian (Price River, Tuscher Formations, major reorganization of drainage patterns. Stratigraphic
right), based on analysis of source terranes and paleo- sequences formed in the basin clearly were strongly influ-
currents (heavy arrows). The mid-Campanian pattern is enced by tectonism. (Lawton 1986b, reprinted by permis-
typical of transverse-oblique drainage from a foreland sion)
from differential uplift of the marginal fold-thrust past position. For example, McMillan (1973) envis-
belt and of the craton on the opposite side of the aged a Tertiary river system draining from the conti-
basin. nental interior of North America into Hudson Bay.
Potter (1978) discussed several similar examples. As
noted in Sect. 11.3.1.1, major river systems may cross
11.3.4 Tectonism and Sediment Supply major tectonic boundaries, feeding sediment of a
petrographic type unrelated to the receiving basin
Sediment supply is primarily driven by tectonism, into the basin at a rate unconnected in any way with
but in geologically simple areas sediment-supply the subsidence history of the basin itself. The mod-
considerations are likely to be subordinate to basin ern Amazon river is a good example. It derives from
subsidence and eustasy as major controls of basin the Andean Mountains, flows across and between
architecture. Such is the case where the basin is fed and is fed from several Precambrian shields, and
directly from the adjacent margins, and source-area debouches onto a major extensional continental
uplift is related to basin subsidence - for example, margin. This is an important point with respect to
through the yoking of subsidence to peripheral questions of sequence stratigraphy, because it means
upwarps or the fold-thrust belts marginal to foreland that large sediment supplies delivered to a shoreline
basins. It is where the basin is supplied by long- may overwhelm the stratigraphic effects of varia-
distance fluvial transportation that complications tions in sea level. A region undergoing a relative or
are likely to arise. eustatic rise in sea level may still experience a major
Many sedimentary basins were filled by river stratigraphic regression if large delta complexes are
systems that have long since disappeared from the being built by major sediment-laden rivers.
continental surface, and it may take considerable A significant illustration of this point is the Ceno-
geological imagination to reconstruct their possible zoic stratigraphic history of the Gulf of Mexico. This
DRAINAGE BASINS
- - _ MIOCENE - HOLOCENE
- - OLIGOCENE
- ' - L. PIlLEOCENE - E. EOCENE
..
!:!
~
I
Fig. 11.27. The major drainage basins which fed delta through which rivers entered the coastal region. (Winker
complexes on the Gulf Coast during the Cenozoic. Arrows 1982; Galloway 1989b, reprinted by permission)
indicate the position of three long-lasting "embayments"
-
RIO GRANDE EMBAYMENT SE
2
G:
E
e "'2
:.:
" ~
"'0-3
• FACIES ASSEMBLAGE
CJ CoostoI-pGl fWel
~ Boy ltoooon
12
o fUoIic ' shore· zone ldeltaic
CJ Shelf and slope
\
4 o krtrasIope basin
16
0
1 .
30mi
5
0 30"'"
I1E.' 401 CRUST
20 CONTINENTAL ATTEf"oaJATEO CONTl.
Fig. 11.28. Generalized dip-oriented stratigraphic cross principal Cenozoic clastic wedges. Many of these thicken
section through the Rio Grande depocenter, in the north- southward across contemporary growth faults. (Galloway
west Gulf Coast (location shown in Fig. 11.27), indicating 1989b, reprinted by permission)
II
EMBAYMENT
\~ Louisiana
N
"..,.. Shelf edge posilions
P Pr incipal drainage axis
____ Focus of off lop
E:3 Sand· rich cantinentol morgin sequence
o IOOmi
I
o 200~m
Fig. 11.29. Progradation of the GuLf Coast continental (structural sags) through which rivers entered the coastal
margin by the development of clastic wedges during the plain are also shown. (Winker 1982; Galloway 1989b, re-
Cenozoic. Locations of the three principal embayments printed by permission) .
CHRONO - LITHO- DEPOSITIONAL

STRATIGRAPHY EPISODES and TECTONIC OCEANOGRAPHY I EUSTATIC
STRATIGRAPHY OUTCROP BASIN OEPOCENTERS EVENTS CLIMATE CURVE
EpeifOQlnic uplil f 01 t ~istoc.,...
1
clitn01ic
Rocky Iotlns. and • c~'-s
Grlat Ploins
t
Stab'- AAlorclic
10 w ice lI""t
z
w Basin and RonQl
(!) •• tension
o
w
z
. 1.. I
Acc,l,rated
.-.Oding rot. ·
20 """ific plat.
MoJO' 'eot~nlz o hon

Act;'" 9' 0"'"
formation - Rio
01 E . Poc;foc Grandt RIft
1 s",fadin9
sysltm
PoIa, boItom
1. .
wot.r formation
t
and Antoret k.
montene qlaeiDlon
If =t
RtQionol ROQIOnO.1 rhyolltoc
st,,,,field b
•
YOk:on ~sm :
reoroanizotion o. OuhuohuO
-5. Roc~1 mln.t ~ J::~o
•
a
1
Flalltninv of
1~~;,,~~Spulse
subducltd
---FIG __ FO,OM""\ plait
OtcoupIinQ of
Colorado
Plotta"
f Mojor PlE - 5W
Laromld. pu'"
RG ' RioGrondt I
H: Houlton
M: Miltislippi
<::!1 1.10' ine shale longue ~Inne, coastal plain unconformity • Abrupt 0'8 increose
V .
Dttp - IOOI" .. tdQI 0<
lubmOtinl canyon
-RG-comporoti.. maVnil"'" and cltpocon_
o! continental ......Vin prDQrQdOl ion
...... Proposed map cQnden,ed
Mctions
Fig. 11.30. Age of the major clastic wedges along the Gulf et aL (1987, 1988) is shown for comparative purposes atthe
Coast, compared with the age range of tectonic events in right. (Galloway 1989b, reprinted by permission)
the North American Interior. The global cycle chart ofHaq
continental margin is fed with sediment by rivers bal cycle chart (Fig. 11.30). The major changes along
that have occupied essentially the same position strike of the thickness of these clastic wedges (Fig.
since the early Tertiary (Fisher and McGowen 1967; 11.29) are also evidence against a control by passive
Fig. 11.23). The rivers feed into the Gulf Coast from sea-level change. Highly suggestive are the correla-
huge drainage basins occupying large areas of the tions with the tectonic events of the North American
North American interior (Fig. 11.27). Progradation Interior, for example, the timing of the Lower and
has extended the continental margin of the Gulf by Upper Wilcox Group wedges relative to the timing of
up to 350 km. This has taken place episodically the Laramide orogenic pulses along the Cordillera. It
in both time and space, developing a series of ma- seems likely that sediment supply, driven by source-
jor clastic wedges some hundreds of meters in thick- area tectonism, is the major control on the location,
ness (Figs. 11.28, 11.29). According to the Exxon timing, and thickness of the Gulf Coast clastic
sequence-stratigraphy models, these clastic wedges wedges. A secondary control is the nature of local
would be interpreted as highstand deposits (Chap. tectonism on the continental margin itself, including
13), but Galloway (1989b) demonstrated that their growth faulting, evaporite diapirism, and gravity
age distribution shows few correlations with the glo- sliding.
In arc-related basins, volcanic control of the sedi- rivers in the area (Tsangpo, Salween, Mekong) are
ment supply may overprint the effects of sea-level known to have entirely switched to different basins
change. For example, Winsemann and Seyfried during the evolution of the orogen. Much work re-
(1991) stated: mains to be done to relate the details of the stratigra-
phy in these various basins to the different controls
"Sediment supply and tectonic activity overprinted the
eustatic effects and enhanced or lessened them. If large of tectonic subsidence, tectonic control of sediment
supplies of clastics or uplift overcame the eustatic effects, supply, and eustatic sea-level changes.
deep marine sands were also deposited during highstand
of sea level, whereas under conditions of low sediment
input, thin-bedded turbidites were deposited even during 11.3.S Intraplate Stress
lowstands of sea level."
A fluvial example of the volcanic control of sedi- "Ridge-push" and "slab-pull" are informal terms
ment supply was given by Kuenzi et al. (1979). Flu- that have been used for some time in the plate-
vial styles in this area seem unusual but may, in fact, tectonics literature to refer to the horizontal (in-
be typical of nonmarine arc-related environments plane) forces associated with, respectively, the
characterized by active intermediate volcanism. Ex- horizontal compressional effects resulting from the
plosive eruptions, coupled with very high rainfall, injection of magma at a spreading center and the
lead to the catastrophic transportation and sedimen- tensional effects on an oceanic plate generated by the
tation of huge quantities of coarse volcaniclastic de- downward movement into a subduction zone, under
bris. This may be confined to the trunk rivers, and gravity, of a cold slab of oceanic crust. In-plane
the result is that the tributaries are dammed at their (horizontal) stress is a central theme of the paper by
junction with the main river and back up to form Molnar and Taponnier (1975) describing a model of
lakes (Fig. 7.15). These then fill with mud and small Himalayan collision, in which it was demonstrated
Gilbertian deltas. Lobate, river-dominated deltas that most of the Cenozoic structural geology of west
form at the coast when sediment supply is high, but China, extending for 3000 km north of the Indus
between major eruptions these may be reworked by suture to the edge of the Siberian craton, including
waves into narrow, arcuate deltas. An ancient ex- that of many hinterland basins filled with non-
ample of this fluvial style was described by Ballance marine sediments (see Sect. 11.4), could be ex-
(1988). plained as the product of deformation resulting
In more general, regional terms, sediment supply from intraplate stresses transmitted into the conti-
in volcanic terrains is markedly influenced by nental interior from the India-Asia collision zone. It
the uplift and subsidence generated by the thermal came to be recognized that tectonic plates are not
effects of volcanic activity. A good example is rigid and can store and transmit horizontal forces
described by Dorsey and Burns (1994). Magmatic many thousands of kilometers from zones of plate-
intrusion generates regional uplift, resulting in margin stress. The presence of residual stress fields
steepening of regional slopes, and the sediment sup- in continental interiors has long been known from
ply is vastly increased by accompanying extrusive such evidence as the development of active joints
activity. Cessation of magmatic activity leads to ther- ("break-outs") in exploration holes (summary in
mal relaxation, subsidence, and transgression, with Cloetingh 1988). Cloetingh (1988, p. 206) suggested
a reduction of sediment supply and the drowing of that
paleovalleys.
"the observed modern stress orientations show a remark-
Other examples of the tectonic control of major ably consistent pattern [in northwest Europe], especially
sedimentary units are provided by the basins within considering the heterogeneity in lithospheric structure in
and adjacent to the Alpine and Himalayan orogens. this area. These stress-orientation data indicate a propaga-
Sediments shed by the rising mountains drain into tion of stresses away from the Alpine collision front over
foreland basins, remnant ocean basins, strike-slip large distances in the platform region."
basins, and other internal basins. But the sediment Cloetingh et al. (1985) were the first to recognize
supply is controlled entirely by uplift and by the the significance of these ideas for their possible ef-
tectonic control of dispersal routes, as described in fects on sedimentary basins. They argued
several places earlier in this chapter. Brookfield
(1992) reported on the beginnings of a project to "that variations in regional stress fields acting within in-
homogeneous lithospheric plates are capable of producing
investigate the shifting of dispersal routes through vertical movements of the Earth's surface or the apparent
basins and fault valleys within the Himalayan orogen sea-level changes ... of a magnitude equal to those deduced
of central and southeast Asia. Some of the major from the stratigraphic record."
The important contribution which Cloetingh et transgressive-regressive cyclicity of Mesozoic de-

al. (1985) made was to demonstrate by numerical posits in Sverdrup Basin to intraplate stresses associ-
modeling that horizontal stresses modify the effects ated with the breakup of Pangea, including changes
of existing, known, vertical stresses on sedimentary in regional sediment dispersal patterns brought
basins (thermal and flexural subsidence, sediment about by regional tilting. Heller et al. (1993) sug-
loading), enlarging or reducing the amplitude of the gested that subtle warping of the Rocky Mountain
resulting flexural deformation. They demonstrated foreland region during the Late Jurassic to Middle
that a horizontal stress of 1-2 kbar, well within the Cretaceous resulted from changing intraplate stress
range of calculated and observed stresses resulting fields throughout this time period. Peper et al. (1992)
from plate motions, may result in a local uplift or suggested that intraplate stresses generated by flex-
subsidence of up to 100 m, at a rate of up to 0.1 m/ka. ural loads in response to episodic shortening on
Compressional stresses generate uplift of the flanks fold-thrust belts could generate tectonic cyclicity on
of a sedimentary basin and increased subsidence at a scale as short as 104 years, a proposal developed by
the center. Extensional stresses have the reverse ef- Yoshida et al. (in press) in their study of the
fect. The magnitude of the effect varies with the Castlegate Sandstone of Utah (see Sect. 13.4.4).
flexural age (rigidity) of the crust, as well as with the Cloetingh and Kooi (1990) reported one attempt
magnitude of the stress itself. to model paleostress in which they superimposed
It now seems likely that intraplate stresses may be short-term intraplate stress on the long-term flex-
transmitted for thousands of kilometers across oce- ural subsidence of the continental margin of the US
anic and continental plate interiors, causing en- Atlantic margins, thereby generating a curve of esti-
hanced or subdued uplift and subsidence of cratonic mated changes in the paleostress field (Fig. 11.31).
and orogenic features. Regional tilts generated by Miall (1991c) explored the implications of this curve
this effect may have important consequences for sea for fluvial sediment supply along the continental
level and sediment supply. For example, Kominz and margin, making use of a study of Appalachian denu-
Bond (1991) attributed a widespread mid-Paleozoic dation rates during the Mesozoic and Cenozoic
sea-level rise in cratonic North America to intraplate (Poag and Sevon 1989). Changes in the stress regime
stresses associated with the assembly of the Pangea of the US Atlantic margin occurred because of
supercontinent and depression of the North-Ameri- the continual change in the plate kinematics of the
can continent. Embry (1990a, 1991) attributed the Atlantic region as the ocean opened and the spread-
Paleo stress field Atlantic p'late motion

from stratigraphic modeling documented changes
US Atlantic Margin (Klitgord & Schouten,1986)
Age (Ma) Atlantic tectonic events
o 2.5 reorganization spreading Allantlc sIs I
17 Jan Mayen ridge Jump i-
28
36
Alrlcan plate boundary jump
Labrador sea-noor spreading abandoned i
o
It) 50
58
major changes In C. Atlantic spreading
break-up Greenland-Rockall - - - - - i
80 Labrador sea·noor spreading starts
119 North Atlantic rifting

(Biscay-Rockall-Norweg,an Sea)
o
...
It) 150 separation 01 Africa-Nova Scotia
175 Gull 01 Mexico-Central Atlantic

sea-lloor spreading starts
o
o
C'I
~
Compression
2 1 0 -1 -2
Stress (kbar)
Fig. 11.31. Paleostress curve inferred from a stratigraphic model of the US Atlantic margins. Timing of tectonic events
in the region is shown at right. (Cloetingh and Kooi 1990)
ing center extended and underwent various changes 11.3.6 Quantitative Models of Sediment Supply,
in configuration. Important events, such as the ini- Transfer, and Accumulation
tiation of spreading in the Labrador Sea, the exten-
sion of spreading between Greenland and Europe, The purpose of this chapter is to address the ques-
and various jumps in ridge position, brought about tion of allogenic controls on sedimentation. How
significant changes in plate trajectories, which do factors of tectonism (uplift, subsidence) and cli-
would have been accompanied by changes in the mate change affect sediment supply, sediment cali-
direction and intensity of intraplate stresses. ber, sediment transfer, and sediment accumulation?
Detailed studies of sea-floor spreading patterns in What does the large-scale architecture of the non-
the north and central Atlantic Ocean by Srivastava marine stratigraphic record tell us about the work-
and Tapscott (1986) and Klitgord and Schouten ing of these controls, and is it possible to relate the
(1986) have yielded a detailed history of plate rates and magnitudes of the controlling processes to
motions. Their reconstruction of a continuous suc- any quantitative, measurable attributes of the sedi-
cession of matched plate margins as the Atlantic mentary product? As stated by Paola (1988, p. 231),
Ocean opened indicates that the rotation poles of the "The transition of basin analysis from a descriptive to an
North American plate relative to Europe and Africa analytical science requires the development of well-
changed at intervals of about 2-16 million years. founded theoretical models for basin development, in-
This episodicity is comparable in magnitude to the cluding both the formation of basins and their inflliing
duration of third-order stratigraphic cycles, a fact with sediments."
that is highly suggestive. Furthermore, some of the Models for basin formation, based on analysis of
changes in rotation pole can be correlated in time the thermal and mechanical behavior of the crust,
with changes in the position of depocenters in the have now reached a sophisticated level of develop-
Mesozoic-Cenozoic stratigraphic record of the US ment (see summaries by Miall1990; Allen and Allen
Atlantic continental margin. Poag and Sevon (1989) 1990). Models of basin filling have been slower to
compiled detailed isopach maps of the post-rift sedi- develop. Much work has been carried out on the
mentary record, and interpreted these in terms of physics of sediment transport (e.g., Middleton and
the shifting pattern of denudation in and sediment Southard 1977), but the quantitative relationships
transport from the Appalachian and Adirondack that have been developed from such work relate only
Mountains. It seems likely that many of the major to instantaneous sedimentary conditions and can-
existing rivers that presently carry sediment into this not be used for the analysis oflong-term (geological)
region have been in existence since the Mesozoic. sediment transfer and accumulation. The alluvial
Poag and Sevon (1989) were able to show many stratigraphy models summarized in Sect. 10.4.3 deal
changes with time in the relative importance of these only with autogenic sedimentary processes. It is only
various sediment sources, and it is suggestive that in the past few years that successful approaches to
several of these changes occurred at times of change this problem have begun to evolve.
in Atlantic plate kinematics. The sea-floor spreading The major developments in this area have been
record reveals at least 14 changes in plate configura- made by C. Paola, P.L. Heller, and their colleagues in
tion in the central Atlantic since the Mid-Jurassic a series of papers (Paola 1988, 1990; Paola et al. 1992;
(Klitgord and Schouten 1986). Seven of these Heller and Paola 1992). This work addresses the
changes, at 2.5,10,17,50,59,67, and 150 Ma, corre- question of mass balance of gravel in nonmarine
spond to times when the major sediment dispersal basins - the transfer, sorting, and accumulation of
routes from the Appalachian Mountains to the con- geologically significant gravel deposits over geologi-
tinental shelf underwent a major shift (data from cally meaningful time periods. There are several in-
Poag and Sevon 1989). dependent factors to consider. Among the major
Tectonism is diachronous and, in the case of the controls are the rate of input of sediment and water,
intraplate stresses discussed by Cloetingh (1988), the the rate of sediment transport and basin subsidence,
effects vary across the plate. Adjacent plates may be the grain-size distribution and composition of the
expected to have dissimilar tectonic histories, except gravel, and its downstream reduction in grain size
where major extensional or collisional events affect through abrasion and selective sorting. As noted by
adjoining plates. However, no such tectonic controls Paola (1988, p. 233):
would be expected to be global in scope.
a
....
1.0
(Mf,Mm,lf) 0 rb
.. 0.8
::J
.Q
I-
~ (Q,C) Ost 0.6

1:11
C
.! (S,lf,C) 0 at ,...
-
"- .....
0 0.4
!II a.
!II
a.
..,
QI
C (Mf,lf,Q) 0 ccp 0.2
U
:s
-.2 (Q) Oecp 0.0
0.4 0.8
...
10-4 0.0 0.2
Ii 0
..
(!f,Q) wip
u
QI QI b
CI.
!II
QI
.r:: 1.0
C 1/1
10- 1 ....._ _ . _ - -......- _ . _ - - . . _ _ -....

0.8
1~ 1~ 1
Sedimentation rate (m/103 yr)
0.6
Fig. 11.32. Gravel body cross-sectional shape as a func-
0 0.4
-
tion of sedimentation rate. Dominant clast types are
shown in parentheses: Mf felsic metamorphic; Mm mafic Cl
metamorphic; If felsic igneous; C carbonate; Q quartz a.
0.2
and chert; S other sdimentary rocks. Basins are rb Ridge
basin, California (Uper Miocene); sf Sevier foreland, Utah
(Albian-Santonian); af Alpine foreland, Switzerland 0.0
(Chattian -Aquitanian); eep California coastal plain (Upper 0.0 0.2 0.4 0.6 0.8
Eocene); eep central Atlantic coastal plain, USA (Pliocene);
wip Western Interior alluvial plain, USA (Pleistocene).
(Diagram and data sources from Paola 1988) Fig. 11.33. Deposited gravel fraction p relative to its initial
value prO), plotted relative to non dimensional transport
distance across the basin X*. In each case three different
values of kL are shown. This is a nondimensional measure
"Subsidence induces deposition, deposition causes down-
of the effectiveness of abrasion relative to selective deposi-
stream fining by selectively removing the coarsest clasts
tion in causing downstream fining. (Paola 1988)
from the flow; and thus areas of rapid subsidence are areas
of rapid downstream fining. In general, basins that subside
slowly near their source areas transport gravels relatively
long distances compared with basins that subside rapidly the other case, symmetric subsidence, deposition of
near their source areas."
gravel is more sensitive to abrasion rate. These ba-
Paola (1988) showed by theoretical argument, sins are more efficient at compositional fraction-
and with data collected from various ancient gravel ation.
deposits, that slow subsidence favored sheet-like In a later paper, Paola et a1. (1992) developed a
gravel bodies composed predominantly of resistant, fully quantitative numerical model for predicting
quartzose clasts (Fig. 11.32). Less resistant clasts the distribution of gravel across an asymmetrically
have higher preservation rates when they are buried subsiding basin, in which four major parameters
more rapidly. Figure 11.33 illustrates calculated could each be independently varied in turn, while
gravel accumulation rates in basins with two differ- keeping the others fixed (Fig. 11.34). In the first case
ent subsidence configurations and three different (Fig. 11.34A), slow variation of sediment flux in a
values for clast durability. The data are shown in sinusoidal manner leads to the development of a
nondimensional form. The first case (Fig. 11.33a) is prograding pattern accompanied by increased
for a basin showing uniform subsidence, the second sedimentation rate (increased spacing between the
case (Fig. 11.33b) for a basin undergoing asymmetric isochrons). Paola et a1. (1992) termed this a "flux-
subsidence, more rapid at the upstream end and zero driven" style of progradation. The second case (Fig.
at the downstream end. The results confirm that the 11.34B) shows that sediment is trapped in more
rate of downstream fining is much greater in the case proximal positions when subsidence rate is in-
of basins which subside asymmetrically, and that in creased. This is the so-called subsidence-driven
A ... model, in which progradation accompanies a dimin-

ished subsidence rate. The third case, termed the
-...
(f)
Q) « distribution-driven" model (Fig. 11.34C), shows the
Q) progradation that takes place when the gravel pro-

E portion is varied while total sediment flux and sub-
0
sidence rate are held constant. The fourth case (Fig.
~
11.34D) shows that varying the amount of water
available for transportation makes little difference
to the gravel distribution in the basin. Paola et al.
B (1992) developed a second set of models in which
these same changes in parameters were imposed at
RELATIVE
AMPLITUDE much more rapid rates, producing some interesting
OF VARIATION
-, variations on the results illustrated here.
Paola et al. (1992) also explored the relationship
between basin scale and the rate of response to the
+ ++++++ + various sedimentary controls. Each basin is charac-
.... .,. .. + ... .. .. 25 km ..
+ + ~ + + + + ~ + ~ + terized by a factor termed the "basin equilibrium
time", which is defined as the square of basin length
c 0 divided by the water flux, or diffusivity. The equilib-
rium time is the time scale over which the basin is
-...
(f)
Q) able to fully respond to periodic or cyclic changes in
controlling parameters (subsidence, sediment flux,
,.
Q)
Increasing
E 10
etc.). In basins with long equilibrium times pro-
decreasing .Q gradation is controlled primarily by changes in
~ + ++ • + + + • + + + +
.. .,. .. .. + + ....
~
25 km .. diffusivity or sediment supply. Where equilibrium

20
+ ~ + + .;. + + + ""~T i " times are relatively short, changes in subsidence rate
appear to be the major control.
0 0 In a companion paper, Heller and Paola (1992)
-
collected some basic data from several actual gravel-
...
(f)
Q)
filled basins to explore the application of the models .
Q) An accurate chronostratigraphic framework is a pre-
E 10
requisite for this kind of testing and is not, in fact,
.Q
:i: available for many basins. Two examples of lithol-
ogy-age plots are illustrated in Figs. 11.35 and 11.36.
With the aid of the numerical modeling approach
summarized above, some useful insights can be
Fig. 11.34. Cross sections of a hypothetical basin showing gained regarding the relationship between sedimen-
the results of slow variations in four parameters: A sedi- tation and tectonics in these basins. The first plot,
ment flux; B subsidence rate; C gravel fraction; and D from Argentina, shows sedimentation rate increas-
diffusivity. Direction of transport is from left to right. Thin
dashed lines are isochrons drawn every 1 million years. ing up section, accompanying a coarsening upward
(Paola et a1. 1992) into gravel. This suggests a flux-driven situation.
In the second case, from Pakistan, sedimentation
Lithology Age (Ma)
18 16 14 12 10 rates decrease upsection, accompanying an upward
5
coarsening into gravel. This suggests an example of
4 subsidence-driven progradation.
The importance of these models is the light they
throw on the temporal relationship between tecton-
ics and sedimentation. It has commonly been as-
sumed that wedges of coarse sediment prograding
o
from a basin margin are "syntectonic", and in many
Fig. 11.35. Generalized stratigraphic section at Las Juntas, earlier studies the dating of such wedges has been
Argentina, showing calculated sedimentation accumula- used to infer the timing of major orogenic episodes
tion rates, corrected for compaction. (Heller and Paola
1992) (see summary and references in Miall 1981b; Rust
and Koster 1984). It can now be seen that this inter-
RATA-DADIAL DINA BASAWAKAS Fig. 11.36. Lithology and age of Siwalik Group, NW
Lithology Age (Mal Age (Mal Himalayan foreland basin, Pakistan. The three sec-
" 3 2 1 0
tions are arranged in order from proximal (left) to
distal (right) over a total distance of 40 km. (Heller
and Paola 1992)
pretation is simplistic. In fact, the relationship be- tation (Fig. 11.34B) is exemplified by the Siwalik
tween tectonism and sedimentation is complex, and sediments of Pakistan (Fig. 11.36), showing that
depends on the balance between a range of controls, coarse clastic progradation occurred at a time of
including sediment supply and sediment type. More decreasing sedimentation rate, following the end of
recent studies have demonstrated that tectonism thrust-induced flexural loading of the basin.
does not necessarily coincide with the progradation There has been considerable debate regarding
of wedges of coarse sediment but may precede such these models, and much additional testing is under-
progradation by a significant period of time (Blair way. For example, widespread Early Cretaceous
and Bilodeau 1988; Jordan et al. 1988; Heller et al. gravels in the Western Interior Basin, particularly in
1988). Tectonism, in the form of basin-margin fault Utah, Colorado and Wyoming, have long been inter-
movement, may lead to increased rates of subsid- preted as syntectonic, related to thrust movements
ence, particularly at the edge of the basin. Foreland in the Sevier Orogen. Heller and Paola (1989) dem-
basins, in particular, are developed as a result of onstrated that the gravels have a stratigraphic distri-
crustal loading by overriding thrust sheets and are bution inconsistent with that which would have been
characterized by syntectonic subsidence of the expected if they simply filled the flexural "moat"
proximal regions of the basin. The immediate basi- developed in front of the Sevier fold-thrust belt.
nal response is likely to be marine or lacustrine They interpreted the gravels as the product of a
incursions, with ponding of coarse debris against regional thermal uplift, distinct from the uplift asso-
the basin margin. Tectonism eventually leads to in- ciated with Sevier tectonism. Burbank et al. (1988,
creased basin-margin relief and hence to a reju- 1989) defended the "syntectonic" model of gravel
venatation of the supply of coarse sediment, but this progradation in the case of the Siwalik sediments of
process take time, and it has been suggested that Pakistan, a model disputed by Heller et al. (1989).
clastic-wedge progradation is largely a post-tectonic The details of the dispute range beyond the scope of
phenomenon (Blair and Bilodeau 1988; Heller et al. this book, involving discussions of the importance of
1988). Increased rates of subsidence accompanying local structural controls, such as folding of the proxi-
tectonism are more likely to be recorded by in- mal foreland leading to uplift and sediment bypass,
creased rates of sedimentation of fine-grained de- the importance of flexural rigidity of the underlying
posits, particularly shallow-marine and lacustrine crust in controlling the flexural behavior during
sediments, at the basin margins, and at this time thrust-sheet loading, etc. It was also pointed out
river systems may actually flow towards the foldbelt (Burbank et al. 1989) that the models of Heller et al.
(Burbank et al. 1988). Heller and Paola (1992) re- (1988) and Blair and Bilodeau (1988) are two-dimen-
ferred to this model of coarse sedimentation as sional and do not acknowledge the importance of
antitectonic. Only in the most fault-proximal re- axial transport of detritus into sedimentary basins, a
gions is tectonic activity likely to be recorded by the very important factor in the case of the Himalayan
rapid development of a clastic wedge. foreland (see further discussion of this basin in Sect.
The differences between the syntectonic and 11.4.6). Gordon and Heller (1993) studied late Ceno-
antitectonic models are explained using the quanti- zoic sedimentation in Pine Valley, Nevada, an exten-
tative models of Paola et al. (1992) and Heller and sional basin in the Basin and Range province of the
Paola (1992). The so-called flux-driven model of western United States. They found that the distance
sedimentation (Fig. 11.34A) represents the tradi- of progradation of alluvial gravels within the basin is
tional syntectonic explanation for conglomerate de- inversely proportional to the contemporaneous sub-
velopment and is illustrated by the example from sidence rate. Jo et al. (1994) argued that local struc-
Argentina (Fig. 11.35). Subsidence-driven sedimen- tural controls and autogenic mechanisms introduce
complications that may have confused the simple Most of the discussion in this section deals with
picture painted by Gordon and Heller (1993), but, in primary plate-tectonic relationships, that is, basins
a reply to this discussion, the authors maintained where the alluvial stratigraphy and architecture
that these are essentially small-scale details which do are directly related to first-order contemporaneous
not obscure the broad patterns predicted by the tectonic controls. However, in many cases contem-
basin-scale model (Heller and Gordon 1994). Lang poraneous tectonism, including long-distance
(1993) attempted to apply the new basin models to a transmission of intraplate stress, reactivates ancient
Devonian alluvial complex in northeast Australia structures, generating topographies and regional
and suggested that examples of both syntectonic and paleo slopes in positions and with orientations that
antitectonic progradation are present. In a study of a would not be predicted from simple plate-tectonic
modern syntectonic fan system in Argentina, models. Many of the most interesting details of local
Damanti (1993) suggested that thrust-proximal fan basin architecture cannot be understood unless the
systems fed by large sediment source areas (large effects of such reactivation are incorporated into
drainage basins for individual fans) are able to re- paleogeographic reconstructions.
spond rapidly to thrust-fault activity by renewed The thickest successions of alluvial deposits occur
basinward progradation, especially where there is in rift basins (including failed rifts and aulacogens),
adequate rainfall to maintain high rates of sediment forearc basins, retroarc and peripheral foreland
transport. G.A. Smith (1994) warned of the ability of basins, and those associated with plate-collisional
rapid climate change to generate stratigraphic ef- hinterland deformation, particularly foreland and
fects (such as changes in grain size and channel style)
very similar to those traditionally thought to be con- Table 11.2. Classification of the types of sedimentary
trolled by tectonism. It is clear that a variety oflocal basins discussed in this chapter. (Miall1990)
factors influence the relationship between tectonics
1 Divergent-margin basins
and sedimentation, and that in each case the nature 1.1 Rift basins
of this relationship needs to be determined by de- 1.1.1 Rifted-arch basins
tailed local studies. 1.1.2 Rim basins
1.1.3 Sag basins
1.1.4 Half-graben
1.2 Continental-margin basins
11.4 Plate-Tectonic Setting of Alluvial Basins 1.2.1 Red Sea type (youthful)
1.2.2 Atlantic type (mature)
1.3 Aulacogens and failed rifts
The purpose of this section is to illustrate the broad 1.4 Oceanic islands, seamounts, plateaus
2 Convergent-margin basins
stratigraphic architecture of alluvial depositional
2.1 Trenches and subduction complexes
systems and their relationship to the structural style 2.2 Forearc basins
and tectonic processes dependent on plate-tectonic 2.3 Interarc and backarc basins
setting. 2.4 Retroarc (foreland) basins
3 Transform- and transcurrent-fault basins
Basin setting:
Plate boundary transform fault
11.4.1 Basin Classification Divergent-margin transform fault
Convergent-margin transcurrent fault
Basin-forming mechanisms and the classification Suture-zone transcurrent fault
of sedimentary basins according to plate-tectonic Basin type:
3.1 Basins in braided fault systems
principles are subjects that have been discussed 3.2 Fault termination basins
by numerous authors (Dickinson 1974; Bally and 3.3 Pull-apart basins in en echelon fault systems
Snelson 1980; Bally 1984; Klein 1987; Ingersoll 1988; 3.4 Transrotational basins
Miall 1990; Allen and Allen 1990; Busby and 4 Basins developed during continental collision and
Ingersoll 1995), and this discussion will not be re- suturing
4.1 Peripheral (foreland or foredeep) basins
peated here. Table 11.2 presents the basin classifica- (on underriding plate)
tion used in this book. A simple numerical code has 4.2 Intrasuture embayment basins (remnant ocean
been added to the classification, and this is used in basins)
Table 11.3, which lists some examples of modern 4.3 Hinterland foreland, strike-slip, and graben basins
depositional systems and ancient stratigraphic ex- (on overriding plate)
5 Cratonic basins
amples.
Plate-Tectonic Setting of Alluvial Basins 377
Table 11.3. Examples of modern and ancient fluvial deposits classified according to tectonic setting
Basin type" Modern examplesb Ancient examples
1.1 East African rifts Sadlerochit Fm, Alaska (Permian-Tr.)
Rio Grande rift, New Mexico Newark Gp., N. America (Triassic)
Bjarni Ss, Labrador (Cretaceous)
1.2.1 Fan deltas, Red Sea margins
1.2.2 Gulf Coast Woodbine-Tuscaloosa, Gulf Coast (Cret.)
Frio Formation, Gulf Coast (Oligocene)
1.3 Benue Trough Statfjord-Brent Ss, North Sea (Tr.-Jur.)
Latrobe Gp., Gippsland Basin (Tert.)
Natal Embayment, S. Africa (Paleozoic)
1.4 Sandurs, Iceland
2.2 Pacific coast, Chile Old Red Ss (Dev), Midland Valley, Scotland
Pacific coast, Guatemala Kenai Gp, Cook Inlet (Tertiary)
Western Trough, Burma (Tertiary)
San Joaquin Basin, California (Olig.-Q.)
Luzon Central Valley, Philippines (Plio.-Q.)
2.3 Cagayan Basin, Luzon Ordovician fluvial-deltaic systems, N. Wales
Rifts in Aegean Sea Carboniferous basins ofN. England
Kayenta Formation, Colorado Plateau (Jur.)
(?) Songliao Basin, China (Cretaceous)
Tertiary ss ofW Japan
Basin-and-Range, USA (Cenozoic)
2.4 Upper Amazon basin Hawkesbury Ss (Triassic), Sydney Basin
Bermejo Basin, Argentina Hutton Ss, Eromanga Basin (Jurassic)
Papuan basin, Papua-New Guinea Rocky Mt. basin, Utah-Colorado (Cret.)
3.1 N. Anatolian basins, Turkey San Andreas system (U. Tertiary)
3.3 Salton Trough, California Ridge basin, California (Tertiary)
Dead Sea
4.1 Indus-Ganges Trough Catskill Delta, New York (Devonian)
Western plains of Taiwan Cherokee Gp., Arkoma Basin (Penn.)
Swiss Molasse Basin (Tertiary)
Siwalik Gp, India-Pakistan (Tertiary)
Aikawa, Ashigara basins, Japan (Tertiary)
4.3 Baikal Rift, Russia Midland Valley, Scotland (Devonian)
Tarim Basin, China Cutler Gp, Ancestral Rockies (Permo-Penn.)
Lake Hazar, Turkey Alberta Basin (Cret.-Tertiary)
Pattani Trough, offshore Thailand (Tertiary)
5 Lake Eyre Late Paleozoic, Illinois Basin
Lake Chad
aNumerical classification as in Table 11.2.
b"Modern" in the sense that the current plate-tectonic classification of the basin is as shown. However, some of these
basins have had a very lengthy geological history, in some cases extending back to the Cretaceous.
strike-slip basins (Fig. 11.37). Fluvial deposits do not 11.4.2 Extensional Basins
occur in some types of basin, such as oceanic
trenches, subduction complexes, and remnant ocean Crustal extension occurs in several distinct tectonic
basins, because these basins are floored by oceanic settings. The stretching and rifting of major conti-
crust, and therefore are unlikely to include any areas nental plates initiates the development of rift basins
of nonmarine deposition. Nonmarine sedimenta- (basin type 1.1 in Table 11.3) and failed rifts (basin
tion is also very limited on oceanic seamounts and type 1.3) and is followed by the development
plateaus for the same reason. of broad continental-margin basins as a result of
Fig. 11.37. Tectonic setting of fluvial deposits.

Note the distinction between transverse drainage,
which is oriented perpendicular to structural
grain, and axial, or longitudinal drainage, which
typically consists of trunk streams flowing along
the basin axis. (Miall1992)
.. ..
direction of
normal fault plate migration
- ,::;::::;==-- strike-slip offset alluvial deposits
HANGING WALL ALLUVIAL

CONES
INTERIOR DRAINAGE
Fig. 11.38. Model for continental half-graben rift basin with interior drainage. Numbers indicate successive fan lobes.
(Leeder and Gawthorpe 1987)
thermally driven flexural subsidence (basin type faults ("escape tectonics") develops pull-apart and
1.2). Rift-type basins also occur in areas of continen- other, more complex, basin types (basin type 4.3).
tal back-arc spreading (basin type 2.3), in transten- The structural and stratigraphic style of rift basins
sional continental transform settings (basin type 3), that develop in these various plate-tectonic settings
and in areas of major plate collision, where lateral are similar.
translation of continental blocks along strike-slip
LARGE CONES SOURCED

IN EN -E CHELON
HANG ING WALL

CONES
CONTINENT AL
HALF-GRABEN WITH
AXIAL THROUGH DRAINAGE
Fig. 11.39. Model for continental half-graben rift basin with axial through drainage. (Leeder and Gawthorpe 1987)
77.4.2.7 Rift Basins similar in both models, the main difference between
this model and the first being the environments and
The broad stratigraphic architecture of rift basins facies developed at the basin center. Axial rivers may
and its relationship to basin tectonics have been be major regional trunk rivers with large channels
studied in detail by Leeder and Gawthorpe (1987) and contrasting facies characteristics to the flanking
and Leeder and Jackson (1993). Numerous modern fans. The position of the axial river within the basin
and ancient examples have been identified in vari- is determined partly by the pattern of subsidence
ous tectonic settings, some of which are noted in and partly by the scale of transverse fans, which may
Table 11.3. be large enough to displace the axial river laterally.
Two useful generalized models for rift basins have Differential subsidence within the basin may cause
been proposed by Leeder and Gawthorpe (1987), the meander belt to migrate laterally, in the manner
based on research in the Basin and Range province of documented by Alexander and Leeder (1987; see Fig.
the United States and in backarc rifts of southern 1l.6) and Alexander et al. (1994). Gradual subsid-
Greece. The first model (Fig. 11.38) is based mainly ence along the marginal fault may lead to rotation of
on studies of the Basin and Range, where this basin the hanging wall and onlap by the floodplain depos-
style is characteristic. Low-gradient hanging-wall its marginal to the axial river system.
dip slopes are the site for broad alluvial cones, while Among the best-known ancient rift systems is
small, coarse alluvial fans prograde from the steeper that filled by the Triassic-Jurassic Newark Super-
slope of the footwall scarp. Lakes characteristically group of eastern North America (Fig. 11.40). These
occupy the basin center; they may be perennial lakes rifts represent the initial rift phase that preceded the
or ephemeral playas, depending on climate. In the development of the modern Atlantic Ocean. They
latter case, evaporite deposits may be important consist of a series of half-grabens, many now buried
(Rosen 1994). A related model, illustrated in Fig. beneath the Jurassic and younger sedimentary cover
11.39, is for a half-graben basin with axial through of the Atlantic continental margin. Many of the
drainage. Flanking transverse alluvial fans may be basins are linked by continuous border faults with
the same sense of dip that represent reactivated dominated, in which fluvial deposits comprise ba-
Paleozoic compressional faults, and in this way sin-margin alluvial fans and small deltas.
they differ from the East African rift system, which The Newark rifts are part of an immense system of
consists of segments of opposite-facing border faults rift basins that developed between North America,
(P.E. Olsen 1990). Newark basins show two different Europe, and Africa during the initial phase of exten-
depositional patterns: one is fluvial dominated, con- sional tectonics that led to the breakup of Gondwana
sisting of basin-wide channel systems, commonly (Ziegler 1988). Numerous individual case studies of
including axial trunk rivers; the other is lacustrine nonmarine rift sedimentation have been carried
out in eastern Canada and the United States, Spain,
Britain, and the North Sea Basin.
Major rift systems developed in Africa in the Me-
Newark sozoic-Cenozoic during the breakup of Gondwana.
Supergroup Rift In some of these transtensional tectonism occurred,
Basins leading to strike-slip components of fault offset
(Fairhead 1986; Genik 1993). Sedimentary patterns
in the East African rift system were documented by
Frostick et al. (l986), and several studies carried out
during petroleum exploration work have now docu-
mented the subsurface geology of these rifts and
other Mesozoic-Cenozoic rift systems in Central and
West Africa (e.g., Genik 1993; Winn et al. 1993).
Figure llA1 illustrates the typical structure of
African rifts, based on an example in Niger. Neo-
comian to Albian sediments in this and adjacent
basins constitute a "phase-I synrift tectono-strati-
CULPEPER
graphic sequence" consisting largely of terrigenous,
TAYLORSVILL~ _ Exposed nonmarine clastics, fining upward from a basal
fluvial feldspathic sandstone through a middle flu-
c:::J Inferred vial deltaic sandstone-shale facies, into an upper
lacustrine shale facies. The phase-II synrift sequence
(Cenomanian-Maastrichtian) is largely marine in
km
the Western African basins but may consist of 1.5-3
o 100 200 300 400 km of fluvial deposits in the Central African rifts,
organized into fining-upward channel-overbank
Fig. 11.40. Distribution of Triassic-Jurassic Newark ba- cycles with interbedded lacustrine shales. In the
sins in eastern North America. (P.E. Olsen 1990, reprinted West African basins phase-III rift deposits consist of
by permission)
sw o ZSKM Ni
DEPTH I I Ji
~ ~
~----------------------~----------------------------------------~
1.
--
--..-
......
.......-::::
......
15
___ ---- -----_.7

---- ---
20
25
3G
Fig. 11.41. Structural cross section through the Termit Basin, part of the West African Rift System in Niger, drawn from
reflection-seismic data. (Genik 1993, reprinted by permission)
up to 2 km of fluvial sandstones overlain by marine centers. This map shows that the coastal plain has
or lacustrine deposits. Miocene-Holocene sediments prograded by up to 350 km since the Cretaceous. In
constitute a post-rift assemblage and consist of a few places the clastic section is at least 15 km thick. Much
tens of meters of fluvial deposits along the basin of this consists of marine deposits, ranging from
margins, with much thicker lacustrine assemblages deltaic and barrier-lagoon deposits to deep-marine
in the center. All the fluvial deposits in these basins submarine fan successions. However, updip equiva-
appear to have accumulated within systems of inter- lents include significant fluvial units. Nonmarine
nal drainage (Genik 1993). stratigraphic units include parts of the Denkman
Sandstone (Upper Jurassic), Travis Peak Formation
(Early Cretaceous), Woodbine-Tuscaloosa Forma-
11.4.2.2 Continental-margin Basins tion (Upper Cretaceous), the Wilcox Group (Pale-
ocene-Eocene), the Frio Formation (Oligocene), and
Extensional continental margins that develop once the Catahoula and Oakville formations (Oligocene-
oceanic crust has evolved along the rift axis are com- Miocene). One of the most interesting features of
monly not characterized by significant thicknesses this continental margin is the extreme long-term
of nonmarine deposits, except where major rivers persistence of many of the sediment transport path-
cross the continental margin. As noted in Sect. ways. As noted by Potter (1978), a river has occupied
11.3.1, many of the world's major rivers enter the sea the course of the Mississippi since the Late Paleozoic.
at divergent plate margins, and their fluvial deltaic All the major rivers draining across the Texas coast
systems form part of the continental-margin basin have remained in essentially the same position since
stratigraphy. Away from such fluvial axes, however, at least the Eocene, as demonstrated by Fisher and
flexural subsidence leads to the development of McGowen (1967), who showed that each is underlain
gently dipping continental shelves which are trans- by a major Wilcox deltaic depocenter. Galloway
gressed' by the sea, with alluvial coastal plains typi- demonstrated the vertical persistence of many of the
cally occupying only the fringes of the basin. As Cenozoic fluvial depositional axes (Fig. 9.15). Many
discussed in Chap. l3, alluvial deposits on these shelf additional details of these fluvial units are discussed
areas may become significant only during times of in Chap. 9.
low sea level, and their stratigraphy may yield much The long-term persistence of fluvial axes on the
useful information regarding local sea-level history. Gulf Coast contrasts with the wide variation in sedi-
The Jurassic-Tertiary section flanking the Gulf mentation rates documented by Galloway (1989b)
Coast of the United States is probably the thickest and Galloway and Williams (1991). Sedimentation
and most extensive suite of alluvial deposits on a was most rapid in the deltaic to shallow-marine envi-
divergent continental margin (Fisher and McGowen ronment of the continental margin, where the gen-
1967; Galloway 1981; Worrall and Snelson 1989). eration of accommodation space by subsidence was
Figure 11.42 is a cross section through the Gulf of most rapid. There, values during the Cenozoic varied
Mexico, Fig. 11.43 shows part of the Cenozoic section by more than an order of magnitude, from 0.03 m/ka
in more detail (see also Figs. 11.23, 11.28), and Fig. to more than 1.4 m/ka (Galloway and Williams 1991;
11.29 shows the position of major Cenozoic depo- the rates were calculated for each chronostrati-
(lv" C(J~II t;'''J),I1C 1 i n •

( Co... t ,J I , I • i n
10.000,"
uOaO,"
Fig. 11.42. Structural cross-section through the northern Reprinted by permission from King PB: The evolution of
margin of the Gulf of Mexico. showing the immense thick- North America, rev. edn. Copyright © 1977 by Princeton
ness of Jurassic to Recent terrigenous clastic sediments. University Press
F LIVE OAK CO. ! SAN PATRICIO CO. F'

I
EXPLANATION
a-
Fig. 11.43. Typical dip-oriented cross section through Catahoula and a major sand-rich fluvial axis of the
the Upper Tertiary clastic succession of south Texas. The Oakville Formation. (Galloway 1981)
section crosses an interfluvial, sand-poor area of the
graphic unit, of 1-2 million years duration. There- fected. Winker (1982), Galloway (I989b), and Gallo-
fore, the sedimentation rates should be compared way and Williams (1991) argued that Laramide tec-
with the group 9 and 10 rates and processes dis- tonics along the Cordilleran orogen were largely
cussed in Chap. 3). Winker (1982) and Galloway responsible for controlling the sediment supply to
(1989b) argued that the Cretaceous-Cenozoic Gulf the Missouri and other rivers draining across the
Coast stratigraphy, with its alternation of sand-rich Great Plains into the Gulf of Mexico (Fig. 11.29). This
clastic wedges and shale-rich intervals, is largely a is discussed further in Sect. 11.3.5.
reflection of the large variations in sediment supply
brought about by changes in tectonic regime in the
continental interior drained by the Gulf Coast rivers. 77.4.2.3 Failed Rifts and Aulacogens
These rivers rise on the flanks of the Appalachian
and Cordilleran orogens and, in the case of the Rio Failed rifts are members of the initial family of rifts
Grande, drain a major, active rift system (Fig. 11.27). associated with continental breakup, along which
As discussed in Sect. 11.3.5, changes in the intraplate much of the initial crustal extension takes place, but
stress regime along the Appalachian Mountains which cease active extensional activity when com-
are probably largely responsible for geomorphic plete continental separation takes place elsewhere
adjustments that controlled changes in sediment and the generation of oceanic crust commences.
supply to the Atlantic continental margin, and the Failed rifts undergo the initial stages of normal fault-
rivers draining from the opposite flank of the moun- ing and crustal extension, but remain floored by
tains into the interior, and hence into the continental crust. The term is normally limited in its
Mississipppi system, would have been similarly af- application to major rifts striking inland at high
angles to the continental margins, which for periods the Viking Graben of the North Sea and its flanking
of a few millions or tens of millions of years act as the East Shetland Basin, Gippsland Basin off southeast
major loci of crustal extension. Aulacogens are failed Australia, Benue Trough, Nigeria, and Kutei Basin,
rifts that have subsequently been affected by closure Borneo (Kalimantan).
of the ocean with which they are tectonically associ- The North Sea is an intracratonic basin which
ated. Plate convergence and collision commonly developed by a complex and lengthy series of exten-
leads to compressional tectonics trending parallel to sional movements beginning in the late Paleozoic
the basin axis. Both the initial rifting and the sub- (Ziegler 1988). The Upper Permian to Jurassic rocks
sequent plate convergence and collision may be of the North Sea were developed during a phase of
associated with some oblique or transcurrent plate rift faulting and thermal subsidence related to the
motion, giving rise to a component of strike-slip on incipient opening of the Atlantic Ocean (Fisher 1984;
the bounding faults. Ziegler 1988; Steel and Ryseth 1990). The Viking-
These basins commonly constitute important Central Graben was the major structure to develop at
petroleum provinces because of the combination of this time (Figs. 9.22, 9.23, 11.44). Regional subsid-
good reservoir characteristics, good petroleum ence that accompanied rifting activated regional
source rocks, appropriate maturation levels, and ex- drainages, leading to the development of major flu-
cellent structural trapping configurations. Examples vial units, the Teist, Lomvi, Lunde, and Statfjord
of failed rifts containing important fluvial units are Formations (Upper Permian-Sinemurian; Steel and
o· 2' 3'
WELLS
• Wells used in study
{ref. wells are labelled}
CORRELATION LINES
Syn-ritt deposits and

post-rift sequences
1 and 2
Post-rift sequence 3
100
o· 1•
Fig. 11.44. Location of Viking Graben and related structures in the North Sea Basin. (Steel and Ryseth 1990)
®
ffl<---+--- 62'
EARL Y TRIASSIC
RIFT SYSTEM
(SCYTHIAN)
,....----------1 6(r-..;;~~_:r.::~~~~~- 60·
S'
~-+- 5S'
.
~,1kE.~~~)\; ~~~~ :'5; ---j'---+_-+-':_~~_--I-_+-_+- 56'
2' 6' 7' S'
~ PROSABLE
~ HINTERLAND
®
~ PROXIMAL
~ ALLUVIUM
[......;...J
':':':":"
MAINLY STREAM AND
STREAM FLOOD DEPOSITS
, .................... , MAINLY FLOODBASIN AND
LACUSTRINE DEPOSITS
P ':G.,;,-,:;::'IINLAND SABKHA OR
'..:..'"",,,-, LAGOONAL DEPOSITS
l' 2' 3' s· 6' 7' S'
Fig. 11.45. Tentative paleogeographic maps for post-rift megasequence 1 in the northern North Sea Basin. (Steel and
Ryseth 1990)
Ryseth 1990). These can be subdivided into a series Faulting and folding occurred in mid to Late Jurassic
of megasequences (Fig. 9.23). The first of these is time, and the fault blocks were tilted, uplifted,
correlated with the rift fault activity, while the three and eroded (Fig. 9.22). Subsequent Late Jurassic
subsequent post-rift sequences are attributed to (Kimmeridgian) and Cretaceous transgressions cov-
post-rift thermal subsidence (Steel and Ryseth 1990). ered the eroded fault blocks with shales, including
With continued subsidence, marine transgression the Kimmeridge Shale, a prolific source rock. The
occurred and the largely deltaic Brent Group shales also provide the seals for the Viking Graben
(Aalenian-Bajocian) was deposited (Brown 1984). oil fields.
Fig. 11.46. Major faults, anticlines, and

ASTUNE oil and gas fields in the Gippsland Basin,
off the southeast coast of Australia.
NORTHERN PLATFORM
(Rahmanian et al. 1990)
~ GAS
• OIL
20KM .
... ...
-'
... ..
]!
-........
II:
Z II:
II; c: 0
;: cm ~ i!I~ f
A~ ,. ~!~; iii::; ~
J
<:I :::0 _~
it'"
u Hi
o ~ ;0
ClC
uz ~ ~ Sii5 ...J B
GIPPSLAND LlMESlONE & LAKES ENlllANCE FORMATION
Marlin ~ene A
2 Channel
aI
LAlllOBE
5 GROUP Fig. 11.47. Regional cross section through
6
Gippsland Basin. The section is oriented
SW-NE, approximately parallel to the
<) l282l GAS coast of Victoria. The location is given by
8 • • OIL the position of Flounder field in Fig. 11.46 .
(Rahmanian et al. 1990)
Speculative paleogeographic maps for the lower Cretaceous. By Oligocene time, subsidence had ex-
and upper part of post-rift megasequence 1 are panded beyond the original confines of the basin,
shown in Fig. 11,45, indicating a complex of fluvial suggesting that it had begun the flexural subsidence
and lacustrine environments. It is not clear whether phase of development. The Gippsland Basin is ori-
these basins drained internally, or whether there ented NW -SE. A typical cross section is shown in Fig.
was a through-going axial drainage. Transgression 11,47. The basin was filled during the Late Creta-
spread a tongue of shallow marine deposits south- ceous-Eocene with 5800 m of nonmarine sediment
ward along the Viking Graben in the Sinemurian assigned to the Latrobe Group (Fig. 1l,48). Sediment
(Steel and Ryseth 1990), and this may have been the was shed into the basin from the north and consists
location for trunk drainage for part or all of the predominantly of coarse-grained, well-sorted sand-
Triassic, with the river system flowing northward stone, with minor fine-grained sandstone, siltstone,
toward the location of the present continental mar- shale, and coal. The basin was uplifted and deeply
gin. incised by fluvial channels late in the Eocene (Fig.
Gippsland Basin is located offshore on the conti- 11,48). Marine mudstones of Oligocene age trans-
nental shelf of southeastern Australia (Fig. 11,46). It gressed this erosion surface, and they provide an
is similar in structure and origin to the Viking-Cen- important regional seal.
tral Graben of the North Sea, having been inter- Kutei Basin in Borneo (Kalimantan) is a failed rift
preted as a failed rift, an arm of the Tasman Sea Basin associated with the short-lived sea-floor spreading
(Veevers 1984; Rahmanian et al. 1990). The basin of Makassar Strait (Hamilton 1979; Hutchison 1989).
commenced faulting and subsidence in the early The basin contains 16 km of Tertiary fluvial-deltaic
NW SE (Gonzalez 1990). Forearc basins with a dominantly
I...
marine fill, but including late-stage nonmarine sedi-
.... I
ments, include parts of the Great Valley and related
basins of the western United States (Nilsen 1987;
I Bartow 1987; Figs. 11.49, 11.50), Luzon Central Val-
ley, Philippines (Bachman et al. 1983), Xigaze basin,
Tibet (Einsele et al. 1994), and the East Coast Depres-
sion of North Island, New Zealand (van der Lingen
1982). Forearc basins with great thicknesses of
nonmarine deposits include the Western Trough of
Burma (Rodolfo 1975; Mitchell and McKerrow 1975;
Hutchison 1989), Cook Inlet, Alaska (Kirschner and
Lyon 1973; Figs. 11.51, 11.52), the Rosario
Embayment, Baja California (Fulford and Busby
1993), and the Old Red Sandstone (Devonian) ofthe
Midland Valley, Scotland (Mitchell and McKerrow
1975; Coward 1990).
Forearc basins are susceptible to tectonic modi-
fication at late stages during plate collsision. For
example, the Western Trough of Burma and the
Fig. ll.48. Schematic stratigraphic diagram of the Gipps- Midland Valley of Scotland were both characterized
land Basin, showing major tectonic events. (Rahmanian et by late-stage strike-slip faulting accompanying plate
al. 1990) collision events. The forearac basin that developed
over the Luzon arc prior to the collision with the
clastics deposited by the Mahakam River and delta, Chinese mainland in the Pliocene evolved into a
and these rocks constitute a significant petroleum collisional basin that filled primarily with marine
province (Verdier et al. 1980). sediment-gravity-flow deposits until it was uplifted
about 0.5 million years ago, at which point fan and
braided stream deposition began (Lundberg and
11.4.3 Convergent-margin Basins Dorsey 1988).
77.4.3.7 Forearc Basins

11.4.3.2 Backarc Basins
Many forearc basins are floored by oceanic crust and
are filled with marine sediments, typically turbidites These basins develop by crustal extension and ther-
and related facies (Dickinson and Seely 1979). mal subsidence in the backarc region. Commonly,
Nonmarine deposits occur where an arc extends they show a stratigraphic evolution that is the in-
onto continental crust, and here sediment thick- verse offorearc basins. Nonmarine sedimentation is
nesses may be considerable. In such settings the most typical of early stages of basin development,
basin typically is floored by accreted continental before crustal extension has led to significant sub-
rocks and is fed by active tectonic highlands of the sidence, with basins deepening through time. Rift
magmatic arc. The subduction complex, on the other basins in two backarc regions, the modern Aegean
side of the basin, may become uplifted and is a Sea and Cenozoic rifts of the Basin and Range, in the
significant sediment source late in basin develop- western United States (Fig. 11.49), formed the basis
ment. Early basin-fill sediments typically are ma- for the work on tectonic geomorphology reported by
rine, with nonmarine deposits forming the youngest Leeder and Gawthorpe (1987), Leeder and Jackson
part of the basin fill. Considerable thicknesses of (1993), and Leeder et al. (1991). This work led to
nonmarine sediment may be present. In Cook Inlet, proposals for generalized paleogeographic models
Alaska, the largely nonmarine Kenai Group is more that are discussed in Sect. 11.4.2.1 (Figs. 11.38,
than 7.6 km thick. 11.39).
Modern examples of nonmarine sedimentation in Two good examples of ancient nonmarine sedi-
forearc basins include the Pacific coast of Guatemala mentation in backarc settings are the mid- to Upper
(Kuenzi et al. 1979) and the Osorno Basin, Chile Carboniferous sediments of northern Britain and
the Jurassic Kayenta Formation of the Colorado A particularly well-studied example of a backarc
Plateau region. During the Namurian-Stephanian basin on continental crust is the Basin and Range
period northern Britain underwent crustal exten- province of the western United States and northern
sion behind the north-dipping Variscan subduc- Mexico (Fig. 11.49). This area consists of a series of
tion zone (Leeder 1982; Besly and Kelling 1988; tilted fault blocks and half-graben basins filled with
Coward 1990), resulting in the development of a up to 3 km of nonmarine upper Cenozoic sediment.
series of graben and half-graben basins. Initially, in Fault-generated relief between uplifted blocks and
the Dinantian, these filled with carbonates and the basement floor of the basins is 2-5 km, and the
basin-margin fan deposits. Later, during the mid entire area has been elevated by 2-3 km (Stewart
to Late Carboniferous post-rift phase, regional sub- 1978) and extended to at least double its original
sidence led to the development of a broad basin width (Hamilton 1987). The province is character-
occupying much of northern Britain, and this ized by an anomalously thin crust and high heat
filled with thick clastic deposits derived from the flow. It contains abundant early Cenozoic calc-alka-
hinterland, to the north (Besly and Kelling 1988; line and silicic volcanics, whereas younger volcanics
Fraser and Gawthorpe 1990; Leeder and Hardman are bimodal, including the voluminous Columbia
1990). These include important coal-bearing deltaic Plateau basalts, of Miocene age. Rifting began locally
deposits (the British "Coal Measures"), delta-top flu- in the Eocene, but the main phase of active extension
vial systems, and prodelta turbidites (Figs. 11.53, and basin subsidence did not commence until the
11.54). mid to late Miocene.
The Lower Jurassic Kayenta Formation of the Songliao Basin, China, may be interpreted as a
Colorado Plateau area developed in the backarc re- backarc basin. Klimetz (1983) interpreted the basin
gion between the Sierra Nevada arc and the Ances- as a backarc rift, but Burke and Sengor (1986) sug-
tral Rocky Mountains of the continental interior gested that the basin may be the product of "escape
(Middleton and Blakey 1983; Luttrell 1993). Again, tectonics" during convergence between Asia and a
the hinterland constituted a major sediment source. North China block (basin type 4.3).
Braided fluvial systems drained southwestward
from this source area into an arid basin center, the
Utah-Idaho Trough, which filled partly with eolian 11.4.3.3 Retroarc (Foreland) Basins
sands (Fig. 11.55). Ordovician fluvial-deltaic sys-
tems in North Wales constitute a third ancient ex- Foreland basins occur in three plate-tectonic set-
ample of an ancient partly nonmarine backarc tings (Miall 1995b). One class of basins, termed
system (Orton 1988). retroarc basins by Dickinson (1974), occurs behind
There are many Cenozoic to Recent examples of (on the continental side of) compressional arcs.
backarc basins in Indonesia and the Philippines. For Hamilton (1985) attributed crustal thickening and
example, Cagayan Basin on Luzon, in the Philippines subsidence in compressional arcs "to the gravita-
(Mathisen and Vondra 1983). In this basin, 900 m tional rise and spread of batholithic plutons."
of nonmarine volcaniclastic sediment accumulated Crustal shortening, in the form of fold-thrust devel-
during the Plio-Pleistocene. The basin deepened and opment, also occurs in compressional arcs (Jordan
widened during sedimentation as a result of increas- 1995). The second and third classes offoreland basin
ing relief of the arc and extensional faulting in the result from collision and suturing, including arc-arc,
basin. Sedimentary styles are similar to those occur- arc-continent, and continent-continent collision,
ring in the nonmarine aprons of some forearc and are described in Sect. 11.4.5. Some sedimentary
basins, such as in Guatemala. Other examples in- and tectonic features common to all classes of
clude the Meervlakte, Sepik, and Ramu basins foreland basin are described in Sect. 11.4.6.
of northern New Guinea, which contain thick Mi- Examples of retroarc basins with significant
ocene to Recent fluvial, peat-bearing successions nonmarine fills occur throughout the length of the
(Hamilton 1979). The Pannonian Basin of the Cordilleran-Andean ranges of the east Pacific conti-
Carpathian region in southeast Europe is another nental margin. They include the Sevier Basin of Utah
example of a backarc basin floored by continental and Idaho (Jordan 1981; Cross 1986) and many
crust, although Burchfiel and Royden (1982) and Andean basins (Jordan 1995), such as the Llanos
Royden et al. (1983) showed that microplate colli- Orientales and Putumayo basins of Columbia
sion and strike-slip faulting complicated the history (Salazar 1990; Stabler 1990), the Marafton and Beni
of basin development. basins of Peru (Dumont and Fournier 1994), the
!l
---..:..----
".·00' _00'
_00'
MONTEREY
FAN
5:1000'
.. ..
I
..
..
MIOCENE
100
100
110 ..
...
aD
a
Fig. 11.49. Miocene paleogeography of the western the forearc toward the Pacific Ocean. Nonmarine basins of
United States, showing nonmarine infill of Great Valley the Basin and Range province are also shown. (Cole and
and westward fluvial drainage from andesitic arc across Armentrout 1979)
IVOLCANlcsl
~ Basalt terrestrial
ILEGEND\ ~ Basalt marine
IMARINEI
r;.71 Andesite terrestrial
Andesite marine
Shoreline ~
~ Rhyolite te"eatrial
Rhyolite marin.
Strondllne deposits ~
\- ·.····=1 Neritic deposit.

[l] Venia
ICONTINENTAL I
1- -I Bathyal a Aby..al depOlIt.
~ Terre.trlal area, I. •.
• oure. area
~
~ Submarine fan depaait.
Alluvial fan depo.it.
~ Deltaic depo.lt.
Ir--.- r--.-I Continental Hdlmentatlon-
--- --- fluvial, lacu.trlne, etc.
~
LlllJJ
Convergence Zane
Ir'l FlUVial depo.ltl
~
~ Paleobathymetry In mete,. 8alln, Ran ... e a Allocl-
ated valcanllm and
Hdlmentatlon
Fig. 11.49b
Cuyo and San Jorge basins of Argentina (Stabler facies transition occurs in reverse toward the east,
1990), and the Magellanes Basin of southern South away from the mountainous source. Syndeposi-
America (Biddle et al. 1986). The Bermejo Valley of tional synclines developed progressively further
northwest Argentina is a small but well-studied ba- from the source, in part as a response to sediment
sin (Johnson et al. 1986; Jordan et al. 1988; Stabler loading in shifting depocenters. Jordan (1981)
1990; Jordan 1995). showed that the isopach pattern of mid-Cretaceous
The Sevier Foreland Basin spanned the Jurassic to strata could be explained by a model of flexural
Early Tertiary, but the main phase of rapid subsid- loading by thrust sheets. Campanian-Maastrichtian
ence occurred during the Albian-Santonian, during isopachs, however, do not correspond to a typical
a phase of rapid convergence between North foreland-basin pattern, and Cross (1986) related
America and the Farallon Plate (Dickinson 1981a; them to a change in subduction patterns and sub-
Jordan 1981; Fouch et al. 1983; Cross 1986). Weimer crustal thermal events, the details of which are be-
(1970) showed that the Cretaceous section under- yond the scope of this book.
went a crude upward coarsening and a facies transi- The Sevier Basin was disrupted by the basement-
tion from marine mudstone and limestone to involved tectonics of the Laramide Orogeny in the
nonmarine sandstone and conglomerate. The same Late Cretaceous-Paleogene, and became segmented
w E
~\:)JSierra Nevada detritus
. .. . . . . .-=======~30km
O
l:;H::~.;,l Coast
"" · ... 1 ·
Range detritus
100 ,·:_-;.-<-::1Lacustrine clay
..
.....
-
"-
III
III
E
c a
-..
0
>
III
W
IIG""'.",, Contact-- hachuree

denota burled 80lla
or oxidized horizon.
-100
Cut and fill --

diagramm a tic
Fig. 11.50. Diagrammatic cross section through the nor- and the Sierra Nevada arc and differences in the composi-
thern San Joaquin valley, showing stratigraphic relation- tion of detritus dervied from these two source areas make
ships of Quaternary alluvial deposits. Differences in correlation of units across the valley difficult. (Bartow
tectonic history for the Coast Range subduction complex 1987)
basins. The basins are entirely nonmarine, contain-

ing thick synorogenic fluvial deposits, commonly
preserving proximal fanglomerates, and lacustrine
successions. Drainage is mainly internal, although
Dickinson et al. (1988) summarized evidence for
spillage between basins and possible drainage east-
ward into the continental interior and westward
across the Sierra Nevada magmatic arc to the Pacific
Ocean. There are many paleogeographic similarities
between these basins and the hinterland basins of
the Himalayan orogen (Sect. 11.5.5.2).
Two studies of basins adjacent to the southern
Andes serve as examples of retroarc foreland basins
developed in the absence of collisional events. The
Neogene Bermejo Basin of northwest Argentina con-
Fig. 11.51. Generalized isopadh map of Kenai Group (Oli- tains a crudely coarsening upward succession of flu-
gocene-Pliocene), Cook Inlet, Alaska.(Kirschner and Lyon vial siltstone, sandstone, and conglomerate. Johnson
1973, reprinted by permission) et al. (1986) and Jordan et al. (1988) discussed the
uses of various facies and stratigraphic criteria in an
into a series of smaller basins separated by fault- attempt to relate sedimentation to tectonic events
bounded uplifts (Dickinson et al. 1988). Ingersoll and to develop an accurate picture of the subsidence
(1988) classified these as foreland intermontane ba- history. Biddle et al. (1986) provided a stratigraphic
sins. Jordan (1995) refers to them as broken foreland synthesis of the Magallanes Basin of southern
NW
SE
20001 FT.
BASE KENAI
GR. UNCONFORMI TY. o OIL PRODUCTION Depocenler KENAI- CHUGACH MTS.
...- - - - - .... PPROX. 60 MIL E S - - - - - - - -... o GAS PRooucrlON SOURCE PREDOMINANT
Fig. 11.52. Schematic cross sections through the Lower Kenai Group (left) and Upper Kenai Group (right). WF West
Foreland Formation; H Hemlock Formation; T Tyonek Formation. (Kirschner and Lyon 1973, reprinted by permission)
Fig. 11.53. Paleofacies map for the Kinders-

cout-Crawshaw delta system, central and north-
ern England. The landmass to the south is the
10 60 10 10 100 .. ",
London-Brabant massif, a microcontinent lo-
• ! ! •
cated to the north of the Variscan subduction

zone. (Fraser and Gawthorpe 1990)
Dtnantlan
r:--"'"' sua ·MAA INE

FAN
~ DEL rA fOP We Tl AflDS
~ SANDSTONES
r==1
~
0 XVGENA TEO SHALE
AHOxtC SHAlE
r---:7
~
CMANNEt.ISED
SANDSTONES
#. CHANNeL Bti:!.. T
Fig. 11.54. Schematic block diagram for Carboniferous extensional basins of the southern North Sea. (Leeder and
Hardman 1990)
_. _.. _.. -
~ PreCQmbrlQn bOSfmfnt core
- .. -.- .- .. - .. -. .,
E!J Volca!ues
A •
o Sedimentary 'errorn
~Bra'ded stream,; Intermlfte"t

7" 10 perennlQI (Assoe t a 2)
r -
Fig. 11.55. Paleogeographic synthesis of the Kayenta Formation (Lower Jurassic), southwest United States. The main
outcrop area of the formation is in western Colorado, southern Utah, and northeastern Arizona. (Luttrell 1993)
Argentina and Chile. This basin developed during developed in the Miocene. Regression occurred
the mid-Cretaceous and Cenozoic above a collapsed everywhere during the Neogene, and these basins are
marginal basin during a major phase of Andean now largely continental. Basement rocks contri-
igneous activity and crustal shortening. The basin buted sediment to the basins at first, but contempo-
fill is primarily marine shale. Coarse sediments de- raneous volcanics became the main sediment source
rived from the Andean uplift are restricted entirely during the younger Neogene. Throughout this pe-
to the Andean margin of the basin. riod the magmatic arc gradually migrated north-
Another series of retroarc basins occurs behind ward. Syndepositional folding commenced first in
the Sunda arc in Sumatra, Java, and the Sunda Shelf the south and is probably continuing, resulting in
(Ben-Avraham and Emery 1973). The foreland ba- deformation that decreases in intensity northward.
sins of Sumatra and Java include the Madura, East Hamilton (1979) attributed the folding to lateral
Java, Tjeribon, West Java, and South Sumatra basins pressures of magmatic intrusion. The sedimentary
(Hamilton 1979; Hutchison 1989). Sediment thick- basins probably owe their origin to downbowing of
nesses reach maxima of about 3-6 km in these ba- the crust beneath the weight of volcanic and plutonic
sins. The Sunda and part of the West Java Basin are rocks in the magmatic belt. A fold-thrust belt has not
graben basins bounded by north-south faults. These yet developed here. The geology of the Sumatra
faults were active during the Cenozoic, but their foreland is similar. Basement control by north to
trend suggests basement control. The West Java- north-northwest trending structures had an impor-
Tjeribon and Madura basins are separated by a base- tant influence on sediment thicknesses and facies
ment high. Sedimentation there began with marine changes. Miocene strata include quartzose deltaic
transgression during the Oligocene. Bathyal condi- sandstones derived from the Malay Peninsula and
tions persisted until the later Neogene in the east, but Sunda Shelf to the north. Drainage in these basins is
thick deltaic clastics accumulated in the West Java typically transverse.
Basin in the Oligocene. A widespread reef limestone
11.4.4 Basins Formed Along Strike-Slip Faults 11.4.4.1 Basins Associated

with Intracontinental Transform Faults
As shown by Miall (1990, Table 9.6), strike-slip faults
occur in a variety of plate-tectonic settings. Of The San Andreas fault system in California and the
concern here are those which constitute intraconti- Alpine fault of New Zealand are the best-known
nental plate-boundary transform faults and the intracontinental transform faults, and much of our
boundaries of divergent plate margins. In both set- knowledge of fault tectonics and basin style is de-
tings, basins with significant nonmarine fills occur. rived from studies in these areas.
Plate boundaries undergoing oblique convergence Ridge Basin, near Los Angeles, is a well-exposed
may also undergo strike-slip displacement, with the example of a basin developed by supracrustal load-
offset typically localized along the axis of the ing in a transpressional setting, and excellent studies
magmatic arc. The Sunda Arc of Sumatra offers an by Crowell (l974b), Link and Osborne (1978),
example of this, and the forearc and backarc basins Crowell and Link (1982), and Link (1984) have
of this area show the influence of wrench tectonics, turned it into a classic of its kind. The basin was
as described by Hamilton (l979) and Hutchison active during Miocene and Pliocene time, when the
(l989). Another important class of strike-slip-fault- San Gabriel Fault was the master fault (it is now
bounded basin is that which occurs as a result of inactive). Salton Trough, at the head of the Gulf of
escape tectonics in the hinterland regions adjacent California, is an example of a pull-apart basin. It was
to continental sutures. These basins are discussed in initiated during the Miocene and is still active, so it
Sect. 11.4.5.2. has not been uplifted and dissected for our examina-
Following the work of Luyendyk and Hornafius tion as has the Ridge Basin. Volcanism, heat flow
(l987), Ingersoll (l988) proposed the definition of a data, and tectonism in the Salton Trough suggest
class of transrotational basins developed by block that it is an incipient spreading center (Crowell
rotation around vertical axes within strike-slip fault 1974a) and may eventually open fully to expose new
zones. Such rotation causes thrust faulting where oceanic crust, as in the Gulf of California. This would
blocks overlap and basins where blocks are rotated accord with what appears to be a pattern of oblique
apart. Luyendyk and Hornafius (1987) demon- spreading in the area (a "leaky transform"). Based
strated that many Neogene basins in southern Cali- on several Californian basins, Crowell (l974b) pro-
fornia are of this type. posed a general model of pull-apart structure and
It is important to realize that strike-slip motion stratigraphy as shown in Fig. 11.56.
may, in fact, be a superficial representation of deep
crustal processes that show quite different structural
style. For example, the construction of the
Carpathian Mountains during the Neogene involved
differential movement of nappes along flat thrust
irregular bosin mor~'n
faults. These nappes, which moved at different Irregular normal- slip faults
times, are separated in places by steep faults, which stretched and attenuated :.t.
show a history of strike-slip movement where marginal floor ~ "/
the nappes were displaced relative to each other. complex unconformities ~
~?lpUIl-apartl
The Vienna Basin is a pull-apart basin that was cre- and overlaps
In subsurface r, f--"', _
ated in this way (Royden 1985). Conversely, it has complex ----.. -E \
now been shown that the Hornelen Basin of the oldest Intersechon ~ ~ 1 C;..-J
intoct / 1 -,.;
Norwegian Caledonides, long cited as a classic ex- basin fill / / \ \'-/
remnants of marQinol '\. .... \.";
ample of a strike-slip basin (e.g., see Steel and rocks within Yolcanics
J
\..I
Gloppen 1980) is soled by a low-angle extensional Talus breCCias and rapid Ibasinl
focles changes bosinward
detachment surface. This surface is scoop shaped.
Where the edges of the scoop reach the surface, Irregular basin
margin - folds i
extensional movement shows a strike-slip motion result of convergence
relative to adjacent undisturbed strata (Hossack between boundary
right -slip faulls
1984; Seranne and Seguret 1987). May et al. (1993) right"slip fault continues On
have demonstrated that the San Gabriel fault, to southeast
bounding the Ridge Basin, is also a low-angle de- Fig. 11.56. A general model for pull-apart basins. (Crowell
tachment fault. 1974b)
sw NE
EXPLANATION
Ip~~ I Breccia
~ Conglomerate
0.' , Sandstone
•
§- . Siltstone
Shale (Mudstone)
1/-,1
/ , Basement Rocks
THICKNESS
17,000 4500
4000
10,000 3000
2000
5000
1000
Feet Meters
0 2 3 Km.
I
!
I
0 2 Miles
Horizontal Scale
Fig. 11.57. Stratigraphic cross section through Ridge Basin. (Crowell and Link 1982)
Ridge Basin contains more than 12 km of marine, 77.4.4.2 Basins Associated with Divergent Plate Boundaries
lacustrine, and fluvial sediments that show numer-
ous rapid local facies changes (Fig. 11.57). The San The mid-Atlantic spreading center includes long
Gabriel Fault bounded the basin to the southwest. It stretches of ridge with few transform offsets, and
was the main active fault of the San Andreas trans- zones of short spreading segments with long trans-
form system during sedimentation and is flanked by form offsets (Wilson and Williams 1979). During the
the Violin Breccia, an alluvial fan wedge 11 km thick. early stages of continental separation, in the Juras-
Much of this great thickness is only apparent and sic-Cretaceous, the transform offsets functioned as
reflects offlapping of the basin-fill to the southeast as intracontinental plate boundaries, and important
the basin subsided and enlarged in that direction. nonmarine basins developed. Transform offsets
The Liebre Fault Zone, on the northeast side, shows commonly continued into the continent as transfer
the typical splayed or flower pattern of strike-slip faults, along which significant dip slip and strike slip
fault zones (Fig. 11.57). Uplift on this fault was inter- occurred. Rift and transtensional basins with non-
mittent and localized, and generated local, short- marine sedimentation were the result. Examples in-
lived basement sediment sources. The basin center is clude the following:
filled with lacustrine mudstones, fluvial-deltaic
sandstones, and minor conglomerates deposited by 1. Central Tertiary basin, Spitzbergen, now bound-
longitudinal and transverse depositional systems ed to the west by the Hornsund Fault and initially
flowing from the north and northwest. formed as a transtensional basin in the zone sepa-
Nilsen and McLaughlin (1985) compared three rating Svalbard from Greenland. The Early Pale-
nonmarine basins bounded by strike-slip faults: ocene Firkanten Formation includes a basal
Ridge Basin, Little Sulphur Creek Basin in northern fluvial-deltaic, coal-bearing member, followed by
California, and Hornelen Basin in Norway. Despite marine deposits (Steel et al. 1981, 1985).
the differences in scale, the basins show remarkably 2. The rifts of Central and West Africa developed by
similar paleogeographic patterns (Fig. 11.58). All dip-slip rift faulting and transtensional strike-slip
three are characterized by rapid lateral facies faulting and include major nonmarine succes-
changes and rapid subsidence. sions. The Benue Trough, Nigeria, is controlled
HORNELEN RI DGE LITTLE SULPHUR

CREE K
EXPLANATI O N
....
~
l andi ll de, and debrIs-
Ta l u~,
flow-dom inated alluv i a l fanl
---=:....
......,..--
Lateral d l recllon~ of
movement , ~tr l ke-~lIp fault
-
Normal or high-ang l e reverse
~
·t·· ·.
Streamflo w - do ml nated
alluvia l fans ~ f aull , barbs on downdropped
I.
block
- -
Direct ion of sed iment tranlport
~ Lacustri n e depoaltl along ma j or Itr lke - sl ip margin
LZJ A ll uvial pla in and

fan - delta deposits ;f Di rection 01 ma j or led iment
transport along other margin
rn
of bal l n
Dl ractlon of sed iment tranl -
.-. Mar i ne turb i dite. port In' b.lln-ax ls reg i on
A'
Approx l ma" strike
and dip 01 bedding <rV Thrust fault, tr i ang l es
on upthrown block
++ /
Genera l trace of d i pp i ng
Ax is and plunge 01 .)lncllne
- -' strata
Fig. 11.58. Comparison of three strike-slip-fault-bound- California, is 13 km long. SGS San Gabriel fault; SAP San
ed basins. The three basins are drawn at different scales. Andreas fault;MP Maacama fault. (Nilsen and McLaughlin
Hornelen Basin, Norway, is 70 km long, Ridge Basin, Cali- 1985)
fornia, is 40 km long, and Little Sulphur Creek Basin,
structurally by landward extensions of the conglomerates and basin-center braidplain de-

Romanche, Chain, and Charcot fracture zones. posits (Tankard et al. 1982; Fig. 11.59).
Nonmarine sediments range from Albian to
Cenozoic age. For example, the Benin Formation
(Eocene-Miocene) is a succession of continental 11.4.S Basins Related to Plate Collision
sandstones more than 2 km thick (BenkhelilI982;
Fairhead 1986; Shannon and Naylor 1989; Genik Nonmarine deposits occur in two classes of colli-
1993). Additional discussion of these rifts is pro- sional basin, peripheral foreland basins and hinter-
vided in Sect. 11.4.2.1. land basins, as defined below. A third class of basin,
3. Basins of southeast South African continental remnant ocean basins, are floored by oceanic crust
margin, of which the Algoa Basin is the largest. and therefore do not accumulate nonmarine sedi-
Paleozoic structural lineaments oriented at an ments.
acute angle to the continental margin were reacti- There are two distinct types of foreland basin
vated during the Late Jurassic by shear accompa- associated with plate collision (Miall 1995b). First,
nying the transform separation of Africa from the collisional retroarc basins are hinterland basins,
Falkland Plateau. A series of transtensional rift located on the overriding plate, behind the arc. At
basins developed, filled by coarse alluvial-fan certain stages of its development, the Alberta basin is
m
o
::: (4)
o km 100
~ ConQlomerote, subOrd lnote sandstone I
D-··'
~ Siltstone, and mudstonl' lenses
: : : .', .' : Sandstone, with IHists of conqtOmerole

' ci~~~' and mudstone
E:I~.g 'SllIstone
50
[ ~ ~ ~3 Mudstone, sha~
MOfIl"lf Invertebrate fossils, Including
ommonltes, bivalveS, Qos1rOOOds
Logs, fol iage

--
Direction of Clecreos1nQ groin size
--
--
Fig. 11.59. Stratigraphic sections in the Gamtoos Basin, a basin-margin fanglomerate succession; sections 2-4 are
rift basin developed in the Late Jurassic during the strike- through finer-grained basin-center deposits. (Tankard et
slip offset between southern Africa and the Falkland al. 1982)
Plateau. Section 1 is through the Enon Conglomerate, a
a good example, and some of the interior basins of Distinguishing the two classes of collision-related
western China, located behind (north of) the colli- foreland basin from each other and from retroarc
sion mountains of Tibet, are also of this type. These basins (Sect. 11.4.3.3) in the ancient record may be
are described in Sect. 11.4.5.2. The second type of difficult because most orogens undergo several
collision-related foreland basin was termed the pe- phases of accretion, changes in subduction polarity
ripheral basin by Dickinson (1974) and is described and changes in convergence angle, leading to such
in the next section. Peripheral basins are initiated by complications as strike-slip displacement of a basin
subduction and loading of the subducting plate in a and its source area(s) and the superposition of ba-
forearc setting. They overlie, adjoin, and may be sins controlled by different plate-tectonic mecha-
overriden by the subduction complex of the arc and nisms. Some features common to all types of
overlap the marginal sediment wedge of the sub- foreland basin are described in Sect. 11.4.6.
ducting plate.
11.4.5.1 Peripheral Foreland Basins the development oflarger peripheral basins. For ex-
ample, during the late Cenozoic, the Luzon arc col-
The western Pacific offers several examples of pe- lided with the Chinese mainland, creating the island
ripheral basins formed by arc-arc collision. For ex- of Taiwan, the western part of which constitutes a
ample, a series of small peripheral basins developed fold-thrust belt and associated peripheral basin
in central Honshu, Japan, as a result of the collision (Covey 1986; Teng 1990).
of the lzu-Bonin arc with Honshu during the late Many of the world's largest and best-known
Cenozoic (Ito and Masuda 1986). Fan deltas and foreland basins (Himalayan foredeep, Appalachian
submarine fans are the dominant types of deposi- basin, Swiss Flysch-Molasse basin, Adriatic fore-
tional system. Arc-continent collisions may lead to land, and Persian Gulf) resulted from continent-
ULTRAHELVETlC
FLYSCH
0
EUROPEAN MARGIN' • 10
E
;(:i~; ; ~t.iljiti lj1tf,~:~i~t.~ ,*f.~}j'i!; "!;, ,\; V!t!liihW;':i~~

20
:!!.
a •SOMa
30
40 .
J:!
ii
0
50
60
--.,
THINNING OF AUSTRo-ALPINE SHEET
BY EXTENSION-LITTLE EROSION
~~~~~~~~~~~~~Y.1'"7~~~0
b ' 38Ma 40
50
60
INSUBAIC PI'IASE
BACKTHAUSTING & FOLDING
--.
2mmlyr (30-25Ma)
..
EXTENSIVE
EROSION
LOWER FRESHWATER
MOLASSE
~ .. . . .... .
'
30
40
50
60
Approximate
Present·Day
Erosion Surface
30
20
d :Present -day
~~~~~~~~~~~~~~~;;~~~~~: 20
30
40
50
60
Fig. 11.60. The structural evolution of the Alpine thrust deposited in the mid to late Oligocene, during the "conti-
wedge and the North Alpine foreland basin, showing the nental wedge" phase of development. (Sinclair and Allen
shallowing of the basin as the accretionary wedge climbed 1992)
the ramp onto the continental margin. The molasse was
continent collision. Typically, these basins form control provided by magnetostratigraphy (e.g., see
above, and rest on, the extensional continental mar- Sects. 11.2.1, 11.3.2, 11.3.6).
gin of the downgoing plate that developed during the The early "collisional facies", the marine Murree
preceding cycle of continental rifting and sea-floor Formation (Cretaceous-Eocene) passes gradation-
spreading. For example, the Himalayan foredeep ally up into the Siwaliks, an entirely nonmarine unit
basin is formed above the Paleozoic through early locally more than 6 km thick (Burbank et al. 1986;
Cenozoic Tethyan continental margin of the Indian Burbank and Raynolds 1988; Burbank and Beck
subcontinent (Gansser 1964). Development of the 1991; Burbank 1992: Fig. 11.61). Depositional envi-
peripheral basin may involve several distinct but ronments were similar to those existing at the
diachronous phases, as first the arc on the overriding present day, namely, giant- alluvial fans draining
plate undergoes compression, and later the main transversely from the rising mountains into longitu-
mass of continental crust reaches the subduction dinal trunk rivers, such as the Indus, Ganges, and
zone. Typically, the basins become shallower with Brahmaputra. Indeed, the major north-south rivers
time, as the colliding arc climbs the ramp formed by are regarded as type examples of antecedent drain-
the subducting continent, and the basin fills with age (Holmes 1965; Seeber and Gornitz 1983). They
sediment (Cant and Stockmal 1989; Stockmal et al. have maintained their course, perhaps since the
1986, 1992; Stockmal and Beaumont 1987; Sinclair Miocene, while cutting through ranges that are now
and Allen 1992). Nonmarine deposits, the typical higher than their source. The Siwaliks have been
"molasse" (Van Houten 1981), constitute the termi- divided into three units. The Lower Siwalik com-
nal basin fill (Figs. 11.24, 11.60). prises a coarsening-upward megacycle consisting of
The Indus-Ganges trough of India and Pakistan sandstone-mudstone alternations passing up into a
is an excellent modern example of a peripheral predominantly sandy sequence. The Middle Siwalik
foredeep basin. It is underlain by the Siwalik Group, consists mainly of medium to coarse sandstones
a molasse-type deposit of Miocene-Pliocene age ac- with interbedded mudstones, and it grades up into a
cumulated in conditions apparently identical to largely conglomeratic Upper Siwalik, the Middle and
those of the present day (Johnson and Vondra 1972; Upper Siwaliks together comprising a second mega-
Parkash et al. 1983). The Siwaliks have been much cycle. Fining-upward cycles 1-35 m thick occur
studied in recent years for the information they yield throughout the sequence. The three subdivisions of
on rates and styles of sedimentary and tectonic pro- the Siwaliks are time transgressive, representing
cesses, information that can be obtained from the southward and southeastward progradation of flu-
rocks because of the excellent chronostratigraphic vial depositional systems into and down the axis of
Fig. 11.61. The Himalayan peripheral foreland basin, showing major drainages, and isopach of Miocene-Recent depos-
its. Major faults are indicated by vertically hatched lines. MBT, main boundary thrust. (Burbank 1992)
the trough. The Siwaliks are folded and faulted near beginning in the Oligocene and not yet completed
the Main Boundary Fault but pass out into flat-lying (Ricci Lucchi 1986). The basin is asymmetric in cross
beds overlain conformably by modern alluvium. The section, with the deepest part of the basin located
alluvium itself is tilted near the mountain front, and adjacent to the Apennine thrusts, in the south. The
Landsat images indicate active folds in parts of the basin has been suplied with terrigenous sediment
trough. Petrographic studies oflight and heavy min- from the northeast, northwest, and southwest sides,
erals and of the conglomerate clasts document the with the largest nonmarine depositional system con-
progressive and rapid unroofing of Tethyan rocks stituting the axial Po River drainage system (Fig.
that were metamorphosed deep in the collision zone 1l.62). Quaternary nonmarine deposits in this basin
(Cerveny et al. 1988). are locally more than 400 m thick (Ori 1993). They
The Swiss molasse basin was initiated as a periph- consist of three main depositional systems (Fig.
eral basin, extending from France to Austria. Its 1l.63): (a) the axial meander belt of the Po River,
strongly deformed basin fill is exposed in the west- up to 25 km wide; (b) transverse drainage from the
ern Alps of Switzerland and France (Van Houten Alps to the north, including large gravel alluvial fans
1981; Homewood et al. 1986; Crumeyrolle et al. 1991; and sandy braided and meandering tributary rivers;
Sinclair and Allen 1992; Fig. 1l.60). The molasse is and (c) relatively small fans and tributary rivers
subdivisible into several marine and freshwater forming an alluvial apron flanking the Apennines to
units reflecting the regional tectonic history. In the the south. These deposits contain intraformational
Digne Basin of France more than 2 km of molasse unconformities, attesting to their syndepositional
deposits are preserved. In Switzerland the basin fill origins.
reaches at least 5 km. The nonmarine part of the A final example of a peripheral basin containing a
Molasse is dominated by thick conglomerate units, significant nonmarine sedimentary fill is the
termed the Nagelfluh. As noted in Sect. 1l.3.2, axial Mesopotamian plain. Baltzer and Purser (1990) clas-
drainage developed during two distinct phases of sified rivers in this basin into four types (Fig. 11.64):
basin evolution, separated by marine transgression. Type 1 - Consequent streams drain active anticlinal
The axial rivers showed opposite drainage directions slopes of the Zagros fold-thrust belt and develop
during the two phases of basin development, reflect- small alluvial fans having low preservation potential.
ing variations in basin subsidence patterns (Fig. Type 2 - Axial rivers flow along the synclines and
11.24). drain into the Gulf where structural trends are ori-
The Adriatic foredeep of northern Italy and the ented obliquely to the present coastline, indicating
Adriatic Sea developed as a result of collision be- axial plunge to the east. Deltas form at the mouths of
tween the Apennines (a spur of Africa) and the Alps, these rivers. Type 3 - A few major antecendent rivers
o Older moraine
o Alluvial fan
D Tributary river
D Main river DeQositional s~stems

D Freshwater wetland
50 Km
D Salty wetland
B
Fig. 11.62. Depositional systems of the modern Po River, Italy. (Ori 1993)
S N
100
51
....
100
200
?
JOO
? Fan
m
deltas
IO,p.1Alluvia l Ian body Downwarping
o
o km 10 0 0-0 r:::t:~:~4
~"'-~"
Main nver sand bod y
1: ::::.1 Tribulary rover sand body Fine grained sedimenl5
.......... Upper li m.1 01 m arLne mlluenCe$

......
Fig. 11.63. Schematic stratigraphic architecture of the Po Basin deposits. (Ori 1993)
Strike-slip faulting is particularly characteristic of

late-stage adjustments to plate collision. Both trans-
pressional foreland-type basins and transtensional
pull-apart basins are common in such settings.
Retroarc-type foreland basins may also be present.
Because these basins are located mainly within con-
tinental plates, they are dominated by nonmarine
depositional systems, and because they are sur-
rounded by tectonically active mountain belts, sub-
sidence and sedimentation tend to be rapid. The type
example of a region dominated by escape tectonics is
southeast Asia, the late Cenozoic geology of which is
dominated by the effects of the collision between
Fig. 11.64. The four main types of river in the Persian Gulf
India and Asia (Tapponnier et al. 1986). Brookfield
foreland basin. (Baltzer and Purser 1990)
(1993) discussed the complex sequence of tectonic
and stratigraphic events that has occurred in the
cut across the fold-thrust belt, linking several elon- Himalayan orogen in the India-China-Afghanistan
gate drainage basins occupied by the first two types area since the collision of India with Asia in the
of river. Type 4 - The Tigris and Euphrates rivers Eocene. The tectonics and sedimentary basins of this
occupy the main structural depression of the region are illustrated in Figs. 11.65 and 11.66. Colli-
foreland basin and form the Mesopotamian Plain sional retroarc basins include Tarim and Jiuquan
and the large deltas at the head of the Gulf. Baltzer basins, while important transtensional pull-apart
and Purser (1990) described the modern deposi- basins include the Baikal and Shansi rift systems and
tional systems of these rivers, but little or no infor- major basins in the South China Sea and the Gulf of
mation is available on the stratigraphy of the Thailand. Nonmarine sediments predominate in all
deposits formed by them. these basins.
Most orogens contain the evidence of late-
stage strike-slip adjustments, with the presence of
11.4.5.2 Hinterland Basins large to small, mainly nonmarine, intermontane
sedimentary basins. Examples include the Cord-
In large orogens, crustal shortening may be accom- illeras of Canada (Long 1981; Eisbacher 1985;
modated by sideways squeezing of major crustal Ridgway and DeCelles 1993), the late Cenozoic and
blocks, with the development of strike-slip faults in Recent development of Turkey (Hempton and
the foreland or the hinterland or both. This has been Dunne 1984; Sengor et al. 1985; Dewey et al. 1986),
termed escape tectonics (Burke and Sengor 1986). the Devonian "Old Red continent" formed by the
o· 130· 150·E •
-
7
,.": "': "'.: ~..,
fi&lJ C,I'IIOIO ' C: •• tenalon
oc: •• n1c CfU..'

o. South ettln. Sea
.nd Andam.n Se.
60
. .
s so'
\ , o·
50
.
o·
30
O'
20
I..,,,· 6
~
10
..
~
..
O' 20
Fig. 11.65. Generalized tectonic map of east Asia, showing movements. Numbers refer to extensional phases: 1, 50-
the effects of the India-Asia collision. Large open arrows 17 Ma; 2, reprinted by permission, 17 Ma to present; 3,
represent major block movements since the Eocene. most recent, including probable future movements.
Smaller solid arrows indicate direction of extensional (Tapponnier et al. 1986, reprinted by permission)
amalgamation of North America and Europe (Dewey Dewey 1982), which provided the basins and sedi-
1982; Friend 1985; McClay et al. 1986; Harris and ment sources (e.g., Paradox Basin, Uncompahgre
Fettes 1988; Coward 1990), the Devonian-Carbonif- Uplift) for the Cutler Group of New Mexico and
erous (Acadian-Alleghanian) orogen of eastern Colorado (see Eberth and Miall 1991). Stanistreet
Canada (Hamblin and Rust 1989; Gibling et al.1992), (1993) compared the history of the Precambrian
and the late Paleozoic Ancestral Rockies and related Witwatersrand Basin of South Africa with the mod-
structures formed by the oblique collision between ern evolution of the Maracaibo Basin of Venezuela
North and South America (Kluth and Coney 1981; and attributed both to escape tectonics in response
FAULT THRUST WRENCH INFERRED FOLD BASIN

FAULT FAULT FAULT AXIS
Fig. 11.66. Basins and major structures in west China and Jiuquan; II Turpan; III Junggar; IV Qaidam; V Tarim; VI
adjacent areas of Russia, Afghanistan, and the Indian sub- Fergana. Faults: 1 Karakoram; 2 Talaso-Fergana; 3 Ertix; 4
continent. Many of the basins are bounded by fold-thrust Changajn; 5 Borohoro; 6 Altun Shan. (Liu 1986)
belts and have the character of foreland basins. Basins: I
to lateral compression within colliding continental orogen of west China, basins may become isolated
blocks. from the world ocean system within surrounding
Complex orogens undergo changes in conver- mountain ranges as a result of regional collisional
gence vectors and multiple terrane-accretion events events (e.g., Tarim, Jiuquan, Qaidam basins: Fig.
during progressive crustal shortening, and this may 11.66), leading to centripetal drainage, changes from
lead to extremely complex stratigraphic and struc- marine to fresh water, lowering of sea level, anoxic
tural histories in hinterland basins. Tectonic styles events, and changes in climate (Sect. 12.12), possibly
may change during basin development, and the including basin evaporation, causing salinity crises
result may be a hybrid basin that is difficult to clas- (Ricci Lucchi 1986). Paleo flow patterns may undergo
sify by plate-tectonic origin. The progressive (dia- major changes, even reversals, as a response to basin
chronous) collision of irregular margins and the tilting, diachronous closure, or the influence of reac-
reactivation of inherited basement structures may tivated basement structures.
lead to repeated pulses of loading and subsidence.
The simple rheological responses predicted by the
geophysical models are then overprinted by the local 11.4.6 Structural and Stratigraphic Patterns
responses to inherited structural anisotropies. Rela- Common to Foreland Basins
tive sea level and sediment source relief are affected
by the resulting tectonism on both local and regional The crustal shortening, thickening, and loading that
scales, with the possibility for developing very com- leads to the development of foreland basins causes
plex patterns of intersecting clastic wedges (e.g., an expulsion and cratonward migration of pore flu-
Ricci Lucchi 1986; Hsu et al. 1990; Cluzel et al. 1990; ids. Earlier-formed faults become locked, and subse-
Brookfield 1993). Within large orogens, such as the quent faults tend to step outward further into the
Mediterranean-Alpine region and the Himalayan basin (Dahlstrom 1970). Syndepositional tectonism
is common, with the development of growth folds, (longitudinal) flow is discussed in Sect. 11.3.2. In
intraformational unconformities (Anadon et al. basins receiving a large sediment supply from the
1986; Sect. 11.2.1.1), and cannibalism of older sedi- fold-thrust belt, the transverse drainage may extend
ments by younger alluvial systems. A wide variety of across the entire basin. Such basins are said to be
basin-fill patterns can develop, depending on the overfilled (Sinclair and Allen 1992; Jordan 1995). In
configuration of the continental hinterland. Trans- contrast, axial drainage may take place along the
verse drainage systems may develop during times of deepest part of the basin. It collects and funnels the
high sea level, when the axis of the basin is occupied
by the sea. During times of low sea level the basin
axis, between the fold-thrust belt and the forebulge, EJtTRASASINA L
may be occupied by a major longitudinal trunk river. STREAMS
The Alberta Basin (which is a hinterland foreland

basin) showed both these conditions during the Cre-
taceous and Cenozoic.
Axial drainage is particularly important in basins A
that develop by oblique closure, with the migration
of the suture longitudinally and the transportation
SUBSIDENCE
of detritus away from the locus of maximum com-
pression and toward any remnant marine basin.
Modern examples of this include (a) the Adriatic EA:OSJOH
foredeep, undergoing filling by the Po river system; lw ,
RSWORKED OEPOSI T S
(b) the Persian Gulf, filling from the northwest end

by the Tigris and Euphrates systems; (c) the Hima- THRUST
layan foredeep; and (d) Huon Gulf and Markham

Valley in Papua-New Guinea, which are the deep-
B
water and nonmarine components, respectively, of a
small peripheral basin currently developing by
diachronous closure of a small ocean basin to the Fig. 11.67. The alternative, "antitectonic" model for fore-
land-basin development. A During active thrust loading
east (Crook 1989; Brierley et al. 1993). gravel deposition is limited to the most proximal parts of
As shown in Sect. 11.3.6, the timing relationship the basin. B Gravel is transported far into the basin during
between tectonism, subsidence, and nonmarine postorogenic tectonic rebound, when the proximal part of
sedimentation in foreland basins is variable, de- the basin may undergo erosion. (Heller et al. 1988)
pending on the rigidity of the underlying crust and
the magnitude and grain-size distribution of the
sediment supply (including that introduced by axial
drainage). Two contrasting models represent end-
member conditions, the conventional "syntectonic"
model of gravel progradation occurring contempo-
raneously with tectonism, and the "antitectonic"
model, in which gravel progradation postdates tec-
:2Fb
~~
u
:E
time--
tonism during isostatic uplift and unroofing of the - overfilled

fold-thrust belt. The latter model is illustrated in Fig.
11.67. Climatic factors may also be important, in- B.
cluding the amount of rainfall, which controls the
rate of sediment dispersal. Uplift of the fold-thrust time--
belt increases orographic rainfall, which increases
~1Il~
III C
erosion and sediment yield, a process discussed in ~ a b
Sect. 12.11.
Drainage patterns within foreland basins have
been discussed by several authors (e.g., Eisbacher et
Fig. 11.68. OverfIlled and underfilled basins. In under-
al. 1974; Graham et al. 1975; Sengor 1976; Dewey filled basins the peripheral bulge is a sediment source,
1977; Mia1l1978d, 1981b; Van Houten 1981; Burbank whereas in overfilled basins the bulge is covered by sedi-
1992). The subdivision into transverse and axial ment. (Jordan 1995)
sediment entering the basin along transverse drain- control (e.g., Posamentier et al. 1988; Miall 1991c;
age systems and therefore tends to be characterized Posamentier and Weimer 1993). The relationship
by larger channels. Such basins are commonly un- between thrust-sheet loading and the development
derfilled, in the terminology ofJordan (1995). These of a sedimentary moat is well established (e.g., Bally
differences are illustrated in Fig. 11.68. Examination et al. 1966) and has been modeled quantitatively by
of these differences in individual cases may throw several workers (e.g., Beaumont 1981; Jordan 1981;
important light on basin development. For example, Jordan and Flemings 1990). Recently, attempts have
Burbank (1992) demonstrated that during the been made to examine the correlation between
Pliocene the Himalayan foreland basin of India un- specific stratigraphic sequences, such as major
derwent a transition from an underfilled condition, molasse pulses, and major regional tectonic events
with an axial ancestral Ganges river system flowing such as terrane collisions (Cant and Stockmal1989;
in a relatively proximal position, to an overfilled Stockmal et al. 1992) and thrust movement episodes
condition, dominated by transverse drainage of gi- (Jordan et al. 1988; Burbank and Raynolds 1988).
ant fan systems, such as the Kosi, and with the axial The nature of the theoretical relationship between
drainage having migrated laterally to a very distal tectonism, subsidence, and sedimentation is exam-
position within the basin. He related this change to ined in detail in Sect. 11.3.6. Tectonic cyclicity is on
post-tectonic uplift of the Himalayan ranges and the several times scales. Of most concern here are the
proximal basin and to the increased sediment supply so-called third-, fourth-, and fifth-order sequences,
that resulted from this unroofing, the initiation of which have indicated durations of 1-10, 0.2-0.5, and
the Indian monsoon, and from the late Cenozoic 0.01-0.2 million years, respectively. Longer-term se-
Alpine glaciation. His data for India therefore sup- quences (> 10 million years) are related to global
port the "antitectonic" model for foreland basin plate-tectonic controls and are beyond the concern
development (Sect. 11.3.6), in constrast to that part of this chapter (see Sect. 13.4 and review in Miall
of the basin located in Pakistan, which Burbank et al. 1990). The relative importance of tectonism and
(1988, 1989) argued developed according to the tra- eustasy in generating these sequences is by no means
ditional "syntectonic" model of sedimentation and clear, either in general global terms or in the case of
tectonics. the specific sequences in foreland basins. The fol-
Eisbacher et al. (1974) demonstrated that trans- lowing paragraphs provide brief discussions of the
verse drainage derives mainly from structural nonmarine cyclicity in several foreland basins
salients along fold-thrust belts, such as the central, around the world.
most uplifted part of thrust plates, where the belt The Oligocene-Miocene molasse in the Digne
tends to have a convex curvature facing the basin, area of southeast France is discussed briefly in Sect.
whereas many major longitudinal streams enter the 11.4.5.1. Crumeyrolle et al. (1991) suggested that the
basin at reentrants, such as at the termination of overall molasse succession is a long-term (second-
thrust plates or at terrane sutures. They cited order?) trangressive-regressive cycle induced by
the modern Himalayan foredeep as an example. Ma- foreland-basin tectonism. The succession is punctu-
jor alluvial fans, such as that of the Kosi River, repre- ated by unconformities and is subdivisibl,e into
sent transverse drainage from structural salients, third-order sequences, which they suggested may be
whereas the Brahmaputra River, one of the major related to eustatic events and can be correlated with
axial drainage channels in the basin, enters at the the Haq et al. (1987, 1988) global cycle chart. How-
major syntaxis that marks the end of the Himalayan ever, recent debate (Miall 1992b) indicates that
ranges, in Assam. Many transverse drainages are this type of correlation is speculative and should be
thought to be antecedent rivers; that is, they predate viewed with caution.
the development of the mountain ranges through The South Pyrenean foreland basin of northeast
which they cut. Examples in the Himalayas are cited Spain contains a mixed carbonate-siliciclastic suc-
by Holmes (1965) and Seeber and Gornitz (1983), cession of Paleocene-Eocene age 3 km thick that has
and in the Zagros Ranges by Baltzer and Purser been subdivided into nearly 20 third-order se-
(1990). quences. Contrasting analyses of different, but over-
Stratigraphic successions in foreland basins lapping portions of this succession were provided by
are strongly cyclic. Regional tectonism has long Puigdefabregas et al. (1986) and Luterbacher et al.
been known to be a primary control, and in recent (1991). In the first of these papers, the authors attrib-
years there has been much debate regarding the uted the development of the sequences to tectonism,
importance of eustasy as a major stratigraphic the gradual southward overthrusting of the fold-
"tl
a-
lb
.:,
Ib
~
o
!:l
6
;:;.
(/J
-- A 11>
~
..''.'
~:;~'
~:,;..~ _.T:0'~.~,-=, ~ .___ ------'"--_ ~ =" ;:,",,,"%;>,;,:U~C;'{~/" ~ S·
ClQ
o
...,
E;
~ 1000", $.
10 k... e:..
I:l::I
I>l
'"
S·
TIME
'"
AGES SEQUENCES FORMATIONS
LINES FACI ES
:i'
AlLUV IAl fAIl CONGlIJI£RATES ANO Al LUVIAl PlA I N FACIES
~'1 IN PASSIVE MARGIN : PONTIlS GR . (SEQ. Z TO 61
Ol IGOCENE IX SOlSOHA .. -,
r_-.-. .'..... I" ACTIVE MARGIN : CCRONES (SEQ. ZI.lIEl~T (SEQ. 61. STII. MAGIlAlENA (SEQ . 71. SOlSONA (SEQ. 9 1
I
VIIICAROONA "
! .~
'~ ....., • • ! ~ .. ~
DEL" fRONT SST; BARCONS - COUBn (SEQ. 61, CEHTELLES - "ILAIIY (SEQ. 71
PRIASONIAN f--® CORAL REEFS; TOSSA (SEQ . 71
~
OUTER PlII1FORM lit) PROO£L fA MARLS; BAHYOLES (SEQ. 6). lCUAlADA (SEQ. 71. (ORONES (SEQ . ZI. SAGHARI (SEQ. 11
VII NILANY
1-----1
BARTONIAN TRAIISGRESSIVE SAIIDSTOHES; COLlBAS - f OLGUEROLES (SEQ. 71
CARBONATE PLATfORM; ORPI - CADI (SEQ. 11. PENYA (SEQ. 3). TAVERTET (SEQ. 6)
~ ~
~ EVAPORITES; BEUDA (SEQ. 5). CARDONA (SEQ . 8) .
LUTH I"" VI BEllMUNT
[ : -=::~~I TE RRIGENOUS TURBIDITES; CAMPDEVANot. (SEQ. 4)
f--12 V 8ElJDA
,
Y IV CAMPDEVANOL SL OP E MARLS AND BRECCIAS (CARBONATE MEGA1U1lBIDITES); AltlWlCIES (SEQ. 3)
I ~.:j
~ CUI SIAN H> III AltlWlCIES
E
S UNDERl YINC PAlEOCENE AND UPPER CRETACEOUS RED BEDS
I II CORONES I 1
A ILERDIAN
H ~ I CADi
A r:,.;;· -: '> BASEMENT • NOSTl Y PAlEOZOIC
...... ,,:'!;":1
Fig. 11.69. Idealized cross section of the Eocene-Lower Oligocene sequence stratigraphy of the South Pyrenean basin. (Puigdefabregas et al. 1986)
>I>-
o
lJ1
thrust belt leading to a migration and offlapping of Kauffman 1984). Most of these are shown in Fig.
depocenters (Fig. 11.69). However, in the second 11.70. High sea level was characterized by the devel-
paper, it is claimed that correlation of the sequences opment of thick and areally extensive mudstone
with the Haq et al. (1987, 1988) chart can be carried units (e.g., Mancos Shale), and by fine-grained lime-
out, and the authors interpreted eustasy as the main stone (including chalk) in areas distant from sedi-
driving mechanism. ment sources (e.g., Niobrara Formation). Lower sea
An idealized cross section of the Eocene-Oli- level gave rise to extensive clastic wedges, in which
gocene section is shown in Fig. 11.69. The overall nonmarine sandstone and conglomerate passed
structure of the sequences is that of a coarse, basinward into shoreline and shelf sand bodies. The
nonmarine clastic wedge prograding southward Indianola and Mesa Verde groups of Utah and Colo-
from the thrust front into basin-center marine rado are good examples of major regressive strati-
marls. Shallow-water carbonate deposits are formed graphic packages which contain numerous minor
on the distal ramp and forebulge of the foreland transgressive-regressive cycles nested within them
basin, away from the influence of the fold-thrust- (Weimer 1960; Fouch et al. 1983; Lawton 1986a,b;
belt clastic source. Carbonates are found in the Swift et al. 1987; Van Wagoner et al. 1990, 1991;
transgressive stage of each sequence. They may be Yoshida et aI., in press; Fig. 11.71).
associated with evaporites formed during the subse- Some of the third-order cycles may be global in
quent regressive stage. Careful dating of these origin and related to eustatic changes in sea level. As
sequences by Burbank et al. (1992) using discussed by Weimer (1986), there is a fair degree of
magnetostratigraphy indicated that "progradation correlation between the sea-level curve for the west-
was controlled primarily by subsidence rather than ern United States and that for northern Europe, al-
sediment supply or base-level changes". though correlations are bedeviled by the difficulties
Ten major third-order trangressive-regressive in reconciling various chronostratigraphic time
cycles have been recognized in the Western Interior scales. Arguments regarding the validity of this type
basin of the United States (Weimer 1960, 1986; of correlation are beyond the scope of this book.
M.Y w. FORT UNION E. TERT. M~

66-
70- 1111~i~tf~1~~~~~~~~i~~i-i~~t~~~;;F?HH~~~;;;;;;=
/!i Be JR
.'.i.i.'.'.!
~ <_
P -70
CAMP.
80- CI -80
80~ ..
87 - ~'i,j:L:j:j:j:.'I"'i~ ~90
90-
I!~~~~~~~~~~~~~~~~~·~'~~~':d~"'!"i~~CEN. . SC
F
ALB.
.K.· 96
UL·K
-100
110 - -110
J!..F:.T __ _
120- G -120
NEOC.
130- -130
135- MORRISON ----mR.
-135
MARINE a TRANSITIONAL NONMARINE []]]J MAJOR HIATUS

DSH 055 §3LSorCHALK IillSS ~SSorSH ~97 ~~~pN:~;xMI6rTE
Fig. 11.70. Diagrammatic west -east cross section through JR Judith River; Be Bearpaw; FH Fox Hills; La Lance. To the
the Western Interior Seaway of the Rocky Mountains, east formations are: L Lytle; LAK Lakota; FR Fall River; SC
showing stratigraphic positions and approximate dates of Skull Creek, J and D sands of Denver basin; G Greenhorn;
major transgressive units and interregional uncon- B Benton; N Niobrara; P Pierre; M and C McMurray and
formities. Formations or groups to the west are: G Gannett; Clearwater of Canada. (Weimer 1986, reprinted by permis-
SC Skull Creek; M Mowry; F Frontier; H Hilliard; MVMesa sion)
Verde; RS Rock Springs; E Ericson; Ea Eagle; CI Claggett;
w BOOKCLIFFS
E
GUNNISON
PLATEAU. COLO.·UTAH
UTAH UNE
--- - - - - - - - - - - -.-- - - --
--=-C'AS'fLEGATE SS-=..c-=--=.c -::=. =-c:=-.=c--'-'--:-'-'.-~
- - -:--- - ----:- -:-:-.-:--:- - -- --
- C--..,. ' . ' - - .~-..,.~_ -"-
- __ . ,-,-"""'"_ -",
___. _-
_-
_ _.
._
. _. _ . , ._.
_. _ DESEATMBR
_ _ _ _ _ _ _ "_ _ _ _ _ ::;;;"';;:;·;;:·-,,¥;;;;":;;~~~,::":GRASSY MBA
KM ~o
I
- UNCONFORMITY
FLUVIAL SS, SH;

LAGOONAL SHALE LJ] MARINE
SANDSTONE LJ] MARINE
SHALE
Fig. 11.71. Regional stratigraphic framework of Upper Cretaceous strata of east-central Utah. (Swift et al. 1987)
Kauffman (1984) claimed that there is a consistent vide the tight correlation between individual tec-
correlation among transgression, thrusting (in tonic and stratigraphic events that is now available
Wyoming and Utah), and volcanism in the Western for parts of the Himalayan foredeep.
Interior, but his data are not convincing to this Are individual clastic tongues within major
writer. However, there is no doubt that some of wedges tectonic or eustatic in origin? Figure 11.71
the transgressions were contemporaneous with illustrates the broad stratigraphic architecture of
thrust-faulting episodes within the Sevier orogen of part of the Upper Cretaceous clastic wedge of Utah.
Utah, possibly including episodes in the Albian, There is no doubting an overriding tectonic control
Santonian, and Maastrichtian (compare Kauffman of sedimentation for many of these successions.
1984; Lawton 1986a,b; and Weimer 1986). These Paleocurrent and petrographic evidence indicates
transgressions were regional in scope, and cannot shifting sediment sources and changes in regional
therefore be attributed to flexural loading by indi- paleoslope during deposition of the Mesa Verde
vidual thrust plates. However, where thrusting is Group and associated units in Utah. Lawton
part of regional compression, possibly related to (1986a,b) documented unroofing of intrabasin up-
terrane accretion or variations in subduction pa- lifts, from which early basin-fill sediments were can-
rameters, the mechanical drag effect described by nibalized, and changes in dispersal patterns related
Mitrovica et al. (1989) may be invoked as a cause of to basin tilting. The volume of sediment within the
regional basin subsidence and transgression. wedge is too great to have been controlled by passive
Other thrusting events within the Sevier orogen changes in sea level (see also Galloway 1989a,b). But
are not clearly correlated with regional changes in the question remains whether tectonically induced
sea level in the basin but are correlated with the sediment input was modulated by sea-level control,
development of major clastic wedges that prograded leading to fourth- and higher-order cyclicity along
across the basin margins. Lawton (1986a,b) indi- the fringes of the clastic wedge (such as in the Book
cated a link between thrusting and clastic-wedge Cliffs of Utah: see Swift et al. 1987; Van Wagoner et
formation in Utah between the mid Albian and the al. 1990, 1991; Yoshida et al., in press). As discussed
late Campanian, although it is not possible to pro- in Sect. 11.3.5, intraplate stresses related to episodic
thrust-sheet loading may be responsible for strati- Tight correlation between individual tectonic and
graphic cyclicity on time scales between 104 and 106 stratigraphic events is now available for some
years. foreland basins, such as parts of the Himalayan
McLean and Jerzykiewicz (1978) examined the foredeep. There, Burbank and Raynolds (1984,1988)
uppermost of the major nonmarine clastic wedges in used magnetostratigraphic techniques to date the
the Alberta Basin, the Brazeau-Paskapoo forma- fluvial deposits of the Himalayan foredeep of Paki-
tions, and showed that this succession, up to 3600 m stan. The technique permits a precision of dating to
thick, encompasses at least four crudely upward- within the nearest 105 years. These results show that,
fining cycles, ranging from 400 to 1600 m in thick- in the Himalayan foredeep, thrusting and uplift epi-
ness. Age control on these cycles is poor, but sodes were very rapid and spasmodic, and that they
deposition of the entire wedge spanned about 20 did not occur in sequence into the basin, as the
million years, and so each cycle represents an aver- classic model (e.g., Dahlstrom 1970) would predict
age of about 5 million years. In the Karoo Basin of (Fig. 11.72). In one case, Burbank and Raynolds
southern Africa, which is also a foreland basin, the (1988) were able to demonstrate the uplift and re-
Beaufort Group comprises three basin-wide up- moval of 3 km of sediment over the crest of an
ward-fining cycles ranging from 160 to 500 m in anticline within the basin over a period of 200000
thickness (Visser and Dukas 1979), whereas the years, an average uplift and erosion rate of 1.5 cm/
overlying Molteno Formation consists of six up- year. In most basins, the precision of dating obtained
ward-fining cycles that reach maximum thicknesses by these authors cannot be attempted, and this
of 140 m (Turner 1983). Tectonic control of these therefore suggests that caution should be used in
foreland basin cycles in Alberta and southern Africa assessing published reconstructions of the rates of
is indicated by sedimentological evidence for shift- convergence, crustal shortening, and subsidence,
ing source terranes (Rahmani and Lerbekmo 1975; except where these are offered as long-term (> 1
Visser and Dukas 1979). million year) averages.
Fig. 11.72. Chronology of fault motions with-

in part of the Himalayan foredeep of Pakistan.
Shaded boxes indicate the time and space
domain over which each thrust is interpreted
to have been active. (Burbank and Raynolds
1988)
A simple satellite basins may assist in unraveling the thrust

kinematics of the fold-thrust belt (e.g., Ori and
:2~-- Friend 1984; Johnson et al. 1986; Ricci Lucchi 1986;

Bentham et al. 1992). Continued crustal shortening
may culminate in the uplift of these minor basins, so
B complex that only remnants may be preserved within the
eroded foldbelt, as in the Swiss Alps (Homewood et
al. 1986; Pfiffner 1986) and the Appalachian orogen
(Lash 1990). Several recent studies of modern thrust-
fault-influenced alluvial-fan environments provide
split into minor basins
excellent analogs for the interpretation of the details
of sedimentary styles and the response of deposi-
tional systems to tectonic tilts and changing sedi-
ment source areas (Jolleyet al. 1990; Allen et al. 1991;
Rlislinen et al. 1992; Dumont 1993; Damanti 1993;
Fig. 11.74).
11.4.7 Sedimentary Basins

E deformed
and Allochthonous Terranes
~ The collision of major continental plates involves

crustal thickening and shortening and late-stage
strike-slip displacement (escape tectonics), as exem-
Fig. 11.73. Variations in the relationship between the plified by the India-Asia collision, discussed in the
fold-thrust belt and the foreland basin, based on seismic
profIles. Minor basins and piggyback basins are varieties preceding section. The basins that form and are
of satellite basins. (Ricci Lucchi 1986) filled during this process are commonly termed in-
termontane or successor basins. They may preserve
much internal evidence of the collision that can be
In the simplest case, a foreland basin is an asym- analyzed to determine its style and timing. Such
metric moat, deepest adjacent to the fold-thrust belt basin analysis may be particularly useful in the case
and shallowing toward the continental interior (Fig. of orogens composed of numerous small plates, tec-
11.73A). However, as a result of the complexities in tonic flakes, or slivers of uncertain affinities, so-
thrust kinematics noted above, the proximal edge of called suspect terranes, for which evidence bearing
the basin may evolve into several separate sub-ba- on the amount and timing of relative motions may
sins, isolated or semi-isolated during sedimentation be limited. However, compression and lateral dis-
by faults growing within the basin (Fig. 11.73B-E). placement of the terranes in the final stages of oro-
Ori and Friend (1984) introduced the term piggy- genic assembly may disrupt and displace the
back basin for basins carried on the backs of thrust successor basins (Fig. 11.75), making their analysis
plates (Fig. 11.73D), but Ricci Lucchi (1986) sug- very difficult. Basins associated with terrane accre-
gested that the specific condition described by Ori tion are not necessarily fluvial, but many are so,
and Friend is only one of several possibilities, which particularly during the final phases of collision,
he preferred to include in a wider category of minor because collision invariably leads to uplift.
or satellite basins (Fig. 11.73C,D). The most useful contribution fluvial basin ana-
Minor satellite or piggyback basins develop as lysis can make to the unraveling of a history of
fault ramps that cut the foreland basin and are up- accretionary tectonics is to throw light on sediment
lifted, isolating part of the foreland and acting as transport directions and detrital provenances
sediment barriers (Figs. 11.73, 11.74). Ponding of (Saleeby 1983; Schermer et al. 1984; Kleinsphehn
sediment takes place within the minor basins, with 1988; Howell 1989; Bluck 1991). The timing of the
diversion of flow along the basin axes toward reen- suturing between terranes is most readily deter-
trants in the main thrust front. Such reentrants com- mined in one of three ways: (a) by the dating of
monly form at terminations or overlaps of thrust intrusive igneous bodies that cross-cut the suture,
plates. Careful stratigraphic work on the fill of these (b) by dating overlap assemblages of sediments that
.......
o
\ ~,
Fig. 11.74. Relationship between paleogeography and structure in an active thrust-faulted basin margin. The basin at lower left is a
satellite basin. (Jolley et al. 1990) (1)
~
'"'
o
::sr;.
n
o
a....
g.
o
....
'TI
[
e.
(J)
(1)
0.-
3·
(1)
...::s~
o·::s
are deposited across the suture, or (c) by identifying cases have been cited by Schermer et al. (1984) and
detritus in a basin-fill that can only have been de- Bluck (1991). Schermer et al. (1984) stated:
rived from the adjacent terrane. In the case of the "An example of an overlap assemblage is the Late Creta-
third criterion, provenance studies may also be used ceous Gravina-Nutzotin belt of southern Alaska, which
to demonstrate post-suture displacement, if the de- links the Wrangellia and Alexander terranes; an example
tritus does not match the petrology of the source of a provenancial linkage is provided by the Middle to
from which paleogeographic analysis indicates the Upper Jurassic strata of the Bowser Basin of British Co-
lumbia, which are deposited on the Stikine terrane and
detritus should have come. Examples of both these contain debris from the Cache Creek terrane."
These relationships are both indicated in the ter-
rane-amalgamation diagram compiled by Saleeby
(1983; Fig. 11.76). Discussing the Devonian sedimen-
tary fill of the Midland Valley of Scotland, Bluck
(1991) stated that sediments banked against the fault
which forms the southeast margin of this basin, the
Southern Uplands Fault, "contain abundant igneous
and metamorphic clasts up to boulder size, but in the
direction from which they were transported lies the
Southern Uplands, a sequence of greywackes and
shales. In this instance a source block has been later-
ally displaced or overthrust by the greywacke se-
quence."
Fig. 11.75. The development of successor basins at the An example of the complex problems in plate
margins of terranes. A and C are terranes which amalgam- kinematics that terrane amalgamation histories
ated along the complex suture zone Z. Arrows indicate can set up was described by Kleinspehn (1985). She
typical dispersal of sediment following terrane amalgam- examined the Tyaughton-Methow Basin in south-
ation. (Bluck 1991)
western British Columbia, a basin formed adjacent
_ _ STiKINIAN SUPER-TERRANE
TON EMERGENT
STIKINE
TERRANE
i5: carbonates
..... clastics
=--: argillite - amalgamation episode
= chert ~ accretion episode
"vv silicic tuffs -;.. pluton link
... & •• basalt-andesite ••••• provenance link
volcaniclastics
"""'" basaltic pillows,
-=- overlap
IllIII1. flOIIS a hypo byssols u unconformity
I I
Fig. 11.76. Amalgamation history of the major terranes ment amalgamation histories. (Saleeby 1983, reproduced
of British Columbia and southern Alaska. Note the use of with permission from the Annu Rev Earth Planet Sci vol 11,
provenance linkages and overlap assemblages to docu- © 1983, by Annual Review Inc.)
to and partly overlapping six Cordilleran terranes. ancient basins and upwarps, with the widespread
Paleocurrent and provenance studies provide partial distribution of thin nonmarine sheets at times when
constraints on the reconstruction of terrane amal- much thicker clastic wedges are being deposited in
gamation and post-suture displacement on a series marginal basins. Veevers (1984, p. 161) stated:
of terrane-bounding faults. Kleinspehn (1985) de- "The Cainozoic record of central Australia is a continua-
scribed four main hypotheses of relative terrane tion back to 60 Ma of the present pattern oflocal sags, such
motion. as Lake Amadeus, within highlands flanked by broad
plains. Today, little sediment is shed by the highlands, and
the wholly carbonate province on the south, in the Great
Australian Bight, is traceable in the Euda Basin back to the
11.4.8 Cratonic Basins Eocene. What little sediment escapes central Australia to-
day, via the coherent stream system on the east and south-
The origin of sedimentary basins and uplifts within east - the Finke River and others to the south are the only
stable areas such as the Canadian Shield and adja- streams that maintain flow out of the region - is concen-
cent stable platform is controversial (Miall 1990, trated in the sump of Lake Eyre and the dunefields of the
Simpson Desert, and this pattern (but not facies) of depo-
Sect. 9.3.6.1). Most hypotheses involve the reactiva- sition persists to the beginning of the Cainozoic in the
tion of ancient lineaments or zones of weakness to form of the Eyre Formation (Wopfner et al. 1974). The
provide loci of subsidence and uplift, although the aeolian sand dunes of the Great Sandy Desert of the Can-
mechanisms for generating vertical movements ning Basin and of the area to the north between 18° and 22°
within stable cratons are not clear. Many cratonic S, in which only a few tens of meters of sediment have
accumulated, are the only other sediment derived from
tectonic elements may be related to plate-margin central Australia."
processes, and intraplate stress is increasingly being
invoked to explain the initiation or reactivation of Sloss (1979) has argued for the existence of verti-
major intracratonic structures at distances far re- cal crustal movements within the craton - true
moved from the main locus of extension or compres- epeirogeny - but mechanisms to explain it have been
sion (e.g., Cloetingh et al. 1985; Sanford et al. 1985; lacking until recently. Gurnis (1990, 1992) proposed
Miall 1986; Heller et al. 1993; Sect. 11.3.5). Embry that mantle convection beneath the crust results in
(1990a, 1991) suggested that during the Mesozoic continental-scale differential heat distribution and
intraplate stress led to tilting on a continental scale consequent heating and cooling of different parts of
and provided the cratonic sediment sources and the overlying earth's surface, resulting in broad re-
paleoslope for the delivery oflarge volumes of terrig- gional uplifts and downwarps because of the effects
enous detritus to Sverdrup Basin, on the northern on crustal densities. These processes maintain what
margin of the North American craton. Miall (1986) is called dynamic topography. The effects of crustal
suggested that intraplate stress transmitted from heating cause uplift. This occurs along the flanks of
Caledonian compression at the suture between new continental rift systems (e.g., parts of present-
North America and Europe generated thrust fault- day East Africa) and above mantle plumes. Subsid-
bounded uplift of the Boothia Uplift, in the Canadian ence takes place over cooling areas of the earth's
Arctic Platform, from which was shed the alluvial fan crust, such as areas of aging oceanic crust distant
and fluvial wedge of the Devonian Peel Sound for- from spreading centers, and over regions of mantle
mation (Miall and Gibling 1978). Dott et al. (1986) downwelling. The stratigraphic effects of this pro-
demonstrated that fluvial and eolian processes cess have yet to be evaluated in detail.
formed widespread but thin clastic sheets in the
Wisconsin area, flanking the southern margin of the
Canadian Shield, during two periods of transgres- 11.5 Basic Paleogeographic Models
sion following low stands of sea level, in the Late
for Nonmarine Basins
Cambrian and again in the Middle Ordovician. Iso-
pachs of the Tertiary Eyre Formation of the Great
Artesian Basin, Australia, indicate rejuvenation of On a basin scale, the architecture of the alluvial fill
Paleozoic structures; basins became deeper and can be classified into a series of distinctive styles,
upwarps were uplifted (Wopfner et al. 1974) governed by the tectonic control of basin shape and
Descriptions of the regional geology oflarge plat- paleoslope (MialI1981b). A basin-fill classification is
form areas such as interior Australia (Veevers 1984; given in Table 11.4, in which depositional systems
Morgan 1993) and cratonic Canada (Stott and are simplified into nine basic types: 1. lacustrine, 2.
Aitken 1993) indicate the long-term persistence of alluvial fan/fan delta, 3. low-sinuosity fluvial, 4.
Basic Paleogeographic Models for Nonmarine Basins 413
high-sinuosity fluvial, 5. river-dominated delta, 6. to distinguish proximal from distal alluvial settings,
wave-dominated delta, 7. tide-dominated delta, 8. although some rivers of exceptionally high discharge
nondeltaic coast (beach and barrier systems), and 9. and sediment load (e.g., Amazon, Ganges, Brahma-
estuarine. putra) do not show significant increases in sinuosity
For the purpose of this analysis, a twofold subdi- near their mouths. The threefold delta classification
vision of fluvial environments into rivers of high and is that of Galloway (1975).
low sinuosity is adequate. This serves, in most cases, An alluvial drainage system can be divided into
four parts: 1. headwaters in upland source areas; 2.
proximal environments, immediately below the fall
Table 11.4. Alluvial basin-fill patterns' line along the margins of sedimentary basins; 3. me- .
Model Proximal Medial Distal dial environments, including axial fluvial systems; '
and 4. distal environments, where the river interacts
T fan T braid Lake margin/non-
plain deltaic coast with some terminating environment, such as a lake,
2 T fan-delta Lake margin/sea coast playa system, eolian dune system, delta front, or
3 T fan/river T river River-dominated delta tidal flat. The headwaters zone is insignificant from
4 T fan/river T river River-dominated delta the point of view of basin analysis, as it is rarely
with barrier lagoon preserved. As discussed in Sect. 11.3.2, rivers are
5 T fan/river T river Wave-dominated delta
6 T fan/river L river Lake margin/estuary classified as either transverse, in which case they
7 T fan/river L river River-dominated delta flow directly from the uplifted source across struc-
8 T fan/river L river Tide-dominated delta tural grain (e.g., the bounding faults), or longitudi-
9 T fan/river L river Wave-dominated delta nal (axial), flow being parallel to strike, along the
T, Transverse; L, longitudinal. basin axis. These generalizations may hold only lo-
• From Miall (l981b). cally, for foldbelts show syntaxes, island arcs are
Fig. 11.77. The nine basin-fIll patterns discussed in this only. Most basin-fIll patterns occur in several tectonic
section. See Table 1104. TTransverse;L longitudinal. Scales settings. (Miall 1981 b)
are variable. Structural controls are given as examples
Fig. 11.78. The Salar de Punta Negra, Chile, illustrating

basin-fill models 1 and 6. (Miall 1981b, after van L---.....J
Overmeeren and Staal 1976) 10km
Fig. 11.79. Part of the North Slope, Alaska and Yukon,

arcuate, and plate margins may be irregular. How- illustrating basin-fill model 2. (Miall1981b)
ever, the generalizations are a useful aid to classifica-
tion of basin paleogeography.
The distinction between transverse and longitu- subdivision of fluvial systems, and the transverse/
dinal rivers is an important one sedimentologically. longitudinal criterion, it is possible to define nine
Commonly, transverse rivers provide the tributaries major basin-fill models (Table 11.4; Fig. 11.77). This
to a longitudinal trunk river, and the two may be classification is not intended to be all-inclusive, na-
characterized by quite different fluvial styles and by ture being highly variable. However, it suggests that
different paleocurrent means and variances. Some these nine models are the most likely to be found in
longitudinal rivers originate as major transverse riv- practice because of the constraints imposed on river
ers which emerge from structural reentrants in the and coastline configurations by plate geometry. A
mountainous source area and curve downstream major complication is that of "big rivers" (Potter
into a longitudinal orientation (Eisbacher et al. 1978), which flow for hundreds or thousands of kilo-
1974). The modern Brahmaputra is an example. Fig- meters, perhaps through or across several tectonic
ure 9.1 illustrates an ancient example of an alluvial zones or terranes, before debouching into the sea.
basin filled by both transverse and longitudinal Alluvial systems that are predominantly erosional in
depositional systems of contrasting fluvial style. The nature are not of concern here.
major sediment packages deposited by each system The first model, that of a transverse fan and
are separated by a bounding surface of sixth order, braidplain system, is a common one in relatively
according to the classification used in this book. small drainage basins, where river length is measur-
Using the simple ninefold environmental classifi- able in tens of kilometers. Runoff may be ephemeral.
cation provided above, the proximal-medial-distal There is no single trunk river, and water and sedi-
ment discharge at base level is inadequate to develop small ocean basins, such as the Red Sea, where fan
major deltas. Rivers may terminate as a network of deltas prograde into the ocean.
shallow distributaries or as a zone of sheetflooding Models 3, 4, and 5 are systems where water and
on a lake margin, playa, or tidal flat. Intermontane sediment discharge are adequate to construct deltas
lake basins (basin type 4.3 in Table 11.2) are com- at base level. They may build major alluvial-deltaic
monly bordered by this type of alluvial pattern (e.g., plains with river lengths measurable in hundreds of
parts of the Salar de Punta Negra, Chile: Fig. 11.78). kilometers. The three models are distinguished on
The pattern is also common in rift basins (basin type the basis of delta type. None of the world's major
1.1), such as Death Valley, California, and the East rivers correspond to any of these three models, be-
African Rift System. Some of the Triassic rift basins cause most of these follow structures, and are there-
of eastern North America also fit this pattern fore longitudinal. The typical pattern is of a broad
(Hubert et al. 1976; P.E. Olsen 1990), as does the coastal plain in front of a foldbelt mountain range, a
Tertiary of Uinta Basin, Utah (basin type 2.4). volcanic are, or a downwarping continental margin.
Ballance (1980) described an example in a modern The coastal plain is crossed by numerous parallel,
wrench-fault basin. Miall and Gibling (1978) de-
scribed a Devonian example of such a pattern flank-
ing the fault-bounded Boothia Uplift in Arctic
Canada. Alluvial fans along the flanks of the moun-
tains drained into a sandy braidplain and an arid
tidal flat.
Model 2 is a variant of the first model, in which the
alluvial fans drain directly into a standing body of
water as fan deltas or braid-deltas (Fig. 11.79).
McPherson et al. (1987) and Nemec and Steel (1988)
provided discussions and classifications of these
types of depositional system. Fan deltas typically are
small, gravel-dominated systems, a few kilometers
across. This pattern occurs on coastlines flanking
Fig. 11.81. Part of the coastal plain of Georgia and North

and South Carolina. An example of basin-fill model 4.
Fig. 11.80. The north Sumatra basin, an example of a Open stippling indicates areas of active sedimentation;
retroarc foreland basin, and drainage model 3. (Miall closed stippling indicates older deposits formed in the
1981b, after Hamilton 1979) same tectonic environment. (MiallI981b)
___.--A' 'Z- 50km
Fig. 11.82. Part of the Guatemala coast, an example of

basin-fill modelS. (Miall1981b)
Fig. 11.84. Part of the East African Rift System, Uganda,

an example of basin-fill model 6. (Miall 1981b)
transverse rivers, commonly flanked by peat

swamps. River termination patterns are controlled
by coastal processes.
Model 3 is well developed on the north coast of
Sumatra (basin type 2.4; Fig. 11.80) and New Guinea.
The petroliferous Cherokee Group of Oklahoma and
Kansas (basin type 4.2, but with sediment source
from the continental side) displays this pattern, as
does the Permian Beaufort Group, South Africa (ba-
sin type 4.1; Hobday 1978).
The Atlantic coastal plain of Georgia and North
and South Carolina (Fig. 11.81) and the Gulf Coast of
Texas are Atlantic-type divergent continental-mar-
gin basins (basin type 1.2.2), and are good examples
of model 4. These are meso tidal and microtidal
coasts, respectively, and are characterized by major
barrier island-lagoon systems into which the coastal
plain rivers drain. The Carboniferous coal-bearing
fluvial, deltaic, and barrier-island deposits of Ken-
tucky and West Virginia are a good ancient example
Fig. 11.83. Jordan Valley and northern Dead Sea Basin, an (peripheral foreland basin, basin type 4.1; Horne et
example of basin-fill model 6. (Miall1981b) al. 1978).
Fig. 11.86. The Central Lowlands of Myanmar (Burma),

an example of basin-fill modelS. (MialI19S l b)
length may be thousands of kilometers. The models

are distinguished mainly on the basis of river termi-
Fig. I1.S5. The Po Basin, Italy, an example of basin-fill
model 7. (MiallI9Slb) nation patterns. In the case of smaller rivers with low
sediment discharge, or those emptying into a rapidly
subsiding basin, there may be inadequate sediment
Model 5 occurs mainly in areas facing open load to form a delta. This is model 6. Rivers may be
oceans, which are characterized by high wave en- ephemeral, terminating in a sheet flood zone along a
ergy. Deltaic deposits are reworked into superim- playa lake margin (Fig. 11.78), or they form an estu-
posed beach ridges flanking the river mouth, which ary with tidal flats. Modern examples of the latter
may be deflected down current into a spit. Modern include the Thames (U.K.), the Amazon, the Rhine,
examples of such alluvial systems include the Pacific the Susitna (Cook Inlet), and the Colorado at the
coast of Guatemala (basin type 2.2; Fig. 11.82) and Gulf of California. Many intermontane basins in
the Sao Francisco delta of Brazil (basin type 1.2.2). California and Nevada (backarc rifts, basin type 2.3)
Ancient examples have been described from the Gulf show longitudinal rivers terminating in a playa or
Coast (Fisher 1969). permanent lake. Some strike-slip basins (e.g., Jordan
Models 6, 7, 8, and 9 are combinations oflongitu- Valley, basin type 3.3; Fig. 11.83) and rift systems
dinal trunk rivers with transverse tributaries. The (e.g., East African rifts, basin type 1.1; Fig. 11.84) also
latter are commonly alluvial fans at the basin mar- show this pattern. Ancient nonmarine examples in-
gins. Most major rivers and large inland basins show clude the Devonian fluvial sediments of Spitsbergen,
one or other of these four patterns. Trunk river which flow into a playa (basin type 4.3; Friend and
Fig. 11.87. Benue Trough, Nigeria, an example of
basin-fill models 8 and 9. (Mial11981b)
rv
/
IOO km
Fig. 11.88. Lake Eyre basin, Australia, an example of centripetal drainage in a cratonic basin. (MialI1981b)
Moody Stuart 1972), and the Permo-Triassic of tively). Hobday and Von Brunn (1979) interpreted.
northwest Scotland (Steel and Wilson 1975). Ancient the Table Mountain Sandstone of eastern South
marine examples are few and far between. Possibly Africa in terms of a longitudinal river flowing south-
Cook Inlet, in the Tertiary, may have consisted of a west to a tide-dominated delta system located near
paleo-Susitna River flowing into an estuary, as it the southern end of the continent, prior to the
does at the present day. breakup of Gondwana (basin type 1.3). Much of the
Model 7 is that of a transverse and longitudinal Cretaceous-Tertiary molasse of the Alberta basin
river system which terminates in a river-dominated (basin type 4.3) was deposited with a model 8 geo-
delta. This includes several of the world's major metry. For example, the Cadomin conglomerate was
present-day alluvial basins, such as those of the a unit deposited by transverse alluvial fans draining
lower Danube, Po (Ori 1993; Fig. 11.85), Tigris- the ancestral Rocky Mountains. These fed a trunk
Euphrates (Baltzer and Purser 1990), Indus, and river flowing along strike, which probably flowed
Mississippi. The Mississippi is an example of a river into an interior epeiric sea through a tide-influenced
flowing along the axis of a grabef! or embayment estuary or delta system (McLean 1977; McLean and
striking at a high angle to the continental margin. Wall 1981).
Some of the best ancient examples include the Ter- Model 9 is that of a transverse-longitudinal river
tiary precursors of the modern rivers listed above terminating in a wave-dominated delta. The best
(e.g., Gulf Coast, basin type 1.2.2; Fisher and examples of this type are the modern Rhone and
McGowen 1967). Other examples include the Alpine Nile, both of which flow along grabens that trend
Lower Freshwater Molasse (basin type 4.1; Van virtually perpendicular to straight coastlines. They
Houten 1981), and the Middle-Upper Devonian are therefore open to the effect of waves oriented
Melville Island Group of Arctic Canada (basin type around a 180 0 arc. Mutch (1968) described a Carbon-
4.3; Embry and Klovan 1976). iferous example in New England.
Tide-dominated deltas are most likely to occur at A tenth model is not listed in Table 11.4 because it
the head of a bay or gulf at the terminus of a longitu- does not fit the transverse-longitudinal drainage
dinal river (model 8). Good examples are the Ganges, pattern discussed here. This is the centripetal drain-
Irrawaddy (Fig. 11.86), and Niger (Fig. 11.87) fluvial- age pattern shown by some cratonic basins, such as
deltaic systems (basin types 4.1, 2.2, and 1.3, respec- Lake Eyre basin (Fig. 11.88).
Chapter 12
What Does Fluvial Lithofacies Reveal About Climate?
12.1 Introduction coarse fluvial channel and bar deposits and the detri-
tal overbank deposits? Are there any characteristics
of the assemblage, vertical succession, and associ-
Integrated analyses of past climates are now becom- ated sedimentary structures of fluvial gravels, sands,
ing commonplace, as attempts are being made to and muds that can be interpreted in terms of local
produce complete reconstructions of ancient envi- climate? Many studies of ancient fluvial deposits
ronments incorporating numerical global climate include an interpretation of the contemporary cli-
models linked to the details of local plate-tectonic mate, but in most cases the interpretation is based on
position and paleogeography. Evidence from the the stratigraphic context, on paleosols, or on associ-
stratigraphic record, including the distribution of ated faunas and floras. Other studies infer climate
certain types of lithology, plus fossil evidence and from regional paleogeography and make predictions
isotopic data are combined in such analyses. This of temperature, rainfall and seasonality from knowl-
chapter is concerned with the nonmarine litho- edge of comparable geographic configurations on
stratigraphic record and the information it can yield the modern earth [e.g., the discussion of monsoonal
regarding climate. conditions in Pangea by Dubiel et al. {1991} and
Several climatic indicators of chemical or bio- Parrish {1993}]. The purpose of this chapter is to
chemical origin are present in the nonmarine sedi- focus on the lithofacies of clastic deposits in fluvial
mentary record, such as coal, evaporite, bauxite, systems to determine whether they contain any use-
laterite, and paleosols, all of which yield limited in- ful independent information regarding climate. This
formation on some of the main attributes of climate, chapter specifically does not focus on paleosols or
particularly rainfall and temperature (Frakes 1979; the paleoecology of nonmarine systems, which have
Parrish and Barron 1986). Some of the clearest cli- received ample coverage elsewhere.
matic signals available in nonmarine sediments are Although the history of global climates during the
the wind patterns that can be deduced from giant Phanerozoic is in general fairly well known (e.g.,
fossil ergs (e.g., Parrish and Peterson 1988), the Frakes 1979; Parrish and Barron 1986; Parrish 1993),
chemical rhythms in lacustrine sediments and the many local details remain to be resolved, and the
climatically driven changes in lake levels recorded in kinds of field criteria described in this chapter may
lacustrine-fluvial interbedding (Katz 1990), redox therefore be useful, especially where the more obvi-
and water-table information from paleosols (Wright ous evidence from lacustrine, eolian, glacial, or other
1986; Retallack 1981, 1988), and the various indica- related facies is absent. In addition, there is increas-
tors of continental glaciation, such as tills, gla- ing interest in climate change, particularly changes
ciolacustrine dropstones, and striated pavements driven by orbital forcing. The recognition of
(Hambrey and Harland 1981). It was once thought Milankovitch rhythms in the nonmarine sedimen-
that continental redbeds were a clear indicator of tary record is a subject barely in its infancy, and
oxidizing and probably arid depositional environ- again the kinds of observational approach reviewed
ments (see Pettijohn 1957; Van Houten 1973), but here may provide useful tools for this type of study
Walker {1967} clearly demonstrated that the red (Sect. 12.12).
color may be produced by diagenesis, requiring that The main body of this chapter (Sects. 12.5-12.7)
color be interpreted with caution. provides a systematic discussion of the lithofacies
Eolian, lacustrine, and continental glacial sedi- evidence that can be obtained from detailed local
ments, and the chemical sediments in fluvial flood- studies of conglomerates, sandstones, and fine-
plain deposits may therefore yield some evidence of grained facies, focusing on the major climatically
climate. What about clastic fluvial deposits - the sensitive attributes for each sediment type. The evi-
422 What Does Fluvial Lithofacies Reveal About Climate?
dence may be complex, however, because sedimen- how paleosol distribution is related to this classifica-
tation at anyone location in a basin may be affected tion. Their four broad climatic zones are wet equato-
by both local and regional factors. For example, the rial, dry subtropical, moist midlatitude, and polar.
channel and floodplain processes of tributary rivers Complications in this zonation are caused by the
may reflect local micro climates, which may vary regional distribution of land and water masses, and
across a basin (e.g., rain-shadow effects), whereas a by topographic effects.
trunk river may bear the discharge and sediment- From the point of view of the preservability of
load signatures of a distant source area that may be different bedform styles, Jones (1977) recognized
characterized by a quite different climate (see also five basic types of fluvial discharge pattern:
Sect. 11.3.1). These and other complexities are intro-
1. Perennial rivers with relatively steady discharge
duced below and are discussed in Sect. 12.10.
and a single maximum peak annual flood; com-
mon in temperate climates, e.g., Rhine
2. Perennial rivers with double or many maximum
12.2 Climatic Variables discharge events; common in tropical and tem-
perate regions with seasonal variations in rainfall,
e.g., Platte
What is climate, and how might it be recorded in
3. Perennial rivers with a pronounced low-dis-
terrigenous detrital sediments? The two most impor-
charge period and a long flood period; common
tant attributes of climate are rainfall and tempera-
in tropical areas with monsoons, e.g., Brahma-
ture. Wind, its strength, direction, and consistency,
putra
are important climatic attributes in the study of
4. Perennial rivers with pronounced low-discharge
eolian deposits, but not of other nonmarine sedi-
period and short flood period; common in high-
ments. The production, dispersal, and deposition of
latitude rivers and those draining alpine moun-
terrigenous detrital sediment depends on water, and
tains, where there is a short spring flood period
therefore on rainfall, its surface runoff in fluvial
and a long period of slowly tapering summer
channels, and in overbank flow. Both the amount of
discharge, e.g. Tana, Norway; sandur rivers in
rainfall and its distribution with time (e.g., seasonal,
Iceland and Alaska
monsoonal, flashy) are climatic factors that have an
5. Ephemeral rivers with flashy discharge; charac-
important influence on sedimentation. Tempera-
teristic of arid regions
ture is important as a control on the physical behav-
ior of water - freezing, thawing, and evaporation. These categories cannot yet be resolved in terms of
Temperature also controls the rate of various chemi- the facies styles described in Chap. 8. Given the obvi-
cal reactions that affect syndepositional diagenesis. ous relationships between discharge regime, fluvial
Rainfall and temperature together control the style, and climate discussed in these paragraphs, it
amount and type of vegetation, which influences is perhaps surprising that climate is not singled
depositional styles in channels and overbank areas out as a significant controlling parameter in the
and, through biochemical processes, the type of major geomorphic classification of fluvial flood-
weathering and erosion processes in fluvial source plains proposed by Nanson and Croke (1992), al-
areas, thereby affecting sediment yield. though several variables affected by climate are
Climatologists have classified global climates in discussed, such as the style of flooding and the influ-
many different ways, recognizing up to as many as 14 ence of vegetation.
different climate types that reflect the ranges of vari- In this chapter the following broad climatic zona-
ability and seasonality of temperature and rainfall tions are used, based on features observable from the
on the present-day earth (Huggett 1991; Barry and lithofacies record:
Chorley 1992), but most of these subtleties are only
- Hot-arid
crudely represented in the lithofacies record.
- Hot-humid
Perlmutter and Matthews (1989) used the following
- Tropical to semitropical with wet-dry seasonality
temperature categories in their discussion of
- Temperate-humid
Milankovitch forcing: tropical, temperate, and po-
- Boreal-paraglacial
lar, and the following rainfall categories: very humid,
humid, subhumid, dry, arid. Mack and James (1994) There are several complications involved in de-
proposed a simple climatic differentiation of the termining climate from the fluvial sedimentary
earth related to latitudinal position and discussed record, as follows:
Distinguishing Tectonic from Climatic Control 423
1. Large variations in fluvial discharge produce rec- rivers and the topographically lower basin-center
ognizable effects in the facies record, such as the channel systems.
interbedding of very coarse and very fine facies, 6. Climate is commonly recorded in the sedimen-
but the time scale of variation may not be obvious tary record by vegetation, by the fossil remains of
and the climatic implications therefore unclear. vegetation itself, and by the effects vegetation has
Seasonal effects, such as those arising from the on sediment erodibility, sediment yield, channel
freeze-thaw cycle, monsoonal variations, and style, and so on. However, the styles of vegetation
those that occur in arid environments character- that characterize the earth today have evolved
ized by occasional violent flash floods, months or with time and were very different in the geological
even years apart, all generate the same wide varia- past. The development of large land plants in
tion in discharge and consequent wide variation the Devonian, of plants that could survive sea-
in facies, but the climatic interpretation of the sonal climate changes in the Mesozoic, and of
resulting facies may be quite ambiguous (consid- grasses in the Miocene all brought about changes
erations of time scale are discussed at length in in fluvial hydrology and channel style (Schumm
Chap. 3). 1968a). Modern analogues therefore have a limi-
2. The climate of the sediment source area and that ted applicability to the distant past, pre-Devonian
of the depositional basin may not be the same. For landscapes probably having functioned sedimen-
example, rivers flowing southward from the Al- tologically much as do arid regions of the present
pine-Himalayan mountain chain are all strongly day, even where rainfall was abundant (Fig. 8.2).
seasonal, reflecting the alpine climate of their
Correct identification of these various effects and
mountainous source areas, but they enter very
separation of them from the sometimes very similar
different climates in the plains to the south. Euro-
effects of tectonic control require meticulous obser-
pean rivers, such as the Rhone and Po, have
vation and very careful deductive reasoning.
coastal plains in the subtropical Mediterranean
region. The Tigris and Euphrates enter the highly
arid Mesopotamian basin. The fluvial plains of
northern India are located in a tropical climate 12.3 Distinguishing Tectonic
where seasonal monsoonal rainfall strongly influ-
ences discharge characteristics (see also Sect.
from Climatic Control
11.3.1 for other characteristics of "big rivers").
3. Related to point 2 above is the fact that flow The most obvious effects of tectonic controls are the
hydraulics of a river, and therefore the facies and changes that tectonism induces in sedimentary ac-
architecture of the bedload deposits, are deter- commodation space and in the quantity and caliber
mined largely by the climate of the source area, of the sediment load. Upward coarsening and fining
whereas floodplains and their finer-grained de- of alluvial successions, in the form of cycles, se-
posits are influenced primarily by the climate of quences, and mega sequences, have long been inter-
the depositional basin. preted in tectonic terms (Steel et al. 1977; Heward
4. Topographic effects may generate local climates 1978a), although the nature of the relationship be-
that complicate the depositional record. For tween tectonic movement and the sedimentary re-
example, mountain ranges rising in the path of sponse is currently undergoing debate, as discussed
prevailing humid winds may generate high oro- in Sect. 11.3.6. However, similar sedimentary effects
graphic rainfalls on their upwind slopes and rain could be generated by climate change. Langbein and
shadows downwind. The topographic diversion Schumm (1958) showed that in the United States
of air masses is an important general effect. Very sediment yield is related to rainfall, varying from a
high plateaus, such as Tibet, are also a significant minimum in arid climates, to a maximum in sub-
source of heat, which has its own effect on air humid climates, to a low in humid climates when the
temperatures and circulation. The possible wider effectiveness of vegetation in binding soil and re-
linkages between tectonics and climate are dis- tarding erosion reaches a maximum. However, later
cussed in Sect. 12.11. work has shown that this is not a simple relationship
5. Even within a fluvial basin, levels of water satura- that applies to all regions of the earth. For example,
tion (the water table) and redox conditions may increased seasonality of the climate also increases
vary from channel to floodplain and between the sediment yield, because the vegetation cover is lim-
proximal, possibly more elevated basin-margin ited by the constraints of the dry season, so that
runoff is more flashy and therefore more competent foreland basin adjacent to the Pyrenees, in Spain.
in the wet season (Walling and Webb 1983). This They demonstrated that in the proximal part of the
effect is particularly marked in monsoonal climates. basin a cyclic alternation of fine and coarse sedimen-
Cooling is induced by uplift, as well as by climate tation persists vertically through about a dozen
change, and the sedimentological effects are the cycles, totaling over 100 m of beds. In the most
same in both cases - changes in fauna and flora, and proximal areas the cycles are entirely nonmarine,
typically an increase in sediment caliber and quan- and in more distal areas entirely marine, with the
tity as a result of a reduction in the density of vegeta- facies boundary (shoreline) remaining fixed verti-
tion cover. What sedimentological criteria can be cally. These are therefore not transgressive-regres-
used to distinguish climatic from tectonic controls sive cycles. The fixing of the facies boundary is
in the generation of these kinds of changes in detrital attributed to the influence of growth anticlines
caliber and volume? which stabilized areas cjf uplift and subsidence, but
In their study of the fan gravels flanking the Dead the absence of transgression and regression suggests
Sea rift, Frostick and Reid (1989) noted a sharp that there were no vertical movements of the basin or
interbedding of coarse and fine deposits, with no of sea level during this time period. It therefore
upward coarsening or fining. This they interpreted seems likely that climatic control of rainfall and
to indicate an absence of progradation or retrogra- sediment yield is the explanation for the observed
dation, which are the typical alluvial responses to cyclicity.
tectonic adjustments (Heward 1978a). They sug- An important debate regarding the distincton be-
gested that the coarse units indicate periods of in- tween global climate change and large-scale uplift in
creased rainfall. the Cenozoic is discussed in Sect. 12.11.
De Boer et al. (1991) studied a complex inter-
fingering of marine and nonmarine deposits in a
Table 12.1. The occurrence of debris flows in gravel deposits

Environment Debris-flow deposits Debris-flow deposits
common to dominant rare to absent
Arid fans Beaty (1963); Bull (1963); Frostick and Reid (1989);
Hooke (1967); Schultz (1984); Jolley et al. (1990)";
Flint (1985); Smith (1990); Harvey (1990);
Hubert and Filipov (1989); Maizels (1990);
Harvey (1990); Hartley (1993)"
Blair and McPherson (1992);
Nemec and Postma (1993)b
Temperate-humid Winder (1965); Ori (1982); Blair (1987)";
fans Pierson (1980); Kochel (1990)";
Kochel and Johnson (1984); DeCelles et al. (1991);
Wells and Harvey (1987); Crews and Ethridge (1993);
Kochel (1990); Ono (1990) Ridgway and DeCelles (1993)
Tropical-humid Heward (1978b); Rust (1984);
fans Iwaniw (1984); Darby et al. (1990);
Ahmad et al. (1993); Wescott (1990);
Brierley et al. (1993) Evans (1991)
Upland periglacial-humid Church and Ryder (1972); Nemec and Postma (1993)'
fans Wasson (1977);
Derbyshire and Owen (1990)
Outwash "fans" Maizels (1993) Le Blanc Smith and Eriksson (1979);
(sandurs) Boothroyd and Nummedal (1978)
" Deposits of hyperconcentrated flows are important.
b Stage 1 fans.
, Stage 2 fans.
Conglomerates: The Significance of Texture and Petrology 425
12.4 Review of Climatic Criteria range of climatic criteria, mostly those preserved in
the overbank environment.
The following sections review the major facies at-
tributes of clastic fluvial lithofacies that convey some
climatic information. For each lithofacies class the 12.5 Conglomerates: The Significance
focus is on the most climatically sensitive criteria, of Texture and Petrology
and these are emphasized in the subheadings. The
data are summarized in accompanying tables and 12.5.1 Mass-flow Versus Traction-current Processes
figures, as follows: Table 12.1 classifies some major
recent descriptions of modern and ancient gravel The textures of gravels reveal a gradation in sedi-
deposits according to climatic setting and according ment transport mechanisms from mass-flow to trac-
to the predominance of mass-flow or stream-flow tion-current processes. Such characteristics as a
deposits. Tables 12.2 and 12.3 summarize the de- matrix-supported framework and the presence of
scriptive criteria of gravel- and sand-dominated flu- grading are typical of mass flow, including debris
vial systems, respectively, subdividing them into flows, mudflows, and hyperconcentrated flows,
various climatic classes and highlighting the at- whereas a clast-supported framework, stratification,
tributes that are climatically sensitive. Figure 12.1 and cross-bedding are typical of traction-current
illustrates two contrasting styles of gravel-domi- deposits (Sect. 5.2). Several factors are important in
nated fluvial deposit, Fig. 12.2 does the same for determining the type of flow that occurs in gravel-
sand-dominated deposits, and Fig. 12.3 illustrates a bed fluvial systems, including the magnitude and
Table 12.2. Field criteria for the interpretation of climate from clastic fluvial lithofacies: gravel-bed alluvial-fan and
braided-channel systems (climatically distinctive criteria are indicated by italics)
Hot-arid systems
Typical clastic lithofacies: Mass-flow deposits, including Gmm, Gmg, Gci, Gcm
Typical architecture: Sheets, lenses and lobes of SG
Major bounding surfaces: Fan lobe deflation surfaces (fifth- and sixth-order) overlying winnowed gravels
Autogenic cyclicity: Debris-flow events, may show winnowed tops
Petrology: Survival of chemically unstable lithologies, such as carbonates
Other structures and associated lithologies: Silcrete, calcrete, dreikanters, eolian lenses, evaporitic cements.
Lenses of plant hash may be present
Hot-humid systems
Typical clastic lithofacies: Stream-flow gravels, typically Gh, Gt, Gp. Mass-flow deposits may be present
Typical architecture: Channel-form CH, sheets and lenses of GB, SB
Major bounding surfaces: Channel-form fourth- and fifth-order cycles, fan-lobe sixth-order surfaces
Autogenic cyclicity: Macroform and channel-fUl cycles, progradation-abandonment cycles of fan lobes
Petrology: Chemical destruction of unstable minerals and lithologies, ghost clasts with diagenetic clay matrix
Other structures and associated lithologies: Bauxite, laterite, coal
Temperate-humid systems
Typical clastic lithofacies: Stream-flow gravels, typically Gh, Gt, Gp. Mass-flow dpeosits may be present
Typical architecture: Channel-form CH, sheets and lenses of GB, SB
Autogenic cyclicity: Macroform and channel-fill cycles, progradation-abandonment cycles offan lobes
Petrology: No distinctive features
Other structures and associated lithologies: Coal may be present
Boreal-paraglacial systems
Typical clastic lithofacies: Stream-flow gravels, typically Gh, Gt, Gp. Mass flow deposits generated by jokulhlaups,
including Gmm, Gmg, Gci, Gcm. Minor traction-current sand lithofacies
Typical architecture: Channel-form CH, sheets and lenses of GB, SB, lobes of SG
Autogenic cyclicity: Macroform and channel-fill cycles, progradation-abandonment cycles of fan lobes
Petrology: Survival of chemically unstable lithologies, such as carbonates. Composition-shape inversion (weak clasts
fractured by frost, resistant clasts retain roundness)
Other structures and associated lithologies: Till, ice-wedge casts, ice-push stone clusters, ice melt depressions,
lacus'trine deposits with dropstones
Table 12.3. Field criteria for the interpretation of climate from clastic fluvial lithofacies: sand-bed fluvial systems, Le.,
meander-belts, braid plains, terminal fans (climatically distinctive criteria are indicated by italics)
Hot-arid systems
Braid plains with playas, terminal fans, flash floods common
Typical clastic lithofacies: St, Sh, Sl, Sr. Large Sp sets rare to absent, Fl, Fm
Typical architecture: Proximal: steep-sided wadis, medial: broad channels with macro forms (elements LA, DA) and
scour hollows (element HO), distal: unchannelized sheet bodies interbedded with playa mudstones
Major bounding surfaces: Channels bounded by fifth-order surfaces, macroforms with abundant second- and
third-order surfaces defining flood-related accretionary bundles; distal fans: Flat fourth- and fifth-order surfaces
Autogenic cyclicity: Flood cycles within macroforms and sheet floods indicating rapid deceleration, commonly from
upper flow regime, plane-bed condition
Petrology: Detrital carbonate sand common
Other structures and associated lithologies: Intraclast siltstone breccias, desiccation cracks, mud curls, silcrete,
calcrete, evaporite lenses, nodules or crystals, dreikanters, eolian lenses
Hot-humid systems
Meandering and anastomosing styles most typical; floodplains with peat bogs typical
Typical clastic lithofacies: Most sand and fine-grained lithofacies present
Typical architecture: Meander-belts consisting of sand-bed channels oflow to high sinuosity, with steep, root-
bounded banks, braiding rare
Major bounding surfaces: Channel-form fifth-order surfaces, macroform-bounding fourth-order surfaces,
accretionary second- and third-order surfaces
Autogenic cyclicity: Macroform and channel-fill cycles
Petrology: Chemical destruction of unstable minerals and lithologies, ghost clasts with diagenetic clay matrix
Other structures and associated lithologies: Peat swamps common, bauxites and laterites in elevated interfluves
Tropical to semitropical systems with wet-dry seasonality
Most fluvial styles; perennial but with seasonally dry floodplains
Typical clastic lithofacies: Most sand and fine-grained lithofacies present. Large Sp sets common, calcrete breccias
Typical architecture: Meander-belts with channels of varying style
Petrology: No distinctive climatic signatures
Other structures and associated lithologies: Common association between peat swamps or drab soils flanking
channels with calcretes, and evidence of desiccation on upland interfluves
Temperate systems
Most fluvial styles. Some seasonal discharge variations
Typical clastic lithofacies: Most sand and fine-grained lithofacies present. Large Sp sets common
Typical architecture: Meander-belts with channels of varying style
Petrology: No distinctive climatic signatures
Other structures and associated lithologies: Peat swamps or drab soils may be present
Boreal-paraglacial systems
Braided outwash plains, sandurs, fans
Typical clastic lithofacies: Most sand and fine-grained lithofacies present
Typical architecture: Complex interfingering of channel lenses and fan lobes
Major bounding surfaces: Channel-form fifth-order surfaces, macroform-bounding 4th-order surfaces, accretion-
ary second- and third-order surfaces
Autogenic cyclicity: Macroform and channel-fill f-u cycles, c-u cycles filling ice melt depressions, fan lobe
progradation and abandonment cycles
Petrology: Survival of chemically unstable detritus, till balls
Other structures and associated lithologies: peat swamps or drab soils may occupy inter-fan lows. Interbedded till or
glaciolacustrine muds and sands with dropstones, ice-wedge casts, ice-push stone clusters, ice-melt depressions
f-u, Fining-upward, c-u, coarsening upward
wo-~ <O~:"_':'~_':'_c:_~_c::>_<:)_~':'~~_:=' a~~g~m~t~d ~e~rfs ~o:;_~_~_~c:'-~_ :_~_ ~_:_o_ ~_<o_::_~_~_:_~_~_

SG
edge of flow lobe
reworked channel lag
~ j"2\ channel/
• • • 1-' ••••••• 5~ • .• '1\' • ••••• • : •• :::f-.;: ...... ::.".: ... ::.: ':"::.'..::.... " ',/ ...... .
. • .' G G G' G eolian sandsheet with rootlets / fan lobe abandonment
~ mm, mg, Cl, cm surface
o~ a
.... _,'"""",IT",·",..... . .. ;'... . ...... .......
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CIaO
~.
2 0 SG
I - - - ----- -- ----------- ----------- -- ----------------------------- ---------- -------------- --- -------------------- ---
o
A 5 m
I I I I I I I I I I I
B
Gh
Gt-------I
• '0. '.
~,__0===-LA
GB
coal or carbonaceous shale floodplain deposit
., . 4
~ longitudinal gravel sheets -'!':-
===-----------= ':.
•.•. 4 .....•..• : ... : •... '.' ...........: ......................... :. ::::::::: St Sr :::.' ••.•.. '.. :-'.'
GB
Gh
Fig. 12.1A,B. Architectural diagrams of conglomerates. interfinger with eolian sand sheets. (Based on Wasson
Numbers in circles are bounding surfaces. A Typical ap- 1977; Hubert and Filipov 1989; Blair and McPherson 1992).
pearance of debris-flow-dominated alluvial-fan deposit, B Gravel assemblage deposited by stream-flow processes,
such as are formed in hot, arid climatic settings. Debris- which are the dominant gravel transport mode in most
flow deposits and their derivatives - winnowed flow tops humid climatic settings. (Based mainly on Ramos and
and channel-floor lags - constitute virtually the entire Sopena 1983; Smith 1990; DeCelles et al. 1991)
deposit. The lateral or distal fringes of fan lobes may
variability of runoff and the quantity and type of important controls on gravel transport processes.
debris available in the catchment area. Blair and McPherson (1994) demonstrated that the
The presence or absence of debris flows, hyper- hydraulic conditions for debris flows and stream
concentrated flows, and sheetfloods has been pro- flows are quite different, and the two types of deposit
posed by McPherson and Blair (1993) and Blair and will therefore tend not to be associated in the same
McPherson (1994) as the major criterion to be used depositional system. They prefer the terms braid
in the definition of alluvial fans, a proposal opposed delta and floodplain delta for dispersal systems built
by Stanistreet and McCarthy (1993) and Galloway by stream-flow (lower flow-regime traction-current)
and Hobday (1995), who preferred an all-embracing processes.
definition that would include the large, sand-domi- Because of the importance of American work in
nated systems deposited by braided and meandering the deserts of the arid southwest US on the nature of
rivers, such as the Kosi and the Okavango (Sect. 8.3). mass-flow processes in particular, and alluvial-fan
The controversy has focused attention on the nature processes in general (see below), there is a tendency
of fluvial processes and has thereby broadened our to associate mass flow with the catastrophic runoff
understanding of the autogenic and allogenic (tec- that occurs in arid and semiarid regions. The ab-
tonic, climatic) controls that govern the develop- sence or sparsity of vegetation in arid regions is an
ment of basin-margin alluvial wedges. Climatic important factor, because vegetation binds the sedi-
factors are commonly taken to be among the more ment and acts as a buffer to rapid runoff during all
o SB
=~
- - -_ _ _DA
A o 5 m
I I II I
B
Fm
LS
Sf
Sh LS flood cycles
~--------~F~m~------------~~-------~
Sh
Sh LS
Fig. 12.2A,B. Architectural diagrams of sandstones. A distal terminal-fan deposits formed under arid conditions,
Typical mobile-channel-belt deposit formed under peren- showing macro forms with accretionary bundles and un-
nial flow conditions, with large macroforms showing few channelized sand sheets, both architectural characteristics
internal sedimentary breaks (third-order surfaces), well- of rivers with flashy discharge. (Based on Tunbridge 1981,
developed bar-top ridge-and-swale features, levees and 1984; Stear 1983; Shepherd 1987; Bromley 1991; Munoz et
swamp deposits. (Based on many examples, including al. 1992; and Kelly and Olsen 1993)
Wizevich 1992b, 1993; and Miall1993, 1994). B Medial to
but the most violent rainstorms. However, it is im- 72.5. 7. 7 Arid Climates.
portant to take account of the significant number of
observations of mass-flow processes occurring in Based on the pioneering work in the desert areas of
other climatic settings, and other studies of arid the southwestern United States by Chawner (1935),
environments where mass-flow deposits are of lim- Sharp and Nobles (1953), Blissenbach (1954), Beaty
ited importance. Some of the more important recent (1963), Bull (1963, 1972), Lustig (1965), Denny
studies are listed in Table 12.1. Interpretations of (1967), and Hooke (1967), it has long been known
climate in each of the ancient studies is based on the that alluvial fans in modern arid and semiarid re-
accumulated evidence from sedimentary features gions are characterized by debris flows (Fig. 12.IA).
and contained fauna and flora, and in most cases are More recent work in this area, including detailed
taken from the conclusions of the authors. A much textural and facies studies, has been reported by
more lengthy tabulation is provided by Blair and Hubert and Filipov (1989), Beaty (1990), and Blair
McPherson (1994, Table l), but it is limited to distin- and McPherson (1992, 1994). Several of the older
guishing desert from nondesert settings for debris papers quoted here include eyewitness accounts of
flows. The fluvial styles under discussion here are debris-flow events, which occur every few decades.
described in Sects. 8.2.1-8.2.5. Much of the time the fan distributaries are dry, with
Fig. 12.3A-C. Some climatic lithofacies indi-

COLD cators of the overbank environment. A Cold
A. Ice-push bank with pebble cluster climates with seasonal ice. (Based on
B. Ice-rafted debris lens Collinson 1971b; Bryant 1983; Martini et al.
C. Kettle hollow 1993). B Hot, seasonal wet-dry climates.
D. Coarsening-upward fill of (Based on Hallam 1984; Jerzykiewicz and
Kettle hollow Sweet 1988). C Hot, arid climates. (Based on
E. Ice-wedge cast Tunbridge 1981, 1984; Stear 1983, 1985;
F. Ice gouge Munoz et aI. 1992)
B
51 HOT, SEASONAL WET-DRY
" ""
. 1Al
A. Thin coals formed in swamps
"- I' /" "-........ B. Desiccation cracks
f1 @ C. Caliche formed on elevated
© interfluves
':.)7~a_-!O~.! _Q ~_Qd:) =:CV _: ~o_~ .
f1
~
c HOT,ARID
( sea.le variable )
A
r:::::' :: h·.~ •••:-·
A. Sheetflood channel-splay cycles
..... " , ........... . B. Wind deflation surfaces with
dreikanters
C. Eolian sand lenses
D. Mud curls
E. Desiccation cracks
~b~~@~~~~~~.o -F. Intraclast breccias
O. Evaporite crystal casts
surface runoff occurring during spring melt and fol- Quaternary fan (braid-delta) deposits in the deserts
lowing cloudbursts. However, it takes a considerable of Oman also do not contain evidence of mass-flow
length of time for weathering to generate enough processes (Maizels 1990). Likewise, some of the Re-
debris for mass flow to be triggered - an example cent and Cretaceous-Tertiary fan (braid-delta) de-
of a significant "geomorphic threshold" that must posits in arid intermontane basins in the Chilean
be passed before a particular type of process can Andes do not contain debris-flow deposits. Jolley et
occur (Schumm 1979). Blair and McPherson (1992) a1. (1990) reported that ephemeral stream-flow de-
emphasized that debris-flow deposits and the prod- posits are dominant in their study of modern
uct of their reworking constitute the entire accumu- syntectonic fan deposits, and concluded that they
lated sedimentary record of the Trollheim fan in were formed by high-concentration turbulent trac-
California (Fig. 8.9), contrary to the interpreted sedi- tional flows. They stated that "[t]he absence of mass
mentary model of Miall (1978c) and others. Traction flow deposits is inspite of favourable depositional
current deposits, such as stratified gravels and cross- slopes, a persistently arid climate, the coarse-
bedded sands, are absent. grained nature of the source material and the avail-
The importance of sediment-gravity flows in the ability of mud in the [source] sediments."
fan deposits of other arid areas is quite variable. The facies composition of ancient arid-climate
Modern fans in the arid intermontane valleys of fan deposits is similarly variable. In an ancient
southeast Spain were subdivided by Harvey (1990) arid-fan system described by Hartley (1993), hyper-
into three types, based on their facies composition. concentrated flood gravels and sheetflood traction-
Debris flows are more common in fans draining current gravels characterize proximal areas. By
small, steep drainage basins and are absent in some contrast, the Tertiary deposits described by Flint
of the fans fed by larger drainage basins. The modern (1985) include significant debris-flow deposits. A
fans draining the flanks of the Dead Sea rift valley, a Triassic conglomerate deposit in SW Britain is inter-
very arid area, do not include significant mass-flow preted as a fan complex by S.A. Smith (1990), who
deposits, according to Frostick and Reid (1989). stated:
"A semi-arid climate during the deposition of at least part indicates an absence of debris-flow deposits, al-
of the Budleigh Salterton Pebble B~ds is suggested by the though he reported evidence of hyperconcentrated
evidence for large variations of discharge (e.g. drapes of the
scour troughs of sand dunes and drapes of the lower por- flows.
tions of gravel macroforms). Also, rare aeolian sandstones It is perhaps significant that several recent
have been found in the upper part of the sequence, where descriptions of ancient alluvial-fan/braid-delta de-
they occur as thin lenses capping bar tops." posits from interpreted temperate-humid climatic
The conglomerates consist entirely of stratified grav- regimes indicate that stream-flow deposits are
els, commonly with cross-bedding, organized into dominant, and that mass-flow deposits are of minor
large in-channel macroforms, and were deposited by importance. This depositional style is illustrated in
traction-currents in gravel-bed braided rivers (Figs. Fig. 12.1B. In the study by Crews and Ethridge
6.18,8.21). The term braid-delta would be more suit- (1993), the dominance of braided stream-flow trac-
able for these deposits, following the recommenda- tion processes was demonstrated, with evidence of
tions of Blair and McPherson (1994). lateral channel migration. Debris flows were deter-
mined to be rare to absent. Likewise, DeCelles et al.
(1991) reported mass-flow deposits in only one of
12.5.1.2 Temperate-Humid Climates numerous localities. In their study, Ridgway and
DeCelles (1993) drew comparisons with the modern
Several studies have documented mass-flow pro- gravel sandurs of southern Alaska (Boothroyd and
cesses in modern temperate-humid climates. Debris Nummedal 1978). Similar clast-supported gravels
flows are particularly important in mountainous re- deposited by stream-flow processes are the domi-
gions of southwestern Japan (Ono 1990) and New nant facies in their rocks. Debris flows are rare. One
Zealand (Pierson 1980), where in many cases they possible inference that could be drawn from the
contribute to the reworking and redeposition of studies noted here is that while mass-flow processes
glacial fan deposits. Kochel and Johnson (1984) de- may be very spectacular in modern humid-tem-
scribed active fans in Virginia. The fan deposits con- perate settings, they may have low preservation
sist of poorly sorted, mud-supported gravels with potential and are therefore of limited geological
inverse grading and poor stratification, with a bed significance. However, this ancient/modern con-
thickness varying from 50 cm to 5 m. The main depo- trast is based on only a few reports and should not
sitional process is interpreted as debris flows and be regarded as definitive.
debris avalanches triggered by intense rainfall. From
20 to 70% of the fan surface is affected at anyone
time. The recurrence interval of flood events is esti- 12.5.1.3 Tropical-Humid Climates
mated at 3-6 ka, based on radiocarbon dating. Collu-
vial processes are also very important. This study was Deposits containing significant mass-flow deposits
expanded to other parts of the Appalachians by and examples with a lack of such deposits have both
Kochel (1990), who confirmed the widespread distri- been described in case studies of the modern and
bution of debris-flow-dominated fans but also docu- ancient record.
mented the local occurrence of fans formed by Modern fan/braid-delta deposits in Papua-New
stream-flow and hyperconcentrated flow processes, Guinea were described by Brierley et al. (1993). The
where lithologic controls on source sediment and area has a monsoonal, savannah-type climate, with
hydrologic controls on runoff prevented the genera- a significant dry season. In addition to stream-flow
tion of debris flows. Wells and Harvey (1987) de- deposits, debris flows and hyperconcentrated flows
scribed the results of a catastrophic cloudburst in are present, as indicated by gravel textures. The
northern England. In a temperate-humid environ- mass flow deposits are limited to proximal reaches
ment such rare catastrophic floods can generate ero- of the fans, where they comprise up to about 30% of
sional and depositional effects that may overshadow the total. However, their preservation potential is
the products of "normal" weather. In this case a single estimated to be low, because of continued incision
flood generated violent stream flows and debris of the deposits, and this is aggravated by syndepo-
flows. In this particular study, which examined nu- sitional tectonic uplift. The modern Yallahs braid
merous fans, the debris-flow deposits were found to delta ofJamaica lacks debris-flow deposits (Wescott
constitute a greater part of the accumulated sediment 1990).
on the fans draining smaller, steeper drainage areas. Several ancient conglomerate deposits have been
The description of a small modern fan by Blair (1987) interpreted to have been formed in a tropical-
humid setting. The Devonian Malbaie Formation of movements that release large volumes of meltwater.
Quebec contains no mass-flow deposits (Rust 1984), In Iceland jokulhlaups are also triggered by volcanic
whereas in a study of Carboniferous fan and paleo- eruptions under ice caps. These floods may do little
valley deposits in Spain, Heward (1978b) reported more than erode and rework pre-existing fluvio-
substantial accumulations of debris-flow deposits. glacial sands and gravels, moraine, or till. Rounded
The paleovalleys are filled with debris-flow con- till balls are common in some outwash deposits
glomerates. The deposits of fan channels and the (Ashley et al. 1985). However, jokulhlaups may trig-
overbank areas of fans consist of interbedded de- ger mass flows, and in Iceland such deposits
bris-flow and stream-flow deposits, with debris consititute a significant proportion of the outwash
flows more important in proximal reaches. Rust succession (Maizels 1993). Another product of cata-
(1984) reported an unusual abundance of cross-bed- strophic floods may be the deposition of trains of
ded gravels (lithofacies Gp) in the Malbaie Forma- giant bedforms, as in the floods that affected the
tion, and attributed this to the presence of deep Columbia River valley during the draining of Lake
braided channels and perennial discharge in a tropi- Missoula (Baker 1978b).
cal, humid setting. In another study in Spain, Iwaniw
(1984) reported that debris-flow deposits are impor-
tant, whereas in a study of a Paleogene fan (braid- 12.5.1.5 Summary and Conclusions
delta) complex in Washington state it was found that
debris-flow deposits are of minor importance and To summarize, it is clear that while mass-flow depos-
can be related to volcanic influence (Evans 1991). its, particularly debris-flow conglomerates, and as-
Late Cenozoic deposits described by Darby et al. sociated upper flow-regime stream-flow deposits are
(1990) contain no evidence of mass-flow processes. very common in arid environments (Trollheim fan
being the type example), they are not always gener-
ated in this climatic setting and can occur under
12.5.1.4 Boreal-Paraglacial Climates other climatic regimes, ranging from paraglacial to
tropical (Table 12.1). The association with an arid
It is not sup rising that mass-flow deposits have been climate is a reflection of the requirement for abun-
found to be a dominant facies in ice-margin settings, dant loose debris, mobilized by rapid runoff, unin-
where abundant glacial debris is available to be hibited by a vegetation cover. However, catastrophic
transported by large volumes of glacial meltwater. floods can occur in all environments. They are trig-
Reports of Late Cenozoic deposits from various Al- gered by cyclonic storms in temperate and tropical
pine environments around the world are consistent settings and by jokulhlaups in glacial settings. Cli-
on this point (Church and Ryder 1972; Wasson 1977; mates characterized by seasonal rainfalls and sea-
Derbyshire and Owen 1990). An interesting excep- sonal dry periods are candidates for flash floods
tion is the complex of Quaternary fan deposits in because of the lower density of vegetation in such
Crete constituting "stage 2" of Nemec and Postma areas than in humid settings. Floods need not be very
(1993). These deposits are interpreted as periglacial frequent to impose a dramatic change on the deposi-
in origin, yet were deposited almost entirely by tional system that takes a long time to eradicate
stream-flow traction-current processes. under "normal" weather conditions.
Glacial outwash deposits provided the first model Many of the studies reported here indicate a con-
for humid fans (Boothroyd and Ashley 1975; Miall siderable variation in the frequency and importance
1977; Sect. 8.2.2), although some have objected to the of mass-flow events and their deposits between one
use of the term "fan" for such deposits (McPherson fan and another within the same study area. The
and Blair 1993; Blair and McPherson 1994). The term steepness of the relief in the drainage basin and the
"sandur" is to be preferred. The deposits are nor- nature of the bedrock are important controls. Mass
mally characterized by stream-flow traction-current flows are favored by steep slopes and by bedrock that
deposits (Boothroyd and Ashley 1975; Boothroyd weathers to produce large volumes of fine-grained
and Nummedal 1978; Leblanc Smith and Erikkson sediment, especially clay. Therefore, within the same
1979; Mia1l1983b; Ashley et al. 1985), although mass- sedimentary basin, under constant tectonic and
flow processes are important in the event of the climatic conditions, the composition of the basin-
catastrophic floods - jokulhlaups - that characterize margin conglomerate wedge may vary along strike
many outwash plains. These floods are triggered by because of controls imposed by the character of the
the failure of ice dams or by other sudden glacial fan drainage basins.
Conglomerates dominated by mass-flow deposits grains eroded from older sedimentary sources (Miall
(Fig. 12.1A) may be somewhat more likely to have et al. 1978). Stanley and Wayne (1972) found that in
been deposited under arid climatic conditions, the the Pleistocene deposits of Nebraska, transport of
Trollheim fan being the extreme example (Blair and mechanically weak clast types (limestone, anortho-
McPherson 1992), but other criteria should be site cobbles, sandstone, hornblende schist, and mica
sought to test a climatic interpretation. Paraglacial schist) was favored during glacial periods, when
alluvial fans may also contain significant debris-flow chemical weathering was at a minimum and trans-
deposits. In most other humid environments lower- port rates were rapid.
flow-regime stream-flow conglomerates with inter- Darby et al. (1990) set out to determine the ways
bedded cross-bedded sandstones are dominant (Fig. in which gravels deposited in humid tropical envi-
12.1B). As discussed in succeeding sections, the ronments might differ from those deposited in other
sandstones, fine-grained sediments, and biochemi- climatic settings, but they were unable to identify
cal sediments that commonly are interbedded with any definitive facies criteria. However, they found
fans and braid deltas contain many other criteria, that deep tropical weathering of clasts produced
none of which are definitive, but which, in combina- much in-place matrix, and clasts tended to be
tion, should provide reasonably certain indications reduced to ghosts. Evans (1991) attributed gravel
of broad climatic characteristics (Fig. 12.3, Tables roundness to spheroidal weathering prior to erosion
12.2, 12.3). and transportation. Iwaniw (1984) interpreted well-
rounded clasts in his deposits as the product of
extreme weathering in a humid tropical climate.
12.5.2 The Influence of Climate Kochel and Johnson (1984) found that during the
on Texture and Composition of Gravels long periods between depositional events on mod-
ern fans in Virginia unstable minerals tend to
Climatic variables affect not only the production and weather, producing clay films around the weathered
transportation of clastic debris, but also the nature clasts and stable minerals. By contrast, in another
of the debris itself (Basu 1985; Johnsson and Basu study of tropical alluvial fans, Brierley et al. (1993)
1993). Styles of weathering and erosion are governed found no evidence of deep weathering of clasts and
largely by levels of water saturation, by temperature, no thick weathered zones, and suggested that this
and by biochemical processes. Mechanical weather- reflects the reduced importance of deep chemical
ing is favored in cold and arid climates; chemical weathering in the seasonally dry, savannah-type cli-
weathering is more important in humid environ- mate of Papua-New Guinea.
ments, especially in tropical settings, where organic Periglacial and boreal environments are charac-
activity is at its most intense. These variations have terized by several distinctive petrological and tex-
measureable effects on the petrology and texture of tural features. As noted above, the chemically more
gravel and sand in fluvial systems. Detrital sand- easily weathered clast types survive transportation
stone composition (including the detrital matrix of in greater abundance in dry or cold climates, where
conglomerates) may contain much climatic infor- rates of chemical weathering are lower (Stanley and
mation. This is a specialized area of study that is Wayne 1972). A composition-shape inversion may
beyond the scope of this book, but some key points develop in channel deposits, whereby resistant clasts
are noted in this section. of plutonic and igneous rock retain their rounded
Several workers have commented on the abun- shapes, whereas softer sedimentary clasts are subject
dance of carbonate sedimentary clasts and sand- to frost breakage and therefore tend to be angular
sized detritus in deposits interpreted to have been (Martini et al. 1993). Clay balls, including till balls,
deposited in cold or arid climates, the inference are common in channel-lag deposits and in the
being that in humid climates chemical weathering lenses of gravel that are commonly pushed by sea-
would have dissolved such material before signifi- sonal ice onto floodplains. Such heterolithic clasts
cant transportation and accumulation could occur are commonly generated by the severe erosion that
{MiallI970a; Besly 1988}. Some arid braidplain and occurs during jokulhlaups (Ashley et al. 1985; Mar-
playa-margin deposits consist largely of detrital tini et al. 1993). Casshyap and Tewari (1982) de-
mud and sand fragments composed almost entirely scribed large disrupted beds of diamictite in Late
of carbonate, including hematite-stained calcareous Paleozoic fluvioglacial gravels in India, which they
granules derived from reworking of soil nodules interpreted as ice-rafted blocks of frozen morainic
(Wells 1983), or detrital limestone and dolomite material. Collinson (1971b) described distinctive
Sandstones: The Significance of Sand Body Architecture and Sedimentary Structures 433
structures formed in the Tana River, Norway, by the banks are stabilized by peat levees. The contrasts in
action of seasonal river ice. They included ice gouge river behavior when forests in upland, catchment
marks up to 1 m wide and several tens of meters long, areas are destroyed by fire or deforestation are
eroded in the gravel and sand of the channel floor marked and well known. Runoff becomes more
and also on finer overbank sediments. A more subtle flashy, the sediment load increases, and a tendency
effect included flattening of current-formed clast may develop for debris flows to occur. The erosion of
imbrication by pressure from stranded ice masses, steep-sided arroyos or box-canyons is typical of
and scour holes arounded stranded ice blocks. Kettle flash-flood processes along basin margins and may
holes are depressions in river bed formed by the melt be an indicator of this type of sedimentation where
of buried ice masses. Some coarsening-upward proximal fluvial deposits are preserved.
gravel-sand cycles in Quaternary outwash (Costello Temporal changes related to climatic causes may
and Walker 1972) have been interpreted as the result be exemplified by several studies of rivers in the
of channel diversion into these depressions, which United States. Nadler and Schumm (1981) and
deepen as melting takes place (Rust 1975). Some of Schumm (1985b) described examples of two rivers,
these criteria are illustrated in Fig. 12.3. the South Platte and Arkansas, that evolved from
braided to meandering patterns within historic
times as a result of damming and irrigation that
12.6 Sandstones: The Significance resulted in lower discharge variability, lessening of
of Sand Body Architecture flood peaks, and stabilizing of banks by the growth
of vegetation. A climatic reduction in seasonality
and Sedimentary Structures would produce the same effect. Conversely, Schumm
and Lichty (1963) described the case of the Cimarron
12.6.1 Fluvial Style in Sand-bed Rivers River, Kansas, that changed from a suspension-load,
meandering morphology to a broad, shallow, bed-
Channel style in sand-bed rivers is controlled by load, braided morphology during a single major
several factors, of which the amount and variability flood in 1914. Channel widening continued until
of discharge, the nature of the sediment load, bank 1942, during a period of below-normal precipitation,
composition, and bank stability, are among the more which inhibited vegetation growth. However, local
important (Sect. 8.1). Some of these variables are factors unrelated to climate may produce the same
strongly affected by climate. Thus, humidity is the effects on channel style, as described in Sect. 8.1.
main control on vegetation, which has an important There may be major proximal-distal differences
effect on discharge characteristics, sediment load, in fluvial style between rivers in humid regions and
and fluvial style. In upland vegetated areas at the those in arid regions. In humid climates flow is pe-
present day, runoff following major rainfalls is rarely rennial, and channel scale increases downstream as
catastrophic because the precipitation is absorbed discharge is added by tributaries, whereas in arid
by soil and plants and released slowly. Also, sedi- climates there may be a loss of discharge down-
ment is stabilized by roots, and sediment yields are stream as a result of infIltration and evaporation,
relatively low. Therefore, as noted in the previous with a consequent reduction in channel scale. Rivers
section, mass-flow processes are less common in may be perennial close to an upland, possibly alpine
humid climatic settings. Vegetated banks are highly source, but become ephemeral downstream toward
resistant to erosion, as demonstrated in an experi- an arid basin center. For example, many of the
ment by Smith (1976). This limits the supply of sedi- present-day rivers draining into the vast inland ba-
ment to the river, and also reduces the bank sins of western China show this characteristic. Sedi-
widening that is necessary for braiding to develop. In mentologically, these differences may be recorded in
vegetated areas, therefore, because of the effects on the scale of preserved bedforms and macro forms,
discharge variability, bank stability, and sediment channel-fill cycles and flood cycles.
supply, braiding is inhibited. Baker (1978a) quoted The terminal-fan model encapsulates many of the
examples in tropical Africa and South America features of arid sandy systems, including the down-
where dense bank vegetation inhibits braiding, even stream decrease of channel scale, absence of channel
where abundant coarse sediment is present in the incision, and the lobate, distributary nature of the
banks. Stanistreet et al. (1993) described the chan- river system (Fig. 8.71). this has now been well docu-
nels of the Okavango fan in Botswana, which are low- mented in several studies, as described in Sect. 3. By
sinuosity to meandering bedload rivers, in which contrast, the downstream increase in channel scale
in perennial systems seems to be taken for granted uct of channel migration and macro form aggrada-
and has not been specifically tested in many studies. tion in low-sinuosity rivers (association 3 of
In a study of a Cretaceous braidplain system in Utah, Lawrence and Williams 1987). Another example was
Miall (1993) demonstrated a downstream increase in the Devonian distal braidplain deposit in the Cana-
sand body thickness over a distance of 140 km and dian Arctic described by Miall and Gibling (1978).
attributed this to a downstream deepening of fluvial These are now interpreted as flood cycles formed in
channels as tributary streams merged into fewer, an arid environment (Miall 1985). The original
larger rivers. South Saskatchewan model of Miall (1978c) there-
fore includes at least two distinct fluvial styles (see
Sects. 8.2.13 and 8.2.15).
12.6.2 Sand Body Architecture Luttrell (1993) suggested that "it takes a medium
to large-sized river with sufficient depth and width
Given a channel form determined by discharge char- before long-term sand flats will develop .... Only
acteristics and bank materials, the main climatic in the large perennial to intermittent rivers do chan-
control on sand body architecture is the ,control nel-macroform assemblages dominate the alluvial
exerted by discharge variability on lithofacies architecture. Under sustained discharge conditions
assemblages and bounding surface characteristics well-developed macro forms show a vertical decrease
(Fig. 12.2). It is to be expected that lithofacies hetero- in both the scale of bedding and grain size." In her
geneity will increase with increasing seasonal or study of the Kayenta Formation (Jurassic) of the
longer-term variability in discharge, and that the Colorado Plateau area she interpreted proximal
frequency of minor internal bounding surfaces (sec- sediments as the deposits of perennial streams
ond- and third-order surfaces) will likewise increase. largely on the basis of the abundance of large
As noted in Sect. 12.2, Jones (1977) classified river macroforms and channel scours indicating water
hydrography into five basic patterns, reflecting cli- depths of up to at least 8 m. In their study of Triassic
matic and topographic controls on discharge vari- sediments in New Mexico, DeLuca and Erikkson
ability. These variations are undoubtedly reflected (1989) also proposed distinguishing deposits formed
in lithofacies and architectural differences within under perennial flow conditions from those formed
fluvial deposits, but the differences are expected to by ephemeral flow on the basis of the occurrence
be subtle and not' amenable to simple interpretation. oflateral-accretion deposits and a comparison with
As noted earlier, the sedimentological differences the Platte- and South-Saskatchewan-type profiles of
between high-stage and low-stage sedimentation Miall (1978c), which were models based partly on
may provide no clear information on the time gap modern perennial rivers (Sects. 8.2.12, 8.2.13).
(diurnal, seasonal, flood-related, or longer term) Other studies of the Kayenta Formation indicate
between these stages. Macroforms and channel-fill that the deposits of the upper flow-regime plane-bed
assemblages typically display upward fining, but this condition, lithofacies Sh, are important in the proxi-
can be generated both by accretionary shallowing at mal regions of the depositional system, in south-
the depositional site and by waning discharge. The western Colorado (Miall 1988b; Bromley 1991a,b).
gradual filling of a channel or accretion of a macro- As discussed later in this chapter (and in Sect.
form under perennial flow conditions generates a 8.2.16), the presence of sand bodies several meters
fining-upward succession, but flood cycles also gen- thick composed predominantly of this lithofacies is
erate upward fining, because most sedimentation normally an indicator of flash-flood sedimentation,
during floods takes place during the waning stages, a style of discharge that occurs most typically in
following the erosion that occurs during peak run- ephemeral rivers. The Triassic deposits described
off. The two causes of upward fining were not at first by DeLuca and Erikkson (1989) also contain an
clearly distinguished from each other in sedimento- ephemeral stream facies, and their descriptions of
logical analyses. When first proposed, the South the deposits seem to confirm aridity throughout the
Saskatchewan model for sandy braided alluvium was succession. It is suggested that neither Luttrell
defined largely on the basis of its cyclicity (Miall (1993) nor DeLuca and Erikkson (1989) have made a
1978c), but the studies of such cyclicity that were convincing case for perennial flow in their studies.
cited include both types. One of these studies In fact, it is quite possible for ephemeral rivers to
described a portion of the Battery Point Formation develop deep channels with large macroforms that
(Devonian) of Quebec (Cant and Walker 1976), de- accrete only spasmodically, during flash floods (Fig.
posits which have now been interpreted as the prod- 12.2B). Shepherd (1987) described examples in the
Sandstones: The Significance of Sand Body Architecture and Sedimentary Structures 435
modern Rio Puerco, New Mexico. The point bars are therefore, be interpreted as necessarily indicating
up to 5 m thick and contain interbedded eolian perennial flow conditions. The presence of numer-
lenses. Turning to the ancient record of ephemeral- ous internal scour surfaces indicates frequent dis-
stream sedimentation, Stear (1983) described ribbon charge fluctuations. Macroforms consisting of
and composite sheet sandstone bodies in the Lower accretionary bundles a few decimeters thick, sepa-
Beaufort Group (Permian) of the Karoo Basin, south rated by third-order erosion surfaces, as described
Africa, that were deposited in channels up to 5 m by Stear (1983), Shepherd (1987), Langford and
deep and 100 m wide. Many of the sandstone bodies Chan (1989), and Munoz et al. (1992), maybe indica-
consist of accretionary bundles of plane laminated tive of ephemeral or strongly seasonal conditions
sandstone (lithofacies Sh) and trough cross-bedded (Fig. 12.2B).
sandstone (St) separated by numerous internal Another possible architectural indicator of high-
scour surfaces. Lateral-accretion surfaces within discharge sedimentation, such as is characteristic of
many of the sheet sandstones indicate the develop- flash floods, is the scour hollow (architectural ele-
ment of point bars. Well-developed fining-upward ment HO). Deep scours form at channel confluences,
cycles are also present (Turner 1978). Various lines as observed in flumes and in modern rivers, and a
of evidence, including numerous scour surfaces, facies model for the deposits of these scours has been
intraclast breccias, and desiccation features, indi- proposed by Bristow et al. (1993). This element is
cate an ephemeral, flash-flood style of sedimen- described and illustrated in Sect. 6.9.
tation. In a later paper, Stear (1985) was able to Numerous studies have confirmed the predomi-
compare these deposits with the products of a recent nance of unchannelized sheet flow in the medial to
flash flood in the arid Karoo Basin. Munoz et al. distal parts of arid braidplains, based on observa-
(1992) described accretionary bundles within Trias- tions of modern floods (McKee et al. 1967; Williams
sic lateral-accretion deposits in Spain. They com- 1971) and ancient deposits (Miall et al. 1978;
monly show a cyclic character, indicating that they Tunbridge 1981, 1984; Smoot 1983; Mertz and
were formed by floods or seasonal high-discharge Hubert 1990; Hirst 1991; Dreyer 1993b; Kelly and
events. The Westwater Canyon Member of the Olsen 1993). The proportion of interbedded fine-
Morrison Formation (Jurassic) in New Mexico con- grained deposits increases distally, depending upon
sists largely of sandstone sheets several meters thick the amount of fines in the sediment load. The
organized into macroforms (Cowan 1991) and con- presence of abundant sheet sandstone deposits deci-
taining well-developed fining-upward cycles of sev- meters to a few meters thick, showing very little
eral types (Godin 1991). This unit has also been evidence of channelization and no macro forms, but
interpreted as having been deposited by flash-flood characterized by fining-upward cycles a few meters
sedimentation in moderate-sized channels (Cowan thick or less (commonly only a few decimeters) and
1991) under arid climatic conditions (Parrish 1993). interbedded with playa mudflats with abundant evi-
In their study of a Devonian unit in Quebec, dence of desiccation (mudcracks, evaporite nod-
Lawrence and Williams (1987) described lateral-ac- ules), is now regarded as a definitive indication of
cretion deposits which they compared with the ephemeral, flash-flood sedimentation (Sects. 8.2.15,
ephemeral Rio Puerco examples described by Shep- 8.2.16). These observations have now been incorpo-
herd (1987). These fluvial styles are described fur- rated into the terminal fan model of Kelly and Olsen
ther in Sects. 8.2.8 and 8.2.14. In their study of (1993; Sect. 8.3). This style of sedimentation is illus-
terminal fans, Kelly and Olsen (1993) erected a trated in Fig. 12.2B.
"feeder zone" lithofacies assemblage for the deposits Feeder streams along the margins of modern arid
of the main channel in the apical regions of the fan. basins commonly occupy narrow, steep-sided can-
This channel may be ephemeral, or perennial with yons. In the arid southwestern United States they
widely fluctuating discharge. The lithofacies are are termed box canyons or arroyos. The distinctive
characterized by channel sand bodies up to several shape is probably generated by erosion during sud-
meters thick (their examples indicate channel den floods in unvegetated areas (Bendix 1992). Such
depths up to 6 m) and may contain the evidence of channels are rarely preserved (an example is illus-
braid-bar complexes (Sect. 8.3). trated in Figs. 6.11 and 6.12), but arroyo erosion may
The presence of channels several meters deep or deliver large quantitites of unconsolidated or loosely
cross-bedded assemblages organized into macro- consolidated sediment, including pedogenic nod-
forms with evidence of lateral or downstream accre- ules, into the dispersal system (Marriott and Wright
tion (architectural elements LA and DA) should not, 1993).
12.6.3 Bedforms and Cycles surface of reactivation during the next flow cycle.
The contrast between large, abandoned forms and
Jones (1977) suggested that the style of preservation the small, low-stage modifications may survive into
of bedforms in fluvial deposits should reflect the the sedimentary record as evidence of this distinc-
discharge regime of the river in which they were tive discharge regime.
formed. He classified discharge regimes into five In cases ofless rapid fall in discharge, a discharge
broad types, as noted earlier in this chapter. Critical style displayed by such large rivers as the Brahma-
parameters are the ratio of maximum to minimum putra and the Mississippi, dune scale may have time
discharge and the rate at which discharge changes. to become adjusted to reduced flow conditions, lead-
Bedforms have a characteristic "lag" or "relaxation ing to the superimposition of smaller forms on the
time", which is the time required for them to re- large dunes. Small eddies in the lee of the superim-
spond to changes in flow conditions. This time will posed forms are likely to erode the crest of the major
be much longer for large forms because of the vol- form, generating convex-up erosion surfaces that
ume of sand to be moved and, in fact, very large are buried by avalanching as the superimposed
dunes in major rivers, such as the Brahmaputra, are forms continue to migrate. In perennial rivers with
unable to undergo continual adjustment to changed slow or minor changes in discharge, such as the
flow conditions and tend simply to be abandoned Platte and Loup rivers, complete reworking of the
during the falling stage. Bedforms are rarely in bedforms can occur in any channel that remains
perfect equilibrium with flow because of constantly active. Channel migration and avulsion events there-
changing conditions, but in general, the type of fore result in the preservation of dunes displaying a
bedform assemblage and the various styles of wide range of scales formed under varying discharge
bedform modification should provide clues as to the conditions, with no clear preserved signal of the
discharge regime. Jones (l977) proposed several discharge regime.
models for field testing, but his work has not been Application of these ideas to ancient fluvial de-
systematically followed up by subsequent research- posits has not, to this writer's knowledge, been at-
ers. Following are some points regarding the genera- tempted. However, the concepts would appear to be
tion and preservation of features sensitive to worth testing in conjunction with the other climatic
discharge regime. criteria discussed in this chapter.
Is evidence of the discharge regime likely to be In channels deep enough for macro forms to be
preserved in the sedimentary record? Rising-stage developed, accretion and gradual shallowing of the
changes are modified by peak flows and have low channel at any given depositional site lead to a pro-
preservation potential. Falling-stage changes are gressive lowering of flow velocity and shear stress,
more preservable, taking into account the effects of resulting in a decrease in grain size and a predictable
bedform lag. Jones (1977) suggested that "sand change in the style of bedform generation. This has
waves" (also termed "linguoid bars") are most likely long been known to occur on point-bar surfaces and
to preserve differences in discharge regime because has formed the basis for the depositional model of
of their relatively long lag time. These correspond the fining-upward cycle, and for computer models of
to what we would now term medium- to large- meandering-river sedimentation under perennial
scale two-dimensional dunes (Ashley 1990; see Sect. flow conditions (Allen 1970b; Bridge 1975; Sect. 6.7).
5.3.1). Most three-dimensional dunes are small Are there patterns in the style of macroform preser-
enough that they are able to respond rapidly to vation that could be interpreted in terms of dis-
changes in discharge, and differences between high charge regime?
and low stage are not as likely to be preserved. The original descriptions of deep, perennial,
Rivers characterized by short, sharp floods, such sand-bed braided rivers, such as the Brahmaputra
as the Tana (Norway), have rapidly falling dis- (Coleman 1969) and the South Saskatchewan (Cant
charge, during which the larger bedforms may be and Walker 1978), did not recognize the three-di-
abandoned. They may become emergent, or may be mensional nature of the large sand flats, and it is now
draped by smaller forms deposited by low-stage run- known that they develop by lateral, oblique, down-
off flowing in small channeLs eroded between the stream, and even upstream accretion (Bristow 1987;
larger forms. Flow around emergent bedforms may Sect. 6.6). Cosets of downstream-climbing bedforms
erode part of some avalanche faces and deposit are common (Banks 1973) and were first observed in
cross-bed sets at a high angle to the dip of the major ancient deposits by Allen (l983a) and Haszeldine
fore set. The final abandonment surface becomes a (l983a,b). Upward fining and an upward decrease in
Overbank Fines: The Significance of Bedding and Minor Sedimentary Structures 437
the scale of cross-bedding have been observed in tinctive "hollow" architectural element formed
some of these deposits. Much work remains to be by scour at channel confluences (Cowan 1991). The
done to document the details of the architecture and three types of cycle are related to macroform accre-
vertical proftle of these macroforms under different tion, channel ftlling, and overall sand body aggrada-
flow conditions, and at present there are inadequate tion (Sect. 1O.3.2). A common lithofacies association
data for Jones' (1977) ideas regarding discharge re- is that between horizontally laminated sandstone
gimes to be tested. (Sh) and low-angle cross-bedded sandstone (Sl). The
Luttrell (1993) suggested that ephemeral streams low-angle cross-bedding is commonly asymptotic
are characterized by a paucity of bedforms formed at with reference to upper and lower set bounding
intermediate flow strengths, notably two-dimen- surfaces. Parting lineation may be present, with an
sional dunes (lithofacies Sp). She suggested that orientation parallel or oblique to the dip of the cross-
"the association of high-energy structures grading bedding. Some occurrences of lithofacies Sl repre-
vertically and laterally into waning-flow structures sent plane beds deposited on initially dipping
reflects ephemeral flood events." It is a matter of surfaces, such as scour hollows. In other cases the
observation that fluvial systems containing abun- lithofacies represents distinct bedform geometries.
dant planar-tabular cross-bedding (Sp) are pre- The washed-out dunes and humpback dunes that
dominantly shallow perennial streams" or at least occur at the transition between sub critical and
show steady summer discharges, the modern Platte supercritical flow typically form this lithofacies
and William rivers being excellent examples (Smith (Saunderson and Lockett 1983).
1970; Blodgett and Stanley 1980; Smith and Smith Observations of modern floods by Frostick and
1984; Sect. 8.2.12). Planar-tabular cross-bedding also Reid (1977) and Stear (1985) indicated that they
forms a significant component of many deep, peren- occur in pulses. The passage of storms across differ-
nial, sand-bed braided rivers, such as the modern ent parts of a catchment area triggers waves of high-
South Saskatchewan, where fields of three-dimen- energy flow from different tributary systems which
sional dunes commonly occupy the deeper channels arrive in the depositional area at different times.
and large, flat-topped two-dimensional dunes de- This may be recorded in the deposits in the form of
velop in shallower reaches (Cant and Walker 1978; changes in the thicknesses of flood laminae (Frostick
Sect. 8.2.13). This river was the focus of a useful and Reid 1977) or as heterogeneous lithologies
discussion of channel aggradation and the paleo- (Stear 1985).
current characteristics of different bedform types by
Cant and Walker (1978). Flume observations indi-
cate that at any given flow depth two-dimensional 12.7 Overbank Fines: The Significance
dunes form at lower flow velocities than three-
dimensional forms (Sect. 5.3.I).
of Bedding and Minor Sedimentary Structures
Flood cycles in ephemeral streams commonly
consist of sets of St overlain by Sh or Sr, reflecting Overbank areas are the main sites for the accumula-
high-energy sheet flow followed by a rapid decline in tion of various climatically sensitive chemical and
discharge. Planar-tabular cross-bed sets (Sp) may be biochemical deposits, which are discussed in a later
present, but the large flat-topped bedforms are not section. However, the detrital sediments can also
as likely to form as in perennial rivers. Reactivation yield some clues as to discharge regime, which de-
surfaces within bedforms are rare, most forms being pends on climate (Fig. 12.3).
deposited rapidly in a single event and then com- Interfluve areas flanking the large channels of
pletely abandoned. Many variations on this succes- perennial, seasonal, or ephemeral rivers are areas
sion have been described following work in various that are only spasmodically affected by high-dis-
modern and ancient ephemeral systems (McKee et charge events. In tropical-humid areas the rivers
al. 1967; Miall and Gibling 1978; Williams 1971; may be flanked by raised peat bogs that keep out
Tunbridge 1981, 1984; Sneh 1983; Glennie 1987; overbank flows (McCabe 1984) or by swamps that
Luttrell 1993; Dreyer 1993b; Sects. 8.2.15, 8.2.16). absorb them (Stanistreet and McCarthy 1993). In
Godin (1991) documented three scales of auto- temperate-humid and arid settings rare flows
genic cyclicity in the Westwater Canyon member of may spread over the floodplains, and normal over-
the Morrison Formation, in northern New Mexico, a bank processes lead to the generation of crevasse
unit probably formed under high-energy ephemeral channels and crevasse splays (Sect.' 7.2), the geo-
flow. As discussed earlier, this unit contains the dis- metry and composition of which convey little useful
information regarding climate. Lenses of gravel may Several types of minor sedimentary structure
be deposited in areas close to the channel by rare carry useful climatic signatures (Fig. 12.3). Desicca-
flood events (Costa 1974) or by ice push (Martini et tion features are common in hot, arid fluvial sys-
al. 1993). tems, including large mudcracks and mud curls on
The most common detrital sediments on flood- bedding planes (Miall et al. 1978; Tunbridge 1981,
plains in all climatic settings are fine silts and muds. 1984; Rust and Legun 1983; Smoot 1983; Flint 1985;
These are normally finely laminated, but commonly Stear 1985; Glennie 1987; Mertz and Hubert 1990;
the lamination is disturbed by pedogenesis, by roots, Hartley 1993). Cracks may be up to at least a meter
or by invertebrate bioturbation. The style oflamina- deep and a decimeter or more in width. Even larger
tion in these fine-grained clastics normally reveals cracks, up to 1.5 m deep, have been recorded in
nothing about climate, although Broadhurst (1988) evaporitic eolian sandstones (Kocurek and Hunter
developed an argument suggesting that laminated 1986). They are similar to ice-wedge casts formed in
siltstone-claystone couplets in a Carboniferous suc- frigid climates (Bryant 1983), and both processes
cession in Britain are seasonal in origin. He recorded may form polygonal crack networks on bedding-
some 45 couplets in 9 m of section (average couplet plane surfaces. Supplementary evidence might be
thickness 20 cm). The siltstone-claystone units are needed to differentiate such features. Rip-up clasts
interpreted as lacustrine or overbank deposits, and and intraclast breccias are very common features of
interbedded sandstones as distal crevasse-splays. A hot, arid environments (even in high intertidal set-
bivalve colony shows evidence of being buried by the tings) because of the rapid lithification of freshly
deposits and not replenished. Given the 20-year av- deposited muds caused by desiccation (Smoot 1983;
erage life span of modern bivalves, the persistence of Stear 1985).
the colony through eight or nine of the overlying Parts of the arid Lake Eyre drainage basin of
couplets suggests that the couplets may be seasonal. Australia are characterized by an unusual combina-
The sedimentation rate calculated from this evi- tion of fluvial styles. A series of seasonally active
dence is 30 cm/year, which Broadhurst (1988) re- anastomosed channels containing permanent pools
garded as high, but in fact it is in agreement with the of water cross a broad system of shallow braided
average rate for seasonal deposits (group 5 of the channels which appeared to Rust (1981) to be relict
depositional hierarchy, Chap. 3). These deposits cor- channels dating from an earlier, wetter, climatic
relate with and are interbedded with coals, which phase. This change in fluvial style has, in fact, long
indicates permanent saturation, and which appears been regarded as a classic example of "river meta-
to conflict with the seasonality suggested by the cou- morphosis" in response to climate change (Schumm
plets. Broadhurst (1988) suggested that the coals 1969). However, detailed studies by Nanson et al.
were probably formed in permanent swamps, while (1986) and Rust and Nanson (1989) showed that the
the rivers were affected by seasonal discharge fluc- braided channels and intervening bars are com-
tuations. posed of mud aggregates forming sand-sized par-
In many arid systems fine-grained sediments ticles that retain their coherence during fluvial
deposited on distal interfluve mudflats are inter- transport and function as sand grains. The high
bedded with lacustrine or playa margin sediments "sand" load causes the rivers to develop the braided
(Figs. 7.27, 7.28). Layers of very fine grained sand- pattern during floods. The clay aggregates are inter-
stone and siltstone may display a millimeter-scale preted as peds formed by volume changes of the
lamination and commonly show graded bedding, floodplain mud in response to wetting and drying
indicating rapid deceleration of flow into very shal- cycles in the arid to semiarid climate. Earlier work
low standing bodies of water (Smoot 1983; Flint had suggested that the presence of salts is required
1985; Mertz and Hubert 1990; Hartley 1993). Evapor- for aggregate formation, but Nanson et al. (1986)
ite crystallization shortly after deposition may disputed this, pointing out that the evaporite con-
disrupt bedding (Mertz and Hubert 1990). Couplets centration is low. Flume experiments indicated
of interbedded clastic and chemical sediments that mud aggregates can form flow-regime type
may reflect seasonal variations in precipitation, bedforms, and it was suggested that mud beds in the
evaporation, and temperature, as in the siltstone- ancient record should be reexamined, as some may
carbonate cycles in parts of the Chinle Formation be mud braids, not floodplain deposits (Rust and
(Triassic) of the Colorado Plateau (Dubiel et al. Nanson 1989). In the modern Lake Eyre basin the
1991). Clastic-chemical cycles also develop in lake- anastomosed channels are deeper channels which
margin settings under the influence of longer-term have cut through lower levels in the muds, where the
climatic rhythms, as discussed in a later section. mud has greater cohesion, and into an underlying
Associated Clastic, Chemical, and Biochemical Sediments 439
sand bed. Ancient examples of pedogenic mud ag- Such a setting favors pedogenic carbonate develop-
gregates have now been described by Rust and ment (Retallack 1981, 1986). A reducing environ-
Nanson (1989), Ekes (1993), and Talbot et al. (1994). ment occurs in saturated environments and is
necessary for the preservation of plant matter. These
differences are commonly preserved following
deposition. In fact, different facies associations with
12.8 The Significance of Color the expected contrasting colors may be interbedded
within the same basin fill, indicating variations in
The significance of color in nonmarine sediments water table across the basin during or shortly after
has been the subject of much debate (Van Houten deposition (e.g., Besly 1988). The existence of these
1973; Walker 1967; Turner 1980) that will not be complex color patterns also indicates that, in such
repeated here. Drab colors (gray, green, brown) are cases at least, colors are set during early diagenesis
generated largely by ferrous oxides or carbonates and not subsequently modified.
(e.g., siderite) in reducing environments, whereas Although colors reflect levels of water saturation,
red, yellow and orange are created by ferric oxides, the climatic implications of this may not be easily
notably hematite, in oxidizing environments. determined, especially where saturation levels vary
Studies in hot arid environments and in humid across a basin. This is discussed further in a later
tropical regions show that red beds can be generated section.
by diagenetic processes following deposition
(Walker 1967, 1974). Yellow and brown iron oxides,
in initial stages of hematite production, are present
above the water table in the modern alluvium of the
12.9 Associated Clastic, Chemical,
Colorado River on the arid shores of the Gulf of and Biochemical Sediments
California, and well below the water table in allu-
vium in Puerto Rico, under hot, humid conditions. Many other nonmarine deposits constitute much
Therefore, red beds are not necessarily an indication clearer climatic indicators than do fluvial lithofacies.
of arid desert climates, as has long been thought. Lakes are sensitive indicators of climate because of
Turner (1980) summarized mineralogical, diage- the delicate balance that commonly exists between
netic, and grain-size data relating to post-deposi- climate and the water level and chemistry of lakes
tional generation of red-bed coloration, drawing on (Katz 1990). Large-scale eolian depositional systems
the work of Walker. He distinguished between (a) a may indicate aridity, and large ancient ergs preserve
desert-evaporite red-bed association in which red evidence of persistent wind patterns that may be
beds are associated with eolian sands, desert fluvial related to the patterns of global atmospheric cir-
sediments, and evaporites formed in playas and in- culation (Parrish and Peterson 1988). Chemical and
land sabkhas, and (b) a moist-climate red-bed asso- biochemical sediments that occur in interfluve envi-
ciation in which red beds are interbedded and ronments, particularly paleosols, coal, and evapor-
interfinger with coal-bearing strata. ites, also contain useful signatures of the climate at
The oxidation state is commonly set during early the time of deposition. It is not the purpose of this
diagenesis rather than at the time of deposition, and chapter to review these climatic indicators at length,
so caution must be used in making environmental most of which have been extensively reviewed else-
interpretations from color. However, many red units where. However, some brief comments are included
contain evidence of an arid or semiarid climate, such here in order to provide the basis for an integrated
as pedogenic carbonates and facies associations in- interpretation of climate from an ancient alluvial
dicating ephemeral deposition, whereas drab units suite.
are typically associated with coals or carbonaceous
shales, indicating saturated environments of deposi-
tion and a humid climate. The preservation of or- 12.9.1 Coal
ganic matter is a function of organic productivity,
burial rate, the depth of the water table, and the rate McCabe (1984) stated:
of groundwater movement.
The main attribute controlling color differences "The necessary conditions for peat accumulation can ... be
regarded as a balance between plant production and or-
is the level of the water table. An oxidizing environ- ganic decay. Both are a function of climate. Plant produc-
ment is favored by a low water table, such as occurs tion is fastest in hot humid areas and is slowest in dry
in arid climates and those with seasonal dry periods. or cold areas. Organic decay is principally a function of
temperature, being slower in cooler climates .... today the overbank flow. Soil types depend on substrate con-
vast majority of peat is accumulating in cool climates - ditions, including chemistry and water saturation,
mostly between 50° and 70° N. Most tropical rain forests
are not sites of peat accumulation because the organic and undergo continual modification with time. Two
matter rapidly deteriorates on the forest floor. The major major concepts are useful in understanding the vari-
area of peat formation in the tropics today occurs in parts ability of paleosols. A "catena" is a soil unit showing
of Southeast Asia where the rainfall is over 2 m year- 1 and lateral variations reflecting differences in relief and
where substantial precipitation occurs throughout the drainage, for example, the differences between the
year. Because evaporation is also a function of tempera-
ture, it is wrong to assume that coal is an indicator of high immature soils developing on a well-drained levee
rainfall; much of the vast peat swamps of Siberia and undergoing periodic addition of increments of sedi-
Canada receive less than 50 cm of precipitation per year.... ment, the soils of a saturated backswamp, and the
Correlation of palaeolatitides with areas of coal deposition mature soils of an adjacent, well-drained, upland
... shows that coal has been deposited at all latitudes from region (Wright 1992). "Maturity" refers to the state
the equator to polar regions, with the majority forming in
mid-latitudes .... Coal itself is not a good palaeoclimatic of completion of the pedogenesis process, and there-
indicator, although characteristics of a coal may be related fore is related to the age of the soil. Bown and Kraus
to palaeoclimate." (1987) pointed out that the concept of the catena
was introduced primarily to encompass variations in
McCabe and Parrish (1992) reviewed the global
modern soils related to their position on hillslopes,
distribution of coal in Cretaceous strata. They noted
whereas most paleosols of interest to sedimentolo-
that coals require nonsaline wetlands for formation.
gists occur within the floodplain, and differences
Continuous wetness is required to permit growth
within and between individual soils depend on
and the replenishment of the nutrient supply for
maturation time and the amount of dilution of the
plants. Wetness is also a necessary condition for the
soil with clastic sediment. They found that these
anoxic environment required for preservation. They
factors depend primarily on proximity to active flu-
stated:
vial channels, and they suggested that a different
"Climate directly influences plant productivity and pres- terminology is required to accommodate such ob-
ervation of plant matter and therefore determines the servational and interpretive differences. They intro-
amount of organic accumulation. High water tables, nec-
essary for the accumulation of peat, tend to occur in cli- duced the term "pedofacies", which they defined to
mates where precipitation exceeds evaporation, especially encompass the "laterally contiguous bodies of sedi-
where rainfall is evenly distributed throughout the year mentary rock that differ in their contained laterally
(Ziegler and others 1987). Seasonal climates, by contrast, contiguous paleosols as a result of their distance
often lead to such large annual fluctuations in the water (during formation) from areas of high sediment ac-
table that organic matter accumulated during the wet sea-
son decays during the dry season, and no net accumulation cumulation." These ideas are discussed further in
of organic matter takes place (Ziegler and others 1987)." Sect. 7.4.2.
The dependency of soil type on substrate and
McCabe and Parrish (1992) pointed out that an ex- groundwater chemistry, on water saturation, and on
cess of precipitation over evaporation does not nec- the rate of groundwater flow through the soil intro-
essarily imply high rainfall. Such a condition may duces climatic factors into soil genesis, although care
develop in cool climates with modest rainfall. must be taken to distinguish depositional from
Very little vegetation cover may suffice to gener- diagenetic conditions. Thus ferric nodules and con-
ate coaly layers in a fluvial deposit. Hubert and cretions typically indicate well-drained, oxidized
Filipov (1989) described the debris-flow-dominated soils; siderite nodules indicate neutral to alkaline,
sediments of Owens Valley, California, an arid area waterlogged soils; calcareous nodules indicate well-
with scattered scrub vegetation. Many of their sec- drained alkaline soils (Retallack 1988). Retallack
tions contain layers of plant debris, which would (1986) reached several generalized conclusions
become coaly streaks following compaction and di- based on his studies of the Upper Eocene and Oli-
agenesis. Therefore, climatic interpretations should gocene succession of Badlands National Park, South
not be attempted from the presence of coal unless Dakota. Several features reflect rainfall and water-
the accumulations are substantial. table factors. For example, calcic soils characterize
dry subtropical climates, where evapotranspiration
exceeds precipitation for most of the year, and the
12.9.2 Paleosols depth to the top of the calcic horizon is related to
mean annual rainfall, becoming deeper in wetter
Pedogenesis takes place on any interfluve surface climates. Clay production in soils increases with
not receiving frequent additions of sediment from humidity. The type of clay also varies with climate,
Associated Clastic, Chemical, and Biochemical Sediments 441
with kaolinite indicating humid climates and illite merges imperceptibly into a playa lake or arid tidal
more common in more arid climates. This reflects flat (Miall et al. 1978; Smoot 1983; Glennie 1987;
the leaching of soluble cations, which decreases with Mertz and Hubert 1990; Dreyer 1993b). Mertz and
aridity. Quartz/feldspar ratios in soils also indicate Hubert (1990) reported a gradual increase in hu-
humidity, with feldspar being destroyed by acidic midity during deposition of the Triassic-Jurassic
weathering in humid climates. Indicators of tem- Blomidon Formation of Nova Scotia, based in part
perature are harder to find. In general, soil becomes on the upward change in intergranular cements
redder in warmer climates, but the amount of rain- from halite and gypsum to calcite.
fall and the age of the soil complicate this relation-
ship.
However, the climatic signature may be very diffi- 12.9.4 Eolian Interbeds
cult to interpret. The difficulties discussed above
with regard to the interpretation of color apply Many fluvial deposits formed in arid settings contain
equally to paleosols. Wright (1990) pointed out, with interbedded layers or lenses of eolian sand. Large-
regard to the calcretes that develop in semiarid scale interbedding may occur at the margins of ergs,
areas, that variations in exposure time (determined or where these are crossed by ephemeral river sys-
by subsidence and channel return time) and local tems, but minor eolian units can occur in any fluvial
variations in rainfall and calcium content all lead to system where aridity and wind strength are suffi-
markedly different rates of soil development. cient for the eolian reworking of fluvially trans-
Given these difficulties, some generalizations ported detritus. Glennie (1970) published a classic
may be made, using the soil-type terms and descrip- study of desert sedimentary environments that con-
tions adapted by Retallack (1988) from the U.S. tains much descriptive information on fluvial-eolian
Department of Agriculture classification. Aridisols interactions, and a detailed local study was pre-
develop in arid climates, and are characterized by sented by Langford (1989).
light coloration and a thin calcareous layer close to Langford and Chan (1989) recognized four major
the surface of the profile, and may include pedogenic facies in mixed eolian-fluvial systems, based on their
gypsum or other evaporite minerals. Vertisols con- study of the Cutler-Ceda Mesa Sandstones (Per-
tain abundant swelling clays (smectite) to depths of a mian) of Utah. 1. Large-scale cross-stratified eolian
meter or more and may display slickensides or clas- sandstone; 2. the deposits of wet interdune areas,
tic dykes. This type of soil develops in climatic set- consisting of interbedded lenses of sandstone and
tings with seasonal variations from saturated to dry. limestones. The sandstones are structureless or re-
Oxisols are a type of soil developed in humid tropical worked into wave ripples. Gypsum crystal casts and
settings and consist of a thick, well-differentiated to bioturbation are common; 3. Fluvial channel sand-
uniform profile with a clayey texture, red, highly stones. Common lithofacies are St and Sh, which
oxidized subsurface horizons, and an absence of may be organized into accretionary macroforms
easily weathered detrital minerals. Many other soil (elements DA and LA), consisting of bundles of
types are listed by Retallack (1988). trough cross-bedded sandstone separated by third-
Mack et al. (1993) developed a paleosol classifica- order surfaces; 4. overbank interdune sandstones
tion better adapted to the study of ancient rocks than and shales in thin sheets and lenses up to 2 m thick.
those developed for agricultural purposes, and Mack Mudcracks and bioturbation are common. Sand-
and James (1994) discussed the paleoclimatic signifi- stones consist of lithofacies Sr, Sh, and Sm. Small
cance of this classification. channel forms up to 2 m thick may be present,
formed from minor overbank flows. Isolated eolian
dune lenses up to 1.5 m thick may also be present,
12.9.3 Evaporites formed by dune migration over desiccated interfluve
surfaces between floods. Langford and Chan (1989)
Evaporite minerals are deposited within fluvial sub- traced some fluvial flooding surfaces over areas of>
strates by the evaporation of groundwaters and on 400 km2, indicating extensive flooding episodes.
the surface of playa mudflats during the evaporation Bioturbation and bleaching indicate incipient pedo-
of sheet floods. Thin beds, laminae, nodules, and genesis, but true calcretes are rare. Dreikanters
individual crystals (or crystal casts) of evaporite, (ventifacts) may be present on eolian deflation sur-
particularly gypsum and halite, are very common in faces, but they require lengthy exposure to the wind
the deposits of arid fluvial systems, especially in the to develop and therefore may not be common in
distal regions where the braidplain or terminal fan areas dominated by fluvial sedimentation.
Bromley (1991a,b) mapped an eolian rippled-silt- A good example of apparently conflicting cli-
stone unit interbedded with braided sandstones in matic indicators occurs in the Late Carboniferous to
the Kayenta Formation (Jurassic) of western Colo- Early Permian coal-bearing succession of central
rado. Similar mixed facies were described by Luttrell England and was highlighted by Besly (1988). He
(1993), based on her regional study of this unit. documented a diachronous interbedding between
Minor eolian lenses interbedded with fluvial con- an alluvial coal-bearing succession, an alluvial-plain
glomerates or sandstones have been recorded in sev- red-bed association, and an alluvial-fan red-bed
eral studies. For example, in his study of the Triassic association. The transition from gray to mottled to
Budleigh Salterton Pebble Beds of SW England, red between these assocations indicates improved
Smith (1990) noted sand lenses with distinct texture drainage of the floodplain. Additional details on the
and petrography. Unlike the fluvial sands in the paleosol catena that indicates a lateral transition
system, they contain no mica and are well sorted and from dry (red) to wet (gray) beds, corresponding to
free of pebbles. Shepherd (1987) documented eolian fan progradation, was provided by Besly and Field-
lenses within the modern lateral-accretion deposits ing (1989). According to Besly (1988), "the transition
of Rio Puerco, New Mexico, distinguishing them from the deposition of coal-bearing sediments to
from fluvial units on the basis of textural analyses. that of red beds resulted from an improvement in
Clemmensen and Tirsgaard (1990) recognized a dis- drainage conditions at the site of deposition. This
tinctive type of bounding surface, which they termed was initially achieved by the progradation of a topo-
"sand-drift surfaces", separating eolian from fluvial graphically elevated, well-drained floodplain area
deposits. over poorly drained alluvial backswamps. Repeated
Flint (1985) noted an absence of eolian sands in progradation led to a progressive drop in groundwa-
his study of the arid mid-Tertiary Pacencia Group of ter table levels, possibly reinforced by seasonal fluc-
Chile. He attributed this to armoring of the sediment tuations ... which eventually led to the accumulation
surface by evaporite crystallization, which prevented of a continuous red-bed sequence." Besly (1988)
eolian deflation. commented that "while the Late Carboniferous red-
bed sequences do allow climatic reconstruction, they
do not in all cases indicate aridity." He noted that the
12.9.5 Palustrine Limestones fairly rapid onset of more arid conditions during the
Westphalian D and Stephanian corresponds to the
Freshwater limestones containing a pedogenic over- acme of coal formation in basins within the moun-
print are present in some alluvial successions, and tain belt to the southeast. This contrast suggests the
indicate the presence of temporary ponds in the growth of a rain shadow in the lee of the Variscan
overbank environment (Platt and Wright 1992). mountains.
They occur in a variety of climatic setttings, the Blakey and Gubitosa (1984) found what appeared
nature of which may be indicated by associated fea- to be conflicting evidence of climate in the Chinle
tures, such as evaporite nodules, desiccation fea- Formation (Triassic) of the Colorado Plateau. An
tures, karst solution, or thin coals. abundant and diverse fossil fauna and flora suggests
a humid climate, whereas the presence of caliches
and other signs of overbank aridity suggest a dry
climate. Their suggested explanation is that the riv-
12.10 Contrasting Climatic Indicators ers were perennial, fed by a humid source area, but
that the depositional basin itself was more arid, with
In many recent regional studies sedimentological at least a seasonal dry period dominating conditions
analysis has suggested that different parts of a sedi- on the floodplains. These suggestions have recently
mentary basin were characterized by different satu- been confirmed and amplified by Dubiel et al.
ration levels, suggesting variations in the balance (1991). The juxtaposition of humid upland source
between rainfall, runoff, infiltration, and evapora- areas and arid basins is a common one, especially
tion. In other cases it can be argued that the basin where the upland region generates orographic rain-
and its fluvial source area were characterized by fall, leading to a permanent rain shadow downwind.
different climates. Such differences are commonly The San Joaquin valley in California and the San Luis
observed in present-day sedimentary basins, but valley of Colorado are good modern examples of
their presence requires subtlety and care in the inter- such contrasting climatic regimes.
pretation of climatic indicators.
The Interrelationship Between Tectonics and Climate 443
DeLuca and Eriksson (1989) distinguished two been in a more humid setting and was the site for the
contrasting fluvial assemblages in the Chinle Forma- accumulation of sandy braided stream deposits with
tion of New Mexico. They suggested that proximal, generally more mature paleosols that show pedo-
basin-margin rivers forming a tranverse drainage facies relationships to the channels.
system were ephemeral, whereas the main basin-
center rivers were perennial. Their evidence for pe-
rennial flow is not strong, as discussed earlier in this 12.11 The Interrelationship
chapter, consisting as it does mainly of comparisons
Between Tectonics and Climate
with the mainly perennial rivers that formed the
basis for the Platte and South Saskatchewan fluvial
models, with which their rocks are compared. The The importance of the orographic effect on the con-
evidence for ephemeral flow and arid conditions is trol of local climates has long been known. For ex-
much stronger, consisting of many of the sedimento- ample, it is common knowledge that the temperate
logical attributes discussed in this chapter. However, rain forests of the west coast of North America de-
the explanation DeLuca and Eriksson (1989) offered pend in large measure on the adiabatic rainfall gen-
for the contrasting fluvial styles is not unreasonable. erated from westerly winds as they rise over the
They suggested that in the basin center the water Cordilleran mountains. The presence of a significant
table was closer to the surface, in part because of rain shadow downwind, in such areas as the upper
groundwater recharge which kept the rivers flowing Fraser Valley (Kamloops) and the Okanogan Valley
during seasonal dry periods. Hobday et al. (1981) of British Columbia, is equally well known. Another
noted a similar climatic differentiation in present- example is the Dead Sea Rift (Manspeizer 1985). The
day south Texas, where "semiarid, caliche-soil, orographic effect of the uptilted rift shoulder west of
chaparal conditions commonly prevail on uplands the rift is the generation of high rainfall; some of this
while nearby contemporaneous riverbottoms sus- flows west in perennial streams draining into the
tain a humid subtropical assemblage of bald cypress Mediterranean, but it also feeds ephemeral streams
and willows, all owing to different microclimate as draining into the Dead Sea rift. The difference in
dictated by water table conditions and associated annual rainfall between the rift crest and the rift
soil moisture regimes." valley is substantial, at > 800 mm/year on the rift
Luttrell (1993) suggested a downstream change shoulder, in the vicinity ofJerusalem, and < 100 mm/
from perennial to intermittent in the rivers that de- year in the Dead Sea area.
posited the Kayenta fluvial system of the Colorado Knowledge of this type of effect is readily applied
Plateau. Again, the evidence for perennial flow to regional paleogeographic and paleoclimatic inter-
seems weak, based mainly on the scale of the chan- pretations. For example, Besly's (1988) suggestion of
nels and macro forms in the proximal part of the the rainshadow effects of the Variscan Mountains
basin (a criterion discussed earlier in this chapter). (Late Paleozoic) in northwest Europe was discussed
However, the downstream change from perennial to earlier in this chapter. Miall (1984b) invoked re-
ephemeral is one that characterizes many arid basins gional tectonically-related climatic variations as a
in modern deserts, for example, the inland basins of possible control on the distribution of fluvial
western China, where snow-fed mountain streams lithofacies assemblages in the Paleogene fluvial de-
enter hot, lowland deserts, such as the Taklamakam posits of Arctic Canada. He suggested that the un-
Desert of Tarim Basin. usual abundance of coals (and of fossil vertebrates)
Miocene fluvial and lacustrine deposits of the in central Ellesmere Island could be an indication of
Madrid Basin, Spain, reveal numerous differences a locally humid climate related to adiabiatic rainfall
between northern and eastern margins of the basin generated by the effects of regional uplifts to the east
(Alonso Zarza et al. 1992). In part, these differences and north of the depositional basin, whereas litho-
reflect differences in sediment sources and tectonic facies in Eclipse Trough, in the area of northwest
activity, but it also appears that there were subtle Baffin Island, indicate a predominance of ephemeral
climatic differences between the two areas that may, rivers and a drier climate, possibly related to the
in part, have been caused by topographic effects. The passage of westerly winds over the broad interior
northern margin may been under the influence of a plains of the Arctic Platform.
relatively dry climate and is characterized by alluvial Evidence from fluvial lithofacies, coupled with
fans with abundant debris flows and immature floral, isotopic, and other evidence, indicates a
calcrete paleosols. The eastern margin may have gradual global cooling during the Cenozoic. Major
regional uplift of such areas as western North Hoffman and Grotzinger (1993) speculated that
America and Tibet has also been documented. This the climatic belt in which a rising orogen develops
has led to a lengthy and complex debate regarding actually influences the tectonic style of the orogen
the role of uplift in generating the cooling trend. This and the architecture of the adjacent foreland basin.
debate is largely beyond the scope of this chapter, Areas of high precipitation, as in monsoonal belts,
but because discussions of fluvial lithofacies assem- are characterized by rapid erosional unroofing, lead-
blages and petrology have formed part of the evi- ing to rapid uplift, deep erosion, and the develop-
dence, some relevant parts of the controversy are ment of a foreland basin overfilled with nonmarine
briefly summarized here. sediments (but of low preservation potential). The
Ruddiman and Kutzbach (1990) summarized ex- Grenville orogen (1.1 Ga) of eastern North America
tensive paleobotanical evidence for widespread glo- may be cited as a possible example. This contrasts
bal cooling during the Cenozoic and developed with orogens such as the Alleghenian, which may
numerical climate models to analyze global climates have developed in arid climates. Erosional unroof-
with and without the effects of large-scale plateau ing would not compensate uplift, leading to preser-
uplift. They pointed out that areas such as southern vation of the fold-thrust belt and an underfilled
China, Burma, and Thailand, which did not undergo foreland basin containing deep-water sediments.
significant uplift during the Cenozoic, show no Sinclair and Allen (1992) pointed out the feedback
botanical evidence of cooling during the Cenozoic, effects of elevating a fold-thrust belt, resulting in
which would seem to confirm the tectonic causes of greater orographic rainfall, increased rates of ero-
cooling. However, they did not point out that these sional unroofing, and consequently enhanced uplift
areas are located in equatorial regions, which are rates.
not necessarily strongly affected by global climate In a recent lengthy review, Eyles (1993) demon-
change. Their field evidence and modeling data sup- strated convincingly the importance of tectonic up-
port the concept of plateau uplift as a cause for global lift in generating the climatic cooling necessary for
cooling, but the data are inadequate to verify this as the triggering of widespread continental glaciation
a cause (or at least a trigger) of the Late Cenozoic throughout geologic time. Many of the major Phan-
glaciation. Regional uplift is also cited as a cause of erozoic and Proterozoic glacial episodes are corre-
increased aridification because of rain-shadow ef- lated with times of regional continental extension
fects and because of changes to global circulation and rifting. However, glaciofluvial deposits repre-
patterns. sent a very small fraction of the glacially generated
Molnar and England (1990) stated: "Late Ceno- clastic successions in the ancient record.
zoic uplift has been inferred for mountain ranges
throughout the world, yet globally synchronous
changes in plate motions, if they have occurred, have
been small. Climate change, on the other hand, has 12.12 Orbital Forcing
been a global phenomenon." Indicators of cooling
climate in sediments include increased grain size, 12.12.1 Sedimentary Evidence of Orbital Forcing
increased detrital volume and sedimentation rates
reflecting increased erosion rates (Stanley and Rhythmicity in glaciation is now attributed to orbital
Wayne 1972; Stanley 1976), changes in fauna and forcing, the so-called Milankovitch effects, named
flora, and changes in oxygen and carbon isotope after the Serbian mathematician who first developed
ratios. These changes could mean regional or global the quantitative relationships between the orbital
climate change, or uplift without climate change. behavior of the earth and its effects on incoming
Rapid uplift in the past 2 million years has been solar radiation. There is now a very considerable
interpreted for the Himalayas and elsewhere, but literature on the theory and sedimentological impli-
Molnar and England (1990) suggested that this may cations of orbital forcing, most of which is beyond
be incorrect. Plate tectonic reconstructions ofIndia- the scope of this book (e.g., Berger et al. 1984; Fischer
Asia convergence suggest that the Tibetan Plateau et al. 1990; de Boer and Smith 1994a). This section
would have risen gradually during the past 50 mil- makes only very brief reference to the theory and
lion years. This would have set off a long-term slow focuses on the few examples of studies in the non-
cooling, amplified by feedback effects, that was then marine record where some evidence of the influence
enhanced in the Late Cenozoic by Milankovitch proof orbital forcing may be demonstrated, or at least
cesses, leading to glaciation. postulated.
Orbital Forcing 445
There are several separate components of orbital the climatic control of rainfall and evaporation,
variation. The present orbital behavior of the earth upon which depends the rate of erosion, the rate of
includes the following cyclic changes (Imbrie 1985): production of detritus and its type, and the rate of
sediment transport. de Boer and Smith (I994b) sug-
1. Variations in orbital eccentricity (the shape of the
gested that orbital forcing in nonmarine settings
earth's orbit around the sun); several "wobbles",
may be particularly important in the 30-40° latitude
which have periods of 413, 100, and 54 ka.
belt, where the effects on precipitation and eva-
2. Changes of up to 3° in the obliquity of the ecliptic,
poration are thought to be the most pronounced.
with a period of 41 ka.
However, the effects are difficult to demonstrate,
3. Precession of the equinoxes. The earth's orbit
particularly where tectonism and sea-level change
rotates like a spinning top, with a period of23 ka.
may have introduced complications into the strati-
This affects the timing of the perihelion (the posi-
graphic record.
tion of closest approach of the earth to the sun on
H. Olsen (1990, 1994) described a Devonian flu-
an elliptical orbit), which changes with a period of
vial section in East Greenland containing prominent
19 ka.
fining-upward cycles that show systematic varia-
Each of these components is capable of causing tions in thickness. Variations in channel sand body
significant climatic fluctuations, given an adequate thickness are shown in Fig. 12.4. The average thick-
degree of global sensitivity to climate forcing. For ness of the fluvial fining-upward cycles is 20 m, with
example, when obliquity is low (rotation axis nearly a range of 13-30 m. Channel sandstones vary from 2
normal to the ecliptic), more energy is delivered to to 13 m in thickness. Bundles of 3-6 cycles can be
the equator and less to the poles, giving rise to a defined, as can thicker trends that trace out mega-
steeper latitudinal temperature gradient and lower cycles. Megacycles range from 79 to 135 m, with an
seasonality. Variations in precession alter the struc- average of 101 m in thickness. H. Olsen (1990, 1994)
ture of the seasonal cycle by moving the perihelion calculated disharge values for the channels using a
point along the orbit. This changes the earth-sun suite of geomorphic equations based on channel
distance at every season, thus changing the intensity dimensions. The results are shown in Fig. 12.5 and
of insolation at each season. "For a given latitude demonstrate the same cyclic variability. H. Olsen
and season typical departures from modern values (1990,1994) argued that the simplest explanation for
are on the order of ±50/0" (Imbrie 1985). Because the the rhythmic variations is changes in discharge
forcing effects have different periods they go in brought about by orbital forcing of climate changes,
and out of phase. One of the major contributions and he attempted to fit the pattern of thickness and
of Milankovitch was to demonstrate these phase discharge variations to patterns of orbital variation.
relationships on the basis of laborious time-series However, Algeo and Wilkinson (1988) offered a
calculations. These can now, of course, be readily warning:
carried out by computer. The success of modern "Despite an often-claimed correspondence between cycle
stratigraphic work has been to demonstrate the ex- and Milankovitch orbital periods, factors independent of
istence of curves of temperature change and other orbital modulation that affect cycle thickness and sedi-
variables in the Cenozoic record that can be corre- mentation rate may be responsible for such coincidence.
lated directly with the orbital variations. For this For example, nearly all common processes of sediment
transport and dispersal give rise to ordered depositional
purpose, sophisticated time-series spectral analysis lithofacies sequences that span a relatively narrow range of
is performed on various measured parameters, such thicknesses ... Further, long-term sediment accumulation
as oxygen-isotope content or cycle thickness. This rates are generally limited by long-term subsidence rates,
approach has led to the development of a special type which converge to a narrow range of values for very differ-
of quantitative analysis termed cyclostratigraphy ent sedimentary and tectonic environments (Sadler 1981).
In essence, the spectra of real-world cycle thicknesses
(Fischer et al. 1990). Many examples of sedimentary and subsidence rates are relatively limited, and this in
cycles generated by orbitally forced sea-level change turn constrains the range of commonly determined cycle
or climate change have now been published. The periods. For many cyclic sequences, calculation of a
record is particularly clear in successions controlled Milankovitch-range period may be a virtual certainty, re-
by Cenozoic glacio eustasy and in certain shallow- gardless of the actual generic mechanism of cycle forma-
tion."
marine platform carbonates and lacustrine rhythms.
In the case of nonmarine deposits, the effects of Algeo and Wilkinson (1988) studied cyclicity in
orbital forcing are thought to be much more subtle more than 200 stratigraphic units, determining
and difficult to prove. Forcing may be felt through periodicities from cycle thickness, sedimentation
m
a m b Fig. 12.4.Variations in sand body
thickness in a Devonian fluvial sec-
450 tion in East Greenland. CR. Olsen
I
1100
1990)
l
1000
!t
BUNDLE G
400 900
~
J t
F
800
t
E
MEGA-CYCLE A
350 700
j
!
D
600
C
300 500
f
I
B
400
250 300 !
a 5 10 m
rate, depositional environment, and age range. They strated that a cyclic alternation of fine and coarse
found that calculated periodicities are randomly dis- sedimentation persists vertically through about a
tributed relative to all the Milankovitch periods ex- dozen cycles totaling over 100 m of beds in a proxi-
cept the 413-ka eccentricity cycle. They concluded mal foreland basin succession (Fig. 12.6). In proxi-
that "if, in fact, a calculated average period within mal areas the cycles are entirely nonmarine and in
the broad range of Milankovitch periodicities is not a more distal areas entirely marine, with the facies
sufficient test of orbital modulation of sedimentary boundary (shoreline) remaining fixed vertically.
cycles, demonstration of such control becomes sig- These are therefore not transgressive-regressive
nificantly more difficult than hitherto appreciated." cycles. The fixing of the facies boundary is attributed
Demonstration of a hierarchy of periodicities may to the influence of growth anticlines which stabilized
be one approach. For example, where there are two areas of uplift and subsidence, but the absence of
orders of cycles with recurrence ratios of 5:1, a transgression and regression suggests that there
precession (20 ka) and eccentricity (100 ka) combi- were no vertical movements of the basin or of sea
nation may be indicated. Fischer (1986) discussed level during this time period. It therefore seems
examples of this, especially from lacustrine environ- likely that climatic control of rainfall and sediment
ments, but much work remains to be done to conclu- yield is the most likely explanation for the observed
sively demonstrate Milankovitch effects in the cyclicity. Calculations of cycle duration suggested an
nonmarine record. approximate correlation with Milankovitch period-
A different type of argument than cycle thickness icity.
was used by de Boer et al. (1991) in their postulation The Late Paleozoic cyclothems ofNW Europe and
of climatic control of sedimentation. They demon- North America have long been attributed to glacio-
Orbital Forcing 447
m eustatic mechanisms generated by the Gondwanan

450 glaciation (see review in Miall 1990, Chap. 8), but
only recently have studies demonstrated climate
change during the formation of the cyclothems
themselves. In Nova Scotia, calcareous paleosols oc-
cur at cyclothem sequence boundaries, formed at
-=- ... - ...
times oflow stands of sea level, indicating seasonally
400
---- ---- dry climates during glacial episodes in the southern
hemisphere. This contrasts with the humid (intergla-
cial) climates indicated by coal-bearing deposits of
the highstand systems tracts (Tandon and Gibling
1994; see also Sect. 13.3.1). Yang and Nio (1993)
demonstrated a similar oscillation between arid con-
350
MEGA - CYCLE A ditions during lowstand and more humid conditions
during highstand in the eolian-fluvial cycles of the
Rotliegend Sandstone of the southern North Sea Ba-
---- sin.
~ --
"- ----
--- ---
300
.... -:.:.:- 12.12.2 Fluvial Response
to the Late Cenozoic Glaciations
The rapid recent developments in sequence stratig-

raphy, and the consequent focusing of interest on
250 ... - -
---- --- --- the response of depositional systems to base-level
change, has renewed interest in the fluvial deposits
o 100 200 o (m J!s) formed during the late Cenozoic glaciation, includ-
Fig. 12.5. Variations in calculated discharge for sandstone ing coastal-plain successions and inland terraces.
bodies in a Devonian section in East Greenland. Dashed These contain the record of substantial sea-level
line connects individual values for each sand body; it changes. Ho'Vever, they also contain the record of
defines a form of short-term cyclicity. Maximum values major climate changes. In coastal fluvial deposits the
for each cycle are connected by solid line and show a
longer-term cyclicity. (H. Olsen 1990) effects of base-level change and climate change are
sw
i i
o 2km
~ Terreslrialline-grained deposrts
D Marine lossililerous mart.
a Slumps and pebbly mudftows

lowards the SW
Fig. 12.6. Cross section through an Eocene cyclic succession in the Tremp-Graus foreland basin, Spain. Note the vertical
persistence of the marine-nonmarine facies boundary. (de Boer et al. 1991)
lEMPERATURE MOISTURE TYPES OF STORMS UPLAND SOILS ALLUVIAL

low high low high Front. Cony. Trop. Min. Max. STRATIGRAPHY
o --.---=- --.---
w, Sp, S S,F
F
Sp,S, --.-
~ Sp,F
..!><: F
...
(I)
~ JO
~\~
..
.
~ c
>. ~
"
u W, Sp, S ---
I F
I
IS I
~ /:
20th Century
Averages
Sp, S,
F
--- ---
20
Fig. 12.7. Late Pleistocene-Holocene climatic and envi- curve represents initial thick red clay formation during the
ronmental changes on the Edwards Plateau, Texas, correlate Pleistocene and its gradual removal during the Ho-
lated with the alluvial stratigraphy of the upper Colorado locene. LP Late Pleistocene; EH early Holocene; LH late
drainage of the same area. Front, mid-latitude cyclones; Holocene; M modern alluvial deposits. (Blum 1994, re-
Conv, convectional storms; Trop, tropical storms. Seasons printed by permission)
are given by W winter; Sp spring; S summer; F fall. Soil
superimposed, and they are complex in nature. The Figure 12.7 shows the correlation between the
major effects of climate change on river systems are alluvial stratigraphy of the Upper Colorado drain-
to alter the discharge regime and, coupled with age, Texas, and independently derived climatic
changes in vegetation cover, to alter the sediment changes. Figure 12.8 shows a model of fluvial pro-
yield. Because of lags in the response to climatic cesses in relationship to glacially controlled changes
forcing and various geomorphic threshold effects, in climate and vegetation, based on the Dutch work.
these changes do not occur in phase with the base- Both these studies deal with periglacial regions,
level changes, which themselves may be affected by where climate change was pronounced but the areas
isostatic responses to ice-sheet loading and unload- were not directly affected by glaciation. It is interest-
ing. ing to note that a major episode of valley incision
In recent years detailed studies of terrace strati- occurred in Texas not during the time of glacio-
graphies and the deposits of buried channel systems eustatic sea-level lowstand but at the beginning of
have been carried out using borehole data and the postglacial sea-level rise, which commenced at
chronostratigraphic correlations based on radiocar- about 15 ka. The Dutch work explains why this
bon dating. The two major areas of study are the Gulf may have been the case. Vandenberghe (1993) and
Coast of Texas (Blum 1994; Blum and Price 1994) and Vandenberghe et al. (1994) demonstrated that a ma-
the valley of the Rhine-Meuse system in the Nether- jor period of incision occurs during the transition
lands (Vandenberghe 1993; Tornqvist 1993, 1994; from cold to warm phases because runoff increases
Tornqvist et al. 1993). Vandenberghe et al. (1994) while sediment yield remains low. Vegetation is
reported a comparative study of the Maas River in quickly able to stabilize river banks, reducing sedi-
the Netherlands and the Warta River in Poland. ment delivery, while evapotranspiration remains
These studies are beginning to unravel the complex low, so that the runoff is high. Fluvial styles in
and out-of-phase responses of river systems to aggrading valleys tend to change from braided dur-
Milankovitch-band external forcing. ing glacial phases to meandering during inter-
Orbital Forcing 449
SYNTHESIS : THE PERIGLACIAL FLUVIAL CYCLE
-6·
r· Sedimentology
Hydrology
MORPHOLOGICAL
ACTIVITY
slartol
INTER- bank stability ~ RIVER INCISION
( meandenng I
GLACIAL E~runoffhi h
E AEOLIAN ACTIVITY
N low runoH ~ (strongly reduced
D fluvial actIvity )
high
GLACIAL peak - discharges RIVEA
~ AGGRADATION
(braiding)
M high
A
sediment - supply
X
( aeolian. slope deposits)
.,
PERIOD
bank stability c:::::;:> AIVER INCISION

G
, E1a;- runoff high
N.
(end 01)
INTER-
GLACIAL
Fig. 12.8. A model showing the relationship between temperature, vegetation density, evapotranspiration (E), precipi-
tation (P), and sedimentary processes in river systems during glacial and interglacial phases. (Vandenberghe 1993)
glacials (Vandenberghe et al. 1994). With increasing on the flanks of the coastal mountains of Oman and
warmth, and consequently increasing vegetation related these to climatically driven changes in runoff
density, rivers of anastomosed or meandering style and sediment load. They found no evidence of tec-
tend to develop, the former particularly in coastal tonic control of this stratigraphy.
areas where the rate of generation of new accommo- It is apparent that fluvial processes inland and
dation space is high during the period of rapidly those along the coast are likely to be completely out
rising base level (Tornqvist 1993; Tornqvist et al. of phase during the climatic and base-level changes
1993; Fig. 13.16). Vandenberghe (1993) also demon- accompanying glaciation. Within a few tens of kilo-
strated that valley incision tends to occur during the meters of the sea, valley incision occurs at times of
transition from warm to cold phases. Reduced eva- base-level lowstand, during cold phases, but the sur-
potranspiration consequent upon the cooling tem- face may be modified and deepened during the sub-
peratures occurs while the vegetation cover is still sequent transgression until it finally becomes
substantial. Therefore, runoff increases, while sedi- buried. Inland, major erosional bounding surfaces
ment yield remains low. With the reduction in veg- correlate to times of climatic transition, from cold to
etation cover as the cold phase becomes established, warm and from warm to cold, that is to say, during
sediment deliveries increase and fluvial aggradation times of rising and falling sea level, respectively.
is reestablished. Minor climatic fluctuations lead to alternations of
Several other studies of the effects of early Ho- aggradation (terrace development) and valley inci-
locene climate change have also revealed important sion, as has now been documented for several Gulf
sedimentary effects (Bull 1991). As noted by G.A. Coast rivers by Blum (1994) and Blum and Price
Smith (1994), "decrease in effective precipitation (1994), as summarized in Fig. 13.ll.
initiated time-transgressive changes in plant com- Although providing a fascinating explanation for
munities and diminished hill-slope vegetation den- fluvial processes during the Late Cenozoic, this body
sities that increased both sediment yield and runoff. of ideas must be used with caution for the interpreta-
The impact of this climate change was so profound tion of more ancient fluvial successions, because of
that it controlled alluvial-fan sedimentation pro- changes in vegetation type. Whether the same model
cesses regardless of regional variations in tectonic can be applied to the interpretation of Late Paleozoic
activity." Abrahams and Chadwick (1994) docu- cyclothems, which were deposited before the advent
mented multiple levels of alluvial fan development of upland vegetation, remains to be examined.
12.12.3 Conclusions Climates may also be different on either side of a

major mountain range. The major criteria are sum-
The question remains whether high-frequency cli- marized in Tables 12.2 and 12.3, and Figs. 12.1-12.3
matic forcing, on the lOS-year time scale, will be are schematic diagrams oflithofacies assemblages in
detectable in the pre-Pleistocene stratigraphic re- which climatically sensitive criteria are identified.
cord. To detect change on such a frequency requires As noted earlier in this chapter, the evolution of
stratigraphic mapping on at least as refined a time land vegetation has probably had a profound effect
scale. This scale of periodicity corresponds to the on fluvial styles. The lack of roots to bind bank
cyclicity of many major geomorphic processes, such sediments in pre-Devonian time would have led to a
as the return period of major channels in large-scale tendency for rapid bank erosion and channel widen-
fluvial systems (sediment groups 8 and 9 of Table ing, with the supply of abundant bedload to the
9.1), and is at or below the resolving power of most system. This would have favored braided over other
chronostratigraphic correlation methods, including fluvial styles, regardless of the discharge regime.
magnetostratigraphy. Long (1978) considered Schumm's (1968a) specula-
tions on this matter, in his review of Proterozoic
fluvial sedimentation. He confirmed that rivers with
an interpreted meandering style are indeed rare in
12.13 Discussion the Precambrian sedimentary record, but much
work has been carried out on Precambrian sedimen-
This review has demonstrated that in very few fluvial tation since the late 1970s, and this subject should be
settings are lithofacies criteria sufficiently distinc- reviewed again. In the meantime, the climatic ambi-
tive that they can be relied upon, in the absence of guity of fluvial lithofacies, and the absence of coal
supplementary evidence, to indicate climatic setting. and of much other relevant fossil material means
Probably the most climatically distinctive lithofacies pre-Devonian climates are very difficult to interpret
assemblage is that of the sandy terminal fan (ephem- from the fluvial record.
eral braid delta of Blair and McPherson 1992), which It may be possible to make quite subtle climatic
is characteristic of arid basin settings. The down- interpretations based on a summation of all the
stream change from channelized to sheet flow and available evidence. For example, Hallam (1984) de-
the decrease in sand body thickness and the scale of fined an "intermediate" type of climate, which shows
hydrodynamic sedimentary structures are particu- evidence of both aridity and humidity. The sedi-
larly characteristic. The architecture of the distal ments contain "thin seams oflignite or coal, modest
deposits, consisting of sandstone sheets composed in thickness and localised in distribution, and minor
of flood cycles and interbedded with fine-grained evaporites, such as scattered thin gypsum lenses and
deposits showing evidence of evaporation, is very seams, [these] could both form in intermediate,
distinctive (Fig. 12.2B). Most other lithofacies and savannah-type environments where wet and dry sea-
architectural features of fluvial deposits have been sons are of roughly equal importance." The coexist-
shown to be nondiagnostic of climate. Mass-flow ence of trees and evaporites indicates a savannah or
deposits and other upper flow-regime fluvial sedi- Mediterranean-type climate with alternating wet
ments, while very characteristic of alluvial fans in and dry seasons.
hot-arid climates (Fig. 12.1A), are not confined to In a general discussion of the distribution of
this climatic setting (Table 12.1), although it is pos- various climatic indicators, especially coal, Ziegler et
sible that their abundance, and therefore their pres- al. (1987) provided this interpretation of the Nubian
ervation potential, may be higher in this type of Sandstone of North Africa, based on many sources:
setting. A quantitative testing of this possibility has Cross-bedding type and paleocurrent variance are
yet to be attempted. interpreted in terms of braided or low-sinuosity
However, a consideration of every aspect of a streams. Flood activity is indicated by repetition of
fluvial assemblage, including features of the over- channel-based scour surfaces and their lag deposits.
bank environment such as chemical and biochemi- Prevalence of soil horizons and common preserva-
cal sediments, soils, minor sedimentary structures, tion of flora suggest well-vegetated alluvial plains,
and bed color, may yield useful clues as to climate. but the common tree Dadoxylon was probably
Conflicting criteria may indicate that the climate of restricted to channel margins and swamps. Floral
the basin center is different from that of its margin, diversity is not high, and the ubiquitous braided
or that at least the water table is at a different level. stream deposits would seem to rule out the possibil-
Discussion 451
ity of forest vegetation and especially the lowland An example of a climatic interpretation of a pre-
rain forest. "We propose a savannah environment Devonian fluvial system is that by Trewin (1993b).
with its typical seasonal rainfall pattern for the He described sand sheets of mixed eolian-fluvial
Nubian Sandstone and its equivalents in NE Africa origin. Discharge is interpreted as flashy on the basis
and Arabia." of a dominance of trough cross-bedded sandstone,
Dubiel et al. (1991) were able to trace a lateral the great lateral extent of thin units, the extreme
gradation from seasonally dry floodplains to per- rarity of channel features, strongly unimodal cur-
manently waterlogged swamps and ponds. In the rent directions, and planar upper bounding sur-
Monitor Butte Member of the Chinle Formation faces. These deposits contain abundant trace fossils.
purple-mottled sandstones contain large, irregular Red grain coatings indicate oxidizing conditions.
mottles of dark purple, lavender, yellow, and white Evidence of desiccation is very rare, but in the ab-
that reflect varying concentrations of iron-bearing sence of significant land vegetation an arid climate is
minerals. These mottles reflect variations in the wa- not necessarily implied by this evidence. A high
ter table, giving rise to alternating oxidizing and water table is suggested by the trace fossils, which
reducing conditions, with redistribution of iron and indicate invertebrate activity. Trewin (1993b) stated:
the formation of gleyed paleosols. The purple- "Before the advent of land plants sandy alluvial
mottled units can be traced laterally into gray and outwash areas were associated with a wider range of
purple siltstones with plant debris, and then into climate than is seen at the present day. Flashy dis-
gray to black, thinly laminated organic-rich shales charge, and the lack of sediment binding resulted in
with fish debris deposited in permanent ponds, lateral expansion of streams in times of flood to
marshes, and lakes. A seasonal climate is suggested, produce laterally extensive thin sandstone domi-
with the ponds occupying permanently saturated nated by trough cross-bedding in medium- to
areas in topographically low parts of the basin. coarse-grained sands." Eolian reworking took place
Jerzykiewicz and Sweet (1988) distinguished between floods or during periods of low discharge.
three climatic regimes in the Upper Cretaceous- In conclusion, it is clear that climatic interpreta-
Paleocene sediments of the Foothills of Alberta. A tions can be made from clastic fluvial lithofacies, but
humid fluvial facies association consists of organic- this is best carried out in a regional context, with due
rich to coal-bearing floodplain deposits associated regard for the tectonic setting and incorporating
with meandering channels of high sinuosity. A semi- data from all interbedded nonmarine facies, espe-
arid fluvial facies association includes varicolored cially chemical and biochemical sediments. Many
floodplain mudstones with mature caliche paleosol features of clastic fluvial lithofacies, and the archi-
horizons. The alluvial plain was drained by broad tectural elements and successions of which they are a
mobile channels. Periodic heavy rainfall modified part, are indicative of a discharge regime, the rela-
the floodplain into a braidplain with numerous tionship of which to climate may be indirect or com-
ephemeral streams eroding the caliche soil. An inter- plex. In many cases climatic and tectonic signatures
mediate fluvial facies association can also be distin- may be ambiguously similar. Nonetheless, a careful
guished, consisting of gray or greenish mudstone, attention to the question of local climate may yield
barren of coal and devoid of caliche, associated with valuable information and insights to assist in re-
a low-sinuosity meander-belt facies. They demon- gional basin studies and paleogeographic recon-
strated the stratigraphic and geographic shifting of struction.
these climatic belts across Alberta through the Late
Cretaceous and early Tertiary.
Chapter 13
Sequence Stratigraphy
13.1 Introduction which various allogenic controls actually work has

stimulated much numerical modeling of basin evo-
lution, aided in recent years by graphical simulation
The currently accepted definition of "sequence" is using computers.
that proposed by Vail et al. (1977, p. 53). It is a Most recent work has focused on the sequence
"stratigraphic unit composed of a relatively con- architecture of shallow marine deposits, because one
formable succession of genetically related strata ... of the major preoccupations of sequence stratigra-
bounded at its top and base by unconformities or phers has been sea-level change, and facies charac-
their correlative conformities". However, the term teristics in the lower coastal-plain to shallow-shelf
sequence has had a checkered history. Sloss (1963) environment are particularly sensitive to changes in
defined stratigraphic sequences as "rock-strati- sea level. Much less is known about the sequence
graphic units of higher rank than group, megagroup, stratigraphy of nonmarine deposits, although this
or supergroup, traceable over major areas of a conti- is now rapidly changing, as summarized in the re-
nent and bounded by unconformities of interre- cent reviews by Shanley and McCabe (1994) and
gional scope." This useage has not received universal Posamentier and Allen (in press).
acceptance. In some of the literature "sequence" has Sequence-stratigraphic analyses of fluvial depos-
been used as a synonym for "succession", and it has its commenced, in effect, with the work of Allen
been employed for cyclic or unconformity-bounded (1974b). As noted in Sects. 2.3.7 and 10.4.3 (Fig.
units of varying dimensions and for time spans rang- 10.32), Allen developed a series of models of alluvial
ing from thousands of years to hundreds of millions stratigraphy based on simple deductive reasoning.
of years. His model 5B is an interpretation of the stratigraphic
Since the ground-breaking publication by Vail et architecture that might develop in an alluvial system
al. (1977), sequence stratigraphy has evolved into characterized by a cycle of dissection and aggrada-
one of the central, theoretical underpinnings of the tion, controlled by base-level changes over a vertical
science of stratigraphy. It comprises a set of pre- interval greater than channel depth. This is precisely
dictions that can be made about the facies and archi- the set of conditions under which nonmarine strati-
tecture of sediments formed in a given range of graphic sequences develop, according to modern
environments under specified allogenic controls. It work. Building on this work and on that of Bridge
is the predictability of sequences that give them their and Leeder (1979), Ross (1990) made some simple
value for regional correlation and for one of the main predictions regarding variations in channel density
practical applications of such correlation - petro- in response to variations in aggradation rate, but he
leum exploration. However, modern work has dem- did not formalize a sequence model. The current
onstrated that sequences are generated by a variety state of knowledge of the effects of base-level change
of causes, including local- to continental-scale tec- on fluvial systems and of the control tectonic move-
tonism, eustatic sea-level change, and climate ments exert on rivers is discussed at length in Chap.
change (see reviews by Miall 1990, Chap. 8; Plint et al. 11 (see in particular Sect. 11.2.2). It is beyond the
1992). More than one generating mechanism may be scope of this book to enter into the current contro-
active at anyone time, and it is not uncommon for versy regarding the relative importance of regional
sequences of several types and scales to be superim- tectonism versus eustatic sea-level change in the
posed on each other within the same succession. generation of base-level change.
Interpreting generating processes from the rock The first modern sequence models, those of Vail
record requires the construction of a detailed strati- et al. (1977), were based essentially on the architec-
graphic framework. Questions about the way in ture of divergent or passive continental margins,
454 Sequence Stratigraphy
although the limitations imposed on the models by table facies successions and architectures? To turn
this fact were not clearly acknowledged by Vail and this question around, what can be deduced regard-
his colleagues until recently (Posamentier and ing allogenic controls from the detailed documenta-
Weimer 1993; Posamentier and Allen 1993). In the tion of such facies and architectures? The research
meantime, working on the Mesa Verde Group of the that has taken place since the late 1970s on the effects
Western Interior foreland basin of Utah, Swift et al. of base-level change on fluvial systems is discussed
(1987) proposed a quite different type of sequence at length in Sect. 11.2.2. The most important such
model. In foreland basins the relationship between research, because of the attention it received, was
subsidence and sediment supply is reversed from that by Posamentier et al. (1988), which was based on
that prevailing in extensional continental-margin extensive computer modeling and observations on
basins. In the latter, sediment enters the basin near small modern rivers. As argued by Miall (1991c), this
the tectonic hinge and is transported toward the research was seriously flawed. In Sect. 11.2.2 the
more rapidly subsiding side of the basin. Sequences work of later researchers, particularly that of
in foreland basins are "clastic wedges", as defined by Schumm (1993) and Wescott (1993), is reviewed, and
Sloss (1962). "They are sediment prisms poured into it is demonstrated how rivers respond to base-level
the deeper side of a linear half-basin undergoing change by incision or aggradation and by changes in
subsidence along its landward margin. Hence it is sinuosity. The most recent work by Posamentier and
the proximal (landward) end of the wedge that is Allen (in press) is touched on in this section. The
most completely preserved" (Swift et al. 1987, p. effects of tectonism on the middle and upper reaches
449). Accommodation space is generated more of river courses are also discussed at length in Chap.
slowly in the deeper part of the basin because of 11. The purpose of the present chapter is to review
slower subsidence there, and so stratigraphic units current attempts to put these data and ideas into a
tend to thin and pinch out toward the center of the sequence framework, that is to say, a predictive
foreland basin, whereas many units thicken toward model.
the basin center in the case of extensional continen- Modern sequence models for fluvial deposits
tal margins. The major architectural differences be- have been proposed by Shanley and McCabe (1991,
tween the two types of basin setting are illustrated in 1993, 1994), Wright and Marriott (1993), and
Fig. l3.1. Posamentier and Allen (in press). Very few modern
Nonmarine deposits are much better preserved in sequence studies of fluvial deposits have been pub-
foreland basins than in extensional basins because of lished. Those by Legarreta and Gulisano (1989),
the relationship between subsidence and sediment Shanley and McCabe (1991, 1993), Van Wagoner et
supply, and many of the recent studies of nonmarine al. (1990, 1991), Hanneman and Wideman (1991),
sequences have been carried out in deposits formed Miall (1993), Gibling and Bird (1994), Gibling
in this tectonic setting (Aubry 1989; Hanneman and Wightman (1994), Olsen et al. (1995), and
and Wideman 1991; Holbrook and Dunbar 1992; Yoshida et al. (in press) are the most important.
Shanley and McCabe 1991, 1993; Martinesen et al. This chapter builds mainly on the work of these
1993; Miall1993; Aitken and Flint 1995; Olsen et aI., researchers.
in press; Yoshida et aI., in press). These studies effec- It is important to note that, unless otherwise
tively answered the question that had been raised stated, no particular time connotation is placed
by Posamentier and Weimer (1993) regarding the upon the sequence models discussed in this chapter.
applicability of sequence concepts to nonmarine de- The examples of alluvial stratigraphy described in
posits. Posamentier and Allen (1993) provided a use- Sects. 9.2 to 9.4 (summarized in Table 9.1) include
ful discussion of the particular characteristics of sequences of third to fifth order (groups 8 to 10, in
sequences formed in foreland basins and made sev- the classification of this book: Tables 3.2, 4.2, 9.1).
eral predictions regarding the architecture of non- These range in duration from a few thousand years
marine deposits in such basins that are discussed to a few million years. These sequences all display a
later in this chapter. certain degree of "self-similarity", indicating that
The purpose of the present chapter is to review the dynamic balance between the various allogenic
recent work on sequence models. Does the response controls can be achieved and maintained over a wide
of fluvial systems to allogenic controls, including range of time scales. The various temporal scales of
source-area uplift (tectonic control) and base-level nonmarine sequences, and their causes, are re-
change (tectonic or eustatic control), lead to predic- viewed in Sect. l3.4.
Introduction 455
A. PASSIVE MARGI N MODEL
CONOENSEO
SECTIO
SUBAERiAl
EAOSION
CONOENSEO
SECTION
lOW I..ANOWAFLO SEAWAAD
se,t,I..EVEl COA.$'TAL QNl A.P
B. FORELAND BASIN MODEL

Q) REGRESSIVE DEPOSITS
-
_ RAVINE"ENT
CONOENSEO
SECTION
CONDENSED
HORIZON
Su8MAA1NE
EROSION
---- --
CONFOA""6LE
-
--
8OUNOARY
,-ow fotlGH lAND BASIN OFFSHORE ""RINE

EROSION SURFACE
SUBAERIAl EROSION
SU~
SENSe OF SttOREllNE
MOVaIENT
Fig. l3.IA,B. Comparison of sequence architecture on ex- sional continental margins, because of the slow subsidence
tensional continental margins (A) and in foreland basins rates on the proximal margins of such basins and the
(B). Note the much thicker record of nonmarine deposits tendency for nonmarine HST deposits to be eroded during
in foreland basins. These deposits also have a higher pres- subsequent base-level falls. (Swift et al. 1987)
ervation potential than the highstand deposits of exten-
13.2 Accommodation Space and McCabe (1994) noted some important addi-
tionallocal "base levels" that may be significant in
The concept of accommodation space is fundamen- nonmarine settings, such as the deflation surface,
tal in sequence stratigraphy. Jervey (1988, p. 47) defined by the water table, in eolian environments.
defined it as "the space made available for potential So how does the concept of accommodation apply in
sediment accumulation [where] in order for sedi- the case of nonmarine deposits?
ments to be preserved, there must be space available Base level is the level to which rivers grade their
below base level (the level above which erosion will course, as discussed in Sect. 11.2.2. Rivers adjust
occur)." The word "accommodation" implies space; their slope, sinuosity, and other hydraulic character-
therefore, the use of the word "space" in the term istics to the power and efficiency necessary to trans-
"accommodation space" is, strictly speaking, super- port the water and sediment supplied from the
fluous. drainage basin. Given equilibrium conditions, a
Jerveywas modeling marine sequences, for which graded profIle will develop, which flattens out to the
base level corresponds to sea level; the sequence horizontal at the precise point at which base level is
architecture of marine deposits is governed mainly reached, at the mouth of the river. Posamentier et al.
by the response of depositional systems to sea-level (1988, p. 135) stated:
change. However, this defmition does not apply in "The fundamental principle which governs sedimentation
the case of the fluvial environments because fluvial in fluvial environments (assuming constant sediment sup-
deposits, by their very nature, are deposited above ply) is that a stream will aggrade if its equilibrium profile
shifts basinward or upward and incise if its equilibrium
sea level. Schumm (1993, p. 279) defined base level as profile shifts downward ... subaerial accommodation
"the imaginary horizontal or level surface to which occurs by basinward shifts of the stream equilibrium pro-
sub-aerial erosion proceeds. It is sea level." Shanley file."
EROSIOH:J...
DEPOSIT~~_J____ j !
!
I I
EROSION.., : FLOODING
PROGRADATION :1
r.Lj : FLOODING -. r -,
I I r"-'
.
I I •
I I
I I
I
I
I
PROGRADATION :
r
I
Fig. 13.2. The creation and reduction of accommodation coastal-plain settings accommodation space is directly re-
space in response to sea-level change and tectonism. The lated to the balance between subsidence and sea-level rise,
composite sea-level curve (bottom) indicates the integra- as shown in this diagram. Where tectonism is the domi-
tion of three rates of sea-level change. Accommodation nant control, the composite sea-level curve could be re-
space is generated when the rate of sea-level rise outpaces garded as a curve tracking the rise and fall of the source
the rate of subsidence (dotted areas in relative sea-level terrane as a result of uplift and erosion. Accommodation
curve). This diagram was adapted by Nummedal et al. space then represents the space between the existing flu-
(1993) from Van Wagoner et al. (1990) as a basis for vial profile and the new, higher profIle required to restore
discussing fourth-order marine cycles, but it can be ap- equilibrium (the same dotted area of the curve)
plied to the development of fluvial cycles, as follows: In
Accommodation Space 457
The problem with the last sentence in this quotation from a few thousand years to hundreds of millions of
is discussed in Sect. 11.2.2. Apart from that, the years. Several of these processes of change may be
quote succinctly explains the importance of the occurring simultaneously at more than one time
equilibrium profile. Accommodation space in fluvial scale. Few rivers are ever, therefore, in a state of true
environments can be defined as the space between equilibrium. The concept of the graded profile is a
an existing graded profile and a new graded profile geomorphic one, which is appropriate for short time
to which the river adjusts in reponse to changes in scales (tens to hundreds of years), but over geologic
base level, tectonic movements, or hydraulic condi- time scales the concept of dynamic equilibrium is
tions (discharge, sediment supply). Accommoda- more appropriate and useful. Local conditions must
tion space is positive, and sedimentation occurs, always be taken into account. For example, along
under the following conditions (Fig. 13.2): much of the present-day coastal plain of South
Island, New Zealand, a slow eustatic rise and
1. At the coast, when the graded profile moves up-
shelf subsidence are underway. This is causing a
ward in response to a rise in sea level. Most sedi-
transgression, but one that is being recorded not by
mentation occurs near the mouth of the river, and
deposition, but by valley incision and erosional
there may be little change in accommodation
shoreface retreat, because of high wave energies
space in upper reaches.
(Leckie 1994).
2. At the coast, when the graded profile moves up-
For marine shoreline and nearshore deposits pre-
ward in response to coastal regression and
dictions of sequence architecture in response to sea-
progradation, under conditions of sea-level
level change are easy to make. When sea level rises
stillstand. This effect is small, and in the case of
the coast is flooded, and transgression takes place.
progradation by lobate deltas it is limited by the
Sea-level fall is accompanied by regression and
tendency for avulsive delta abandonment when
progradation. Such predictions were the basis for the
the profile of the main channel is raised by aggra-
first sequence models in the modern sense, those of
dation.
Frazier (1974), followed by Vail et al. (1977). How-
3. Inland, when the graded profile moves upward in
ever, in the case of rivers, the response to change is
response to uplift of the source area, for example
more complex. For example, as discussed in Sect.
along a basin-margin fault. New accommodation
11.2.2, a lowering of sea level and the extension of a
space is generated adjacent to the zone of uplift
river profile out to a new mouth, lower out and
and, given adequate sediment supply, prograda-
further down on the continental shelf, may lead to
tion may shift the equilibrium profile basinward.
incision, progradation, or simple changes in sinuos-
4. Throughout the profile when, because of tectonic
ity with neither enhanced aggradation or erosion,
or climatic changes in the source area, there is an
depending on a variety of factors, of which the slope
increase in discharge and/or sediment supply.
of the newly exposed shelf is one of the most critical.
Accommodation space is negative, and incision oc- The development of sequence models for fluvial sys-
curs, under the reverse of these conditions: lowering tems has therefore lagged behind that for marine
of sea level, reduction in source-area elevation, e.g., environments, and the subject could be said to have
by regional erosion, and reduction in discharge or arrived at a state of maturity only recently, with the
sediment supply. publication of the review by Shanley and McCabe
All these allogenic controls fluctuate constantly (1994).
in all sedimentary basins. Sea level has been ris- In the case of tectonic control, the effects of uplift,
ing and falling eustatically at least throughout the subsidence, and erosion at a faulted basin margin
Phanerozoic (see review by Miall 1990, Chap. 8). have long been known [see the discussion of the
Tectonic subsidence and uplift, driven by crustal important paper by Heward (1978a) in Sect. 11.2.1.].
stresses and subcrustal thermal changes, is in a con- Only in recent years, however, have we begun to
tinual state of dynamic balance with uplift driven by appreciate the complexities of tectonism as a sedi-
erosional down cutting of uplifted areas and conse- mentary control. For example, in foreland basins it
quent isostatic buoyancy. Climatic changes, driven now seems likely that regional crustal shortening
by orbital forcing and by the drift of continental and loading on the one hand, and the movement of
plates through latitudinal climatic zones, bring individual thrust or nappe plates, on the other hand,
about changes in discharge and sediment yield. All may generate pulsed, irregular subsidence on both
these changes have been occurring throughout geo- long and short time scales at the same time (Sects.
logic time over a wide range of time scales, ranging 11.3.6, 11.4.6). Uplift and subsidence may also be
Relative
role
of major
controls
I
I
I
I
I
I
Coastal plain ~
Fig. 13.3. The relative importance of the major controls montane basins (left) are, of course, unaffected by sea-
on fluvial sequence architecture, showing how the balance level change. (Shanley and McCabe 1994, reprinted by
of control varies between source area and shoreline. Inter- permission)
thermal in ongm, related to volcanic activity or processes in relationship to position within a fluvial
plutonism (Dickinson 1981b; Dorsey and Burns system.
1994). Major changes in sediment supply (Sect. It is ironic that interpretations of some fluvial
11.3.4) and regional paleoslopes are likely to be re- cycles may be coming full circle. As summarized in
flected in the resulting stratigraphy. Patterns of ba- Sect. 2.3.6.2, in the 1940s some fluvial fining-upward
sin subsidence and the sedimentary response are cycles were interpreted as the result of pulsed sub-
complex in these various tectonic settings and have sidence of a sedimentary basin. For example, Bersier
yet to be fully resolved. Studies of the sequence (1948) expressed this interpretation in his study of
stratigraphy of nonmarine deposits can play an im- the Cenozoic molasse cycles of the Alpine foreland
portant part in this research. basin. With the advent offacies models in the 1960s,
In general the importance of base-level change the emphasis shifted, and cyclic sedimentary pro-
diminishes upstream, as discussed by Miall (1991c) cesses were attributed almost entirely to autogenic
and Shanley and McCabe (1994). In large rivers, such controls (Chap. 2). The importance of tectonism was
as the Mississippi, the evidence from the Quaternary downplayed in the enthusiasm for such processes as
record indicates that sea-level changes affect aggra- delta-lobe switching and fluvial meander migration.
dation and degradation as far upstream as the region We are now returning to a recognition of the impor-
of Natchez, Mississippi, about 220 km upstream tance of allogenic controls and, as pointed out by
from the present mouth (Saucier, in Autin et al. Shanley and McCabe (1994), there is a danger that
1991). Farther upstream than this, source-area ef- the importance of autogenic processes may be for-
fects, including changes in discharge and sediment gotten in the new enthusiasm for allogenic mecha-
supply, resulting from tectonism and climate change nisms, particularly eustasy. This researcher shares
are much more important. In the Colorado River of the uneasiness expressed by Shanley and McCabe
Texas base-level influence extends about 90 km up- (1994) regarding the interpretation of all observed
stream, beyond which point the river has been af- cyclicity in terms of sequence concepts based on
fected primarily by the climate changes of the Late allogenic controls, such as in the study of some Gulf
Cenozoic glaciations (Blum 1994; see Chap. 12). Coast deposits by Mitchum and Van Wagoner
Based on his detailed work on the Gulf Coast rivers, (1991). Nevertheless, the appearance of the new se-
Blum (1994, p. 275) stated: "At some point upstream quence concepts, and a more sophisticated under-
rivers become completely independent of higher- standing of tectonic subsidence processes, have
order relative changes in base level, and are respond- raised anew the possibility that many cyclic succes-
ing to a tectonically controlled long-term average sions currently attributed to autogenic processes
base level of erosion." Figure 13.3 illustrates a con- may in fact be allogenic in origin. This may be the
ceptual estimate of the relative importance of these case particularly in foreland basins, where the evi-
Main Components of the Fluvial Sequence Model 459
dence for short-term, irregular or pulsed subsidence The main photographic illustrations are drawn from
driven by thrust-sheet loading, and for rapid uplift the work of the writer and his students Shuji Yoshida
and erosional unroofing, is now abundant (see Sect. and Andrew Willis on the Castlegate Sandstone of
11.4.6). Utah (Fig. 9.18). Standard abbreviations used in
Weare beginning to realize that more than one many sequence diagrams are as follows: SB, se-
process may control sequence development. These quence boundary; LST or L WST, lowstand systems
processes can act over different time scales and may tract; TST, transgressive systems tract; MFS, maxi-
vary in importance from source area to shoreline. mum flooding surface; HST, highstand systems
The result can be a complex stratigraphy. This is tract; FSST, falling-stage systems tract [=forced-re-
exemplified by the Castlegate Sandstone of Utah, gressive-wedge systems tract of Hunt and Tucker
which is one of the units discussed in the following (1992)]. A composite sequence model showing the
sections. relationships between the sequence terminology,
allostratigraphic terms, and the bounding-surface
hierarchy of this book is illustrated in Fig. 13.6.
13.3 Main Components
of the Fluvial Sequence Model
13.3.1 Sequence Boundary
This discussion uses the standard sequence ter- In the original work of Vail et al. (1977), sequence
minology, as set out by Vail (1987) and Van Wagoner boundaries were defined as the regional uncon-
et al. (1987) and as emended by Hunt and Tucker formities that separate the stratigraphic record into
(1992). Their definition of sequences, including the successions of conformable, genetically related
placement of sequence boundaries at subaerial ero- strata. It was recognized that in some areas, such as
sion surfaces formed during base-level lowstand, is zones of especially rapid subsidence, or in the deep
preferred to that of Galloway (1989a), who recom- sea, below the level to which sea level could fall
mended use of the maximum flooding surface as the eustatically, an unconformity may pass laterally into
sequence boundary. However, this latter work con- a "correlative conformity." In later work (Vail and
tains many important ideas and concepts that are Todd 1981; Vail et al. 1984) the differences between
discussed later in this chapter, and in at least one unconformable and conformable sequence bound-
recent study (Gibling and Bird 1994) it was found aries were clarified, and two types were defined.
easier to follow Galloway'S approach to the defi- A type 1 unconformity develops where sea-level
nition of sequences, because of the difficulty of fall is rapid, more rapid than tectonic subsidence.
correlating sequence boundaries associated with The coastline may move out to near the shelf edge,
discontinuous paleovalley fills [a difficulty antici- and extensive subaerial erosion takes place, with the
pated by Galloway (1989a)]. Reference is made here development of incised fluvial valleys on the shelf
to the two main sequence models for fluvial deposits and the deepening of submarine canyons on the
(Figs. 13.4, 13.5) published by Shanley and McCabe continental slope. Clastic detritus is transported
(1991, 1993, 1994) and Wright and Marriott (1993). down these fluvial and canyon systems to the base of
HST ~ ,;:.:.:: ..:::.. ~ ~

-="" TIL
~~-,-rrr- ~ "TTT'T' ~ ~ ~
m
TST
LST /
gullying
:,"
.......
I
--rrr-mT ""7I'f'I'?'"" ~ c:::::J Fig. 13.4. The fluvial sequence model
Increasing soil hydromorphic channel floodplain
maturity soil deposit deposit of Wright and Marriott (1993).
Fluvial strata Base level Shoreface strata

Isolated, high sinuosity fluvial channels High Low
D
~~~~~~~;= :::=::~::~:~:~~:~::~~'::::::~:~~~~:
~","","'","""''''''''''''''''''''''''''''''''''''''''''''·A'''''''''.'''. .. b ,t if' .
.. ~
.................................... A ............ "' ..... "' ... A ....................... A ... "' ... ,. ..... A ................... .
~
. ..
c
B
Amalgamated fluvial channel deposits
VaHey incision and terrace fonnation

A
~ L I_ _ _ _ _. . , . , - . , -_ _ _ _ _
J- IO's km
Fig, 13.5, The fluvial sequence model of Shanley and tract, with development of incised valley and fluvial ter-
McCabe (1991, 1993, 1994, reprinted by permission), races; B lowstand systems tract; C tidal influence indicates
showing the relationship between shoreface and fluvial the beginning of the transgressive systems tract; D
architecture and base-level change. A Falling stage systems highstand systems tract
1;:;:;::::\ fluvial channel-fills

1'::;;:;:::':'1 tidal channels and bars marine shale
-
=
~ deltaic distributaries and mouth bars marine limestone
~
~
nonmarine floodplain deposits coal
Fig, 13.6. Composite sequence model amalgamating the tions explained in text), the bounding-surface hierarchy of
concepts of Wright and Marriott (1993), Shanley and this book (numerals within circles), and allostratigraphic
McCabe (1994), and Gibling and Bird (1994), showing the terms
relationship between sequence terminology (abbrevia-
the continental slope, forming extensive falling- A type 2 unconformity develops when relative sea
stage and lowstand systems tracts. At type 1 level falls slowly, resulting in a gradual seaward shift
unconformities facies belts undergo a substantial of facies tracts but only minor subaerial exposure
basinward shift. Highstand deposits below the and erosion. According to Vail et al. (1987, 1991), a
unconformity may be deeply eroded. shelf-margin systems tract is formed under these
conditions. Type 2 unconformities are much more ance between the rate of subsidence and the rate of
difficult to indentify in seismic and outcrop records sea-level change. As the sea rises and falls and the
because they are not characterized by deep erosion rate of sea-level change fluctuates against a back-
or major facies shifts. ground of steady tectonic subsidence, this point
On extensional continental margins, where the moves in and out across the basin margin, as shown
nonmarine environment occupies the proximal part in Fig. 13.7. Landward of the equilbrium point, an
of the basin, close to the hinge line, subsidence rates area termed zone A, is a region in which subsidence
are relatively slow, and type 1 unconformities are is always faster than the rate of sea-level change and
common. The Quaternary record of modern conti- sedimentation is continuous. The sequence bound-
nental margins has been characterized by rapid ary here should be of type 2. Seaward of the equilib-
glacioeustatic fluctuations, and the record of fluvial rium point sea-level change is at times the faster
deposition is condensed, fragmentary, and cut by process, and sequence boundaries will be type l.
numerous erosional unconformities (e.g., Atlantic Posamentier and Allen (1993) developed a simple
Coastal Plain: Pazzaglia 1993), except where the hin- stratigraphic model that incorporates these ideas
terland provides an unusually large sediment sup- (Fig. 13.8).
ply, as along the Gulf Coast (Galloway 1989b; Sect. This analysis invokes only two processes, regional
11.3.4). foreland-basin subsidence and eustasy. However,
In foreland basins the pattern of subsidence and much detailed stratigraphic and modeling work on
sedimentation is quite different, and Posamentier foreland basins is demonstrating that tectonic and
and Allen (1993) suggested that the margins of such sedimentary processes there may be much more
basins could be subdivided into two zones, reflecting complex (e.g., Jordan and Flemings 1991; Plint et al.
the relative rates of subsidence and base-level 1993). Rates of tectonic uplift and subsidence in
change (Fig. 13.7). Drawing on the work of Jervey convergent tectonic settings can be very high, as
(1988), they defined the position of an equilibrium much as 10 m/ka over intervals as long as a few
point, the point on the basin margin marking a bal- hundred thousand years (Burbank and Raynolds
Fig. 13.7. Schematic subsidence profile across a

foreland basin, showing the rate of subsidence
along a dip-oriented profile. The position of the
equilibrium point is shown at time T7, with other
positions given by the numbered dots below the
profile. The equilibrium profile marks the bound-
ary between zones A and B, as discussed in the text.
(Posamentier and Allen 1993)
Zone A --I-zone B
[Con.<lnuous Deposition] I.,."... .... &ooI~J Fig. 13.8. Theoretical correlation between
1 stratigraphic sections located in zones A and
~stratlgra~hIC Markers B of a foreland basin. In zone B the sequence
boundary is expressed as an unconformity,
Increased [ ~ Type 1
possibly with an incised valley and a
Channel Clustering - - Sequence lowstand fluvial deposit. Continuous sub-
[Approximate position
of Sequence Boundary)
Boundary sidence and sedimentation may occur in
zone A, leading only to clustering of channel
deposits at the time the unconformity devel-
ops in zone B. (Posamentier and Allen 1993)
Fig. 13.9. Sharp, and probably erosional sequence boundary between the Blackhawk Formation and overlying
Castlegate Sandstone at Price Canyon, Utah, indicated by arrow. (Miall1993; Olsen et aI. 1995; Yoshida et aI., in press)
JOES PRICE TRAIL ruSHER TIlOMPSON

VALLEY CANYON CANYON CANYON CANYON
I 3OOfiOO
ft m I I I I
VERTICAL 200
SCALE 100
__ 0 0
DESERT
M.EMBER
LEGEND
- - - - Type 1 Sequence Boundary
Paradox Basin
- - - - Migrating Zonl\ A / Zone B Boundary Margin
Fig. 13.10. Schematic synthesis of the sequence stratigra- high-frequency « 1 million years cyclicity) base-level
phy of the upper Mesa Verde Group in the Book Cliffs, changes of probable tectonic origin on a long-term (5
Utah, showing the interpreted migration of the zone AI million years) cycle of flexural subsidence. Additional
zone B boundary of Posamentier and Allen (1993), and the stratigraphic details of the Castlegate Sandstone and
distribution of type 1 sequence boundaries. This architec- related units are shown in Fig. 9.18. (Yoshida et aI., in
ture is interpreted as the product of the superimposition of press)
1988; Fortuin and de Smet 1991), possibly even ex- and Gibling and Bird (1994). Hanneman et al. (1994)
ceeding the rates of sea-level change induced by discussed the identification of sequence-bounding
glacio eustasy. Long-term rates of vertical tectonic paleosols in the subsurface using seismic and well-
movement reflecting regional plate-tectonic pro- log data (see Sects. 9.5.2, 9.5.4). In the Breathitt
cesses may be punctuated by short-term rates ofvery Group of eastern Kentucky interfluves are underlain
rapid subsidence, reflecting movements on indi- by carbonaceous siltstones, underclays, and rooted
vidual thrust sheets or nappes. Therefore, there may surfaces (Aitken and Flint 1995). There is little to
not be a zone A in some foreland basins. In fact, distinguish such deposits from floodplain deposits
erosional sequence boundaries are present in all the of the transgressive and highstand systems tracts,
studies of nonmarine sequences in foreland basins and Aitken and Flint (1995) were able to designate
published to date (e.g., Fig. 13.9), suggesting an ab- sequence boundaries in the interfluve areas only by
sence of zone A. It is not yet known whether the physically tracing them from the base of incised
channel clustering effect suggested in Fig. 13.8 actu- valleys.
ally occurs. Tandon and Gibling (1994) argued for a climatic
Figure 13.10 illustrates an interpretation of the control on paleosols at sequence boundaries in the
sequence architecture of the Castle gate Sandstone late Paleozoic cyclothems they studied in Nova
and associated units in the Book Cliffs, based on the Scotia. They interpreted calcareous paleosols as in-
work of Yoshida et al. (in press; Fig. 9.18). It is dicating seasonally dry climates, contrasting with
suggested that the basin was affected by a pattern of the humid climates indicated by coal-bearing depos-
long-term flexural subsidence which was greatest in its of the highstand systems tracts. These cyclothems
proximal regions (sections toward the left side of the are, of course, ascribed to glacioeustatic controls
diagram) and decreased distally, and which fluctu- (Crowell 1978). During major episodes of continen-
ated in rate over a time scale of a few millions of tal glaciation large-scale shifts in the earth's climatic
years. Superimposed on this was a pattern ofhigher- belts occur, and it is to be expected that the effects of
frequency base-level fluctuation which may have eustasy will be overlayered with the sedimentary
been tectonic or eustatic in origin. Erosional (type-I) effects of climate change. In this case, a northern
sequence boundaries in distal areas help to define hemisphere seasonal dryness is correlated with
the position of zone B of the foreland basin, but, as southern hemisphere glacial coldness (sea-level
shown in Fig. 13.10, the lateral extent of these ero- lowstand), and a northern hemisphere warm, humid
sion surfaces varies considerably. The boundary climate is correlated with a southern hemisphere
between zones A and B appears to have moved proxi- warm interglacial climate (sea-level highstand). As
mally and distally. During deposition of the Lower discussed in Sect. 12.12.2, with reference to the late
Castlegate Sandstone and the Bluecastle Sandstone Cenozoic record, the complexities of these linked
the long-term rate of subsidence was slow and zone changes in the controlling mechanisms are only just
A was absent, or was restricted to the most proximal beginning to be properly understood.
areas west of the project area. Throughout the earlier studies of sequence
Away from the main meander belt, in interfluve stratigraphy the assumption has been made that the
areas, type 2 unconformities may be characterized regional unconformities that constitute sequence
by well-developed paleosols, which require stability boundaries have chronostratigraphic significance
of the land surface and low sediment influx for their and can be used for regional, even global, correla-
development (Sect. 7.4.2). Posamentier and Allen tion. A discussion of this premise is beyond the scope
(1993) suggested the Boulder Creek Formation of of this book. It should be noted, however, that it has
British Columbia as a possible example. In this unit long been known that the processes that generate
15 paleosols are associated with a major regional nonmarine erosion surfaces are not instantaneous
unconformity (Leckie et al. 1989). Superimposed or regionally synchronous in their effect. For ex-
paleosols also mark unconformity surfaces in ample, an erosion surface generated by a fall in base
Eocene-Oligocene deposits of South Dakota and level develops by headward erosion as the longitudi-
have been interpreted as indicating two episodes of nal river profiles adjust to the new lower level. A
either base-level fall and rise, or changes in sediment change in slope from the old to the new, lower slope
supply relating to volcanism or climate change may generate a "nickpoint", which migrates land-
(Evans and Terry 1994). Other examples of paleosols ward with time (Butcher 1990; see Sect. 1l.2.2). The
marking sequence-boundary disconformities have time required for this migration is on the order of at
been described by Hanneman and Wideman (1991) least a few thousand years, which is a significant
length of time in the context of high-frequency last 1000 yrs

• floodplain :abandonment and temcf! formadon
cycles, such as those generated by glacio eustasy. The • sediment supply extremely Umited
response of fluvial systems to tectonic and climatic • development-of modem dl2nne~ and
changes of all kinds may be characterized by a sig- f100dplain-rela1ed depositional envtronments
nificant lag time (Shanley and McCabe 1994), which

is likely to increase with distance from the source of
the change and may therefore introduce significant
complexities in the regional correlation of non-
marine sequences.
Within successions that are entirely nonmarine, ca. 2500-1000 yrs BP
• sediment supply exreeds transport capadty
especially within inland basins, where correlation • continued dfPOsidon of Iale Holocene liUs
with marine units cannot be attempted, the defini- • IIoodplaln-rellled f:ldes domilWlt
• deep flooding and burial of low tern«
tion of sequences and the recognition of sequence
boundaries may be a challenging task. The response
of fluvial systems to tectonic and climatic changes is
complex, and the resulting sequences may not dis-
play the simple succession of depositional-systems
tracts that are now becoming familiar and perhaps
ca. 5000-2500 yrs BP
even standardized features of the shallow-marine
• IIoodplain abandonment and lemct formation
sedimentary record. The following examples of • sedimenl supply exceeds transport apadty
nonmarine sequence boundaries illustrate this • initial deposition of l:ue Holocene flUs
• channe~l'tl:ued f:u:les domlnanl
point.
The adjustment of longitudinal fluvial profiles to
tectonic and climatic change may not be complete if
the ability of the system to respond to change is
outpaced by the rate of change. The existence of river
terraces flanking major alluvial valleys is a good
example of this. Terrace complexes are present in ca. 11.000-5000 yrs BP
most modern river valleys and have commonly been • sediment supply exceeds transport apadty
• slow lateral migration and valley widening
attributed to base-level changes following the last ice • dfPOsition of luesl Pleistoane 10 middle
age. However, in a detailed review of the alluvial Holocene fills
· channel-rel:ued facies domlnanl
valley of the Colorado River, Texas, Blum (1992)
showed that postglacial terraces represent periods of
increased alluviation in response to climatically
driven increases in sediment supply. Each alluvial
level underwent dissection, leaving terrace rem-
nants, at times of reduced sediment supply (Fig. ca. 14.000-11,000 yrs BP
13.11). The nature of the climatic forcing process • sedimenl supply much less than transport
_ capacl\V
that controls this pattern is discussed in Sect. 12.12.2. • I100dplaUi ab2l1donment and terr:u:e formalion
Blum (1994) demonstrated that nowhere within \ followed by deep Indslon of lled.nxk valley
" . no prt5Cn'ed depositional record
coastal fluvial systems is there a single erosion sur-
face that can be related to lowstand erosion. Such
surfaces are continually modified by channel scour,
even during transgression, because episodes of
channel incision may reflect climatically controlled
ca. 20,000-14,000 yrs BP

• sediment supply exce~ transport capadry
Fig. 13.11. History of alluvial incision and aggradation in • slow l:ueral migr.ation and valley widening
• deposition of late PleiSlocene fills
response to postglacial climate change, Colorado River, · channe~rela1ed f:u:ies domin2l11
Texas (Shanley and McCabe 1994, modified from Blum
1992, reprinted by permission). As discussed in Chap. 12,
the major phase of valley incision within the 14-11 ka
period corelates to the beginning of the last postglacial
transgression
times of low sediment load, which are not synchro- paleosols, suggesting more rapid alluviation during
nous with changes in base level. This is particularly episodes of raised base level.
evident landward of the limit of base-level influence. Fluvial deposits are, by their very nature, charac-
Postglacial terraces within inland river valleys reveal terized by multiple channel scours. One of the diffi-
a history of alternating aggradation and channel culties that must be tackled in a sequence analysis of
incision, reflecting climate changes, all of which oc- entirely nonmarine deposits is the distinction of se-
curred during the last postglacial rise in sea level quence boundary erosion surfaces from autogenic
(Blum 1994). The implications of this have yet to be channel scours. Major changes in fluvial style or
resolved for inland basins where aggradation occurs grain size across the boundary may be indicative, as
(because of tectonic subsidence), rather than inci- in the sequences described by Steel and Ryseth
sion and terrace formation. However, it would seem (1990; see Figs. 9.21-9.23) and Olsen et al. (1995).
to suggest that no simple relationship between major Rogers (1994) recognized two major erosion sur-
bounding surfaces and base-level change should be faces in a nonmarine succession in Montana based
expected. on erosional relief, changes in grain size, and the
Patterns of terrace fragments and incised valleys presence of coarse channel-lag deposits with evi-
are probably associated with many nonmarine se- dence of pervasive oxidation. However, such sedi-
quence boundaries in the geological record, where mentological changes may not occur. A much more
they may be related to tectonic, eustatic, or climatic subtle sequence boundary was recognized by
processes. For example, the Mannville Group (Lower Bromley (1991 b) within a monotonous succession
Cretaceous) of Alberta is characterized by several of braid-bar sandstones in the Kayenta Formation
regional unconformity surfaces with incised valleys, (Jurassic) of western Colorado. Cut-and-fill ero-
many of them containing producing oil fields sional features along this surface are unusually steep
(Hopkins 1981, 1985; Chap. IS). Stratigraphic sepa- sided, suggesting that the sand was at least partially
ration of the various channel and terrace fragments, consolidated at the time of incision. Petrographic
which also include Jurassic remnants, is a difficult examination of the sandstone immediately below
task in the subsurface. this surface revealed limited grain-to-grain contacts
In another part of the Alberta Basin careful and abundant euhedral, syntaxial quartz over-
observation and interpretation made it possible growths, suggesting cementation prior to compac-
to demonstrate the importance of fluvial incision tion, in contrast to other sandstones within this de-
and sediment transport during falling base level, posit. Cementation can probably be attributed to
even in the absence of any recognizable preser- silcrete development on an exposed desert surface.
ved fluvial deposits. Conglomerates of the Upper Paleocurrent directions are markedly different
Cretaceous Cardium Formation are interpreted above and below this surface. Bromley (1991 b) clas-
as shoreface deposits (Plint et al. 1986), reworked sified the surface as sixth-order in rank (in the classi-
from original fluvial channel deposits by wave pro- fication system used in this book) and attributed it to
cesses during transgression (Arnott 1992). Depres- tectonic adjustments of the alluvial plain. In this
sions in the transgressive surface, which here particular case salt tectonics, originating in the Pa-
corresponds to the sequence boundary, are inter- leozoic evaporites of the underlying Paradox Basin,
preted as incised fluvial valleys widened by marine was thought to have caused shifts in the paleoslope
processes, which also reworked all remaining fluvial and changes in fluvial dispersal, resulting in lengthy
deposits. exposure of the alluvial plain in the area under study.
Good outcrops in Petrified Forest National Park In another study of a monotonous braided sheet
of Arizona allowed Kraus and Middleton (1987) to sandstone unit Lopez-Gomez and Arche (1993) were
map a major intraformational disconformity surface able to distinguish sixth-order surfaces on the basis
within the Upper Triassic Chinle Formation. At least of detailed mapping of large cliff sections. They sub-
two periods of incision and aggradation are present, divided their Triassic unit in Spain into six se-
comparable in geometry to those shown in Fig. quences bounded by these surfaces and related the
13.11 but probably of different origin. Kraus and development of the sequences to periodic basin-
Middleton (1987) attributed the erosion to lowering margin fault movement.
of base level, which led to gullying of a mature These studies by Kraus and Middleton (1987),
alluvial landscape characterized by well-developed Bromley (1991b), Lopez-Gomez and Arche (1993),
paleosols. The fill of the incised valley is mud domi- and Rogers (1994) identified sequence boundaries
nated. It contains several generations ofless mature located entirely within nonmarine deposits and
revealed the subtle nature of such boundaries. They case; for example, studies by Hopkins (1981, 1985),
reflect major tectonic, and possibly climatic changes Wood and Hopkins (1989, 1992), Shanley and
that may have no relationship to changes in sea level. McCabe (1991, 1993), Legarreta and Uliana (1991),
Kvale and Vondra (1993), Gibling and Bird (1994),
Gibling and Wightman (1994), and Aitken and Flint
13.3.2 Lowstand Systems Tract (1995) all described laterally limited lowstand
paleovalley fills. A recent book on incised-valley sys-
The lowstand systems tract is defined as the succes- tems (Dalrymple et al. 1994) contains numerous
sion of beds between the sequence boundary and examples of fluvial to estuarine lowstand deposits.
the first widespread transgressive surface. Most se- Ribbon sandstone bodies occupying linear, com-
quence models, including those developed for monly sinuous paleovalleys incised into uncon-
nonmarine deposits (Figs. 13.4, 13.5), indicate that formities are an important type of fluvial oil and gas
lowstand deposits occupy incised valleys and are reservoir (Harms 1966; Wood and Hopkins 1989,
laterally limited in extent. This may indeed be the 1992; Weimer 1986; Chaps. 14, 15). In Ordos basin,
LOUISIANA
,
\
1
I
/
/
I
".",'
o
/
10 20 Miles /
o 30 Kilomet ers
,I
Fig. 13.12. Network of paleovalleys forming a lowstand systems tract developed during glacio eustatic sea-level low, Gul
Coast (Suter et ai. 1987). See also Fig. 9.12
China, incised valleys up to 300 m deep have been strike and lack mappable interfluves. The lower
mapped (Moore et al. 1986; Song Guochu 1988). Castlegate Sandstone of Utah (Fig. 9.18) is a good
Most of the world's large rivers occupy the fill of example (Fouch et al. 1983; Yoshida et al., in press).
canyons incised into the continental margin during Van Wagoner et al. (1991) mapped a network of
glacial sea-levellowstands (e.g., Fisk 1944). incised channels at the base of the Castlegate Sand-
Individual paleovalleys can commonly be traced stone that define the initial braidplain system (Fig.
along the valley axis for many kilometers in the 13.13), but aggradation resulted in these channels
subsurface, as in the case of the Mannville Group merging laterally, and most of the lower Castlegate
valleys of Alberta (Wood and Hopkins 1989, 1992; Sandstone consists of amalgamated bar and channel
Fig. 15.1) and those filled with the "J" sandstone in deposits (Fig. 13.14) in which individual channel
the Denver Basin (Harms 1966). Paleovalleys are forms can rarely be traced for more than a few tens of
associated with considerable stratigraphic relief and meters (Miall 1994). The Lower Cretaceous Mesa
lateral facies changes, and this may make for diffi- Rica Sandstone of northeastern New Mexico is a
culties in correlation between valley fills in both the similar lowstand sheet sandstone consisting of amal-
surface and the subsurface (e.g., Gibling and Bird gamated channel deposits (Holbrook and Dunbar
1994). As Galloway (1989a) pointed out, this is one 1992).
good reason for defining sequence boundaries in The Castlegate Sandstone and the overlying and
nonmarine deposits at the nonmarine equivalent of underlying beds east of Green River, Utah, comprise
the maximum flooding surface, a proposal followed a good example of a mixed marine-nonmarine suc-
by Gibling and Bird (1994; see Sect. 13.3.4). In many cession in which fluvial deposits characteristically
cases, extensive networks of channels may be form the lowstand systems tract and are not well
present on unconformity surfaces, in which indi- represented elsewhere in the succession (Fig. 9.18).
vidual channels are commonly incised tens of meters Another good example is the Early Jurassic succes-
into the substrate (Hopkins 1981; Aubry 1989). In sion of Poland and Sweden (Pienkowski 1991; Fig.
some cases the channel fills merge laterally to form 13.15). Highstand deposits might be expected to in-
extensive sheet sandstone bodies separated by inter- clude deltaic and fluvial deposits, as shown in some
fluves, such as the lowstand deposits of the Gulf standard sequence models (Fig. 13.1), but com-
Coast Shelf off Texas and Louisiana (Suter et al. monly insufficient time is available for coastal
1987). Here, the subsurface data base is adequate for progradation during the highstand, or the deposits
the mapping of individual valleys (Fig. 13.12). are removed by the erosion that generates the next
In some cases cases the mobile channel belts in sequence boundary.
which the lowstand deposits are formed merge later- Another of the prevailing assumptions about
ally, to form broad sheets that are continuous across lowstand fluvial deposits is that they are likely to
11
112 MILE
Fig. 13.13. Network of incised channels mapped at the base of the Castlegate Sandstone, near Green River, Utah. (Van
Wagoner et al. 1991, reprinted by permission)
0'1
00
""
Fig. 13.14. Panorama of Castlegate Sandstone, Price Canyon, Utah, showing upper and lower sequence boundaries (5B) and the subdivision into Lower and Upper
units (Kc/, Kcu), corresponding to LST and TST-HST deposits
en
11>
.0
~
11>
::s
n
11>
...~
~
ciQ'
...
~
::r
'<
Fig. 13.15. Schematic stratigraphic cross

section of the Lower Jurassic deposits
from Poland (left) to Sweden (right),
showing systems tracts and sequence
boundaries. Facies types are 1 fluvial; 2
lacustrine; 3 deltaic-fluvial-lagoonal; 4
deltaic and nearshore; 5 nearshore with
storm and tide deposits; 6 fully marine.
(Pienkowski 1991)
s a
2000
6000
fli@ • •~1~18000
l.m~ :~:~ :~g: i~ :i: :~:I 10000
10000
yr B.P. ' - - - - -_ _ _ _ =:=~ ______ ~
==I....__
1o;,;".;;.;.;.;...;.:..;.:..;.;..;.:..J.._ _...... ...I:.;;.;.;....;;.;.;;.;~.;.;:.:J yr B.P.
.---------50~--------~~~
........
........
__------150km-------I~~
Fig. 13.16. Time-space model showing the Holocene development of channel patterns in the Rhine-Meuse delta. W-E
diagram is longitudinal, S-N diagram is a transect across the alluvial valley. (Tornqvist 1993)
consist of amalgamated channel deposits exhibiting in which a greater proportion of floodbasin fine
high net-to-gross sandstone ratios, in constrast to deposits are preserved, and the channels exhibit a
the deposits of the transgressive and highstand sys- greater sinuosity. Most of the examples of
tems tracts, which current models postulate should nonmarine sequences referred to in this chapter ex-
consist of more isolated channel bodies encased hibit this pattern (see in particular Gibling and
within thicker floodplain fines (Figs. 13.4, 13.5). This Wightman 1994), but the nature of the sedimentary
assumption has a long history; it may stem from the record depends entirely on the balance between sub-
work of Fisk (1944) on the Mississippi River valley, as sidence, base-level change, and sediment supply.
repeated by many later workers (e.g., Weimer 1986). Where accommodation space is generated more rap-
The oldest lowstand deposits, resting on the incised idly than it can be filled by sediment, incised valleys
valley floor, consist primarily of a braided sandstone will be invaded by arms of the sea, and the environ-
sheet. These pass upward into transgressive deposits ment will be estuarine, not fluvial (Zaitlin et al.
(.1
~
~
~ Meandering
~ &tuarineIr1dal-influence
D MoudI BarlBay Delta
r-:::::::I Lwr. Mouth BarlBay Delta I HighsWld deposi.fs

~
fbI
( A)
(.1
Braided ~ Estuarinetrldal influence
Meandering D Mouth BarlBay Delta
(':::::1 Lwr. Mouth BarlBay Delta I Highstand DepositS
(bl
(B) Distan(c
Fig. 13.17A,B. Longitudinal stratigraphic cross section of A rapid rise of base level and B slow rise of base level.
(a) and chronostratigraphic diagram (b)(Wheeler chart) Based on studies of Upper Cretaceous sequences in south-
of fluvial-coastal sequences developed during conditions ern Utah. (Shanley et al. 1992)
1994). Such deposits are better classified with the incision of river valleys that drain into the retreating
transgressive systems tract and are discussed in the coastline. This is occurring because of high coastal
next section. wave energies, and is resulting in an extension of the
marine ravinement surface inland up the major river
valleys (Leckie 1994).
13.3.3 Transgressive Systems Tract The key to the recognition of the transgressive
systems tract and the maximum flooding surface
The transgressive systems tract is defined as the beds may be the recognition of marine influence in other-
between the first widespread transgressive surface wise nonmarine successions (Shanley and McCabe
below and the maximum flooding surface above 1994). The evidence may consist of tidal or wave-
(Posamentier and Vail 1988; Zaitlin et al. 1994). In formed sedimentary structures or marine faunas
practice, there is considerable variation in the com- (Figs. 13.5, 13.18). The landward extent of this ma-
position of the lower part of stratigraphic sequences. rine influence depends on the balance of such factors
As noted above, in some cases little or no fluvial as wave power, tide range, sediment supply, and
deposit may be preserved at the base of the sequence, subsidence rate. A detailed description of tidal fea-
although there may be evidence for their former tures in transgressive deposits of Upper Cretaceous
presence, as in the example discussed by Arnott age in southern Utah is given by Shanleyet al. (1992),
(1992). In this case a marine transgressive systems who mapped tidal facies some 65 km inland of coeval
tract rests directly on the sequence boundary. In- shoreline deposits. Similar features are present in
cised valleys may be filled with fluvial or estuarine the Castlegate Sandstone of central Utah and are
strata, depending on the balance between the rate illustrated here. The most prominent sedimentary
of generation of accommodation space and the rate indicators of tidal influence are sand waves (Allen
of sediment supply (Zaitlin et al. 1994). There may 1980), tidal bedding (Reineck and Wunderlich
be an upward transition from fluvial to estuarine 1968), and inclined heterolithic strata (Thomas et al.
, deposits corresponding to the boundary between the 1987), examples of which are illustrated in Fig. 13.18.
LST and the TST, although this is likely to be Tidal sand waves are commonly characterized by
diachronous, as the rate of generation of accommo- sigmoidal bedding and rhythmically spaced mud
dation space increases, and the coastline undergoes drapes, which are distinctive structures that may be
transgression and retrogression (Fig. 13.17). Alter- recognizable even in small outcrops or core.
natively, there may be simply a change in fluvial Marine influence may extend inland for tens of
style, as suggested in some hypothetical fluvial se- kilometers, but a limit is reached at the time of
quence models (Fig. 13.4). In the Holocene deposits highest base level, when the maximum flooding sur-
of the Rhine-Meuse delta Tornqvist (1993,1994) and face is formed. This marine limit may be well within
Tornqvist et al. (1993) demonstrated that a rapid rise the depositional basin, especially in foreland basins,
in base level led to the development of an anasto- where such proximal deposits have a high preserva-
mosed fluvial style with a high avulsion frequency, tion potential. Updip from the marine limit, the
whereas a lower rate resulted in the formation of a transgressive systems tract may be recognizable only
meandering channel system (Fig. 13.16). by a change in fluvial style. The study by Shanley et
The thickness of the transgressive systems tract al. (1992) in southern Utah (Fig. 13.17) and that of
depends on the balance between sediment supply the Castlegate Sandstone (Olsen et al. 1995; and as
and the rate of generation of accommodation space reported here: Fig. 9.18) both show upward changes
(Fig. 13.17). A rapid rise in base level may result in a in fluvial style. In Price Canyon the Castlegate Sand-
very thin TST. In the case of the Castlegate Sandstone stone can be subdivided into two members (Olsen
of Utah, there is virtually no recognizable TST in et al. 1995). There is a lower sandstone member,
distal regions, east of Green River, where the marine which consists of a succession of superimposed
shales of the Buck Tongue overlie the LST sand- braided sandstone sheets with very little in the way
stones with a sharp contact (Fig. 9.18), whereas in of preserved overbank deposits [e.g., Castlegate-A
updip areas, such as near Sunnyside, much of the location of Miall (1993)] and an upper member
Upper Castle gate Sandstone is assigned to the trans- containing thicker and more extensive mudstone
gressive systems tract (Fig. 9.18). In the modern deposits (Fig. 13.14). There is little evidence of a
Canturbury Plains of New Zealand, transgression, change in channel style [contrary to the interpreta-
resulting from tectonic subsidence and eustatic sea- tions of Olsen et al. (1995)], and the change may
level rise, is leading to shoreface erosion and the simply reflect an increase in the rate of generation of
Fig. 13.18A,B. Examples of tidal sedimentary structures dined heterolithic stratification (IHS). SB indicates the
in transgressive systems tracts within dominantly fluvial sequence boundary at the top of the Castlegate Sandstone;
sequences, Castlegate Sandstone, near Price, Utah. A In- B tidal bedding
accommodation space from the LST to the TST, re- clastic sedimentation. An additional factor may be
sulting in greater vertical separation of channel the reduction in clastic influx into backswamp areas
bodies (a lower net-to-gross sandstone ratio). It is associated with certain regional allogenic condi-
not clear whether any of the upper Castlegate Sand- tions. Many coals are associated with transgressions,
stone corresponds to a highstand systems tract. Pre- and this has long formed part of the interpretation
sent evidence suggests that the HST of the Castlegate of the classic Carboniferous cyclothems of Europe
sandstone is not well preserved. and North America (e.g., Moore 1964; Wilson 1975;
In the absence of marine-influenced deposits, Crowell 1978; Gibling and Bird 1994). In these depos-
the distribution of coal may provide a useful indica- its coals typically occur above the lowstand fluvial
tion of transgressive to highstand depositional con- sandstones and below marine shales and limestones.
ditions. Coal is formed in various low-energy During the transgressive phase of sequence forma-
environments, notably alluvial backswamps and tion the rate of generation of accommodation space
back-barrier lagoons, but, as McCabe (1984) has is at its highest, while clastic depositional systems
pointed out, in many such environments a continu- are undergoing flooding and retrogradation. These
ous clastic influx (from overbank flooding or barrier are ideal conditions for coal development, and the
washover) inhibits the development of true coals, thickest and most areally extensive coals are typi-
and carbonaceous shales are the result. The exist- cally formed at such times (Ryer 1984; Hamilton and
ence of widespread, ash-free coals therefore indi- Tadros 1994). As noted above, fluvial deposits com-
cates that some additional process has occurred. prising the TST commonly contain a higher propor-
McCabe (1984) emphasized the importance of raised tion of floodplain deposits than those formed during
mires, which are organic swamps in which the rapid lowstand or highstand, and so, given appropriate
rate of organic growth and accumulation outpaces conditions for restricting clastic influx, the more
Fig. 13.18B
widespread distribution of coal in this part of a base- will be observed in fluvial sequences. It is unlikely, in
level cycle is to be expected. the first place, that in a fluvial system any deposi-
Still further updip from the coastal plain, fluvial tional response to changes in accommodation space
style may remain constant through the LST, TST, will generate a single, simple bedding surface that
and HST. The effects of base-level change are likely could be traced for long distances on a seismic
to be minimal at a distance of 100 km or more record. Stratigraphic architecture is complex in flu-
upstream from the river mouth at lowstand (Fig. vial systems and may be difficult to trace for very far
13.3), and in a foreland basin the more rapid subsid- on seismic records (Sect. 9.5.4.). Modern seismic
ence rate prevailing in proximal regions is likely to acquisition and processing methods are capable of
be the dominant control. Shanley et al. (1992) indi- yielding a much higher degree of resolution than
cated a continuous succession of braided fluvial de- they did when the models of Vail et al. (1977) were
posits through their sequences (Fig. 13.17), and the first published, and Cartwright et al. (1993) have
westernmost (most proximal) exposures yet exam- pointed out the difficulty of "forcing through" seis-
ined in the Castlegate Sandstone (Franczyk and Pit- mic correlations in many clastic settings that are
man 1991) reveal that the type of fluvial facies that chaiacterized by rapid lateral facies change. In the
characterizes the lower Castlegate at Price Canyon second place, the response to the addition of fluvial
extends through the entire thickness of the unit accommodation space during transgression may be
(Fig. 9.18). to generate a "backwater effect"> whereby the river
The onlap of sequence boundaries by the trans- responds to a rising relative base level by the aggra-
gressive systems tract is one of the major lines of dation of a wedge of sediment that gradually tapers
architectural evidence used by Vail et al. (1977) for upstream to zero thickness over tens of kilometers,
the definition of sequences in the subsurface, espe- yielding no clear-cut onlap geometry (Fig. 13.17; the
ciallywhen using seismic data. However, it is consid- effects of base-level change on rivers are discussed in
ered very unlikely that simple transgressive onlap detail in Sect. 11.2.2).
13.3.4 Equivalent of Maximum Flooding Surface limestones and shales with limited faunas, deposited
in restricted bay environments.
The maximum flooding surface (MFS) is repre- Interfluve areas may be characterized by slow
sented by the maximum landward extent of marine rates of aggradation and the development of exten-
deposits; it is immediately followed by the highstand sive paleosols. During times of base-level lowstand
systems tract (Vail et al. 1977; Vail 1987). In marine water tables are at their lowest, and such soils are
deposits the surface may be represented by a con- likely to be oxidized types (Shanley and McCabe
densed section (Loutit et al. 1988). Typically, 1994; Gibling and Bird 1994). During the transgres-
highstand deposits such as regressive delta and sive phase, and culminating in the MFS, the water
coastal barrier-strandplain systems prograde sea- table rises, and soils may undergo a change from
ward, developing a downlap relationship to the MFS. oxidized types to those formed in reducing condi-
This downlap architecture is a distinctive feature of tions (see Sect. 7.4.2 for paleosol description and
seismic cross sections (Vail et al. 1977; Vail 1987). classification). However, such changes may be con-
Galloway (1989a) argued that the MFS is commonly a fused with those resulting from changes in climate,
more readily mappable surface than a subaerial ero- which are to be expected in cycles caused by
sion surface, and he recommended using the MFS as glacioeustatic mechanisms (see Sect. 12.12.2).
the sequence boundary. To distinguish such se- As in the case of the transgressive systems tract,
quences from those of Vail and his colleagues he well inland from the coastal plain, beyond the reach
termed them genetic stratigraphic sequences. This of marine processes, there may be no facies indica-
usage has been followed by some recent workers. For tors of maximum flooding. In these proximal zones,
example, Gibling and Bird (1994), who studied the basin-margin influences, such as rapid rates of sub-
coal-bearing cyclothems of the Sydney Basin, Nova sidence resulting from extensional faulting or
Scotia, found that coal beds which form at or close to crustal loading, are the dominant control on sedi-
the time of maximum flooding are much easier to mentary style (Fig. 13.3).
correlate in the subsurface than the unconformity
surfaces marking the lowstand. The latter are cut by
laterally impersistent valley-fill deposits and may 13.3.S Highstand Systems Tract
juxtapose similar facies, making correlation diffi-
cult. A similar approach was recommended by The highstand systems tract develops when base-
Hamilton and Tadros (1994). Gibling and Bird level rise slows down and the rate of generation of
(1994) placed the boundaries of their cyclothems accommodation space decreases to a minimum
(sequences) at the base of the maximum flooding (Figs. 13.2, 13.5). There are two possible depositional
surface (Fig. 13.19). The most well-developed coals scenarios for this phase of sequence development.
occur in the TST, while the maximum flooding sur- Retrogradation of the river systems during trans-
face itself is typically represented by carbonaceous gression will have led to reduced slopes and to a low-
! ! '
TST
Fig. 13.19. Stratigraphic model for Carboniferous cyclothems of Sydney Basin, Nova Scotia. (Gibling and Bird 1994)
Time Scales of Nonmarine Sequences and Their Causes 475
energy landscape undergoing slow accumulation of and will increase only when transgression has
floodplain deposits, limited channel aggradation, brought the coastline farther inland, where the base-
and closely-spaced, well-developed soil profiles level rise on the lower reaches of the river produces a
(Shanley and McCabe 1994). Given no change in more rapid increase in accommodation. In this
source-area conditions, however, the sediment sup- model the rate of generation of accommodation is
ply into the basin will continue, and vigorous chan- greater during the highstand than during transgres-
nel systems will eventually be re-established. Under sion and results in low net-to-gross sandstone ratios.
these conditions, channel bodies will form that show However, this line of reasoning omits the influence
reduced vertical separation relative to the TST, lead- of upstream factors, and must therefore not be fol-
ing to lateral amalgamation of sandstone units and lowed dogmatically. One must also be cautious in
high net-to-gross sandstone ratios (Wright and using systems-tract terminology derived from ma-
Marriott 1993; Olsen et al. 1995; Fig. 13.4). Basinward rine processes for the labeling of nonmarine events.
progradation of coastal depositional systems leads There may be a considerable lag in the transmission
to downlap of deltaic and barrier-strandplain depos- of a transgression upstream to inland positions (Fig.
its onto the maximum flooding surface (Fig. 13.17). 13.17). The inland reaches of the river will not
A good nonmarine example of this was described by "know" that a transgression is occurring, and it is
Ray (1982), who mapped the progradation of an questionable, therefore, whether the deposits
alluvial plain and deltaic system into lake deposits formed during the initial stages of the marine trans-
(Fig. 9.44). gression should be included with the TST.
It seems likely that the HST will be poorly repre-
sented in most nonmarine basins, because the
highstand is immediately followed by the next cycle 13.3.6 Falling-stage Systems Tract
of falling base level, which may result in the removal
of much or all of the just-formed HST deposits by This systems tract has not yet been identified in any
subaerial erosion. A minor increase in the sand- ancient fluvial deposits. In nonmarine settings a fall
shale ratio immediately below the sequence bound- in base level is a time when the generation of accom-
ary may be the only indication of the highstand modation space is negative, and erosion occurs. This
phase, as in the Castlegate Sandstone of Utah (Olsen constrasts with shoreline environments, in which
et al. 1995; Fig. 9.18). "sharp-based shoreface bodies" form "stranded
Care must be taken to evaluate all the evidence in parasequences" that have a high preservation poten-
interpreting such data as net-to-gross sandstone ra- tial (Plint 1988; Hunt and Tucker 1992; Posamentier
tios. Changes in this parameter may not always be et al. 1992).
attributable to changes in the rate of generation of Terraces flanking incised valleys may correspond
accommodation space. G.A. Smith (1994) described to the FSST. The origin of fluvial terraces is complex,
a case where an increase in the proportion of channel however. Blum (1992; Fig. 13.11) argued that climate
sandstones in a section seems to be related not to changes may have been more important than base-
changes in the rate of generation of accommodation level changes in the development of postglacial ter-
space, but to increased sediment runoff resulting races of the Colorado River, Texas (this topic is
from increased rainfall. In the case of sequences discussed at greater length in Chap. 12). As argued
driven by orbital forcing mechanisms, where both by Arnott (1992), detritus generated by fluvial ero-
base-level change and climate change may be in- sion during the falling stage may provide an impor-
volved, unraveling the complexity of causes and ef- tant sediment supply for shoreline sedimentation,
fects is likely to be a continuing challenge. Some even when all traces of the river system are removed
aspects of this problem are discussed in Sect. 12.12.2. by the falling-stage erosion itself.
In the model of Shanley and McCabe (1994) a greater
degree of channel amalgamation is shown in the TST
than in the HST (Fig. 13.5), the opposite of that 13.4 Time Scales of Nonmarine Sequences
shown in the model of Wright and MarQ-ott (Fig.
and Their Causes
13.4). Shanley and McCabe (1994) suggested that
where rising base level is the main control on the rate
and style of channel stacking, the rate of generation Stratigraphic sequences occur over time scales rang-
of accommodation will be small during transgres- ing through five orders of magnitude, from tens of
sion in inland areas while the coastline is still distant, thousands of years to hundreds of millions of years
Table 13.1. Stratigraphic cycles and their causes

Type" Terminology Duration, Probable causes
(million years)
First-order 200-400 Major eustatic cycles caused by formation and
breakup of supercontinents
Second-order supercycle (Vail et al. 1977); 10-100 l. Eustatic cycles induced by volume changes in global
sequence (Sloss 1963) midocean spreading centers
2. Regional extensional downwarp and crustal loading
Third-order mesothem (Ramsbottom 1979); 1-10 Regional cycles caused by intraplate stresses. Most are
megacyclothem (Heckel 1986) probably not of global extent
Fourth-order Cyclothem (Wanless 0.2-0.S l. Milankovitch glacio eustatic cycles,
and Weller 1932); astronomical forcing
major cycle 2. Regional cycles caused by flexural loading,
(Heckel 1986) especially in foreland basins
Fifth-order minor cycle 0.01-0.2 l. Milankovitch glacioeustatic cycles,
(Heckel 1986) astronomical forcing
2. Regional cycles caused by flexural
loading, especially in foreland basins
a Hierarchy modified from Vail et al. (1977).
(Vail et al.1977; Miall1990, Chap. 8). They have been thermally elevated and displace ocean waters onto
conveniently subdivided into a fivefold hierarchy the continents. The breakup and dispersion of su-
(Table l3.1), but the subdivision into the five groups percontinents therefore tends to be accompanied by
is very approximate and classifies sequences to- high sea levels, such as those that characterized the
gether that probably have very different origins. Ex- earth during the rapid opening of the Atlantic, In-
amples of nonmarine stratigraphic units that can be dian, and Southern Oceans during the Cretaceous.
classified into these various ranks, and the corre- These cycles have only a marginal, long-term rel-
sponding "groups" used in this book, are listed in evance to fluvial sedimentation, however, and are
Table 9.1 and are described and illustrated in Chap. not discussed further, except for the following point:
9. The purpose of this section is to provide a sum- Ager (1981) noted the widespread distribution of
mary of these five types of sequence, together with a certain facies at certain geological times, and this can
brief discussion of the various allogenic processes be attributed to control by these long-term global
responsible for their formation, processes that are cycles. For example, nonmarine deposits were
discussed extensively in Chaps. 11 and 12. The focus particularly widespread during the Permian and Tri-
here is on nonmarine deposits, marine sequences assic (the New Red Sandstone of Europe). At this
having received extensive discussion elsewhere in time Pangea was beginning to undergo extension,
the geological literature. and rift basins developed in virtually all continents.
Well-known examples include the Newark rift
basins flanking the Atlantic margins of eastern
13.4.1 First-order Cycles North America (Fig. 11.40) and northwest Africa, the
Karoo rifts of southern Africa, and numerous rift
First-order cycles are related to the formation and systems in Europe (e.g., the Buntsandstein deposits
breakup of supercontinents over periods of several of Germany).
hundreds of millions of years. Worsley et al. (1984,
1986) described two Phanerozoic cycles related to
the breakup and dispersal of Pangea, and Hoffman 13.4.2 Second-order Cycles
(1991) developed a tentative plate-tectonic model
that extended the cycle back into the Proterozoic. Second-order cycles are the original interregional
The most important stratigraphic effect of these cycles of Sloss (1963). There is increasinglyconvinc-
long-term plate-tectonic movements is the variation ing evidence that many of these are global in scope
in the long-term rate and global extent of sea-floor and are generated by cycles of eustatic sea-level
spreading. Lengthy, active spreading centers are change lasting several tens of millions of years (Miall
Discussion 477
1990, Chap. 8). The primary mechanism is probably tectonically active basins, such as foreland basins,
variations in global sea-floor spreading rates super- are caused by short-term episodes of crustal loading
imposed on the first-order cycle discussed above. (Peper et al. 1992; Plint et al. 1993; Yoshida et aI., in
Such variations reflect continental-scale adjust- press). If this is the case, these cycles are likely to be
ments to spreading patterns in response to plate localized in extent and confined to individual basins.
rifting and collision events. The mechanism was first The mechanics of basin movement and consequent
suggested by Hallam (1963), and Pitman (1978) is changes in sediment delivery, depositional slope,
credited with the theoretical research that lead to and accumulation rates are the subject of much theo-
general acceptance of the idea. retical study at the present time (e.g., Jordan and
Cycles of base-level change can also be generated Flemings 1991).
by crustal movements at continental margins. On Other cycles are driven by orbital forcing of
extensional margins crustal thinning and post-rift climate. Minor climatic cycles may simply lead to
cooling lead to subsidence on a time scale of tens of variations in discharge within a river system,
millions of years (McKenzie 1978; Watts 1981), and whereas major cycles may cause regional or hemi-
rift-fill and extensional-margins deposits may in- spheric continental glaciation and consequent
clude significant thicknesses of fluvial deposits changes in base level. This has long been thought to
(Sect. 11.4.2). Clastic wedges are also generated by have been the cause of the classic cyclothems of
convergent and strike-slip tectonics, especially in Europe and North America. Minor cycles may have
foreland basins (Sect. 11.4.6), where wedges of non- occurred at any time, whereas those involving
marine "molasse" spanning several millions of years glacio eustatic base-level change probably are con-
to more than 10 million years are common. fined to the Neogene, the Late Paleozoic (the time of
the Gondwana glaciation), and various times in
the Precambrian. An additional discussion of orbital
13.4.3 Third-order Cycles forcing is given in Sect. 12.12, where examples of
minor and major cycles are described.
The nonmarine sequences and sequence models de-
scribed earlier in this chapter are based mainly on
examples of third- and fourth-order cycles that have
been described in the outcrop or subsurface.
13.5 Discussion
It is this writer's view that most, perhaps even all
third-order cycles are tectonic in origin (Mialll991c; Sequence stratigraphy is a subject undergoing rapid
Yoshida et aI., in press). Regional tectonic episodes evolution at this time, and there is a danger that this
localized along major faults generate basin-margin chapter in the book, in particular, may be out of date
uplifts and cycles ofbasin subsidence on time scales before it is published. Although a consensus seems
of several millions of years. Tectonism is triggered by to be emerging regarding the application of se-
the regional response to plate-tectonic movements quence concepts to the nonmarine record, very few
and may include minor changes in sea-floor spread- well-documented examples of nonmarine sequences
ing vectors and the adjustment of major fault blocks have yet been described. There is room for a consid-
and terrane fragments on the continents. Stresses erable amount of additional research in this area.
are transmitted laterally by intraplate stress (Sect. Quantitative modeling is also required, of the type
11.3.5), and in this way may cause the development discussed in Sect. 11.3.6, to more fully explore
of synchronous cycles of uplift and subsidence over the relationship between tectonic processes, basin
areas as large as entire continental plates. Several movements, sediment supply, sediment delivery,
examples are given in Sect. 9.4 (group 10 deposits of and facies changes within sedimentary basins.
Table 9.1), and the plate-tectonic settings of these The example of the Castlegate Sandstone (Fig.
various types of sequence are discussed in Sect. 11.4. 9.18) illustrates some of the difficulties in applying
sequence concepts to fluvial deposits (Yoshida et
aI., in press). Where marine deposits interfinger
13.4.4 Fourth- and Fifth-order Cycles with the fluvial succession, a,s in the area east of
Trail Canyon, recognition of sequence boundaries,
Fourth- and fifth-order cycles correspond to groups flooding surfaces, and systems tracts is relatively
9 and 8, respectively, in Table 9.1, where several straightforward (Van Wagoner et al. 1990, 1991).
nonmarine examples are listed. It seems increasingly The fact that some of these features are difficult to
likely that many fourth- and fifth-order cycles in trace up dip, away from the coastal plain, raises an
important question regarding the sensitivity of flu- process, such as third-order tectonism (fluvial reju-
vial systems to base-level change. In the most proxi- venation, changes in sediment supply), especially in
mal exposures of the Castlegate Sandstone (Fig. 9.18; updip regions, and a fourth-order cycle of base-level
Joes Valley, Price Canyon) there appears to be only a change, of tectonic or eustatic origin, which is most
single sequence, of probable third order, according pronounced in coastal areas. Local complications
to the chronostratigraphic reconstruction of Fouch arise in this particular case, because the downdip
et al. (1983). These beds appear to correlate with a areas (Trail Canyon and areas to the east) are under-
succession of higher-order sequences in the coastal lain by the Late Paleozoic evaporites of Paradox
plain area (Trail Canyon and points east). The updip Basin, and the effects of contemporaneous salt tec-
merging of these higher order sequences into a single tonics may have been significant. This should serve
undifferentiated succession may be interpreted in to remind us that in each real-life case study, the
one of at least two ways: (a) Fluvial systems are lessons to be learned from theoretical or numerical
relatively insensitive to base-level change, the effects models must be tempered by the practicalities of the
of which disappear updip; or (b) the Castlegate given basin setting, with its inherited crustal hetero-
stratigraphy may be controlled by more than one geneity and history of subsidence and tectonism.
Chapter 14
Stratigraphic and Tectonic Controls on the Distribution

and Architecture of Fluvial Oil and Gas Reservoirs
14.1 Introduction poses of this chapter is to develop a threefold de-

scriptive grouping of fluvial reservoirs based on
Significant volumes of oil and gas are trapped in these criteria. The three reservoir types tend to occur
fluvial sandstones. Major reservoirs include the in particular tectonic settings, but the tectonic con-
Statfjord Formation (Triassic-Jurassic) of the North trol is varied enough that this cannot form part of the
Sea Basin, the Sadlerochit Group (Permian-Triassic) classification. This chapter deals with only part of
of Prudhoe Bay on the Alaskan North Slope, the what constitutes a complete "play". Other aspects
Lower Cretaceous reservoirs of the giant Daqing including source rocks, migration history, trap tim-
field of the Songliao Basin, China, the heavy-oil ing, etc. (White 1980, 1988) are not discussed here.
sands of the Cretaceous Athabasca and related de- Some of the ideas regarding the various styles of
posits in Alberta, Canada, and numerous large to fluvial traps were described briefly by Cant (l982).
small fields in mature areas such as the Alberta Basin Blackbourn (l984) discussed the facies variability of
(Mannville fields, Cretaceous-age reservoirs), the fluvial deposits but not the details of petroleum oc-
southern Midcontinent (Cherokee fields, Pennsylva- currence or reservoir types. Davies et al. (l991, 1993)
nian), and Gulf Coast (Cretaceous Tuscaloosa and offered two models for fluvial channel-fill reservoirs
Travis Peak fields, and Oligocene Frio fields). The and suggested that their classification has a broad
tectonic setting and reservoir geometries of these applicability, but they did not develop this idea fur-
and other fluvial reservoirs are varied, but this chap- ther. Conybeare (1976) described numerous fluvial
ter demonstrates that oil and gas fields in fluvial oil and gas fields but did not offer any useful ideas
sandstones fall into three broad classes, on the basis regarding classification.
of two major descriptive criteria. An extensive re- Detailed published information on the facies ar-
view of fields around the world is drawn upon to chitecture of producing fluvial reservoirs is available
illustrate this classification. for many fields in North America and northwest
The two criteria are: (a) geometry and origin of Europe, and for some in Australia. Most of this paper
the depositional system and (b) geometry of the is based on data from these areas. For other parts of
reservoir bodies (Table 14.1). One of the major pur- the world much less information is available, in part
Table 14.1. Criteria for classification of petroleum reservoirs in fluvial sandstones

Criterion Classes Comments
Geometry of 1. Clastic wedge Active tectonism
depo. system 2. Paleovalley Low base level
Geometry of 1. Sheet Mainly braided systems

reservoir 2. Ribbon or lens Meandering, anastomosed
Tectonic setting' 1. Rift basin Assoc. with lake beds

2. Extensional-margin basin
3. Backarc basin
4. Retroarc foreland basin
S. Forearc basin May be volcaniclastic
6. Strike-slip basin Assoc. with collision
7. Peripheral foreland basin Assoc. with collision
• Tectonic setting is included for discussion purposes but does not form a primary part of the classification.
480 Stratigraphic and Tectonic Controls on the Distribution and Architecture of Fluvial Oil and Gas Reservoirs
because petroleum exploration and development 74.2.7.7 Clastic Wedges

are still in relatively early stages, and the details of
reservoir characteristics are not yet well known (or Major clastic wedges and thick sandstone sheets oc-
have not been published). This applies to much of cur wherever tectonic conditions maintain steep
Africa (Gondwanan and Cenozoic rifts), Andean paleo slopes and significant relief. Such wedges are
South America, southeast Asia, India, and China. In therefore commonly associated with rift basins and
all these areas nonmarine reservoirs are significant foreland basins, which are generally bounded by up-
and may even be the most important reservoir type. faulted mountainous source areas (Figs. 14.1A,B,
The classification proposed here, with the accom- 14.2). Clastic wedges may be hundreds of meters to a
panying extensive referencing to published ex- few thousand meters thick, and may extend along
amples, is offered as an aid to exploration and strike for tens to hundreds of kilometers. They typi-
production. The numerous published case studies cally pass downslope into major delta or strandplain
cited here may provide useful analogs for future field systems (Fig. 14.1A) that may themselves be signifi-
studies. cant petroleum reservoirs. Alternatively, they may
grade into lacustrine deposits (Fig. 14.1B). Whether
the distal margins are marine or lacustrine, base-
14.2 The Geometry of Fluvial Reservoirs level change is a significant control on the distribu-
tion and geometry of the fluvial wedge. Tongues or
14.2.1 Geometry and Origin of Depositional Systems sheets of fluvial deposits, formed during times oflow
sea level or during periods of slow subsidence, may
There are two broad architectural styles of impor- be interbedded with fine-grained sediments formed
tance for petroleum trapping, the clastic wedge, and in subaqueous settings (lake basin, marine shelf, or
the paleovalley fill (Table 14.1, Fig. 14.1). inter deltaic bay). Autogenic switching of delta dis-
o 50 100 150 km
I I I I
Fig. 14.1A-C. Typical geometries of fluvial depositional active faulted uplift and bordered by an arid lake system.
systems. A Typical foreland basin clastic wedge, showing Based on the Cretaceous rifted margin of Brazil (Mello and
the Western Interior Basin, Utah-Colorado, during accu- Maxwell 1990; modified from Eugster and Hardie 1975). C
mulation of the Mesa Verde Group (Upper Cretaceous) A typical paleovalley system. Early Cretaceous paleogeog-
(Cole and Friberg 1989). B Clastic wedge bounded by raphy of west-central Alberta. (Rosenthal 1988)
The Geometry of Fluvial Reservoirs 481
fault block mountains
mud flats
mountain with carbonate
precipitation
shallow alkaline lake

with alternations of
salt, carbonate and
lipidic Ol]!anic-rich
calcareous black shn1e
deposits
ARID BASIN MARGIN.

BRAZILIAN CONTINENTAL MARGIN
B UINTA BASIN. etc.
PALEOGEOGRAPHIC RECONSTRUCTION
GLAUCONITE FORMATION
WEST-CENTRAL ALBERTA
c
Fig. 14.1B,C
tributaries may also be an important factor in the (Triassic-Jurassic) of the North Viking Graben
interbedding of fluvial-deltaic and marine deposits. and East Shetland Basin, Triassic-Jurassic rocks of
Clastic wedges with a nonmarine-deltaic compo- Ordos Basin, China, the Jurassic-Cretaceous succes-
nent that are important as petroleum reservoirs in- sion of Eromanga and Surat basins, Australia, early
clude some of the molasse wedges of the Cordilleran Tertiary rocks of Gippsland Basin, Australia, and
foreland basin (e.g., Upper Cretaceous Mesa Verde the thick, mainly Cenozoic sections of the Hima-
Group of the Colorado Plateau; Fig. 14.2), the Chero- layan and Andean foreland basins. Along the US
kee Group (Pennsylvanian) of the Arkoma Basin, the Gulf Coast, clastic wedges of fluvial and fluvial
Sadlerochit Group (Permian-Triassic) and Kekiktuk delta-plain origin include the Denkman Sandstone
Formation (Mississippian) of the North Slope of (Upper Jurassic), the Travis Peak formation (Lower
Alaska, the Statfjord Formation and Brent Group Cretaceous), the Woodbine-Tuscaloosa Formation
UTAH ICOLORADO
WASATCH BOOK CLIFFS FRONT RANGE UPLFT DENVER BASIN
PLATEAU
metres
1000
800
600
400
~ Nonmarine rocks
_
o Calcareous Shale
200
Marine and coastal·barrler ~lime.tone
o
o
sandstone
_ _ _ _ Conglomerate
Marine shale and s41'stone
e. ... ApptOXimate time line Vert Exag. .. 151
Fig. 14.2. Stratigraphic cross section of Cretaceous-Ter- foreland basin of the North American Western Interior.
tiary rocks from central Utah to northeastern Colorado. Note interfingering of marine and nonmarine strata.
Clastic wedges of this type are very common in retroarc (Molenaar and Rice 1988)
and peripheral foreland basins. This example is from the
(Upper Cretaceous), the Wilcox Group (Paleocene- 74.2.7.2 Paleovalley Fills

Eocene), and the Frio Formation (Oligocene) (re-
ferences to all these examples are provided later in A quite different style of fluvial depositional system
this chapter). In all of these cases separating deltaic is the paleovalley fill (Fig. 14.1C). Such valleys are
production from strictly fluvial production is a diffi- typically a few kilometers wide, a few tens of meters
cult and sometimes an arbitrary exercise. Wherever deep, and may be traced for a few tens of kilometers.
fluvial reservoir sandstones are interbedded with These valleys are formed by subaerial channel inci-
fine-grained marine sediments deposited in inter- sion during times of falling base level, and are filled
deltaic or interdistributary bays, they have been during subsequent lowstand to trangressive stages
omitted from this review. The geology of such fields (Sect. 13.3.2). They may therefore be associated with
is distinctly different from that of purely fluvial pro- a major regional or interregional unconformity.
duction. For this reason, some major producing re- They may be filled entirely with fluvial deposits, or in
gions such as the Niger Delta and many of the major part with estuarine sediments, depending on local
clastic units of the US Gulf Coast are not discussed. conditions of paleoslope, sediment supply, and rate
Fluvial deposits that are interbedded with lacus- of base-level change (Zaitlin et al. 1994). Fluvial
trine facies include the Green River deposits of the depositional systems of this type occur primarily in
Green River and Uinta basins of the United States areas of low regional slope, such as cratonic basins,
(Paleocene-Eocene), the Tulare Formation in South extensional continental margins, and foreland ba-
Belridge field, California (Pleistocene), the Bahia sins distant from the mountainous source area [the
Supergroup of the Reconcavo Basin (Jurassic-Lower "B" zone of Posamentier and Allen (1993)]. For ex-
Cretaceous), and the fill of most of the Chinese ample, the Quaternary continental shelf of the Gulf
basins (Songliao, Jiangsu, Ordos, Bohai Bay). Coast is crossed by numerous incised valley systems
LEGEND
wmJ Pre-Pleistocene surface
~ Sangamonian Iluvial-deltaic trends
H?i·:.::] Sangamonian beach and strandplain
~lt Wisconsin ri vcr courses
~ Late Wisconsinan shelf-margin delta
Late Wisconsinan-Holocene reefs
present shoreline
transgressed Holocene PI'

beaches and '"
sand ridges ",0'0
~
.,,~
Brazos-Colorado ~~~~Y Sabioc-Calcasieu

delta delta
o 50 100 ISO Iun

I I I I
QUATERNARY DEPOSITIONAL SYSTEMS,

TEXAS COASTAL PLAIN
Fig. 14.3. Late Pleistocene and Holocene depositional systems of the Texas coast, showing location of incised fluvial
systems and lowstand deltas formed during glacial phases. (Simplifed from Morton and Price 1987)
Alluvial valley
Fig. 14.4. Schematic block diagram of the valley-fill deposits of the

Mississippi River. The basal bounding discontinuity of the valley
formed during the Pleistocene lowstand of sea level, when the river
incised bedrock to adjust itself to a lower base level. As sea level rose
following glaciation, the valley filled first with coarse sands deposited
in a braided-stream environment and then with lower-energy mean-
dering-stream deposits. Many ancient valley-fill alluvial successions
formed in this way. In some, the uppermost deposits are estuarine to
marine in depositional environment and may be followed by wide-
spread transgressive-marine blanket deposits. (Weimer 1986; based
Channel scour on the work of H. Fisk)
484 Stratigraphic and Tectonic Controls on the Distribution and Architecture of FluVial Oil and Gas Reservoirs
formed during glacial lowstands of sea level (Figs. Alaska (Melvin 1993), the Lower Cretaceous Sarir
14.3, 14.4). Sandstone of Libya (Clifford et al. 1980), the
Several important petroleum plays in the Cordil- unnamed Cambrian sandstone of the Oued Mya ba-
leran foreland basin produce from paleovalleys in- sin, Algeria (Balducchi and Pommier 1970), the
cised into regional unconformities generated during Sherwood sandstone (Triassic) at South Morecambe
the Early Cretaceous (Mannville fields, Alberta; "J" field in the Irish Sea (Stuart and Cowan 1991; G.
and "D" sandstones, Denver Basin). Deep, fluvial, Cowan 1993), and many of the Hutton Sandstone
incised valley fills of Lower Jurassic age are impor- fields of Eromanga Basin, Australia (Moore et al.
tant reservoirs in the Ordos Basin of China. There is 1986; Wecker 1989). In most of these examples broad
much ongoing debate about whether the uncon- sheets are limited at their margins by lateral
formities in the Western Interior Basin are tectonic pinchout or channel incision. The Peco oilfield of
or eustatic in origin (Weimer 1986; Dolson et al. Alberta produces from a single braided channel
1991). complex 20 m deep and up to 3 km wide enclosed in
impermeable floodplain deposits (Gardiner et al.
1990; Putnam 1993). The Travis Peak formation of
14.2.2 Geometry of Reservoir Bodies East Texas includes several separate channel belts up
to 10 km wide (Tye 1991).
The second criterion for the classification of fluvial Sheet sandstones can also accumulate where
petroleum reservoirs is that of the geometry of the lower-slope and lower-energy rivers, such as some
reservoir body. Two similar classifications have been meandering systems, develop in areas oflow subsid-
used. Friend (1983) suggested that fluvial sandstones ence rate, so that channel deposits are incised into
occur primarily in two modes, the laterally extensive and superimposed upon each other and the preser-
sheet and the more laterally restricted ribbon or lens. vation potential of fine-grained floodplain deposits
Galloway (1981) adopted the geomorphic classifica- is low (Miall1980). Examples include at least some of
tion of Schumm (1977) in his review of Gulf Coast the Permian fluvial reservoirs of Cooper Basin, Aus-
fluvial systems, and this approach has been used tralia (Stuart et al. 1988; Stanmore and Johnstone
by several other workers (e.g., Davies et al. 1991, 1988).
1993). The classification suggests three architectural Slightly different is the style of the lake-margin
modes, (a) multilateral channel deposits formed by fluvial reservoirs in the Green River Formation of
bedload rivers, with high net-to-gross sandstone ra- Red Wash field, Uinta Basin. Here, individual pro-
tios and good interconnectedness; (b) multistory, ducing sandstones are extensive, but much less so
mixed sandstone-shale deposits with intermediate than the area of the entire field. Isopachs of the
values for interconnectedness, formed in mixed- producing sandstones suggest that they include thin,
load fluvial systems; and (c) multistory ribbon fluvially dominated, deltaic sheets (Chatfield 1972;
bodies with low net-to-gross ratios and poor Castle 1990). Examples of fields producing from
interconnectedness, formed in suspended-load riv- sheet-like reservoir bodies are described in Sect.
ers. The first and third types are similar to the two 15.3.
classes of Friend (1983), which are the subdivisions
used here.
74.2.2.2 Sandstone Ribbons and Lenses
14.2.2.1 Sheet Sandstones These form by the accumulation of sandstone bodies

within channel systems that are isolated within the
Deposits of this type are commonly formed by steep- fine-grained sediments of the overbank environ-
gradient, bed-load systems such as braided rivers, ment. Point bars are the best known of such deposits;
where channels comb across broad areas of the val- they form distinctive lens-shaped reservoir bodies,
ley floor. Sheet sandstones typically form laterally many of which preserve the fine-grained fill of the
extensive reservoir bodies that may require a struc- abandoned channel around the margin of the lens
tural component to generate a petroleum trap. Ex- (Berg 1968; Busch 1974; Werren et al. 1990; Davies et
cellent examples are parts of the Permian-Triassic al. 1991, 1993; Chapin and Mayer 1991). Not all
Sadlerochit Group of Prudhoe Bay, Alaska (Lawton braided fluvial systems form sandstones with a sheet
et al. 1987; Atkinson et al. 1990), the braided-fluvial geometry. In Wytch Farm oilfield, England, sand-
zone 2 of the Kekiktuk Formation in Endicott field, stones interpreted as braided channel deposits and
"sheet-flood" units are lenticular, and form part of a which is an important gas producer (Galloway et al.
heterogeneous clastic succession (Dranfield et al. 1982; Jirik 1990; Kerr 1990; Kerr and Jirik 1990); and
1987). According to Bentham et al. (1993), braided (c) the Lower Cretaceous Donovan/Sligo Formation
systems consisting of channel sandstone bodies in Texas. The Citronelle field, which produces from
interbedded with significant proportions of flood- this unit, is an excellent example, illustrated later
plain fine-grained deposits constitute a distinctive (Eaves 1976).
type of fluvial assemblage that is much more impor- Most of the producing zones in the Songliao and
tant in the geological record than has hitherto been Ordos basins, China, are lenticular or ribbon-shaped
recognized. in geometry, and a detailed knowledge of their sub-
Many fluvial sand bodies are mapped and de- surface distribution and facies is required for the
scribed as ribbons and may represent the fill of high- development of efficient production programs (Jin
or low-sinuosity channels. Examples are described et al. 1985; Moore et al. 1986; Yinan et al. 1987).
by Davies et al. (1991) and Yin an et al. (1987). Likewise, fluvial-deltaic channel-and-bar sands of
Hastings (1990) illustrated a complex of linear the Oligocene Barail Sands of northeast India pro-
sandbodies forming a low-sinuosity meander belt in vide numerous ribbon-shaped and lenticular reser-
the Tyler Formation of North Dakota. Many may, voir bodies in combined structural-stratigraphic
however, represent the amalgamation of point bars traps (Murty 1983; Desikachar 1984). Tertiary pro-
or other bar deposits along the channel trend, production in the Green River Basin (McDonald 1976)
ducing an overall ribbon geometry (Ebanks and also provides excellent examples of this type of reser-
Weber 1982). Where sufficient well control is avail- voir geometry. It may be very difficult to produce
able, such ribbons may be broken down into their accurate maps of these ribbon sandstones. A tight
individual bar units. Examples of these various con- well spacing is required, as otherwise spurious corre-
ditions are described later. Ribbon and lens reser- lations may result. This is discussed further below.
voirs may form purely stratigraphic traps. The Examples of fields producing from ribbons and
ribbon shape of these sandstones should not be con- lenses are described in Sect. 15.4.
fused with the ribbon shape of many paleovalley-fill
sandstones, which are typically much wider and
thicker and represent the fill of entire, incised mean- 14.2.2.3 Stratigraphic Variations in Reservoir Geometry
der belts.
The Cherokee Group (Pennsylvanian) of Okla- In many fields the geometry of the reservoirs varies
homa, Kansas, and Missouri has provided many stratigraphically, reflecting allogenic influences on
classic examples of shoestring or ribbon sandstones. fluvial style or autogenic migration of the fluvial
Examples of producing sandstones in this group, system. For example, in South Belridge field, Califor-
particularly the Bartlesville Sandstone (Desmoine- nia, the Tulare Formation consists of a braid-delta
sian), have been known since the 1920s and have complex that prograded into a lake. Lower Tulare
been illustrated in many classic textbooks (e.g., reservoir sands are lenticular or ribbon bodies de-
Levorsen 1967). posited in meandering fluvial systems, whereas Up-
Another classic area of production from fields per Tulare sandstones were deposited in wider, more
consisting largely of stratigraphically trapped petro- proximal, braided channels and are more laterally
leum in ribbons and lenses is the Gulf Coast. Major continuous (Miller et al. 1990). Similar stratigraphic
fluvial producers there are (a) the Upper Cretaceous variations have been described in Prudhoe Bay and
Lower Tuscaloosa Sandstone of southern Missis- Endicott fields, Alaska (Lawton et al. 1987; Melvin
sippi and Louisiana, which constitutes a transgres- 1993). In the Gullfaks field, in the Norwegian sector
sive, retrograde, fluvial-deltaic depositional systems of the East Shetlands Basin, the Statfjord Formation
tract (Karges 1962; Berg and Cook 1968; Stan cliffe is divided into three members comprising at least six
and Adams 1986; Hamilton and Cameron 1986; reservoir units, reflecting stratigraphic variations in
Hamlin and Cameron 1987; Wiygul and Young 1987; fluvial style (Petterson et al. 1990). In the nearby
Hersch 1987; Klicman et al. 1988; Hogg 1988; Miller Brent field, eight reservoir units have been mapped
and Groth 1990; Werren et al. 1990; Garrison and in the Statfjord Formation (Struijk and Green 1991).
Chancellor 1991); (b) the Oligocene Frio Formation Sequence concepts which may explain these varia-
of southwest Texas, another fluvial-deltaic system tions are discussed below.
14.3 Tectonic Setting of Fluvial Reservoirs 14.3.2 Backarc Basins
The Songliao Basin of eastern China, where the giant

The thickest and most widespread fluvial deposits Daqing field is reservoired in Lower Cretaceous flu-
accumulate in a few specific tectonic settings. These vial deposits, is a possible backarc basin, although it
group into two main categories - extensional basins,
may have been influenced by escape tectonics
and basins associated with plate collision, including (Burke and Seng6r 1986). The Jiangsu Basin, to the
foreland basins, forearc basins, and strike-slip ba- south, has a similar tectonic origin, and includes
sins (Fig. 11.37). Petroleum plays in fluvial sand- some early Tertiary fluvial reservoirs forming parts
stones reflect this distribution, most of the major of lacustrine delta complexes (Moore et al. 1986).
plays occurring in rift basins and foreland basins, The North China Basin, which includes the im-
with a few major plays in extensional-margin basins portant producing region of Bohai Bay, is also a
and forearc basins. This localization of the petro-
Cenozoic backarc basin and contains fluvial-deltaic
leum is not sup rising, because it is only in these
wedges and lacustrine deposits forming numerous
specific types of tectonic setting that fluvial deposits structural and stratigraphic trap types (Shaui Defu et
are volumetrically important enough to constitute a al. 1988; Matsuzawa 1988; Zhao Xueping et al. 1988).
distinctive, major category of reservoir body. Fluvial Some of the productive Andean basins of Argen-
successions up to 8 km thick occur in some of these tina, notably Cuyo and San Jorge basins, were
basins. backarc basins during much of the Cretaceous, at the
The following paragraphs describe the tectonic time the main reservoir units were deposited
setting of selected producing units. Other examples (Mpodozis and Ramos 1990; Stabler 1990).
of the tectonic setting of fluvial deposits are given in
Sect. 11.4.
14.3.3 Forearc Basins

14.3.1 Retroarc (Backarc) Foreland Basins
Cook Inlet, Alaska, is a forearc basin with significant
production from the upper Tertiary Kenai Group, a
The Jurassic-Tertiary foreland basin adjacent to the
predominantly fluvial succession more than 7 km
Cordilleran Orogen of North America includes
thick (Kirschner and Lyon 1973; Hayes et al. 1976;
many petroleum-bearing fluvial sandstones.
Boss et al. 1976; Figs. 11.51, 11.52).
Foreland basins associated with the Paleozoic Appa-
lachian Orogen and its southward extension (e.g.,
Arkoma Basin) are also important, although in the
case of the Arkoma Basin, the major producing unit, 14.3.4 Collision-Related Basins
the Pennsylvanian Cherokee Group, was derived
from cratonic uplifts to the north of the basin The nonmarine Cenozoic basins of western China,
(Rascoe and Adler 1983), and not from uplifts asso- including Tarim, Djunggar, Jiuxi, and Qaidam ba-
ciated with the fold-thrust belt adjacent to the basin, sins, all include important production from fluvial
which is usually the case for foreland-basin deposits. sandstones [e.g., Oligocene-Miocene reservoirs in
Cenozoic fluvial deposits in the foreland basins Jiuxi Basin: Wang and Coward (1993)]. They are
flanking the Andes of Colombia (Llanos Orientales basins associated with plate collision - their main
and Putumayo basins) are also productive (Simlote features were developed during the Himalayan orog-
et al. 1985; Salazar 1990; Stabler 1990). For example, eny, and they have the characteristics of foreland or
the giant Cafto Limon oil field of Colombia produces strike-slip basins (Figs. 11.65, 11.66). Many show
mainly from Eocene and Oligocene channel sands in lateral block movements, a characteristic of escape
fluvial-deltaic complexes (Gabela 1990; Cleveland tectonics (Burke and Seng6r 1986).
and Molina 1990). The Ordos Basin in China is a foreland basin of
The Eromanga and Surat basins of Australia are Triassic-Jurassic age which has important petroleum
Jurassic-Cretaceous foreland basins that contain sig- production from fluvial reservoirs (Moore et al.
nificant fluvial production. However, the Hutton 1986). Cenozoic fluvial deposits in the peripheral
sandstone, which is the major reservoir unit, was foreland basins flanking the Himalayan Mountains
derived from uplifts on the cratonic sides of the ofIndia are also productive (Murty 1983; Desikachar
basins (Moore et al. 1986; Watts 1987). 1984).
Tectonic Setting of Fluvial Reservoirs 487
Cenozoic basins in southeast Asia were generated gas, but most of these basins are relatively unex-
by tectonism associated with the India-Asia collision plored and the full potential remains to be deter-
(Tapponier et al. 1986; Fig. 11.65). These include rift mined (Genik 1993; Winn et al. 1993; Derksen and
and transtensional basins in offshore Malaysia, Thai- McLean-Hodgson 1988).
land, Cambodia, Vietnam, and the South China Sea
and some onshore basins in these same countries
[grouped together as the Sundaland rifts by Derksen 14.3.7 Basins on Extensional Continental Margins
and McLean-Hodgson (1988)]. Fluvial-deltaic pro-
gradation from adjacent contemporaneous uplifts Fluvial reservoirs are very important in some exten-
provided numerous potential reservoir units, and sional, craton-margin basins. The Gulf Coast of the
offshore exploration is now discovering many impor- United States (Fig. 1l.42) is among the most well-
tant oil and gas fields in these basins (Lee 1982; known of such basins, as it has produced huge
Crostella 1983; Roberts 1988; Derksen and McLean- volumes of oil and gas from terrigenous clastic res-
Hodgson 1988; Blanche 1990). For example, in ervoirs (Tyler et al. 1984). Most of the reservoirs are
Pattani trough, offshore Thailand, a Tertiary-Quater- marine in origin, but significant production has
nary fluvial-deltaic section up to 8 km thick is present also been obtained from fluvial units forming part
and contains important gas accumulations (Blanche of inland alluvial systems or delta plains distant
1990). In the Phitsanulok Basin of central Thailand, 8 from the coast. Nonmarine stratigraphic units in-
m 3 x 106 of recoverable oil have been discovered to clude parts of the Denkman Sandstone (Upper Ju-
date from an 8 km Oligocene-Lower Miocene fluvial- rassic), Travis Peak Formation (Lower Cretaceous),
lacustrine succession (Derksen and McLean- Woodbine-Tuscaloosa Formation (Upper Creta-
Hodgson 1988). ceous), the Wilcox Group (Paleocene-Eocene), and
the Frio Formation (Oligocene).
The Triassic-Tertiary extensional continental
14.3.S Basins in Continental-Transform Settings margins of western and northwestern Australia
have yielded significant gas production from fluvial
Few producing fluvial reservoirs are located in this reservoirs (e.g., Rankin field: Beston 1986; Posaner
type of tectonic setting. The South Belridge field, and Goldthorpe 1986; Warris 1988).
California, is an example. The reservoir is a Pliocene The Upper Paleozoic to Triassic reservoirs of the
fluvial-deltaic unit derived from uplifts associated North Slope of Alaska were deposited on the south-
with the San Andreas fault and trapped in associated facing margins of a craton that was rifted and dis-
anticlinal structures (Miller et al. 1990). persed during the Cretaceous when the Arctic
Ocean basin was generated (Morgridge and Smith
1972). Other important extensional-margin basins
14.3.6 Rift Basins occur in North Africa (Sirte Basin, of Cretaceous
age, and Cambrian Oued Mya Basin; e.g., Clifford et
Several major fluvial exploration plays occur in rift al. 1980).
basins, including East Shetland Basin and Viking
Graben, North Sea (Triassic-Tertiary; Parsley 1984;
Figs. 11.44, 1l.45), and the geologically very similar 14.3.8 Intracratonic Basins
Gippsland Basin, Australia (Late Cretaceous-Ter-
tiary; Rahmanian et al. 1990; Figs. 11.46-11.48), as Intracratonic basins do not constitute a major class
well as the Reconcavo Basin of Brazil (Jurassic-Creta- of basin from the point of view of nonmarine petro-
ceous) (Figueiredo et al. 1994). leum fields, but a few significant discoveries have
Triassic rift-fill fluvial sediments host a significant been made in this setting. The Morrow Sandstone
oil field at Wytch Farm in southern Britain (Dranfield (Mississipian) of eastern Colorado and western
et al. 1987) and gas off the west coast, at Morecambe Kansas produces from paleovalleys crossing an
field (Stuart and Cowan 1991; G. Cowan 1993; Mead- intracratonic basin at the margin of the Anadarko
ows and Beach 1993). basin. The valleys were incised during glacio-
The rift basins of Africa, including the Karoo eustatic falls in sea level. The Clifford-Siaana trend
rifts and those originating during the breakup of in eastern Colorado, which is still undergoing ac-
Gondwana in the Mesozoic-Cenozoic, all contain sig- tive exploration, is estimated to contain up to
nificant fluvial clastic units that have yielded oil and 9.5 m 3 x 106 of recoverable oil (Bowen et al. 1993).
geometry of depositional system Fig. 14.5. Classification of nonmarine reservoirs ac-

- -~
paleovalley fill
cording to the geometry of the depositional system
and the geometry of the reservoir body
geometry of reservoir
sheets~ ~ ribbons and lenses
s:::: b;/ . : .....::::z::=:= -=-":. :. . <s=-SO· ::':.....::?=-=-
·z· ..
4? ·<>.::5.:
.~
=~=--G2:;-=~
SH-TYPE CB -TYPE
Table 14.2. The three major types of petroleum reservoir in fluvial sandstones
Type Code Depositional Reservoir Most typical
system geometry tectonic setting
Paleovalley PV Incised valley Ribbon Foreland basin
Sheet SH Clastic wedge Sheet Rift basin, extensional margin
Channel and bar CB Clastic wedge Multiple lenses, ribbons Foreland basin, extensional margin
Table 14.3. Examples of petroleum production from fluvial sandstones in North America
Location Tectonic Stratigraphy Age Reser- Reserves Reserves
setting voir oil gas
type (m3 x 106 ) (m3xIQ9)
Alberta Retroarc Mannville Cretaceous PV 305 962

Basin foreland Group PV-CB
Alberta Retroarc Athabasca Cretaceous CB 144000
Basin foreland Oil Sands (HO)
Denver Retroarc "1" and Cretaceous PV 127
Basin foreland "0" Ss
Uinta Retroarc Mesa Verde, Cretaceous CB 80 422
Basin foreland Green River Tertiary
Green River Retroarc Hoback, Tertiary CB 12 70
Basin foreland Green River
Arkoma Retroarc Cherokee Penn. CB 436
Basin foreland Group +320(HO)
East Texas Extensional Woodbine Cretaceous CB 252
Basin margin
Texas Gulf Extensional Frio Oligocene CB 59
Coast margin
Prudhoe Bay Extensional Sadlerochit Perm-Tr SH 1525 732
Alaska" margin Formation
Endicott Extensional Kekiktuk Miss. SH 56 22
Alaska" margin
Cook Inlet Forearc Kenai Tertiary SH,CB 238 99
Alaska basin Group
HOHeavyoil
a Reserve values based on one field.
Styles of Fluvial Reservoir 489
Table 14.4. Examples of petroleum production from fluvial sandstones in Europe, Africa, Asia, Australia, and South
America
Location Tectonic Stratigraphy Age Reser- Reserves Reserves
setting voir oil gas
type (m 3x10 6 ) (mlx 10")
N. Viking Rift Statfjord, Triassic- SH-CB 2170
Graben basin Brent Jurassic
Sirte B Extensional Sarir Cretaceous SH 240
Libya' margin Sandstone
Oued Mya B Extensional unnamed Cambrian SH 397
Algeria' margin
Songliao B Rift basin various Cretaceous CB 1460
China (backarc?)
Sundaland Collisional various Tertiary SH? 3813 1943
S. E. Asia rifts
Gippsland B Rift Latrobe Tertiary CB 665 234
Australia basin Group
Eromanga B Retroarc Hutton Ss Jurasic SH 40
Australia foreland
Cooper B Intracratonic Gidgealpa Gp Permian SH? 87
Australia rift
Reconcavo B rift Sergi U. Jurassic SH 302 28
Brazil basin Formation
B Basin; HO heavy oil
, Reserve values based on one field.
The Main Consolidated field of the Illinois Basin is of lected examples of producing units are listed in
similar age and origin (Howard and Whitaker 1990). Tables 14.3 and 14.4.
An intracratonic Gondwana basin located in what The primary criterion used in this classification
is now Oman contains the Permian Gharif Forma- is the type of depositional system - clastic wedge
tion, a fluvial coastal-plain deposit up to 2.5 km thick versus incised paleovalley. The second criterion is
that contains more than 30 producing fields (Focke reservoir geometry. Incised-valley fills are ribbon
and van Popta 1989). bodies, but they may be broad in scale; clastic wedges
The Cooper Basin, Australia's major onshore consist of either sheet bodies or multiple lenses and
petroleum producer, is an intracratonic basin with ribbons enclosed in impermeable fine-grained sedi-
several largely nonmarine reservoir units of Per- ment.
mian-Triassic age that have yielded significant gas Weber and van Geuns (1990) classified reservoir
production (Cosgrove 1987; Heath 1989). heterogeneity into three basic types, all of which are
represented by fluvial deposits. Some braided sys-
tems and sheet-flood deposits are examples of their
14.4 Styles of Fluvial Reservoir "layercake" reservoirs, which are laterally extensive
and can be treated as tabular units for mapping and
modeling purposes. "Jigsaw-puzzle" reservoirs con-
The stratigraphic criteria discussed in this chapter sist of separate sand bodies such as channel sands,
can be grouped to define three major types of that fit together without major gaps but with occa-
nonmarine reservoir, as shown in Fig. 14.5 and Table sionallow-permeability barriers and baffles. Many
14.2. Each of the three types of reservoir occurs in braided and coarse-grained meander-belt fluvial
•
more than one tectonic setting; therefore, this does systems are of this type. "Labyrinth" reservoirs con-
not form part of the classification, although some sist of complex arrangements of lenses and pods
typical tectonic settings are listed in Table 14.2. Se- which are poorly interconnected and difficult to
map. Fine-grained meandering and anastomosing or it may contain an estuarine component {e.g.,
systems are commonly of this type. Wood and Hopkins 1989, 1992; Howard and
Whitaker 1990}. Internally, the reservoir may be
homogeneous, or it may contain important facies
14.4.1 Paleovalley Bodies (PV Type) variations, such as the channel-and-bar geometries
described below under the heading of the CB type of
Paleovalley bodies are distinguished by the ribbon or reservoir. Some traps are therefore combination PV-
shoestring shape of the reservoir and by their asso- CB types. Examples include some of the Mannville
ciation with regional unconformities. The reservoirs fields of Alberta (Little Bow field, Figs. 14.6, 14.7;
are typically tens of kilometers in length, up to a few Sect. 15.2.1), Cut Bank field of Montana (Shelton
kilometers in width, and several tens of meters in 1967; Sect. 15.2.2), and Blackwell field, Texas
thickness (Sect. 15.2). In Ordos basin incised valleys (Bloomer 1977).
up to 300 m deep have been mapped (Moore et al. PV -type fields can occur only where the paleo-
1986; Song Guochu 1988). Paleovalleys define the valley fill is incised into impermeable strata, and
former courses of incised meander belts and may be they also may require a top seal and a structural
virtually straight, or more or less sinous (e.g., Harms component to generate a trap. For example, the
1966; Weimer 1986). The fIll may be entirely fluvial, paleovalley may be oriented parallel to strike, and
•
•
• .••
• • •
•
•
Fig. 14.6. Isopach map (contours in meters) of valley-
fill sandstone, Little Bow Field, Mannville Group,
1km Alberta. Cross section A-A' is shown in Fig. 14.7. (Wood
contours in m.
and Hopkins 1989)
A AI
++.--=.::....:..:;=-. + +
G 0 G 0
~+I~~---fi~~ Coa3------~ftt·~----11~~--------tH~~--~~~
=Y-~~~==~+~=c~1~c~oa~I~2~~~~~~~~========~~~~~!*F~
Mississippian
m
G = gamma ray log o= density log
Fig. 14.7. Stratigraphic cross sections across the valley fill The location of the cross section is given in Fig. 14.6.
at Little Bow field. Note the presence of three separate (Wood and Hopkins 1989)
pools in the valley fill, separated by shale-fIlled channels.
the trap occurs where bends in the paleovalley in- duction is located in a fluvial-deltaic complex that
clude a convexity facing updip. South Ceres field, overlies the valley fill. Incised valley complexes are
Oklahoma (Lyons and Dobrin 1972; Sect. 15.2.4), an important component of the Wilcox Group in
Recluse field, in Powder River Basin (Woncik 1972), Texas, but they are mainly mud filled (Devine and
and Midland field, Kentucky (Reynolds and Vincent Wheeler 1989). There is believed to be minor fluvial
1972) are good examples. Another type of trap is sandstone reservoir potential in the base of some of
where the paleovalley is folded over an anticline. the channels, but production from them is not
Glenrock field, Wyoming (Curry and Curry 1972), known to this writer. Similarly, the Latrobe Group in
and Lane field, Colorado (Harms 1966), are ex- the Gippsland Basin contains incised valley fills
amples. Cheyenne Valley field, Oklahoma (Withrow that are largely mud dominated, but some isolated
1968), occurs where a paleovalley is oriented oblique channel-sand deposits are present which may have
to regional dip but porosity decreases updip, provid- future potential as exploration targets, given im-
ing a trapping mechanism. provements in seismic methods (Mebberson 1989;
A subtle paleogeomorphic trapping mechanism Rahmanian et al. 1990). Exceptions to this generali-
may be created by the erosional topography over a zation include the Jiyang Depression of Bohai Bay
bar deposit. Such is the case in the Bellshill Lake and basin, China, and Mississippian-Pennsylvanian
Hughenden fields, Alberta (Martin 1966; Conybeare fields of the u.S. Midcontinent. The former is an
1976), in which the actual trap occurs where pene- extensional backarc basin, in which sandstone
contemporaneous fluvial erosion sculpted the top of pinchouts within erosional valleys constitute one
the sandstone body into a series of terraces. of four types of "lithologic" trap, providing about 3%
Nearly all PV-type fields documented to date of total oil reserves (Shaui Defu 1988). The u.S.
occur in foreland-basin settings, including the those Midcontinent fields include the Morrow Sandstone
in the Cretaceous of the Western Interior Basin trend of Colorado and Kansas (Bowen et al. 1993)
(Mannville fields, Alberta; "J" and "D" Sandstone, and Main Consolidated field in the Illinois Basin
Denver Basin; South Glenrock and Recluse fields, (Howard and Whitaker 1990).
Powder River Basin, Wyoming), Arkoma Basin
(South Ceres and Cheyenne Valley fields), and those
in Ordos Basin, China. Paleovalleys are, of course, 14.4.2 Sheet Bodies (SH Type)
common in other structural settings, but most do
not appear to be associated with significant petro- Sheet sandstones reflect high source-area relief and
leum trapped in fluvial reservoirs. For example, a steep paleoslopes, with the development in most
major unconformity with incised valleys underlies cases of a broad braidplain. Such conditions do not
the Tuscaloosa Sandstone of Mississippi, in the ex- occur in any particular tectonic setting, although
tensional-margin setting of the Gulf Coast (Hogg they might be expected to be rare within cratonic
1988; Garrison and Chancellor 1991), but fluvial pro- areas, except in association with an active local
Fig. 14.8. Isopermeability map of the

(-, . . §J~§~
.. braided channel complex in the
, , ="i ' Sadlerochit Formation, Alaska North
,, . . , ...
~
,'" . . 1\........
,~-
Slope, in millidarcies. The Prudhoe Bay
\&I,.~~~"
· ..\ ..I,.,
field is outlined by the dashed line. The
wavy line indicates the faulted margin of
~:: : <
>
• . , the field. High permeability values are
yielded by the coarse, texturally more ma-
I · .. . .
· · ... .. . . .
. · ....... ..... ..
· · .. .. . ... . .
ture deposits formed in proximal reaches
of the complex. The crudely lobate shape of
the area of high permeability corresponds
. . .. ... . . . . to the distribution of the sand sheet, in the
Permeability in millidarcies
o 10km form of a large, fan-shaped braided delta
complex. (Wadman et al. 1979)
structure. The large fields of the North Viking Gra- Tyler 1991; Ashton 1992; Flint and Bryant 1993; Mar-
ben (e.g., Kirk 1980; Parsley 1984; Sect. 15.3.3) and tin 1993), the details of which are beyond the scope of
the North Slope of Alaska (Morgridge and Smith this chapter (some aspects are discussed in Sect.
1972; Wadman et al. 1979; Bloch et al. 1990; Fig. 14.8; 10.4.3). Similarly, developments in the use of 3-D
Sect. 15.3.1) include good examples ofSH-type reser- seismic mapping for channel definition are proceed-
voirs, as does Messla field, Libya (Clifford et al. 1980; ing at a rapid pace (e.g., A.R. Brown 1991; D. Brown
Sect. 15.3.2). All these are located in extensional- 1993; Weber 1993). Techniques of "surveillance geol-
margin tectonic settings, the North Viking Graben ogy", including pressure tests and fluid-flow moni-
being a major rift structure. The Hutton Sandstone toring, may prove to be extremely powerful tools for
of Eromanga Basin, Australia, forms the reservoir in subsurface mapping as field development proceeds
many structurally trapped SH-type fields. The basin (Sect. 9.5.9).
is a foreland basin, but the sandstone was derived
from structural upwarps on the cratonic side of
the basin (Moore et al. 1986; Watts 1987; John and 14.4.3 Channel-and-Bar Bodies (CB Type)
Almond 1987; Wecker 1989).
These fields require a structural trapping mecha- CB-type fields are characterized by the small size of
nism because, by definition, the reservoir consists of the individual reservoir body, although in some
a sheetlike porous unit providing little or no con- cases there may be hundreds to thousands of indi-
straints on fluid movement. Most of the cases re- vidual reservoir units which, if all filled with petro-
ferred to here consist of large faulted anticlines leum (as they are in some cases), can add up to
in which the reservoir sandstone is overlain by an significant, even supergiant accumulations (Figs.
impermeable shale. The Hassi Messaoud field of 14.9, 14.10). Development of these fields may be
Algeria produces from a breached anticline, in which
the reservoir sandstone has been eroded from the R 68 W
crest of the anticline (Balducchi and Pommier 1970). •
A different type of structural SH-type field is
exemplified by Red Earth field, Alberta, which pro-
duces from the basal Paleozoic Granite Wash sand-
stone. The reservoir is a sheetlike unit, and the trap is
provided by drape over paleotopographic highs, re- T
flecting the original erosional topography that was 35 49
covered by Granite Wash sedimentation (Hunter N
1966).
The detailed geological and engineering informa-
tion required for production modeling may result in Channel
the definition of considerable internal heterogeneity Fill T
within reservoir bodies that, at an early exploration 4 2 48
stage, might appear fairly uniform. Fields that are Oil-water N
originally classified as SH type might later be better contact
considered CB type, at least in terms of the style of 3000ft
......
.,
research required to further develop them. The con- t---~-'--
siderable amount of production modeling carried 9

out on the braided "sheet" of the Sadlerochit For-
mation at Prudhoe Bay field (Wadman et al. 1979;
Geehan et al. 1986; Atkinson et al. 1990), the Contours in teet
Kekiktuk Formation of Endicott field (Woidneck et
al. 1987; Melvin 1993), and the Statfjord Formation Fig. 14.9. Isopach map (contours in feet) of the Coyote
of Statfjord field (MacDonald and Halland 1993; Creek field, Wyoming. Regional dip is toward the west. Oil
Martin 1993) are good examples. Recent studies, is trapped against the updip flank of the point bar, which is
particularly of North Sea fields, have stimulated a bordered by the impermeable shale of the abandoned
channel. Detailed wireline-log cross sections show that the
considerable body of work in the area of computer point-bar lens is probably a composite of more than one
modeling of fluvial systems and comparisons with bar, with the mapped channel representing the final chan-
outcrop analogs (e.g., Buller et al. 1990; Miall and nel position at the time of abandonment. (Berg 1968)
® a detailed knowledge of depositional style. For

example, many small fields occur within point-bar
lenses, particularly where the point bar is flanked on
012 its updip side by the shale fill of the channel, formed
LLJ when the channel was abandoned and point-bar
km sedimentation ceased. In such cases, subsurface pre-
diction for field extension and development may
require recognition of the scale and orientation of
the channel and its associated bars. Good examples
are Miller and Coyote Creek fields, Wyoming (Berg
1968, Fig. 10), Mallalieu and Little Creek fields, Mis-
sissippi (Busch 1974; Berg 1986; Werren et al. 1990;
S <1.2m Sect. 15.4.1), Berwick field, Mississippi (Garrison
1.2 - 2.4 m and Chancellor 1991), Peoria field, Colorado
_ 2.4- 3.6m (Chapin and Mayer 1991), and Eastburn field, Mis-
souri (Ebanks and Weber 1982). In Citronelle (Fig.
14.10) and Daqing fields, where multiple reservoir
horizons occur, several types of reservoir geometry
are present, indicating subtle local variations in flu-
vial style.
The classic shoestring or ribbon sandstones of the
Cherokee Group (Bartlesville Sandstone) of Kansas,
Oklahoma and Missouri are a different category of
CB-type fluvial oil field. Here the reservoir body is
lenticular rather than lens-shaped (e.g., Charles
1941). These bodies may represent the fill of rela-
tively straight channels, as suggested by the regional
Fig. 14.10. Isopach maps of typical sandstone bodies, isopach mapping of Visher et al. (1971). Such chan-
Citronelle field, Alabama. Contours in meters. Vertical nels are common in some delta-plain environments
channel interconnectedness has created continuous fluid-
(e.g., modern Mississippi delta), as is interpreted to
flow paths. Very dense well spacing is required to map
such narrow ribbon sandstone bodies. Dots indicate wells. be the case for the Cherokee Group. However, some
(Eaves 1976) of the ribbons may in fact consist of laterally coa-
lesced bar lenses. The very detailed drilling carried
extremely difficult. They constitute examples of out in the area of the Eastburn field, Missouri, sug-
what Weber and Van Geuns (1990) termed the "laby- gests that this is the case there (Ebanks and Weber
rinth" and "jigsaw" types of reservoir, reflecting the 1982).
tortuous flow paths that the stratigraphy imposes on Seismic-stratigraphic mapping is becoming in-
the fluids. Examples include Citronelle field, Texas creasingly important in the delineation of small
(Eaves 1976), and Daqing field, China (Jin et al. 1985; structural-stratigraphic traps in many CB-type flu-
Yinan et al. 1987; Sect. 15.4.2). Examples where vial oil plays. This technique has formed a key to the
significant accumulations occur in numerous small location of most of the Tuscaloosa fields in Missis-
to medium-sized fields include those of the Chero- sippi (e.g., Hersch 1987; Garrison and Chancellor
kee Group of Kansas-Oklahoma and the Lower 1991) and is becoming of increasing importance in
Tuscaloosa sandstone of Mississippi-Lousiana (Sect. the Gippsland, Surat, Cooper, and other Australian
15.4.1), references to which are given above. Traps basins, which are beginning to enter the mature
that are entirely stratigraphic in character are typical exploration phase (e.g., Clark and Thomas 1988;
of all these examples, although initial discoveries in Stanmore and Johnstone 1988; Elliott 1989).
the Cherokee and Tuscaloosa were all in domal An important aspect of production modeling
structures, and most fields occur in areas of gentle may be to ascertain the degree of interconnected-
regional dip, which affects migration routes and trap ness of the reservoir units, and the definition of
configuration. separate flow units within the field, as discussed in
Prospecting for and definition of this type of trap, Sect. 10.4.
and the development of production models, require
14.5 Conclusions geostatistical techniques for predicting their scale

are essential. Developments along the u.S. Gulf
Coast, in the North American Western Interior ba-
The potential for future discoveries of SH - and CB- sin, and in inland Australia are at this stage.
type fields in clastic wedges would seem to be great- Much of the best work now being carried out on
est in the collision-related basins of southeast Asia, the integration of geological and engineering skills
where clastic terrigenous sediment supply during for enhanced production purposes is being applied
the Cenozoic was very high and fluvial conditions to fluvial oil fields, such as those in the North Sea
were widespread. Some giant discoveries may be basin, where the expense of offshore operations re-
expected. quires the use of the most sophisticated prediction
Considerable potential remains for discovery of and modeling tools. Such work is bringing petro-
smaller fields in mature areas, where detailed sub- leum geologists back to the field to collect ever more
surface information provides the basis for cali- detailed data on the characteristics of outrop ana-
bration of seismic-reflection data. Seismic mapping logs, and is providing scope for specialized numeri-
of channels, sedimentological techniques for inter- cal modeling of sand body geometries.
preting fluvial channel-and-bar geometries, and
Chapter 15
Case Studies of Oil and Gas Fields in Fluvial Reservoirs
15.1 Introduction nal, regional unconformities. The internal stratigra-

phy of the unit is very complex, because one or more
of the members have locally been removed by intra-
Selected examples of petroleum plays in fluvial sand-
Mannville erosion. Recognition and correlation of
stones are summarized in Tables 14.3, 14.4. Some of
the members is difficult, because they are present
the major oil fields from these plays are summarized
only in the subsurface, and detailed stratigraphic
in Tables 15.1 and 15.2. A large number of additional
work must rely on precise lithostratigraphic correla-
fields are described briefly in the literature cited in
tion of closely spaced wireline logs and petrographic
Chap. 14 (particularly in the SE Asia and Australia
analysis of the sandstones. The unit and its com-
areas), but reservoir and production data are inad-
ponent formations/members have been assigned a
equate for complete tabulation. Gas fields have not
large number of alternative local names, reflecting
been tabulated for this book. The purpose of this
the fact that much of the outcrop and subsurface
chapter is to provide a suite of documented case
stratigraphic work represents local studies carried
studies of selected examples of these fields, in order
out in isolation from one another. A thorough re-
to illustrate the three main types of reservoir con-
gional reevaluation of the Mannville incorporating
figuration identified in Chap. 14.
concepts of sequence stratigraphy still needs to be
carried out [the most recent synthesis is that by
Hayes et al. (1994)].
15.2 Paleovalley Fields (PV Type) The Alberta Basin generally is characterized by a
gentle dip to the southwest, toward the fold-thrust
15.2.1 Little Bow Area, Alberta belt of the Rocky Mountains. In the southeast corner
of the province is the buried Sweetgrass Arch; re-
The Mannville Sandstone (Lower Cretaceous) is a gionally the beds dip to the north and northwest
fluvial, lacustrine, coastal deltaic and barrier, and away from this subtle feature. These dips impose an
open marine unit that extends throughout central overriding structural control on all stratigraphic
and southern Alberta. This play is located within the traps, in that it is necessary for there to be an updip
Alberta Basin, a large foreland basin that extends seal on the east side of each trap. In most cases in the
throughout Western Canada and was formed by Mannville Group, this is provided by the juxtaposi-
terrane collision and suturing activity between the tion of porous Mannville strata filling a paleovalley
Jurassic and the mid-Cenozoic (Sect. 11.4.6). The against impermeable beds of the host rocks (gener-
Mannville and underlying Jurassic deposits contain ally older Mannville beds).
many small oil fields trapped in fluvial and deltaic In southern Alberta (Fig. 15.1) the Glauconitic
channel sandstones in central and southern Alberta. Sandstone of the Upper Mannville Group contains
Many are formed in paleovalleys. Wood and numerous sandstone ribbons that are interpreted as
Hopkins (1992) identified eight levels of valley cut- valley fills (Hopkins et al. 1982; Hopkins 1985).
and-fill in the Glauconitic Sandstone Member of These rocks exemplify a problem common in the
the Upper Mannville. The Mannville Sandstone is in subsurface analysis of fluvial channels associated
part correlative with the McMurray Formation with marginal-marine rocks: Are the channels con-
(Athabasca Oil Sands) of northern Alberta, the temporaneous with the host strata, or are they
Cutbank Sandstone of Montana, and the "J" and "D" younger deposits formed following a period of base-
sandstones of the Denver Basin, all of which contain level lowering and incision? The presence of a deeply
similar PV -type fields. The group consists of several scoured contact between the channel and the host
formations and members and contains several inter- rocks does not demonstrate that the channels are
Table 15.1. Summary of oil fields in North America
Name Location Basin name Tectonic Basin Trap Reservoir Age Typeb Area Ultimate References
setting code" type Recovery
(km 2 ) (m 3xl06 )
Bantry Alberta Alberta Foreland 4.3 Strat Mannville Cretaceous PV 41 10.5 Farshori and Hopkins (1989)
Bellshill Lk Alberta Alberta Foreland 4.3 Strat Mannville Cretaceous PV-CB 15 12.4 Martin (1966); Conybeare (1976)
Medicine River Alberta Alberta Foreland 4.3 Strat J2 SS Jurassic PV 7 1.3 Hopkins (1981)
Little Bow Alberta Alberta Foreland 4.3 Strat Mannville Cretaceous PV-CB 10 2.6 Hopkins et al. (1982); Wood and
Hopkins (1989)
Peco Alberta Alberta Foreland 4.3 Struct/strat Belly River Cretaceous SH 19 0.5 Gardiner et al. (1990)
Cut Bank Montana Rocky Mt. Foreland 4.3 Strat Kootenai Cretaceous PV-CB 930 4.8 Shelton (1967)
S. Glenrock Wyoming Powder River Foreland 4.3 Struct/strat L. Muddy Cretaceous PV 36 11.9' Curry and Curry (1972)
Recluse Wyoming Powder River Foreland 4.3 Strat Muddy Cretaceous PV 36 10.0 Forgotson and Stark (1972)
Coyote Creek Wyoming Powder River Foreland 4.3 Strat Fall R. Cretaceous CB 8 3.0 Stapp (1967); Berg (1968)
Lane Colorado Denver Foreland 4.3 Strat ''1'' Ss Cretaceous PV 25 Harms (1966)
Zenith Colorado Denver Foreland 4.3 Strat "D" Ss Cretaceous PV 9 Sonnenberg (1987)
Red Wash Utah Uinta Foreland 4.3 Struct/strat Green R. Eocene CB 185 9.0 Chatfield (1972); Pitman et al.
(1982)
Altamont-Bluebell Utah Uinta Foreland 4.3 Struct/strat Green R. Eocene CB 907 39.7 Lucas and Drexler (1976)
McDonald Draw Wyoming Green River Foreland 4.3 Struct/strat Hoback Paleocene CB 4 1.1 McDonald (1976)
South Belridge California San Joaquin Transform fault 3.1 Struct Tulare Pleistocene CB-SH 50 190.7 Miller et al. (1990)
Citronelle Alabama Gulf Coast Ex. margin 1.2.2 Struct/strat Donovan Cretaceous CB 66 19.8 Eaves (1976)
Mallalieu Mississippi Gulf Coast Ex. margin 1.2.2 Struct/strat Tuscaloosa Cretaceous CB 12 > 5.6 Berg (1986)
Little Creek Mississippi Gulf Coast Ex. margin 1.2.2 Struct/strat Tuscaloosa Cretaceous CB 47 11.0 Busch (1974); Werren et al. (1990)
Berwick Mississippi Gulf Coast Ex. margin 1.2.2 Strat Tuscaloosa Cretaceous CB -5 0.6 Garrison and Chancellor (1991)
Cushing Oklahoma Arkoma Peri ph. frland. 4.1 Anticline Bartlesville Penn. CB 72 > 74.8' Weirich (1929); Visher et al. (1971)
Bush City Kansas Arkoma Periph. frland. 4.1 Strat Squirrel Penn. CB 8 1.0 Charles (1941)
Cheyenne Valley Oklahoma Arkoma Periph. frland. 4.1 Strat Red Fork Penn. PV 9 1.0 Withrow (1968)
South Ceres Oklahoma Arkoma Periph. frland. 4.1 Strat Red Fork Penn. PV 80 1.6 Lyons and Dobrin (1972)
Blackwell Texas Midland Ex. margin 1.2.2 Struct/strat Cook Permian PV-CB 1 > 0.25 Bloomer (1977)
Big Wall Montana Ex. margin 1.2.2 Struct/strat Tyler Penn. PV 10 0.9 Kranzler (1966)
Rocky Ridge North Dakota Williston Cratonic 5 Struct/strat Tyler Penn. PV-CB 21 0.5 Hastings (1990)
Main Consolidated lllinois Illinois Cratonic 5 Struct/strat Caseyville Penn. PV 3 0.2 Howard and Whitaker (1990)
Clifford -Siaana Colorado Oklahoma Cratonic 5 Strat Morrow Miss PV 80 9.5 Bowen et al. (1993)
Red Earth Alberta Alberta Ex. margin 1.2.2 Struct Granite Wsh Devonian SH 29 3.5 Hunter (1966); Conybeare (1976)
Endicott N. Slope, North Slope Ex. margin 1.2.2 Anticline Kekiktuk Miss SH 3580 56.0 Melvin 1987; Woidneck et al.
Alaska (1987)
Prudhoe Bay N. Slope, North Slope Ex. margin 1.2.2 Anticline Sadlerochit Perm-Tr SH 506 1700.0' Morgridgeand Smith (1972);
Alaska Eckelmann et al. (1975)
Atkinson et al. (1990)
Swanson River Cook Inlet, Cook Inlet Forearc 2.2 Anticline Kenai Tertiary SH 40.0 Kirschner and Lyons (1973)
Alaska
Middle Grnd. Cook Inlet, Cook Inlet Forearc 2.2 Anticline Kenai Tertiary CB 16 > 13.8 Kirschner and Lyons (1973);
Shoal Alaska Boss et al. (1976)
a Basin type, from Table 11.2.
b Trap type, from Table 14.2: PV, paleovalley; CB, channel-bar form; SH, sheet.
, Data for all pools, fluvial portion not separated; >=only cumulative production data are available. Some fields have significant gas reserves; these are not listed in this table.
Table 15.2. Summary of fluvial oil fields in the remainder of the world
Name Location Basin name Tectonic Basin Trap Reservoir Age Type b Area Ultimate References
setting code' type Recovery
(km 2 ) (m3 x 106 )
Statfjord North Sea North Viking Graben Rift 1.1 Anticline Statfjord Tr-Jur SH 36 120.0 Kirk (1980)
Ninian North Sea North Viking Graben Rift 1.1 Anticline Brent Jurassic SH 580 190.0' Albright et al. (1980)
Brent North Sea North Viking Graben Rift 1.1 Anticline Statfjord+ Tr-Jur SH? 52 318.0 Bowen (1975)
Brent
Snorre North Sea North Viking Graben Rift 1.1 Struct Statfjord Tr-Jur CB 30 115.0 Stanley et al. (1990);
Martin (1993)
Beatrice North Sea Moray Firth Rift 1.1 Anticline Unnamed Jurassic cm 17 26.0 Linsley et al. (1980)
Wytch Farm S. England W. Approaches Trough Rift 1.1 Struct Sherwood Ss Triassic CB 34 55.6 Dranfield et al. (1987);
Martin (1993)
Messla Libya Sirte Rift 1.1 Struct Sarir Cretaceous SH 200 240.0 Clifford et al. (1980)
Sarir C-Main Libya Sirte Rift 1.1 Struct Sarir Cretaceous SH 800 1271.3 Sanford (1970); Martin (1993)
Hassi Messaoud Algeria OuedMya Craton 5 Anticline Unnamed Cambrian SH l300 397.0 Balducchi and Pommier (1970)
Daqing China Songliao Rift 1.1 or Struct/strat Various Cretaceous CB 2800 1460.0 Tang (1982); Jin et al. (1985);
2.3? Yang (1985)
Halibut Australia Gippsland Rift 1.1 Anticline Latrobe Eocene CB 27 >4.0 Franklin and Clifton (1971);
Rahmanian et al. (1990)
Jackson Australia Eromanga Foreland 2.4 Struct Hutton Jurassic SH 21 14.3 Wecker (1989)
Agua Grande Brazil Reconcavo Rift 1.1 Anticline Bahia Jur-Cret SH 80 41.0 Ghignone and Andrade (1970)
Aracas Brazil Reconcavo Rift 1.1 Sergi Jurassic SH? 8 18.0 Nascimento et al. (1982);
Martin (1993)
Vacas Muertas Argentina Cuyo Foreland 2.4 Struct Barrancas Cretaceous SH -84 25.0 Simlote et al. (1985); Martin (1993)
, Basin type, from Table 11.2.
b Trap type, from Table 14.2: PV, paleovalley; CB, channel-bar form; SR, sheet.
, Data for all pools, fluvial portion not separated; >=only cumulative production data are available. Some fields have significant gas reserves; these are not listed in this table.
498 Case Studies of Oil and Gas Fields in Fluvial Reservoirs
a b
R18 R17W4 R19 R18 R17W4
N N
t T16 t T16
T14 T14
T13 T13
T12 T12
T11 --..""·,··.··..··. 1
T11
Fig. 15.1. Channels in the Glauconitic Sandstone, Upper Mannville Group, southern Alberta. (Wood and Hopkins 1992,
100/09-26
~"..,..,.,.,.nrtr'
0.11
g
.!!
w 0.8
~
~
1.0
200 m
0.7
r« l~ r/t't', I t't'. 'I&. ('(\ ',.I, l"c<'
~
¥
-
MNVL
I'>' .'j ~') :'"c r..< ~~ ~
!,!! 'Pr!- ~ :J)
;y>'f?,: Kf"
;;; 0.8
"f.
! ~~ MSSP
0.9
~.)m) i) ~)'O: ~lr{1 ')J[
'W,'(t'i
\)~ I)S)~~<
O( If ~?~ I/, ~?, ~,,; ;5c ~~ I)?~~l(
SANDSTONE EfbDY SANDSTONE BODY EDGE OF
OF U POOL OF W POOL VALLEY FLL
Fig. 15.2. North-south seismic section across Little Bow field. Location is close to that of the section in Fig. 14.7. (Wood
and Hopkins 1989, reprinted by permission)
Paleovalley Fields (PV Type) 499
younger, because fluvial channels are by their very filling. Wood and Hopkins (1989) preferred the first
nature erosive. Modern delta distributaries are of these interpretations and pointed to the simulta-
known to scour for several meters below base level neous, juxtaposed development of sand bars and
into associated delta-front deposits. The scale of the mud banks in the Gironde Estuary. Figure 9.72 illus-
incised channel may be suggestive. Individual active trates fluid-flow paths through a cross section of
channels, such as delta distributaries, are typically a this field, inferred from data obtained during water-
few hundred meters wide, whereas paleovalleys may flooding of the field.
be significantly larger (a few kilometers wide) and
contain composite fills indicating several periods of
incision and infilling. A close comparison of facies
and petrography between the fluvial channels and
the host strata might suggest a close genetic associa-
tion (e.g., delta distributaries within an active delta)
but does not conclusively prove such a relationship.
Consideration of the succession in a regional se-
quence-stratigraphic context is probably the most
reliable method of interpretation.
Eight separate valley-fill levels have been identi-
fied in the Little Bow area (Fig. 15.1). Initially, these
were interpreted by Hopkins et al. (1982) as being
delta distributaries, but later work has indicated that
an estuarine interpretation is more realistic (Wood
and Hopkins 1989, 1992). Some of the valley fills
have been traced for more than 60 km. They are 2-2.5
km wide and up to 30 m deep (Fig. 15.1) They contain
three main facies: sandstone, sandstone-shale inter-
beds, and shale. The shale contains evidence of
marine trace fossils, and the sandstone includes
sand-mud couplets interpreted as tidal in origin.
Wood and Hopkins (1989) compared the facies and
distribution of the beds to those of the modern es-
tuarine sediments of the Gironde Estuary, France.
They suggested that the valley filled by fluvial and
subsequent estuarine sedimentation following a pe-
riod of base-level lowering and downcutting.
Details of one of the valley fills at Little Bow
(valley 2 in Fig. 15.1 b), at the bend in the channel, are
shown in the detailed map of Fig. 14.6, and Fig. 14.7
provides a stratigraphic cross section through the
Little Bow field. Three separate pools (G, U, and W)
have been delineated. Proof of their separation is
based on two lines of evidence: (a) The shale caprock
that separates the pool has been penetrated by cores
in several wells and has been shown to occupy the
entire stratigraphic interval of the valley fill. (b) Each
pool has different pressures and oil-gas contact el-
evations (Wood and Hopkins 1989). The pools may
also be separated on the basis of seismic evidence, as
shown in Fig. 15.2. The pools each represent major
Fig. 15.3. Isopach map of Cut Bank Sandstone in Cut Bank
sand banks that formed within the estuary. They field area. Producing area is outlined. Note sinuous ribbon
may have been contemporaneous, or they may pattern of thick (> 60 ft) sandstone distribution. (Shelton
represent separate periods of incision and valley 1967, reprinted by permission)
UNION PHILLIPS UNION Vl
KULLBERG SKELLY JUNIPER o
Tribal I-G Hawkin s No.4 Forme" 51. Bk. No.8 Kouthlen No. I "midon No. I "I.e-Stat ..... 1 o
NE NE 3-32N-6W SWSW 17-33N-5W C NE 17- 33N- 5W ISWSWNWI8- 34N-4W SE SE NW 5-33N-4W NE SW 16-34N- 4W
sw I NE
i
A
o
KOOTENAI
()
I»
/1)
'"
V>
8'
0..
iii'
'"o
......
o
:::..:
I»
::s
0..
C"l
I»
'"'TJ
iii'
0:
'"
S·
ir;s.
a
::0
/1)
/1)
'"
Fig. 15.4. Cross section through Cut Bank field. Line of section is shown on Fig. 15.3. (Shelton 1967, reprinted by permission) ~.
....
CI>
R 83 W R 82 W 15.2.2 Cut Bank Sandstone, Montana

~ ~ ~
~N "
The Cut Bank field is located in northernmost Mon-
tana, adjacent to the border with Alberta. The geo-
0 ~ logical setting is that of the Rocky Mountain
./
foreland basin, a southward extension of the Alberta
T
3 Basin. The Cut Bank field produces from meander-
S ing channels filling a broad paleovalley at the base of
the Kootenai Formation, which is the Montana
~
equivalent of the Lower Mannville Group of Alberta.
The producing beds are incised into Jurassic strata.
The field is located on the west flank of the
Sweetgrass Arch. Regional drainage was to the north.
A ISOPACH HUNTSMU C.I.= 10ft This paleovalley is considered to be a southward
extension of the major drainage systems that have
been identified in Alberta [the Spirit River Channel
R 83 W R 82 W
of McLean and Wall (1981); see Fig. 11.25].
-¢- ~ ~
The valley is more than 50 km wide (Blixt 1941;
IN " Conybeare 1976). The Cut Bank Sandstone is up to 70
m thick in outcrop, to the west of the field, but within
0 ~
the field itself it has a maximum thickness of 25 m.
T
The dominant lithology is medium-grained, cross-
3 bedded sandstone showing an upward decrease in
S
grain size. An isopach map of the sandstone in the
field area is shown in Fig. 15.3. Here the thickness
-¢- • FIELD /-¢- ~
-~-- contours suggest a pattern of sinous sandstone rib-
. bons up to 5 km wide. According to Conybeare

/
(0 • ./
~" ./"
(1976), individual channels are 450-1200 m wide,
~ IJ -¢-
but the evidence for this is not presented. Shelton
B ISOPACH D SUDSTOIIE C.I.= 10ft
(1967) and Conybeare (1976) interpreted these beds
as point-bar deposits within a broad meandering
R 83 W R 82 W
river system.
.
~ ~ ~ A cross section through the area confirms the
incision of the valley into Jurassic strata which pro-
1N " CROSS
vide the updip seal to the field (Fig. 15.4). Well logs
,,'
~ SECTION
through the Cut Bank Sandstone are serrated in

0 ~
character and do not contain picks correlatable from
T well to well. This line of evidence, coupled with the
3
S
overall thickness and areal extent of the valley fill,
.
suggests that the fluvial succession is a composite
FIELD /~ ~
one, consisting of multiple, multistory meander-belt
sandstones.
ISOPACH
c HUIITSMAII a D SAIIDSTOIIE C.I. 10ft
Fig. 15.5. Isopachs of the Huntsman Shale and "D" sand-

stone. The Huntsman is thin along the paleovalley trend, as
. LEGEND a result of valley incision. When combined, both units
............
, @ 0 Producer
show a linear trend of increased thickness, probably
"---- I., Z)
,
J Producer because these units fIll a paleostructural depression.
~ Dry Hole (Sonnenberg 1987, reprinted by permission)
B B'
• • •
NwNw Sec. 20. T3S. R82W SwSwSec.17. T3S.R82W NwNw Sec. 17. T3S. R62W SwNw Sec. 17. T3S. R82W
{)
... ...
-~- ... -~-
o
o -~-
GRANEROS o
c c·
• •...
SwNw Sec. 24. T3S. R83W NwNw Sec. 24, T3S. R83W SeSw Sec. 13. T3S. R83W
<)
--~-----------;=-~:-r--------------r--~----------
o
GRANEROS ...
-g-
Fig. 15.6. Stratigraphic cross sections across the Zenith field. Locations are given in Fig. 15.5. Note the depression of the
Huntsman beneath the "D" paleovalley. (Sonnenberg 1987, reprinted by permission)
15.2.3 Zenith Field, Colorado Dakota Sandstone, which is comparable in age, fa-
cies, and interpreted depositional environments to
The Zenith field is one of an important suite of fields the Mannville Group of Alberta (Sect. 15.2.1), and
producing from paleovalleys in the Denver Basin, a the Lower Muddy Sandstone of Wyoming. Accord-
basin formed by deformation of the Rocky Mountain ing to McCaslin (1983), the ultimate recovery for the
foreland basin during the Late Cretaceous-Cenozoic Denver Basin is expected to be 127 m 3 x 106, most of
Laramide orogeny (Weimer 1986). The major pro- which will be derived from stratigraphic traps in the
ducing trends in the Denver Basin are the valley- "J" and "D" sandstones.
fill deposits of the "J" and "D" Sandstones (Harms The valley fills incise a regional unconformity,
1966). These units are part of the Lower Cretaceous just as the producing sandstones of the Mannville do
ZENITH FIELD
FORMATIONS
TWO-WAY
I I
TIME (SEC)
0.3 NIOBRARA
CARLILE
GREENHORN
GRANEROS
0.4 _;t~~~~~~~~~;;~~~~~~~D SANDSTONE
i~~~1~~~~NE
Trac'e Interval =220 ft Sampling Rate = 1 msec
Wavelet = Ricker 60 Hz
B A
Fig. 15.7. Synthetic seismic section (above) and actual cross section (below) through the Zenith field. (Sonnenberg 1987,
(indeed, the comparisons between the "J" and "D" Mowry Shale, deposited during a regional transgres-
sandstones and the Mannville in terms of their pe- sion. The Mowry Shale is, in turn, overlain by the
troleum geology are quite striking). Weimer (1986) Huntsman Shale, and this is incised by paleovalleys,
showed that the age of the regional unconformity which are part of a regional unconformity developed
underlying the "J" sandstone is 97 Ma (late Albian; at the end of the Albian, at about 95 Ma (Weimer
Fig. 11.70). The valley fill cuts down into a regressive 1986; Sonnenberg 1987; Fig. 11.70). The "D" sand-
unit [II and J2 units of Harms (1966); Fort Collins stone infills these valleys, with a fluvial-estuarine
Member of Weimer (1986)] consisting of fine- succession similar to that of the "J" sandstone, and is
grained sandstones deposited in deltaic and shore- followed by another widespread shale unit.
line environments. The valley fill itself consists of The Denver Basin is an asymmetric trough. Im-
fluvial and estuarine strata, indicating increasing portant secondary structures provide a structural
marine influence upward through the section. these component to the traps in the "J" and "D" sand-
beds are themselves then overlain by the widespread stones. A series of broad, NE-SW trending arches
·-r- -I --- R lW
------ RIE
I '
I i
I ~
I ,
---L---------2---tw;'n---\;---'tt--;';--~~~__+_tt_+__\___l_\___\__1
._--1-__
I T
·~~~~~~~_+~~~~~,_~~4_-~+_~23
N
T
~~~--~~~-+~~-K-~~~~~~~~~~~~~~~~~~~~22
N
1'-'--.'----!.-!/,I"--.!.-'---.;.-----1k--~--+'.......-~-+-"!n_~I_'_-_+-____1 SOU T H CE RES POOL AREA

Nobl. Co" Oklahoma
Con/ours -Pink Lime
Col. - 20'
=0'.t. trOttI, 71
Fig. 15.8. Structure contour map, South Ceres Pool area. a result of structural drape. (Lyons and Dobrin 1972, re-
Contours are drawn on a marker overlying the Red Fork printed by permission)
sandstone. Note the "nosing" of contours over the pool, as
cross the north end of the Denver Basin. These devel- duces primarily from the "D" Sandstone. Isopach
oped in response to recurrent movement on buried maps of the (CD" sandstone and the underlying
Precambrian lineaments throughout the Cretaceous. Huntsman Shale are shown in Fig. 15.5, and these
Weimer (1986) grouped them into the broad struc- delineate an E-W trending paleovalley 518-790 m
tural feature termed the Transcontinental Arch. wide. The "D" Sandstone ranges from 3.6 to 20.4 m in
During the major late Albian regression, this arch thickness in the Zenith field area. Cross sections
became an exposed drainage divide, with valleys through the field show that the "D" sandstone is
incised across it, draining to the north and south. incised into the underlying Huntsman Shale, and
The Zenith field is located near the center of the that the base of the shale is also depressed beneath
Denver Basin, some 45 km east of Denver. It pro- the field (Fig. 15.6).
R IW RIE
LINE of ECT/ON_.-T V T • ••
A. ~.
CEJlfS[J
~: .: ~ : : :-
--'-.'--'I.-.'----'-.-'.'---'t.---+---------~--+_--l--
I-'-'L' SOUTH CERES POOL AREA
.,• " I, i Nobl. Co., Oklahoma
_, # • i Red Fork Sands/one
: ISOPACH
--..:,.--
C.I. - 20'
=cJ
I'.l.lYOII.71
Fig. 15.9. Isopach map of Red Fork Sandstone, South Ceres Pool, Oklahoma. (Lyons and Dobrin 1972, reprinted by
permission)
The "D" sandstone trend could be intepreted as a sandstone is that both units were deposited in a
deltaic distributary comparable in age to the beds subtle structural depression developed during the
into which it is incised, but Sonnenberg (1987) Laramide reactivation of basement fault blocks.
rejected this intepretation on several grounds. Sonnenberg (1987) demonstrated that "D" sand-
First, the contact between the valley fill and the host stone paleovalleys could probably be mapped using
strata is sharp. No gradational facies change can be seismic techniques. He developed a synthetic seis-
mapped. Second, the top surface of the valley-fill and mic model by determining reflection coefficients
adjacent strata is flat, whereas a prograding distribu- from a sonic well-log, and then building a seismic
tary of this size would be expected to develop pro- cross section using a paleovalley up to 15.8 m deep.
nounced levees. Sonnenberg's (1987) interpretation Figure 15.7 compares the synthetic section with an
of the depressed Huntsman underlying the "D" actual seismic cross section through the field. In the
Fig. 15.10. Structure contour map of Prudhoe Bay field. (Morgridge and Smith 1972)
actual section the "D" sandstone is 17.6 m and 4.8 m the sinuosity of the incised valley rather than the
thick in wells A and B, respectively. Note the slight sinuosity of individual channels.
velocity pull-up beneath the "D" sandstone, and the
presence of a listric fault that does not offset the "D"
sandstone.
1S.3 Sheet Reservoirs (SH Type)
15.2.4 South Ceres Field, Oklahoma 15.3.1 Prudhoe Bay Field, Alaska
South Ceres field is a classic example of a paleovalley Prudhoe Bay field and Endicott field are located on
field; it was discovered by seismic methods in the the North Slope of Alaska, adjacent to the Beaufort
1940s, a time when seismic acquisition and process- Sea. At present this area can be classified as a
ing techniques were very primitive by today's stan- foreland basin (Colville Trough) that developed in
dards. The field is located on the north flank of the response to Cretaceous tectonism in the Brooks
Anadarko Basin, a peripheral foreland basin that Range area, to the south. However, during the late
developed during the early and midle Paleozoic Paleozoic and early Mesozoic, when the reservoir
(Rascoe and Adler 1983). The Cherokee Group, of units were deposited, the tectonic environment of
Desmoinesian age, is a major producing unit depos- the region was quite different. The most widely ac-
ited on the north flank of this basin and the adjacent cepted hypothesis for the evolution of northern
Arkoma Basin in Missouri, Kansas, and Oklahoma. Alaska interprets the North Slope as a fragment of
It contains many small to giant fields, many discov- the continental margin of North America that was
ered in the early part of this century. The South Ceres formerly continguous with Arctic Canada, but was
field is one of a suite of fields in which the reservoir is rotated anticlockwise away from Canada during the
the Red Fork Sandstone, one of the component flu- Cretaceous (Embry 1990b). In this tectonic model
vial-deltaic units of the Cherokee. A structure con- the North Slope formed part of an extensional conti-
tour map and an isopach map of the producing nental margin that now faces south in the subsurface
interval are shown in Figs. 15.8 and 15.9. Figure 9.45 beneath Mesozoic cover.
illustrates a seismic cross section through the field. The dominant structure in the region of the
The giant U-shape of the sandstone isopach reflects Endicott and Prudhoe Bay fields is Barrow Arch, a
Sheet Reservoirs (SH Type) 507
SUBSEA
DEPTH LlTH- PERMEABILITY DEPOSITIONAL
(FT.) fB OLOGY POROSITY ENVIRONMENT
Transgressive
Marine
DepOSits
Stacked, Distal
Fluvial Channel
Deposits
Stacked. Mld·Braided Stream

Fluvial Channel Deposits
Mld·Dlstal Fluvial
Channel Deposits
Stacked.... Mld·Bralded Stream

Fluvial l,;hannel Deposits
Floodplaln Deposits
with Isolated
'Channel Units
Distal Fluvial I Delta Mouth·

bar DepoSits
Delta·Front Deposits
13048 ",1 ,02 2000

(6( % ) 2. 3.02 1LO 2000
6 16
-.. . ···· ·······LLS···· ..

"' . • • • • • • • • • • u .• •
. ••• • • _ •.•••••• ~' U """""" " "" "" '·"···I·--"''''''''.''''''''''''''' -t
110. RHOS 50 0.2 ILM 2000
10. CALI 110. , ~------01"----- ··_··-i
GR
Fig. 15.11. Typical suite oflogs through the Ivishak Mem- can be mapped throughout the field. G Conglomerate; gS
ber of the Sadlerochit Formation. The zonation of the pebbly sandstone; IG intraformational conglomerate; S
Ivishak Member is based on petrophysical properties and sandstone; M shale. (Atkinson et al. 1990)
buried structural high, trending WNW-ESE. This undergo subsidence beneath the sediments of
arch was an emergent landmass and a sediment Colville Trough.
source for the main reservoirs in the oil fields, in- The Prudhoe Bay structure is a large faulted anti-
cluding the Kekiktuk and Sadlerochit Formations. It cline (Fig. 15.10). There are several producing hori-
seems likely that Barrow Arch was an intracratonic zons in this field, including the Lisburne Group and
uplift that was active throughout much of the Paleo- the Sadlerochit Formation, but the most important
zoic, and that it functioned as part of a rift system is the latter. This unit is truncated beneath a Creta-
when the Canada Basin (Arctic Ocean) opened to the ceous shale on the flanks of Barrow Arch. The shale
north. Beginning in the Jurassic, it then began to acts as the seal. The northern limit of the field is
WELL WELL
J SANOY
BRAIDED
ft m FWVIAL
FLOODPLAIN .... DEPOSITS
100] 30
} PEBBLY
BRAIDED
FWVIAL
DEPOSITS
o 0
} DI~AL
FLUVIAL I
DELTA
DEPOSITS
2500 FT.
+NORTH
760 M.
SOUTH +"I • CONTINUOUS
SHALE
0 DISCONTI NUOUS
SHALE Fig. 15.12
M£SSLA FIELD
IJ£rAIL OF
SARIR SANDS
LIMITED BY
PERMEABILITY
OIly HOLE
OBSERVATION WELL
Fig. 15.13
HH8-65 HH3-65 HH21-65 HH12-65

2.22 ktn 3 krn 2krn
West ••• East
-2500 ~~~~~~~~~~~~~~~~~~~~~~~~T-2500
rn m
-2560 -2560
-2620 -2620
-2680 -2680
-2740 Index Map -2740

Messla Field
-2800 -2800
-2860
m datum =sea level
~l....L-__--'___ ..JT km
-2920
Fig. 15.14. Structural cross section through southern truncation by the Rakb Shale, and the position of the oill
part of Messla field, showiDg regional stratigraphy and water contact. (Clifford et al. 1980, reprinted by permis-
petrophysical log character of the Sarir Sandstone, its sion)
along a fault which drops the reservoir rock to the delta-plain complex. Figure 15.12 is a reconstructed
north. interpretation of the depositional environment and
The Sadlerochit Formation consists of a south- schematic architecture of zone 2 of the Ivishak
ward-prograding deltaic complex, in which the Member, based on log response and core data. In
main producing horizons constitute a thick braided terms of primary production, the reservoir units can
channel complex of the upper delta plain (Morgridge be regarded as homogeneous (typical SH-type ge-
and Smith 1972; Eckehnann et al. 1975; Fig. 15.11). ometries), but enhanced recovery work, including
Braided stream deposits are estimated to constitute surveillance geology, has detected a considerable
55% of the total rock volume of the Ivishak Member degree of internal heterogeneity of the reservoir,
of the Sadlerochit Formation. The lower portion of based particularly on the presence of shale beds of
the interval consists of repeated, stacked channel varying lateral persistence (Geehan et al. 1986;
sequences 3-6 m thick, each grading upward from Atkinson et al. 1990). This is typical of many braided
conglomerate to medium-grained sandstone. In the "sheet" reservoirs (Martin 1993).
upper part of the Ivishak Member conglomerates are
rare.
Very detailed reservoir studies have been under- 15.3.2 Messla Field, Libya
taken on the Sadlerochit Formation to permit
efficient enhanced recovery (Eckelmann et al. 1975; Messla oil field is located in the southeast part of
Wadman et al. 1979; Geehan et al. 1986; Atkinson et Sirte Basin, in central Libya. The Sirte Basin is a Late
al. 1990). A permeability map for the braided steam Mesozoic-Tertiary rift basin, resulting from exten-
complex [zone 4 of Wadman et al. (1979)] is shown sion of the north Africa craton (Clifford et al. 1980).
in Fig. 14.8. Note the variation from a high of more More than 7600 m of Cretaceous-Tertiary sediment
than 700 millidarcy in the north to less than 50 md in are present, most derived from land areas to the
the south. The contours are thought to outline, in a south.
crude sense, the distribution of the braided upper
Fig. 15.12. Interpreted depositional environment and generalized depositional architecture of zone 2 of the Ivishak
Member at Prudhoe Bay. (Atkinson et al. 1990; based on Geehan et al. 1986)
Fig. 15.13. Structure contour map, top Sarir Sandstone, Messla field, Libya. East-trending lows on this surface may be
the result of subaerial erosion preceding the Late Cretaceous transgression. (Clifford et al. 1980, reprinted by permission)
,' E
,' N
UTSIRA
~ OILFIELDS ' HIGH
fiO "N
GAS /CO
FIELDS
Fig. 15.15. Major fault trends and location of main oil fields in the north Viking Graben and East Shetland Basin area.
(Parsley 1984)
w CORMORANT HUTTON ST A TFJORD
6.7 x VERTICAL EXAGGERATION
E
A'
o 20 40 60 80 100
r=-=-=t 4' 2'

~ OLIGOCENE - PLEISTOCENE
PALEOCENE - EOCENE A ....._.}]I-=~_

61'
~
UPPER CnETACEOUS
tf.
SHETLAND IS • EAST
LOWER CP<;TACEOUS SHETlAND
...
60' PLATFORM
JURASSIC
ORKNEY IS .
~'"
"
TRIASSIC
59° ~.
'ilc
BASEMENT
.0' 4'
Fig, 15.16. Typical structural cross section through the north Viking Graben, showing position of Statfjord and other
fields, on tilted fault blocks. (Kirk 1980, reprinted by permission)
Fig. 15.17. Structure contour map, top of Statfjord Forma-

tion, Statfjord field, North Sea. (Kirk 1980, reprinted by
permission)
TRUNCATION OF
STATFJORD FM.
o, ,
3
Miles
C.l. = 200 Feet
Rift faulting began in the Early Cretaceous, as the Main Pay, is more than 150 m thick in the south,
evidenced by basin-fill sediments of that age, which but was truncated to the north and west by intra-
consist of poorly sorted nonmarine sandstones and Cretaceous erosion (Fig. 15.14). The Main Pay is a
siltstones. This succession is overlain by the Sarir lithologically homogeneous unit of fine- to medium-
Sandstone, a clean fluvial sandstone deposited by grained, cross-bedded sandstone, interpreted to
northward-flowing river systems. The beds were have been deposited predominantly in braided
tilted and partially eroded during an episode of stream environments. The underlying Red Shale is a
intracratonic tectonism in the mid-Cretaceous; then playa-lake or flood-basin deposit. Almost all the oil
the basin subsided once again, and a marine trans- production is from the Main Pay unit. Petrophysical
gression from Tethys, to the north, led to the deposi- properties and production characteristics indicate a
tion of evaporites, shales, and carbonates of Upper remarkable lack of heterogeneity in this reservoir.
Cretaceous to Miocene age. Tilting of the area The beds probably represent the constant channel-
occurred during later extensional tectonism in the ing and reworking of relatively well-sorted sands
Tertiary (Clifford et al. 1980). within an are ally extensive braided fluvial system, in
In the Messla field area the result of this basin which overbank or floodplain areas had minimal
evolution was the development of an east-dipping preservation potential.
wedge of Sarir Sandstone, overlain and truncated to
the west by Upper Cretaceous shales. In the field area
the beds dip northeast at less than 1°. Faulting within 15.3.3 Statfjord Field, North Sea
the field area is minimal, except at the southern end,
where the field.is truncated by extensional faults, The North Sea is an intracratonic basin which
with downthrows to the south (Fig. 15.l3). developed by a complex and lengthy series of exten-
The Messla Sandstone includes two sandstone sional movements beginning in the late Paleozoic
members, separated by an extensive tight unit (Ziegler 1988). Fluvial-deltaic sandstones of Upper
named the Red Shale. The upper sandstone, called Triassic to Middle Jurassic age comprise 1 of 11
B8 A38 A37 A33
Fig. 15.18. Conceptual model of the large-scale architec- north-south, parallel to assumed drainage direction.
ture of the Statfjord Formation, Statfjord field. Well spac- (MacDonald and Halland 1993, reprinted by permission)
ing is approximately 1 km. The section is oriented
major petroleum plays in this area (Parsley 1984). and eroded (Figs. 15.15, 15.16). Subsequent Late Ju-
Several giant oil fields contain significant reserves in rassic (Kimmeridgian) and Cretaceous transgres-
these fluvial sandstones, including Brent, Snorre, sions covered the eroded fault blocks with shales,
Ninian, and Statfjord in the north Viking Graben including the Kimmeridge Shale, a prolific source
and Beatrice in the Moray Firth Basin. rock. The shales also provide the seals for these north
The Triassic and Jurassic rocks of the North Sea Viking Graben oil fields.
were developed during a phase of repeated rift fault- Petroleum occurs in two main fluvial units, the
ing, related to the incipient opening of the Atlantic Statfjord Formation, the sandstones of which are
Ocean (Ziegler 1988). The Viking-Central Graben relatively widely distributed and characterized by
was the major structure to develop at this time (Fig. good porosity, and the Brent Group, where this unit
11.44). Regional subsidence that accompanied rift- includes sandstones of fluvial, rather than deltaic
ing activated regional drainages, leading to the de- origin. The Cormorant Formation typically is thin
velopment of major fluvial units, the Cormorant and and impersistent, and displays poor reservoir char-
Statfjord Formations (Rhaetic-Hettangian) (Fisher acteristics, although it has yielded some oil and gas
1984). With continued subsidence marine transgres- shows, notably in the Beryl field (Fisher 1984).
sion occurred, and the largely deltaic Brent Group Petroleum occurrences in Triassic-Jurassic strata
(Aalenian-Bajocian) was deposited (Brown 1984). of the northern North Sea are trapped in fault-
Faulting and folding occurred in mid- to Late Juras- bounded anticlinal structures (Fig. 15.17). Strati-
sic time, and the fault blocks were tilted, uplifted, graphic trapping mechanisms are secondary. The
Channel-and-Bar Reservoirs (CB Type) 513
R8E
28 25
T
5
N
12
c=J 0 - 10' (0 - 3.0 ml
h\N:! 10 - 20' (3.0 - 6.1 """--+16
~\\\\\\\\i 20 - 30' (6.1 - 9.1 m)
WW11 30 - 40' (9.1 - 12.2 m)

T
w'iMj > 40' (>12.2 m) 4
N
-'
o feet 3000
o-------
meters 900
Fig. 15.19, Isopach of the upper producing unit (the Q sandstone) at Little Creek field. The three main pay areas
constitute three separate point bars formed inside the meander bends of a highly sinuous river. (Werren et al. 1990)
major exploration method for these fields has been 15.4 Channel-and-Bar Reservoirs (CB Type)
seismic structural mapping. The reservoirs are sheet-
like in their overall geometry (Fig. 15.18), but consid- 15.4.1 Little Creek Field, Mississippi
erable modeling work has now been done on this and
nearby fields in attempts to provide detailed descrip- Little Creek field is a classic example of a field
tions of the heterogeneities within the channel sheets defined largely by the stratigraphic configuration of
(Stanleyet al. 1990; Nybniten et al.1990; H0imyr et al. a local depositional system. It was one of the first to
1993; MacDonald and Halland 1993; Martin 1993). be explored using facies concepts, in the late 1950s
This is particularly important for enhanced recovery (Werren et al. 1990), and has been widely figured in
of mature fields such as Statfjord, which is now being textbooks (e.g., Busch 1974).
produced by high-pressure miscible gas drive, be- The field is located within the Upper Cretaceous
cause of the potential for gas breakthroughs. Tuscaloosa Sandstone, one of numerous productive
SWEETWATER FIELD
,::::::.7::::::1:::::::: :'
co,
LITTLE CREEK FIELD 4-

4- 4-
DEPTH BELOW 0 MARKER
o- 5'
B ---9'~ 5 - 10'
20 - ,~ ._-..:
~ 10-15'
.. 15-20'
o SANDS
SHALED OUT
CON TOUR INTERVAL S'
o 5000'
I
o 1500m
Fig. 15.20. Isopach of fine-grained sandstone, siltstone, underlying a regional marker that defines the top of the
and mudstone, constituting the flil of the abandoned chan- fluvial interval. (Werren et ai. 1990)
nel overlying the Q sandstone at Little Creek field and
sandstone units comprising the Mesozoic-Cenozoic of the upper producing sandstone (Fig. 9.39 is an
clastic wedge of the Gulf Coast (Sect. 11.4.2.2). The older version of this map). The configuration of
field consists of point-bar lenses occupying the the top surface of this point bar was formed by
crestal region of a low-relief anticline. Floodplain abandonment, and the shape of this surface is shown
deposits provide lateral seals, and the point-bar by Figs. 15.20 and 15.21. These diagrams indicate a
sandstones are overlain by transgressive tidal depos- northeast-dipping point-bar surface prior to aban-
its that constitute the top seal. Two major sandstones donment along the line of section A-A', and a west-
are present, reaching maximum thicknesses of 16.8 ward-dipping surface in the opposite-facing point
and 9.1 m, respectively. Figure 15.19 is an isopach bar along the line of section B-B'.
A
SW~-+~--~~~-----*~--~r-----~~r-~--~~Tr,-NE
~_- LAST POSITION OF RIVER
B B'
WEST--~~.I--~~~~~~~~----~~--_7r<7, - EAST
, ""
:1;
Q'(OATUMl
'50"L o
m
!500m
o 15 00 '
Fig. 15.21. Cross sections through the Q sandstone and overlying beds at Little Creek field. The thickness of the unit
between the Q datum and the Q sandstone is shown in Fig. 15.20. (Werren et al. 1990)
15.4.2 Daqing Field, China the center of the basin. Reserves are now estimated at
1460 m 3 x 106 (Soeparjadi et al. 1986).
Daqing field is the largest in China, and one of the Production is from four nonmarine Lower Cre-
few giant fields in the world producing oil of non- taceous formations, the Quanton, Qingshankou,
marine origin (Yang 1985). It is located in Songliao Yaojia, and Nenjiang formations, in order from base
Basin, an extensional basin in northeast China (Fig. to top (Tang 1982). These units developed as a series
15.22), which first appeared as a small rift structure of fluvial-deltaic complexes that prograded into the
in the Jurassic and underwent active enlargement, basin from the north. The total reservoir thickness is
and faulting and folding, during the Cretaceous and approximately 3 km, but this includes thousands of
Cenozoic. A major period of subsidence occurred separate oil-producing bodies. When waterflooding
during the Early Cretaceous, when the main reser- was initiated, at an early stage of field development,
voir rocks were formed in a complex of fluvial, del- the field was divided into as many as 50 flow units to
taic, and lacustrine environments. A lake extended permit separate production and pressure mainte-
though the basin in the Late Cretaceous, and an nance (Jin et al. 1985). Individual producing sand-
episode of folding, leading to the development of the stones are less than 10 m thick.
Daqing structure, occurred before the end of the Eight different types of sand bodies have been
Cretaceous (Tang 1982; Liu 1986). distinguished (Jin et al. 1985). These include fluvial
The tectonic interpretation of the Songliao Basin meander-belt and various deltaic and shoreline
is uncertain. Klimetz (1983) constructed a model of facies. Meander-belt sandstones are the thickest
regional subduction and microplate collision in (4-10 m). They range from 2 to 4.5 km in width, and
which the Songliao Basin is interpreted as a backarc individual units have been traced for more than 3
rift. However, Burke and Sengor (1986) suggested km. Waterflooding experience has shown that the
that, rather than extension, there may have been meander-belt sandstones are typically "bottom-
convergence between Asia and a North China block, flooding"; that is, injected water advances rapidly
with the Songliao basin developing by lateral move- along the bottom of the zone, because this is where
ment along strike-slip faults, a process termed "es- the sandstones are the coarsest and best sorted, with
cape tectonics". The field is located within a complex the best permeabilities (Jin et al. 1985).
anticlinal structure, which extends NNE-SSW along
Fig. 15.22. Structural setting of Songliao Basin,

east China. The Songliao Basin is lettered I at the
north end of the map area. It is bounded by exten-
N sional faults oriented NNE-SSW, forming part of
the Cathaysian Rift System. (Liu 1986, reprinted by
t
permission)
0 200 400 KM
I I I
o FAULTS AND INfl:I!RED FAULT
~ WRENCH FAUL'r
o lSOI'ACHS Ot· CKU~T

!OJ IIOUNDARY OF BASIN
An example of a meander-belt sandstone body is indicates a channel width of 90-120 m. The meander
illustrated in Fig. 15.23. This unit consists of a fining- belt is therefore 10-15 times the width of the chan-
upward succession of medium- to fine-grained, nel, which is a typical ratio for meander belts of
cross-bedded sandstone, with a basal gravel lag. It is highly sinuous rivers. Yinan et al. (1987) also de-
interpreted as a succession of amalgamated point- scribed braided sandstone units from Daqing oil
bar deposits (Yinan et al. 1987). The meander belt field.
is 800-1000 m wide, with a width/depth ration of Reservoir connectedness is considered by Jin et
130-170. The complete point-bar sequence is esti- al. (1985). They illustrated an example of a section
mated from core studies to be 5-7 m thick. Given a containing three flow units: a meandering-channel
5°-10° dip on the point-bar accretion surface, this unit, consisting of fine sandstone, overlying a
55 56 64 6
IU
481+----+-
46+---+-
Lelend
o 200 400 M
~ Boundary
~ meander
of
beltl
Fig. 15.23. Geometry of a meander-belt sandstone body, Daqing oil field. (Yinan et al. 1987)
, } Meandering channel
~~~~~~¥>~~~"",",,::::"';:;:~~~"t""""""'::i:3;-o:::::::=-~-1 '"l sandstone

Braided cl)annel
sandstone
'=) Sheet·like sandstone

~§::§'§§S:t"':::'~
==========::::i =====4 = of lacustrine facies
with distributary
sandstone in parts
Fig. 15.24. Interconnection of three types of sandstone body: meandering channels; 2 braided channels; 3 core
analyses. (J in et al. 1985)
VI
......
00
-+-... -_+_-:1':-_-"'-
_ i-..- _T_+_+-_+-:"';_!.+:;:_!_!TT+"'_-:-_!+=+!...- .T_
~
~
~ ~ -*.
~~~~~~~~~~~~~~~~~ ...r...r ....r...r-....r...;-....r-r..J"--..r-...r-~.-r-...J""..r...r
~
:?::&'5'3><
~ ~
~ ~~
=<:
~~-...r-- ___ ~~
- ~ ~~ ~ -
~ <S:
:.r--r~:....r""
"
T'- + - ..... ... +~~-~!~-~....!.:=+-:... .....±. ...

.'L> :er:~ _-+-+-• . ,. . 0
+"+ ... - ...... +-~ +...~ .+--.+--±+-+±+--"-O-+~C+"'oC+· ~-,,+- +- .-.+---+. - ~] I
:::z.:; ~
++--++ +-.-.. .-+

50 0
(")
I
SANDSTONE CARBONATE CLAYSTONE (1)
'"on
METERS
CI>
:,P\..-
~ Small Mud'supported --
[g) Red
2~
Scale ~ n;'
rti9 on
o
.....,
~ Medium Stratification Grain·supported o
E<-j Green I::..:
Scale ~ ~
~
Cl
~ Stromato"1ic ~~~ Gray
~= '"on
::n
~
Fig. 15.25. Schematic stratigraphic cross section showing interfingering of fluvial sandstones and lacustrine carbonates and ~
on
claystones in Uinta Basin. (Pitman et al. 1982, reprinted by permission) s·
::n
.:
:J.
e..
~
(1)
on
(1)
~
::;.
'"
R 2l[ R 24[. "25£.
!
i
~:
slo
:0
~~77~~~------r---------~--~~~----~------~-------,--~ct~~
I!ct
• 1-:0
:::>!....J
~:o
io
51
~
•
i
I
I
T
e:
i
s I
L
R 21E. .. 22E
"25£.
!
'II'ID • .lSlo4 ,u" ..
.
TOTAL SANDSTONE ISOLITH OF MA IN PRODUC ING ZONE : DOUGLAS CREEK MEMBER
,
LW '
CI -20'
. -
,
..J
Fig. 15.26. Total sandstone isolith of producing zones in basin. In detail individual sandstone lenses occupy only a
Red Wash field, Utah. Shape of sandstone thickness con- small part of this area. (Chatfield 1972, reprinted by per-
tours suggests a coastal plain or deltaic complex thicken- mission)
ing toward the source area on the northern margin of the
braided, coarse-grained sandstone with a high-relief age disconformably overly Cretaceous clastic strata
scoured base, which in turn overlies a succession of of the foreland-basin clastic wedge. Cenozoic depos-
thin, sheetlike lacustrine sand bodies (Fig. 15.24). its were formed in alluvial and lacustrine deposi-
Permeability barriers, such as shales draping third- tional systems, with the main sediment source
and fourth-order surfaces, may be as little as 10- 40 located in Uinta Uplift, to the north of the basin.
cm thick but may effectively separate sand bodies More than 80 m 3 x 10 6 of recoverable oil and
into distinct flow zones. 422000 m 3 x 106 of recoverable gas have been discov-
ered in the Upper Cretaceous and Paleocene-Eocene
rocks of Uinta Basin (Fouch 1983). Most of this pro-
15.4.3 Red Wash Field, Utah duction is from lenticular fluvial sandstone bodies
that were deposited in coastal plain depositional
During the late Maaastrichtian and early Paleocene, systems. In the Paleocene-Eocene the Uinta Basin
Laramide tectonism in the Rocky Mountain foreland was part of a large lacustrine system. Sandstone bod-
basin created new uplifts in the Colorado Plateau ies were formed as channel and point-bar lenses
area, and the foreland basin was partitioned (Lawton encased within widespread, low-energy floodplain
1986b; Fouch et al. 1983; Dickinson et al. 1988; Fig. and marinellacustrine bay-fill deposits (Fig. 15.25)
11.26). The Uinta Basin, situated in northeastern and shaling out to the south. These coastal com-
Utah and westernmost Colorado, is one of the inter- plexes have been referred to as deltas (e.g., Fouch
montane basins generated by this tectonic episode. 1983), but they do not show the thick (hundreds of
In northern Uinta Basin, where there is significant meters), prograding, coarsening-upward character
oil and gas production, rocks of Paleocene-Eocene of typical major delta systems and are interpreted as
B PALEOENVIRONMENTAL MAP
T
7
S
T
8
S
R 24 E
HIGH-ENERG,. .AIIRIER-SEACH
c::::J (UPPER SHOIlfFACE AND SWASH ~ONE)
LACUSTRINE
MUDFLAT-LAGOON
LOW-ENERG" BARRIER-lEACH. BAR. AND o 3 M'LfS
ImiI NEAR' HOllE I
C3 FLUVIAL-DEJ..TAlC o & Km .
Fig. 15.27. Paleoenvironmental map of one of the producing units in the Red Wash field. Compare with Fig. 15.26.
(Castle 1990, reprinted by permission)
R 2 1E R 22( R 23E R 250[

f . ... ..
!
i
I:
510
:0
:x::.q
~~~~~~~----~~~~~~~~~uL~~~~------~------------~.q ~
1-' 0
::li...J
,
T,1
0
7, U
51
~
I
I
I
Ti
,:
S!
Ir-
ft ZU:
I
R 25[.
-
tJD . ,A I" A"C.,
..
STRUCTURE - TOP OF MAI N PRODUCING Z ONE ' DOUGLAS CREEK MEMBE R
C.I. - 100'
Fig. 15.28. Structure contour map of top of producing zone, Red Wash field, Utah, (Chatfield 1972, reprinted by
permission)
thin coastal-plain alluvial sheet sandstone bodies. present where the shale-out takes place updip within
Figure 15.26 shows an old isopach map of the pro- Laramide structures (Chatfield 1972; Pitman et al.
ducing zones, which suggests a lobate, deltaic form 1982; Fouch 1983; Fig. 15.28). Reservoir quality in
to the deposits. However, modern, more detailed most of the fields has been enhanced by the develop-
facies analysis has yielded the map shown in Fig. ment of fractures and secondary porosity, and the
15.27, in which a complex range of interfingering traps therefore also have a diagenetic component to
depositional systems has been identified. Traps are them (Keighin and Fouch 1981).
Chapter 16
Future Research Trends
In this brief concluding chapter some attempt is ground-penetrating radar promises to provide
made to assess the present state of knowledge ofthe much valuable information on the three-dimen-
geology of fluvial deposits, and to predict the kinds sional structure of both modern and ancient fluvial
of research developments that might be expected in systems in the near future, to further improve our
the next decade. data base of architectural information.
Facies Analysis. Although some workers still prefer to The Architecture of Overbank Regions. There is room for
establish new facies classifications with each new considerable additional work on the regional map-
field study, the classification and coding system ping of floodplain units, using catena and pedofacies
described in Chaps. 4 and 5 are by now virtually concepts. Such work is needed in order to under-
standard methodologies. Room remains for the rec- stand the long-term dynamics of channel avulsion
ognition and better description of facies that are and flooding mechanisms, as well as floodplain con-
rarely formed or preserved in the sedimentary struction.
record, such as those formed under very high shear
stresses. There continues to be a need for additional Fluvial Styles. Most sedimentological studies of mod-
case studies of ancient units to serve as a basis for a ern river systems have been carried out in temperate,
wide variety of interpretive studies. arctic, and hot arid regions. There is a need for
comparable research in tropical humid regions, such
In-channel Depositional Processes. Observations of natural as the rivers of the African and South American rain
sand-bed and gravel-bed rivers and flume observa- forests. Very few detailed studies of the ancient
tions and models have provided a thorough under- record have been completed, and a wide variety of
standing of the processes of formation of most architectural case studies is needed to complement
bedforms and macroforms. Work in this area con- our knowledge of modern rivers.
tinues, with the use of flume experiments, flow visu-
alization techniques, and numerical models. Stratigraphic Architecture. One of the most interesting
areas of research in recent years has been the
The Architecture of Channel-and-Bar Deposits. The develop- application of high-resolution chronostratigraphic
ment of the architectural-element method of de- techniques, particularly magnetostratigraphy, to the
scription has facilitated the documentation of the study of alluvial stratigraphies. The availability of
complex three-dimensional nature of the channel- correlation frameworks of this type permits much
and-bar deposits that characterize many modern more detailed reconstructions of regional sedimen-
and ancient fluvial systems. Classifications of archi- tary controls, particularly tectonism, and allows for
tectural elements and bounding surfaces have been the quantitative examination of the rates of such
developed, but there is scope for considerable devel- processes as basin accumulation rates and channel
opment and refinement. This is a case where work on avulsion frequency. Such data are essential for the
large outcrops of ancient deposits has forced a reex- modeling of the autogenic and tectonic control of
amination of many concepts derived from studies of alluvial stratigraphy, many studies of which are now
modern rivers that were restricted to data derived appearing, particularly as an aid to the reconstruc-
from shallow trenches dug above the water table. For tion and interpretation of fluvial petroleum reser-
example, it is only relatively recently that the accre- voirs.
tionary geometry of most bar complexes in braided
systems has been recognized. The application of Sequence Stratigraphy. Basic sequence models for flu-
high-resolution seismic reflection techniques and vial systems are now available and are being applied
524 Future Research Trends
to a wide array of case examples. Some of the most knowledge of fluvial styles, sequence stratigraphy,
interesting work underway at the time of writing and the tectonic control of fluvial depositional sys-
is the attempt to determine the relative importance tems has provided a range of tools for the discovery
of climate change and base-level change in the and mapping of fluvial reservoirs. Many additional
development of Quaternary fluvial valley fills. An discoveries remain to be made, particularly in fron-
alternation between valley incision and channel tier areas. At the exploitation stage, seismic mapping
aggradation has been documented that appears to be of channel bodies and numerical modeling of chan-
a complex response to changes in discharge regime nel architecture are both areas of active research.
and sediment supply associated with changes in Dipmeter data, cross-well seismic tomography, hori-
rainfall and vegetation cover. Not only is such work zontal drilling (two areas of research not discussed
of importance in the understanding of the recent in this book), and surveillance methods are all being
(postglacial) history of the earth, but it will also used in conjunction with outcrop analogue studies
provide models for the interpretation of earlier peri- as input to architectural models that then form the
ods in earth history when orbital-forcing mecha- basis for detailed flow studies. Such work is essential,
nisms may have been responsible for regular or both for primary production and for enhanced re-
episodic climate change. covery from old fields.
Quantitative Basin Models. One of the most active areas Hydrogeology. Few workers have applied architectural
of research in basin analysis is the development of concepts to problems in hydrogeology, and this is
numerical forward and inverse models of basin sub- an area that is likely to become considerably more
sidence and fill. Much attention has been paid to the important in the next few years. Geophysical tech-
tectonic and eustatic control of stratigraphic archi- niques for imaging the shallow subsurface, particu-
tecture of shallow-marine and continental-margin larly ground-penetrating radar, can provide a wealth
deposits, and recent work has also focused on the of architectural detail on the structure of aquifers
nature of the balance between subsidence and sedi- and aquitards, which can now be more accurately
ment supply at tectonically active nonmarine basin interpreted using sedimentological knowledge de-
margins. There is scope for much additional work rived from architectural techniques. Such know-
on the bringing together of models describing the ledge is likely to be of particular value in the study of
physics of sedimentation, models describing the the flow of contaminants through surficial sedi-
autogenic construction of alluvial plains, and mod- ments, such as the alluvial floodplains upon which
els that examine large-scale basin architecture and much of the developed world's industrial plant has
incorporate differential subsidence and base-level been constructed. Architectural details can be used
change. to constrain mathematical flow models in order to
provide precise, site-specific information on fluid
Petroleum Geology. Patterns of fluvial reservoir devel- transport paths.
opment are now relatively well understood. Our
References
Abdullatif OM (1989) Channel-fill and sheet-flood facies Allen JRL (1962) Petrology, origin and deposition of the
sequences in the ephemeral terminal River Gash, highest Lower Old Red Sandstone of Shropshire, En-
Kassala, Sudan. Sediment Geo163: 171-184 gland. J Sediment Petrol 32: 657-697
Abrahams MJ, Chadwick OH (1994) Tectonic and climatic Allen JRL (1963a) The classification of cross-stratified
implications of alluvial fans sequences along the units, with notes on their origin. Sedimentology 2:
Batinah coast, Oman. J Geol Soc (Lond) 151: 51-58 93-114
Ager DV (1981) The nature of the stratigraphical record, Allen JRL (1963b) Henry Clifton Sorby and the sedimen-
2nd edn. Halsted, Somerset tary structures of sands and sandstones in relation to
Ager DV (1993) The new catastrophism: the importance of flow conditions. Geol Mijnbouw 42: 223-228
the rare event in geological history. Cambridge Univer- Allen JRL (1964) Studies in fluviatile sedimentation: six
sity Press, Cambridge cyclothems from the Lower Old Red Sandstone, Anglo-
Ahmad R, Scatena FN, Gupta A (1993) Morphology and Welsh basin. Sedimentology 3: 163-198
sedimentation in Caribbean montane streams: ex- Allen JRL (1965a) A review of the origin and characteristics
amples from Jamaica and Puerto Rico. Sediment Geol of recent alluvial sediments. Sedimentology 5: 89-191
85: 157-169 Allen JRL (1965b) The sedimentation and palaeogeo-
Aitken JD (1966) Middle Cambrian to Middle Ordovician graphy of the Old Red Sandstone of Anglesey, North
cyclic sedimentation, southern Rocky Mountains, Wales. Yorkshire Geol Soc Proc 35: 139-185
Canada. Bull Can Petrol Geo114: 405-441 Allen JRL (1966) On bed forms and paleocurrents. Sedi-
Aitken JD (1978) Revised models for depositional grand mentology 6: 153-190
cycles, Cambrian of the southern Rocky Mountains, Allen JRL (1967) Notes of some fundamentals of paleo-
Canada. Bull Can Petrol Geo126: 515-542 current analysis, with reference to preservation poten-
Aitken JF, Flint SS (1995) The application of high-resolu- tial and sources of variance. Sedimentology 9: 75-88
tion sequence stratigraphy to fluvial systems: a case Allen JRL (1968) Current ripples. North Holland,
study from the Upper Carboniferous Breathitt Group, Amsterdam
eastern Kentucky, USA. Sedimentology 42: 3-30 Allen JRL (1970a) Physical processes of sedimentation.
Albright WA, Turner WL, Williamson KR (1980) Ninian Allen and Unwin, London
field, UK sector, North Sea. In: Halbouty MT (ed) Giant Allen JRL (1970b) Studies in fluviatile sedimentation: a
oil and gas fields of the decade: 1968-1978. Am Assoc comparison of fining-upward cyclothems, with special
Petrol Geol Mem 30: 173-193 reference to coarse-member composition and interpre-
Alexander J (1986) Idealized flow models to predict allu- tation. J Sediment Petrol 40: 298-323
vial sandstone body distribution in the Middle Jurassic Allen JRL (1973) Phase differences between bed configura-
Yorkshire Basin. Mar Petrol Geol 3: 298-305 tions and flow in natural environments, and their geo-
Alexander J (1992) Nature and origin of a laterally exten- logical relevance. Sedimentology 20: 323-329
sive alluvial sandstone body in the Middle Scalby For- Allen JRL (1974a) Reaction, relaxation and lag in natural
mation. J Geol Soc (Lond) 149: 431-441 sedimentary systems: general principles, examples and
Alexander J, Gawthorpe RL (1993) The complex nature lessons. Earth Sci Rev 10: 263-342
of a Jurassic multistorey, alluvial sandstone body, Allen JRL (1974b) Studies in fluviatile sedimentation: im-
Whitby, North Yorkshire. In: North CP, Prosser DJ plications of pedogenic carbonate units, Lower Old Red
(eds) Characterization of fluvial and aeolian reservoirs. Sandstone, Anglo-Welsh outcrop. Geol J 9: 181-208
Geological Society, London, pp 123-142 (Spec Pub173) Allen JRL (1977) The plan shape of current ripples in
Alexander J, Leeder MR (1987) Active tectonic control on relation to flow conditions. Sedimentology 24: 53-62
alluvial architecture. In: Ethridge FG, Flores RM, Allen JRL (1978) Studies in fluviatile sedimentation: an
Harvey MD (eds) Recent developments in fluvial sedi- exploratory quantitative model for the architecture of
mentology. Soc Econ Paleontol Mineral Spec Publ 39: avulsion-controlled suites. Sediment Geo121: 129-147
243-252 Allen JRL (1980) Sand waves: a model of origin and inter-
Alexander J, Bridge JS, Leeder MR, Collier EL, Gawthorpe nal structure. Sediment Geo126: 281-328
RL (1994) Holocene meander-belt evolution in an ac- Allen JRL (1983a) Studies in fluviatile sedimentation: bars,
tive extensional basin, southwestern Montana. J Sedi- bar complexes and sandstone sheets (low-sinuosity
ment Res B64: 542-559 braided streams) in the Brownstones (1. Devonian),
Algeo TJ, Wilkinson BH (1988) Periodicity of mesoscale Welsh Borders. Sediment Geo133: 237-293
phanerozoic sedimentary cycles and the role of Allen JRL (1983b) River bedforms: progress and problems.
Milankovitch orbital modulation. J Geo196: 313-322 In: Collinson JD, Lewin J (eds) Modern and ancient
526 References
fluvial systems. Int Assoc Sedimentol Spec Publ 6: Ashton M (ed) (1992) Advances in reservoir geology. Geol
19-33 Soc Lond Spec Publ 69: 240
Allen JRL (1983c) Gravel overpassing on humpback bars Atkinson CD (1986) Tectonic control on alluvial sedimen-
supplied with mixed sediment: examples from the tation as revealed by an ancient catena. in the Capella
Lower Old Red Sandstone, southern Britain. Sedimen- Formation (Eocene) of northern Spain. In: Wright VP
tology 30: 285-294 (ed) Paleosols: their recognition and interpretation.
Allen JRL (1984) Sedimentary structures: their character Blackwell Scientific, Oxford, pp 139-179
and physical basis. Elsevier, Amsterdam, 663 p (Devel- Atkinson CD, McGowen JH, Bloch S, Lundell LL, Trumbly
opments in sedimentology, vol 30) PN (1990) Braidplain and deltaic reservoir, Prudhoe
Allen JRL, Williams BPJ (1982) The architecture of an Bay field, Alaska. In: Barwis JH, McPherson JG,
alluvial suite: rocks between the Townsend Tuff and Studlick RJ (eds) Sandstone petroleum reservoirs.
Pickard bay Tuff Beds (Early Devonian), southwest Springer, Berlin Heidelberg New York, pp 7-29
Wales. Philos Transact R Soc Lond [BioI] 297: 51-89 Aubry WM (1989) Mid-Cretaceous incision related to
Allen MB, Windley BF, Zhang C (1991) Active alluvial eustasy, southern Colorado Plateau. Geol Soc Am Bull
systems in the Korla Basin, Tien Shan, northwest 101:443-449
China, sedimentation in a complex foreland basin. Geol Autin WJ (1992) Use of alloformations for definition
Mag 128: 661-666 of Holocene meander belts in the middle Amite
Allen PA, Allen JR (1990) Basin analysis: principles and River, southeastern Louisiana. Geol Soc Am Bull lO4:
applications. Blackwell Scientific Publications, Oxford 233-241
Allen PA, Matter A (1982) Oligocene meandering stream Autin WJ, Burns SF, Miller BJ, Sucier RT, Snead JI (1991)
sedimentation in the eastern Ebro basin, Spain. Eclog Quaternary geology of the Lower Mississippi valley. In:
Geol Helv 75: 33-50 Morrison RB (ed) The geology of North America, vol
Alonso Zarza MA, Wright VP, Calvo JP, Garcia del Cura K-2: Quaternary nonglacial geology conterminous US.
MA (1992) Soil-landscape and climatic relationships in Geological Society of America, pp 547-581
the Middle Miocene of the Madrid Basin: Sedimento- Bachman SB, Lewis SD, Schweller WJ (1983) Evolution of a
logy 39: 17-35 forearc basin, Luzon Central Valley, Philippines. Am
Ambrose WA, Tyler N, Parsley MJ (1991) Facies heteroge- Assoc Petrol Geol Bull 67: 1143-1162
neity, pay continuity, and infill potential in barrier- Badgley C, Tauxe L (1990) Paleomagnetic stratigraphy and
island, fluvial, and submarine-fan reservoirs: examples time in sediments: studies in alluvial Siwalik rocks of
from the Texas Gulf Coast and Midland Basin. In: Miall Pakistan. J Geo198: 457-477
AD, Tyler N (eds) The three-dimensional facies archi- Bailey EB (1962) Charles Lyell. Nelson, London
tecture of terrigenous clastic sediments, and its impli- Bailey EB (1967) James Hutton - the founder of modern
cations for hydrocarbon discovery and recovery. Soc geology. Elsevier, Amsterdam
Econ Paleontol Mineral Concepts Models Ser 3: 13-21 Baker VR (1973) Paleohydrology and sedimentology of
Anad6n P, Cabrera L, Colombo F, Marzo M, Riba 0 (1986) Lake Missoula flooding in eastern Washington. Geo-
Syntectonic intraformational unconformities in allu- logical Society of America, Spec Pap 144
vial fan deposits, eastern Ebro basin margins (NE Baker VR (1978a) Adjustment of fluvial systems to climate
Spain). In: Allen PA, Homewood P (eds) Foreland ba- and source terrain in tropical and subtropical environ-
sins. Int Assoc Sedimentol Spec Publ8: 259-27l ments. In: Miall AD (ed) Fluvial sedimentology. Can
Anderson B (1961) The Rufiji basin, Tanganyika, vol 7: Soc Petrol Geol Mem 5: 211-230
soils of the main irrigable area. FAO report to Govern- Baker VR (l978b) Large-scale erosional and depositional
ment of Tanganyika features of the Channeled Scabland. In: Baker VR,
Anstey NA (1980) Seismic exploration for sandstone reser- Nummedal D (eds) The channeled scabland. National
voirs. International Human Resources Development Aeronautics and Space Administration, Washington,
Corporation, Boston pp 81-115
Arche A (1983) Coarse-grained meander lobe deposits in Balducchi A, Pommier G (1970) Cambrian oil field of Hassi
the Jarama River, Madrid, Spain. In: Collinson JD, Messaoud, Algeria. In: Halbouty MT (ed) Geology of
Lewin J (eds) Modern and ancient fluvial systems. Int giant petroleum fields. Am Assoc Petrol Geol Mem 14:
Assoc Sedimentol Spec Publ6: 313-321 477-488
Arnott RWC (1992) The role of fluvial processes during Ballance PF (1988) The Huriwai braidplain delta of New
deposition of the (Cardium) Carrott Creek/Cyn-Pem Zealand: a Late Jurassic, coarse-grained, volcanic-fed
conglomerates. Bull Can Petrol Geo140: 356-362 depositional system in Gondwana forearc basin. In:
Ashley GM (1990) Classification oflarge-scale subaqueous Nemec W, Steel RJ (eds) Fan deltas: sedimentology and
bedforms: a new look at an old problem. J Sediment tectonic settings. Blackie, Glasgow, pp 430-444
Petrol 60: 160-172 Ballance PF, Reading HG (eds) (1980) Sedimentation in
Ashley GM, Shaw J, Smith ND (1985) Glacial sedimentary oblique-slip mobile zones. Int Assoc Sedimentol Spec
environments. Soc Econ Paleontol Mineral Short Pub14:265
Course 16 Bally AW (1984) Structural styles and the evolution of
Ashmore PE (1991) How do gravel-bed rivers braid? Can J sedimentary basins. American Association of Petro-
Earth Sci 28: 326-341 leum Geologists Short Course, Tulsa
Ashmore PE (1993) Anabranch confluence kinetics and Bally AW, GordyPL, Stewart GA (1966) Structure, seismic
sedimentation processes in gravel-bed streams. In: Best data and orogenic evolution of southern Canadian
JL, Bristow CS (eds) Braided rivers. Geol Soc Lond Spec Rocky Mountains. Bull Can Petrol Geo114: 337-381
Publ 75: 129-146
References 527
Bally AW, Snelson S (1980) Realms of subsidence. In: Miall Belt ES (1993) Tectonically induced sediment diversion
AD (ed) Facts and principles of world petroleum occur- and the origin of thick, widespread coal beds (Appala-
rence. Can Soc Petrol Geol Mem 6: 9-94 chian and Williston Basins, USA). In: Frostick LE, Steel
Baltzer F, Purser BH (1990) Modern alluvial fan and deltaic RJ (eds) Tectonic controls and signatures in sedimen-
sedimentation in a foreland tectonic setting: the lower tary successions. Int Assoc Sedimentol Spec Publ 20:
Mesopotamian Plain and the Arabian Gulf. Sediment 337-397
Geol67: 175-197 Ben-Avraham Z, Emery KO (1973) Structural frame-
Banks NL (1973) The origin and significance of some down work of Sunda Shelf. Am Assoc Petrol Geol Bull 57:
current dipping cross stratified sets. J Sediment Petrol 2323-2366
43:423-427 Bendix J (1992) Fluvial adjustments on varied timescales in
Bardaji T, Dabrio cr, Goy JL, Somoza L, Zazo C (eds) Bear Creek Arroyo, Utah, USA. Z Geomorph NF 36:
(1990) Pleistocene fan deltas in southeasern Iberian 141-163
peninsula: sedmentary controls and sea-level changes. Benkhelil J (1982) Benue trough and Benue chain. Geol
In: Colella A, Prior DB (eds) Coarse-grained deltas. Int Mag 119: 155-168
Assoc Sedimentol Spec PubllO: 129-151 Bentham PA, Burbank DW, Puigdefabregas C (1992) Tem-
Barrell J (1912) Criteria for the recognition of ancient delta poral and spatial controls on the alluvial architecture of
deposits. Geol Soc Am Bull 23: 377-446 an axial drainage system: late Eocene Escanilla Forma-
Barrell J (19l3/1914) The Upper Devonian delta of the tion, southern Pyrenean foreland basin, Spain. Basin
Appalachian geosyncline. Am J Sci 36: 429-472; 37: Res 4: 335-352
87-109,229-253 Bentham PA, Talling pJ, Burbank DW (1993) Braided
Barrell I (1917) Rhythms and the measurement of geologi- stream and flood-plain deposition in a rapidly aggrad-
cal time. Geol Soc Am Bull 28: 745-904 ing basin: the Escanilla formation, Spanish Pyrenees.
Barrell J (1925) Marine and terrestrial conglomerates. Geol In: Best JL, Bristow CS (eds) Braided rivers. Geol Soc
Soc Am Bull 36: 279-342 Lond Spec Publ 75: 177-194
Barry RG, Chorley RJ (1992) Atmosphere, weather and Berg RR (1968) Point-bar origin of Fall River Sandstone
climate, 6th edn. Routledge, London reservoirs. Am Assoc Petrol Geol Bull 52: 2116-2122
Bartow JA (1987) Cenozoic nonmarine sedimentation in Berg RR (1986) Reservoir sandstones. Prentice Hall,
the San Joaquin Basin, central California. In: Ingersoll Englewood Cliffs, New Jersey
RV, Ernst WG (eds) Cenozoic basin development Berg RR, Cook BC (1968) Petrography and origin of Lower
of coastal California, Rubey, vol VI. Prentice-Hall, Tuscaloosa sandstones, Mallalieu field, Lincoln
Englewood Cliffs, pp 146-171 County, Mississippi. Trans Gulf Coast Assoc Geol Soc
Basu A (1985) Influence of climate and relief on composi- 18:242-255
tions of sands released at source areas. In: Zuffa GG (ed) Berger A, Imbrie J, Hays J, Kukla G, Saltzman B (eds)
Provenance of arenites. Reidel, Dordrecht, pp 1-18 (1984) Milankovitch and climate. NATO ASI Ser D.
Bates RL, Jackson JA (eds) (1987) Glossary of geology, 3rd Reidel, Dordrecht, 2 vols
edn. American Geological Institute, Alexandria cr
Bernard HA, Major (1963) Recent meander belt depos-
Beaty CB (1963) Origin of alluvial fans, White Mountains, its of the Brazos River; an alluvial "sand" model (ab-
California and Nevada. Assoc Am Geogr Ann 53: stract). Am Assoc Petrol Geol Bull 47: 350
516-535 Bernard HA, Leblanc RJ, Major cr (1962) Recent and
Beaty CB (1990) Anatomy of a White Mountain debris flow pleistocene geology of southeast Texas. In: Rainwater
- the making of an alluvial fan. In: Rachocki AH, EH, Zingula RP (eds) Geology of the Gulf Coast and
Church M (eds) Alluvial fans: a field approach. Wiley, central Texas. Geological Society of America guide-
Chichester, pp 69-89 book for 1962 annual meeting, pp 175-224
Beaumont C (1981) Foreland basins. Geophys J R Astron Berne S, Auffret J-p, Walker P (1988) Internal structure of
Soc 65: 291-329 subtidal sandwaves revealed by high-resolution seis-
Beer JA (1990) Steady sedimentation and lithologic com- mic reflection. Sedimentology 35: 5-20
pleteness, Bermejo basin, Argentina. J Geol98: 501-517 Berne S, Durand J, Weber 0 (1991), Architecture of
Beerbower JR (1964) Cyclothems and cyclic depositional modern subtidal dunes (sand waves), Bay of
mechanisms in alluvial plain sedimentation. Geol Surv Bourgneuf, France. In: Miall AD, Tyler N (eds) The
Kansas Bull 169 (1): 31-42 three-dimensional facies architecture of terrigenous
Begin ZB (1981) Development of longitudinal profiles of clastic sediments, and its implications for hydrocarbon
alluvial channels in response to baselevel lowering. discovery and recovery. Soc Econ Paleontol Mineral
Earth Surface Proc Landforms 6: 49-68 Conc Sedimentol Paleontol3: 245-260
Begin ZB, Meyer DF, Schumm SA (1980) Knickpoint mi- Bersier A (1948) Les sedimentation rhythmiques syn-
gration in alluvial channels due to baselevellowering. orogeniques dans l'avant-fosse molassique alpine. 18th
Am Soc Civil Eng J Waterway Port Coast Ocean Div 106: international geological congress, part IV, pp 83-93
369-388 Bersier A (1958) Sequences detritiques et divagations flu-
Behrensmeyer AK (1987) Miocene fluvial facies and verte- viales. Eclog Geol Helv 51: 854-893
brate taphonomy in northern Pakistan. In: Ethridge Besly BM (1988) Palaeogeographic implications of late
FG, Flores RM, Harvey MD (eds) Recent developments Westphalian to early Permian red-beds, central En-
in fluvial sedimentology. Soc Econ Paleontol Mineral gland. In: Besly BM, Kelling G (eds) Sedimentation in a
Spec Pub139: 169-176 synorogenic basin complex: the Upper Carboniferous
Behrensmeyer AK, Tauxe L (1982) Isochronous fluvial of northwest Europe. Blackie, Glasgow, pp 200-221
systems in Miocene deposits of northern Pakistan. Belsy BM, Fielding CR (1989) Palaeosols in Westphalian
Sedimentology 29: 331-352 coal-bearing and red-bed sequences central and north-
528 References
ern England. Palaeogeogr Palaeoclimatol Palaeoecol Blair TC, Baker FG, Turner JB (1991) Cenozoic fluvial-
70:303-330 facies architecture and aquifer heterogeneity, Oroville,
Besly BM, Kelling G (eds) (1988) Sedimentation in a syn- California, superfund site and vicinity. In: Miall AD,
orogenic basin complex: the upper Carboniferous of Tyler N (eds) The three-dimensional facies architec-
northwest Europe. Blackie, Glasgow, 276 p ture of terrigenous clastic sediments and its implica-
Best JL (1987) Flow dynamics at river channel confluences: tions for hydrocarbon discovery and recovery. Soc
implications for sediment transport and bed morphol- Econ Paleontol Mineral Cone Sedimentol Paleontol 3:
ogy. In: Ethridge FG, Flores RM (eds) Recent and an- 148-159
cient nonmarine depositional environments. Soc Econ Blair TC, Bilodeau WL (1988) Development of tectonic
Paleontol Mineral Spec Pub131: 27-35 cyclothems in rift, pull-apart, and foreland basins: sedi-
Best JL (1988) Sediment transport and bed morphology at mentary response to episodic tectonism. Geology 16:
river channel confluences. Sedimentology 35: 481-498 517-520
Best JL, Brayshaw AC (1985) Flow separation - a physical Blair TC, McPherson JG (1992) The Trollheim alluvial fan
process for the concentration of heavy minerals within and facies model revisited. Geol Soc Am Bull 104:
alluvial channels. J Geol Soc Lond 142: 747-755 762-769
Best JL, Bristow CS (eds) (1993) Braided rivers. Geol Soc Blair TC, McPherson JG (1994) Alluvial fans and their
Lond Spec Publ 75: 419 natural distinction from rivers based on morphology,
Best JL, Bristow CS, Roy AG (1989) The morphology of hydraulic processes, sedimentary processes, and facies
river channel confluences: scales and dynamics, pro- assemblages. J Sediment Res A64: 450-489
gram and abstracts of the 4th international conference Blakey RC, Gubitosa R (1984) Controls of sandstone body
on fluvial sedimentology, Barcelona, Sitges, p 75 geometry and architecture in the Chinle Formation
Beston NB (1986) Reservoir geological modelling of the (Upper Triassic), Colorado Plateau. Sediment Geo138:
North Rankin field, northwest Australia. Aust Petrol 51-86
Explor Assoc J 26: 375-389 Blanche JB (1990) An overview of the exploration history
Beutner EC, Flueckinger LA, Gard TM (1967) Bedding and hydrocarbon potential of Cambodia and Laos.
geometry in a Pennsylvanian channel sandstone. Geol SEAPEX Proc 9: 89-99
Soc Am Bull 78: 911-916 Blissenbach E (1954) Geology of alluvial fans in semi-arid
Bhattacharyya DP, Lorenz JC (1983) Different depositional regions. Geol Soc Am Bull 65: 175-190
settings of the Nubian lithofacies in Libya and southern Blixt JE (1941) Cut Bank oil and gas field, Glacier County,
Egypt. In: Collinson JD, Lewin J (eds) Modern and Montana. In: Levorsen AI (ed) Stratigraphic type oil
ancient fluvial systems. Int Assoc Sedimentol Spec Publ fields. American Association of Petroleum Geologists,
6:435-448 Tulsa, pp 327-381
Bhattacharya J (1991) Regional to sub-regional facies ar- Bloch S, McGowen JH, Duncan JR, Brizzolara DW (1990)
chitecture of river-dominated deltas, Upper Creta- Porosity prediction, prior to drilling. In: Sandstones of
ceous Dunvegan Formation, Alberta subsurface. In: the Kekiktuk Formation (Mississippian), North Slope
Miall AD, Tyler N (eds) The three-dimensional facies of Alaska. Am Assoc Petrol Geol Bull 74: 1371-1385
architecture of terrigenous clastic sediments and its Blodgett RH, Stanley KO (1980) Stratification, bedforms,
implications for hydrocarbon discovery and recovery. and discharge relations of the Platte braided river sys-
Soc Econ Paleontol Mineral Cone Sedimentol Paleontol tem, Nebraska. J Sediment Petrol 50: 139-148
3: 189-206 Bloomer RR (1977) Depositional environments of a reser-
Biddle KT, Christie-Blick N (eds) (1985) Strike-slip defor- voir sandstone in west-central Texas. Am Assoc Petrol
mation, basin formation and sedimentation. Soc Econ Geol Bull 61: 344-359
Paleontol Mineral Spec Pub137: 386 Bluck BJ (1967) Deposition of some Upper Old Red Sand-
Biddle KT, Uliana MA, Mitchum RM Jr, Fitzgerald MG, stone conglomerates in the Clyde area: a study in the
Wright RC (1986) The stratigraphic and structural evo- significance of bedding. Scott J Geol3: 139-167
lution of the central and eastern Magallanes Basin, Bluck BJ (1971) Sedimentation in the meandering River
southern South America. In: Allen PA, Homewood P Endrick. Scott J Geol7: 93-138
(eds) Foreland basins. Int Assoc Sedimentol Spec Publ Bluck BJ (1979) Structure of coarse grained braided stream
8:41-61 alluvium. Trans R Soc Edinb 70: 181-221
Billi P, Magi M, Sagri M (1987) Coarse-grained low-sinuos- Bluck BJ (1980) Structure, generation and preservation of
ity river deposits: examples from Plio-Pleistocene upward fining, braided stream cycles in the Old Red
Valdarno Basin, Italy. In: Ethridge FG, Flores RM, Sandstone of Scotland. Trans R Soc Edinb 71: 29-46
Harvey MD (eds) Recent developments in fluvial sedi- Bluck BJ (1991) Terrane provenance and amalgamation:
mentology. Soc Econ Paleontol Mineral Spec Publ 39: examples from the Caledonides. In: Dewey JF, Gass IG,
197-203 Curry GB, Harris NBW, Sengor AMC (eds)
Blackwelder E (1928) Mudflow as a geologic agent in semi- Allochthonous terranes. Cambridge University Press,
arid mountains. Geol Soc Am Bull 39: 465-484 Cambridge, pp 143-153
Blackbourn GA (1984) Sedimentary facies variation and Blum MD (1992) Modern depositional environments and
hydrocarbon reservoirs in continental sediments - a recent alluvial history of the lower Colorado River, Gulf
predictive model. J Petrol Geol 7: 67~76 Coastal Plain, Texas, PhD dissertation, University of
Blair TC (1987) Sedimentary processes, vertical stratifica- Texas, Austin, Texas, 304 p
tion sequences, and geomorphology of the Roaring Blum MD (1994) Genesis and architecture of incised valley
River alluvial fan, Rocky Mountain National Park. J fill sequences: a late Quaternary example from the
Sediment Petrol 57: 1-18 Colorado River, Gulf Coastal Plain of Texas. In: Weimer
References 529
P, Posamentier HW (eds) Siliclastic sequence stratigra- Bridge JS, Diemer JA (1983) Quantitative interpretation of
phy: recent developments and applications. Am Assoc an evolving ancient river system. Sedimentology 30:
Petrol Geol Mem 58: 259-283 599-623
Blum MD, Price DM (1994) Glacio-eustatic and climatic Bridge JS, Gordon EA (1985) The Catskill magnafacies of
controls on Quaternary alluvial plain deposition, Texas New York State. In: Flores RM, Harvey MD (eds) Field
coastal plain. Gulf Coast Assoc Geol Soc Trans 44: 1-9 guide to modern and ancient fluvial systems in the
Boothroyd JC, Ashley GM (1975) Process, bar morphology, United States. 3rd Int Sedimentology Conf, Fort
and sedimentary structures on braided outwash fans, Collins, Colorado, pp 3-17
northeastern Gulf of Alaska. In: Jopling AV, McDonald Bridge JS, Jarvis J (1976) Flow and sedimentary processes
BC (eds) Glacio-fluvial and glaciolacustrine sedimen- in the meandering River South Esk, Glen Clova, Scot-
tation. Soc Econ Paleontol Mineral Spec Paper 23: land. Earth Surface Proc 1: 303-336
193-222 Bridge JS, Leeder MR (1979) A simulation model of alluvial
Boothroyd JC, Nummedal D (1978) Proglacial braided stratigraphy. Sedimentology 26: 617-644
outwash: a model for humid alluvial-fan deposits. In: Bridge JS, Mackey SD (1993a) A revised alluvial stratigra-
Miall AD (ed) Fluvial sedimentology. Can Soc Petrol phy model. In: Marzo M, Puigdefabregas C (eds) Allu-
Geol Mem 5: 641-668 vial sedimentation. Int Assoc Sedimentol Spec Pub117:
Boss RF, Lennon RB, Wilson BW (1976) Middle Ground 319-336
Shoal field, Alaska. In: Braunstein J (ed) North Ameri- Bridge JS, Mackey SD (1993b) A theoretical study of fluvial
can oil and gas fields. Am Assoc Petrol Geol Mem 24: sandstone body dimensions. In: Flint SS, Bryant ID
1-22 (eds) The geological modelling of hydrocarbon reser-
Botvinkina LN, Feofilova AP, Yablokov VS (1954) Alluvial voirs and outcrop analogues. Int Assoc Sedimentol
deposits for the Middle Carboniferous coal-bearing Spec Pub115: 213-236
beds of the Donets Basin. Trans Inst Geol Nauk Akad Bridge JS, Smith ND, Trent F, Gabel SL, Bernstein P (1986)
Nauk SSSR Geol Ser 151: 30-89 Sedimentology and morphology of a low-sinuosity
Bowen DW, Weimer P, ScottAJ (1993) The relative success river: Calamus River, Nebraska Sand Hills. Sedimento-
of siliciclastic sequence stratigraphic concepts in ex- logy 33: 851-870
ploration: examples from incised valley fill and turbid- Brierley GJ (1991) Floodplain sedimentology of the
ite systems reservoirs. In: Weimer P, Posamentier HW Squamish River, British Columbia: relevance of ele-
(eds) Siliciclastic sequence stratigraphy. Am Assoc ment analysis. Sedimentology 38: 735-750
Petrol Geol Mem 58: 15-42 Brierley GJ (1993) Facies analysis of fine-grained fluvial
Bowen JM (1975) The Brent oil-field. In: Woodland AW systems: a constructivist approach: keynote address,
(ed) Petroleum and the continental shelf of north-west 5th international conference on fluvial sedimentology,
Europe, vol 1: geology. Wiley, New York, pp 353-361 Brisbane, Australia, July 1993, keynote addresses and
Bown TM, Kraus MJ (1981) Lower Eocene alluvial paleo- abstracts, pp K5-K13
sols (Wilwood Formation, northwest Wyoming, USA) Brierley GJ, Hickin EJ (1991) Channel planform as a non-
and their significance for paleoecology, paleoclimato- controlling factor in fluvial sedimentology: the case of
logy, and basin analysis. Palaeogeogr Palaeoclimatol the Squamish River floodplain, British Columbia. Sedi-
Palaeoecol34: 1-30 ment Geol 75: 67-83
Bown TM, Kraus MJ (1987) Integration of channel and Brierley GJ, Liu K, Crook KAW (1993) Sedimentology of
floodplain suites, I. Development of sequence and lat- coarse-grained alluvial fans in the Markham Valley,
eral relations of alluvial paleosols. J Sediment Petrol 57: Papua New Guinea. Sediment Geo186: 297-324
587-601 Brinkman R (1933) Uber Kreuzschichtung im deutschen
Brenner RL, Swift DJP, Gaynor GC (1985) Re-evaluation of Buntsandsteinbecken. Nachrichten von der Gesell-
coquinoid sandstone depositional model, Upper Juras- schaft der Wissenschaften zu Gottingen, Math Phys KI
sic of central Wyoming and south-central Montana. Fachgruppe IV, no 32
Sedimentology 32: 363-372 Bristow CS (1987) Brahmaputra river: channel migration
Brice JC (1964) Channel patterns and terraces of the Loup and deposition. In: Ethridge FG, Flores RM, Harvey MD
Rivers in Nebraska. US Geol Survey, Prof Pap 422-D (eds) Recent developments in fluvial sedimentology.
Bridge JS (1975) Computer simulation of sedimentation in Soc Econ Paleontol Mineral Spec Pub139: 63-74
meandering streams. Sedimentology 22: 3-44 Bristow CS (1988) Controls on the sedimentation of the
Bridge JS (1984) Large scale facies sequences in alluvial Rough Rock Group (Namurian) from the Pennine
overbank environment. J Sediment Petrol 54: 583-588 Basin of northern England. In: Besly BM, Kelling G
Bridge JS (1985) Paleochannel patterns inferred from allu- (eds) Sedimentation in a synorogenic basin complex:
vial deposits: a critical evaluation. J Sediment Petrol 55: the Upper Carboniferous of northwest Europe. Blackie,
579-589 Glasgow, pp 114-131
Bridge JS (1993a) Description and interpretation of fluvial Bristow CS (1993a) Sedimentary structures exposed in bar
deposits: a critical perspective. Sedimentology 40: tops in the Brahmaputra River, Bangladesh. In: Best JL,
801-810 Bristow CS (eds) Braided rivers. Geol Soc Lond Spec
Bridge JS (1993b) The interaction between channel geom- Publ 75: 277-289
etry, water flow, sediment transport and deposition in Bristow CS (1993b) Sedimentology of the Rough Rock: a
braided rivers. In: Best JL, Bristow CS (eds) Braided carboniferous braided sheet sandstone in northern
rivers. Geological Society, London, pp 13-71 (Spec Publ England. In: Best JL, Bristow CS (eds) Braided rivers.
75) Geol Soc Lond Spec Publ 75: 291-304
530 References
Bristow CS, Best JL (1993) Braided rivers: perspectives and South Africa. In: Collinson JD, Lewin J (eds) Modern
problems. In: Best JL, Bristow CS (eds) Braided rivers. and ancient fluvial systems. Int Assoc Sedimentol Spec
Geol Soc Lond Spec Pub175: 1-11 PubI6:549-562
Bristow CS, Best JL, Roy AG (1993) Morphology and Bull WB (1963) Alluvial-fan deposits in western Fresno
facies models of channel confluences. In: Marzo M, County, California. J Geol 71: 243-251
Puigdefabregas C (eds) Alluvial sedimentation. Int Bull WB (1964) Alluvial fans and near surface subsidence
Assoc Sedimentol Spec Pub117: 91-100 in western Fresno County, California. US geological
Broadhurst FM (1988) Seasons and tides in the West- survey professional paper 437-A
phalian. In: Besly BM, Kelling G (eds) Sedimentation in Bull WB (1972) Recognition of alluvial fan deposits in the
a synorogenic basin complex: the Upper Carboniferous stratigraphic record. In: Rigby JK, Hamblin WK (eds)
of northwest Europe. Blackie, Glasgow, pp 264-272 Recognition of ancient sedimentary environments. Soc
Bromley MH (1991a) Variations in fluvial style as revealed Econ Paleontol Mineral Spec Pub116: 63-83
by architectural elements, Kayenta Formation, Mesa Bull WB (1977) The alluvial fan environment. Prog Phys
Creek, Colorado, USA: evidence for both ephemeral Geogr 1: 222-270
and perennial fluvial processes. In: Miall AD, Tyler N Bull WB (1991) Geomorphic responses to climate change.
(eds) The three-dimensional facies architecture of Oxford University Press, New York
terrigenous clastic sediments and its implications Buller AT, Berg E, Hjelmeland 0, Kleppe J, Torsaeter 0,
for hydrocarbon. discovery and recovery. Soc Econ Aasen JO (eds) (1990) North Sea oil and gas reservoirs
Paleontol Mineral Cone Sedimentol Paleontol3: 94-102 II. Graham and Trotman, London, 453 p
Bromley MH (1991 b) Architectural features ofthe Kayenta Burbank DW (1992) Causes of recent Himalayan uplift
Formation (Lower Jurassic), Colorado Plateau, USA: deduced from deposited patterns in the Ganges basin.
relationship to salt tectonics in the Paradox Basin. Sedi- Nature 357: 680-683
ment Geol 73: 77-99 Burbank DW, Beck RA (1991) Models of aggradation ver-
Brookfield ME (1977) The origin of bounding surfaces in sus progradation in the Himalayan foreland. Geol
ancient aeolian sandstones. Sedimentology 24: 303-332 Rundsch 80: 623-638
Brookfield ME (1992) The paleorivers of central Asia: the Burbank DW, Raynolds RGH (1984) Sequential late Ceno-
interrelationship of Cenozoic tectonism, erosion and zoic structural disruption of the northern Himalayan
sedimentation. 29th Int Geol Congr, Kyoto, Japan, Ab- foredeep. Nature 311: 114-118
stracts, vol 2, p 292 Burbank DW, Raynolds RGH, Johnson GD (1986) Late
Brookfield ME (1993) The interrelations of post-collision Cenozoic tectonics and sedimentation in the north-
tectonics and sedimentation in central Asia. In: western Himalayan foredeep: II. Eastern limb of the
Frostick LE, Steel RJ (eds) Tectonic controls and signa- Northwest Syntaxis and regional synthesis. In: Allen
tures in sedimentary successions. Int Assoc Sedimentol PA, Homewood P (eds) Foreland basins. Int Assoc
Spec Pub12D: 13-35 Sedimentol Spec Publ 8: 293-306
Brown AR (1985) The role of horizontal seismic sections in Burbank DW, Beck RA, Raynolds RGH, Hobbs R,
stratigraphic intepretation. In: Berg OR, Woolverton Tahirkheli RAK (1988a) Thrusting and gravel progra-
DG (eds) Seismic stratigraphy II. Am Assoc Petrol Geol dation in foreland basins: a test of post -thrusting gravel
Mem 39: 37-47 dispersal. Geology 16: 1143-1146
Brown AR (1991) Interpretation of three-dimensional Burbank DW, Raynolds RGH (1988b) Stratigraphic keys to
seismic data, 3rd edn. Am Assoc Petrol Geol Mem 42: the timing of thrusting in terrestrial foreland basins:
341 applications to the northwestern Himalaya. In:
Brown D (1993) Make room! It's 3-D for your Pc. Ameri- Kleinspehn KL, Paola C (eds) New perspectives in basin
can Association of Petroleum Geologists Explorer, analysis: Springer, Berlin Heidelberg New York, pp
Tulsa, June 1993, p 14 331-351
Brown LF Jr, Fisher WL (1977) Seismic-stratigraphic Burbank DW, Beck RA, Raynolds RGH (1989) Reply to
interpretation of depositional systems: examples from comment by Heller et al. on "Thrusting and gravel
Brazilian rift and pull-apart basins. In: Payton CE (ed) progradation in foreland basins: a test of post-thrust-
Seismic stratigraphy - applications to hydrocarbon ex- ing gravel dispersal". Geology 17: 960-961
ploration. Am Assoc Petrol Geol Mem 26: 213-248 Burbank DW, Verges J, Munoz JA, Bentham P (1992) Co-
Brown S (1984) Jurassic. In: Glennie KW (ed) Introduction eval hindward- and forward-imbricating thrusting in
to the petroleum geology of the North Sea. Blackwell the south-central Pyrenees, Spain: timing and rates of
Scientific, Oxford, pp 103-131 shortening and deposition. Geol Soc Am Bull 104: 3-17
Brush LM Jr (1958) Study of stratification in a large labora- Burchfiel BC, Royden L (1982) Carpathian fold and thrust
tory flume (abstract). Geol Soc Am Bull 69: 1542 belt and its relation to Pannonian and other basins. Am
Bryant ID (1983) Facies sequences associated with some Assoc Petrol Geol Bull 66: 1179-1195
braided river deposits of late Pleistocene age from Biirgisser HM (1984) A unique mass flow marker bed in a
southern Britain. In: Collinson JD, Lewin J (eds) Mod- Miocene streamflow molasse sequence, Switzerland.
ern and ancient fluvial systems. Int Assoc Sedimentol In: Koster EH, Steel RJ (eds) Sedimentology of gravels
Spec Publ6: 267-275 and conglomerates. Can Soc Petrol Geol Mem 10:
Bucher WH (1919) On ripples and related sedimentary 147-163
surface forms and their paleogeographic interpreta- Burke K, Sengor C (1986) Tectonic escape in the evolution
tion. Am J Sci 47: 149-210,241-269 of the continental crust. In: Barazangi M, Brown L (eds)
Buck SG (1983) The Saiplaas Quartzite Member: a braided Reflection seismology: the continental crust. Am
system of gold and uranium bearing channel placers Geophys Union Geodyn Ser 14: 41-53
within the Proterozoic WitwatersraDd Supergroup of
References 531
BurnettAW, Schumm SA (1983) Alluvial-river response to Carson (1984c) Observations on the meandering-braided
neotectonic deformation in Louisiana and Mississippi. river transition, the Canterbury Plains, New Zealand. N
Science 222: 49-50 Z Geogr 40 (2): 89-99
Busby q, Ingersoll RV (1995) Tectonics of sedimentary Carson MA (1986) Characteristics of high-energy "mean-
basins. Blackwell Science, Oxford dering rivers" the Canterbury Plains, New Zealand.
Busch DA (1974) Stratigraphic traps in sandstones - explo- Geol Soc Am Bull 97: 886-895
ration techniques. Am Assoc Petrol Geol Mem 21: 174 Cartwright JA, Haddock RC, Pinheiro LM (1993) The lat-
Butcher SW (1990) The nickpoint concept and its implica- eral extent of sequence boundaries. In: Williams GD,
tions regarding onlap to the stratigraphic record. Dobbs A (eds) Tectonics and seismic sequence stratig-
In: Cross TA (ed) Quantitative dynamic stratigraphy. raphy. Geol Soc Lond Spec Publ 71: 15-34
Prentice-Hall, Englewood Cliffs, pp 375-385 Casshyap SM, Tewari RC (1982) Facies analysis and paleo-
Cairncross B (1980) Anastomosing river deposits: paleogeographic implications of a Late Paleozoic glacial
environmental control on coal quality and distribution, outwash deposit, Bihar, India. J Sediment Petrol 52:
northern Karoo Basin. Trans Geol Soc South Afr 83: 1243-1256
327-332 Castle JW (1990) Sedimentation in Eocene Lake Uinta
Cameron GIF, Collinson JD, Rider MH, Li Xu Analogue (Lower Green River Formation), northeastern Uinta
dipmeter logs through a prograding deltaic sandbody. Basin, Utah. In: Katz BJ (ed) Lacustrine basin explora-
In: Ashton M (ed) Advances in reservoir geology. Geo- tion. Am Assoc Petrol Geol Mem 50: 243-263
logical Society, London, pp 195-217 (Spec Pub169) Cavazza W (1989) Sedimentation pattern of a rift-filling
Campbell CV (1976) Reservoir geometry of a fluvial sheet unit, Tesuque Formastion (Miocene), Espanola Basin,
sandstone. Am Assoc Petrol Geol Bull 60: 1009-1020 Rio Grande Rift, New Mexico. J Sediment Petrol 59:
Campbell JE, Hendry HE (1987) Anatomy of a gravelly 287-296
meander lobe in the Saskatchewan River, near Cecil CB (1990) Paleoclimate controls on stratigraphic
Nipawin, Canada. In: Ethridge FG, Flores RM, Harvey repetition of chemical and siliciclastic rocks. Geology
MD (eds) Recent developments in fluvial sedimentol- 18:533-536
ogy. Soc Econ Paleontol Mineral Spec Pub139: 179-189 Cerveny PF, Naeser ND, Zeitler PK, Naeser CW, Johnson
Cant DJ (1976) Braided stream sedimentation in the South NM (1988) History of uplift and relief of the Himalaya
Saskatchewan River. PhD Thesis, McMaster Univer- during the past 18 million years: evidence from fission-
sity, Hamilton track ages of detrital zircons from sandstones of the
Cant DJ (1978) Development of a facies model for sandy Siwalik Group. In: Kleinspehn KL, Paola C (eds) New
braided river sedimentation: comparison of the South perspectives in basin analysis. Springer, Berlin Heidel-
Saskatchewan River and Battery Point Formation. In: berg New York, pp 43-61
Miall AD (ed) Fluvial sedimentology. Can Soc Petrol Chamberlin TC, Salisbury RD (1909) Geology: processes
Geol Mem 5: 627-640 and their results, second edition. Murray, London
Cant DJ (1982) Fluvialfacies models. In: Scholle PA, Spear- Chapin MA, Mayer DF (1991) Constructing a three-dimen-
ing D (eds) Sandstone depositional environments. Am sional rock-property model of fluvial sandstones in the
Assoc Petrol Geol Mem 31: 115-137 Peoria field, Colorado. In: Miall AD, Tyler N (eds) The
Cant DJ, Stockmal GS (1989) The Alberta foreland basin: three-dimensional facies architecture of terrigenous
relationship between stratigraphy and terrane-accre- clastic sediments, and its implications for hydrocarbon
tion events. Can J Earth Sci 26: 1964-1975 discovery and recovery. Soc Econ Paleontol Mineral
Cant DJ, Walker RG (1976) Development of a braided- Conc Models Ser 3: 160-171
fluvial facies model for the Devonian Battery Point Charles HH (1941) Bush City oil field, Anderson County,
Sandstone, Quebec. Can J Earth Sci 13: 102-119 Kansas. In: Levorsen AI (ed) Stratigraphic type oil
Cant DJ, Walker RG (1978) Fluvial processes and facies fields. American Association of Petroleum Geologists,
sequences in the sandy braided South Saskatchewan Tulsa, pp 43-56
River, Canada. Sedimentology 25: 625-648 Chatfield J (1972) Case history of Red Wash field, Uintah
Carey WC, Keller MD (1957) Systematic changes in the County, Utah. In: King RE (ed) Stratigraphic oil and gas
beds of alluvial rivers. Am Soc Civil Eng Proc 83 (HY 4): fields. Am Assoc Petrol Geol Mem 16: 342-353
1-24 Chawner WD (1935) Alluvial fan flooding, the Montrose,
Carling PA (1990) Particle over-passing on depth-limited California, flood of 1934. Geogr Rev 25: 225-263
gravel bars. Sedimentology 37: 345-355 Chow N, James NP (1987) Cambrian grand cycles: a north-
Carling PA, Glaister MS (1987) Rapid deposition of sand ern Appalachian perspective. Geol Soc Am Bull 98:
and gravel mixtures downstream of a negative step: the 418-429
role of matrix-infilling and particle-overpassing in the Church M (1983) Pattern of instability in a wandering
process of bar-front accretion. J Geol Soc Lond 144: gravel bed channel. In: Collinson JD, Lewin J(eds)
543-551 Modern and ancient fluvial systems. Int Assoc Sedi-
Carlston CW (1965) The relation offree meander geometry mentol Spec Publ6: 169-180
to stream discharge and its geomorphic implications. Church M, Gilbert R (1975) Proglacial fluvial and lacus-
Am J Sci 263: 864-885 trine environments. In: Jopling AV, McDonald BC
Carson MA (1984a) The meandering-braided river thresh- (eds) Glaciofluvial and glaciolacustrine sedimentation.
old: a reappraisal. J Hydrol 73: 315-334 Soc Econ Paleontol Mineral Spec Publ 23: 22-100
Carson MA (1984b) Observations on the meandering- Church M, Rood K (1983) Catalogue of alluvial river chan-
braided river transition, the Cant~rbury Plains, New nel regime data. Department of Geography, University
Zealand. N Z Geogr 40 (1): 12-17 of British Columbia, 99 p
532 References
Church M, Ryder JM (1972) Paraglacial sedimentation: a Collinson JD (1971b) Some effects of ice on a river bed. J
consideration of fluvial processes conditioned by gla- Sediment Petrol 41: 557-564
ciation. Geol Soc Am Bull 83: 3059-3072 Collinson JD (1978) Vertical sequence and sand body
Clark ABS, Thomas BM (1988) The intra-Latrobe playa: a shape in alluvial sequences. In: Miall AD (ed) Fluvial
case history from the Basker/Manta Block (Vic/PI9), sedimentology. Can Soc Petrol Geol Mem 5: 577-586
Gippsland Basin. Aust Petrol Explor Assoc J 28: Collinson JD (1986) Alluvial sediments. In: Reading HG
100-112 (ed) Sedimentary environments and facies. Blackwell
Clark SL (1987) Seismic stratigraphy of Early Pennsylva- Scientific, Oxford, pp 20-62
nian Morrowan sandstones, Minneola Complex, Ford Collinson JD, Lewin J (eds) (1983) Modern and ancient
and Clark Counties, Kansas. Am Assoc Petrol Geol Bull fluvial systems. Int Assoc Sedimentol Spec Publ6: 575
71: 1329-1341 Collinson JD, Thompson DB (1982) Sedimentary struc-
Clement WA (1977) A case history of geoseismic modeling tures. Allen and Unwin, London, 194 p
of basal Morrow-Springer Sandstones, Watonga- Conaghan PJ, Jones JG (1975) The Hawkesbury Sandstone
Chickasha trend: Geary, Oklahoma - T13N, RlO W. In: and the Brahmaputra: a depositional model for conti-
Payton CE (ed) Seismic stratigraphy - applications to nental sheet sandstones. J Geol Soc Aust 22: 275-283
hydrocarbon exploration. Am Assoc Petrol Geol Mem Conybeare CEB (1976) Geomorphology of oil and gas
26:451-476 fields in sandstone bodies. Elsevier Scientific, Amster-
Clemmensen LB, Tirsgaard H (1990) Sand-drift surfaces: a dam (Developments in petroleum science, vol 4)
neglected type of bounding surface. Geology 18: Conybeare CEB, Crook KAW (1968) Manual of sedimen-
1142-1145 tary structures. Aust Bur Miner Resources Geol
Clem mens en LB, 0xnevad lEI, De Boer PL (1994) Climatic Geophys Bull 102: 327
controls on ancient desert sedimentation: some late Conybeare WD, Phillips W (1824) Outline of the geology of
Palaeozoic and Mesozoic examples from NW Europe England and Wales. Am J Sci 7: 203
and the Western Interior of the USA. In: De Boer PL, Cornish V (1899) On kinematology. The study of waves
Smith DG (eds) Orbital forcing and cyclic sequences. and waves structures of the atmosphere, hydrosphere
Int Assoc Sedimentol Spec Pub119: 439-457 and lithosphere. Geogr J 13: 624-626
Cleveland MN, Molina J (1990) Deltaic reservoirs of the Cosgrove JL (1987) South-west Queensland gas - a re-
Cano Limon field, Colombia, South America. In: Barwis source for the future. Austral Petrol Explor Assoc J 27:
JH, McPherson JG, Studlick RJ (eds) Sandstone petro- 245-262
leum reservoirs: Springer, Berlin Heidelberg New York, Costa JE (1974) Stratigraphic, morphologic, and pedo-
p 281-315 genic evidence of large floods in humid environments.
Clifford HJ, Grund R, Musrati H (1980) Geology of a strati- Geology 2: 301-303
graphic giant: Messla oil field, Libya. In: Halbouty MT Costello WR, Walker RG (1972) Pleistocene sedimento-
(ed) Giant oil and gas fields of the decade: 1968-1978. logy, Credit River, southern Ontario: a new component
Am Assoc Petrol Geol Mem 30: 507-524 of the braided river model. J Sediment Petrol 42:
Cloetingh S (1988) Intraplate stress: a new element in basin 389-400
analysis. In: Kleinspehn K, Paola C (eds) New Perspec- Cotter E (1971) Sedimentary structures and the interpreta·
tives in basin analysis. Springer, Berlin Heidelberg New tion of paleo flow characteristics of the Ferron Sand··
York, pp 205-230 stone (Upper Cretaceous), Utah. J Sediment Petrol 41:
Cloetingh S, Kooi H (1990) Intraplate stresses: a new per- 129-138
spective on QDS and Vail's third-order cycles. In: Cross Cotter E (1978) The evolution of fluvial style, with special
TA (ed) Quantitative dynamic stratigraphy. Prentice- reference to the central Appalachian Paleozoic. In:
Hall, Englewood Cliffs, pp 127-148 Miall AD (ed) Fluvial sedimentology. Can Soc Petrol
Cloetingh S, McQueen H, Lambeck K (1985) On a tectonic Geol Mem 5: 361-383
mechanism for regional sea-level variations. Earth Cotter E, Graham JR (1991) Coastal plain sedimentation
Planet Sci Lett 75: 157-166 in the late Devonian of southern Ireland: hummocky
Cluzel D, Cadet J-p, Lapierre H (1990) Geodynamics ofthe cross-stratification in fluvial deposits. Sediment Geol
Ogcheon Belt (South Korea). Tectonophysics 183: 72:201-224
41-56 Covey M (1986) The evolution of foreland basins to steady
Cole RD, Friberg JF (1989) Stratigraphy and sedimentation state: the foreland basin of the Banda Orogen. In: Allen
of the Book Cliffs, Utah. In: Nummedal D, Wright R PA, Homewood P (eds) Foreland basins. Int Assoc
(eds) Cretaceous shelf sandstones and shelf deposi- Sedimentol (Spec PubI8), pp 77-90
tional sequences, Western Interior Basin, Utah, Colo- Cowan EJ (1991) The large-scale architecture of the fluvial
rado and New Mexico. 28th international geological Westwater Canyon Member, Morrison Formation (Ju-
congress, Washington DC, Field Trip Guidebook T119, rassic), San Juan Basin, New Mexico. In: Miall AD, Tyler
pp 13-24 N (eds) The three-dimensional facies architecture of
Colella A, Prior DB (eds) (1990) Coarse-grained deltas. Int terrigenous clastic sediments, and its implications for
Assoc Sedimentol Spec PubllO: 357 hydrocarbon discovery and recovery. Soc Econ
Coleman JM (1969) Brahmaputra river: channel processes Paleontol Mineral Conc Sedimentol Paleontol 3: 80-93
and sedimentation. Sediment Geol3: 129-239 Cowan EJ (1993) Longitudinal fluvial drainage patterns
Collinson JD (1970) Bedforms of the Tana river, Norway. within a foreland basin-fill: Permo-Triassic Sydney ba-
Geogr Ann 52A: 31-55 sin, Australia. Sediment Geo185: 557-577
Collinson JD (1971a) Current vector dispersion in a river Cowan G (1993) Identification and significance of aeolian
of fluctuating discharge. Geol Mijnbouw 50: 671-678 deposits within the dominantly fluvial Sherwood Sand-
References 533
stone Group of the East Irish Sea Basin, UK. In: North Dalrymple RW (1984) Morphology and internal structure
CP, Prosser DJ (eds) Characterization of fluvial and of sandwaves in the Bay of Fundy. Sedimentology 31:
aeolian reservoirs. Geol Soc Lond Spec Pub173: 231-245 365-382
Coward MP (1990) the Precambrian, Caledonian and Dalrymple RW, Boyd R, Zaitlin BA (eds) (1994) Incised-
Variscan framework to NW Europe. In: Hardman RFP, valley systems: origin and sedimentary sequences. Soc
Brooks J (eds) Tectonic events responsible for Britain's Econ Paleontol Mineral Spec Publ51
oil and gas reserves. Geol Soc Lond Special Pub155: 1-34 Dalrymple RW, Makino Y (1989) Description and genesis
Crews SG, Ethridge FG (1993) Laramide tectonics and hu- of tidal bedding in the Cobequid Bay-Salmon River
mid alluvial fan sedimentation, NE Uinta Uplift, Utah Estuary, Bay of Fundy, Canada. In: Taira A, Masuda F
and Wyoming. J Sediment Petrol 63: 420-436 (eds) Sedimentary facies in the active plate margin.
Crook KAW (1989) Suturing history of an allochthonous Terra Scientific, Tokyo, pp 151-177
terrane at a modern plate boundary traced by flysch-to- Damanti JF (1993) Geomorphic and structural controls
molasse transitions. Sediment Geo161: 49-80 on facies patterns and sediment composition in a
Cross TA (1986) Tectonic controls offoreland basin subsid- modern foreland basin. In: Marzo M, Puigdefabregas
ence and Laramide style deformation, western United C (eds) Alluvial sedimentation. Int Assoc Sedimentol
States. In: Allen PA, Homewood P (eds) Foreland basins. Spec Pub117: 221-233
Int Assoc Sedimentol Spec Publ8: 15-39 Dana JD (1850a) On denudation in the Pacific. Am J Sci 9:
Cross TA (ed) (1990) Quantitative dynamic stratigraphy. 48-62
Prentice Hall, Englewood Cliffs, New Jersey, 625 p Dana JD (1850b) On the degradation of the rocks of New
Crostella A (1983) Malacca Strait wrench fault controlled South Wales and formation of valleys. Am J Sci 9:
Laland and Mengkapan oil fields. SEAPEX Proc 6: 24-34 289-294
Crowell JC (1974a) Origin oflate Cenozoic basins in south- Dana JD (1862) Manual of geology. Ivison, Blakeman and
ern California. In: Dickinson WR (ed) Tectonics and Taylor, New York
sedimentation. Soc Econ Paleontol Mineral Spec Publ Darby DA, Whittecar GR, Barringer RA, Garrett JR (1990)
22: 190-204 Alluvial lithofacies recognition in a humid-tropical
Crowell JC (1974b) Sedimentation along the San Andreas setting. Sediment Geo167: 161-174
Fault, California. In: Dott RH, Shaver RH (eds) Modern Darwin GH (1883) On the formation of ripple marks. R
and ancient geosynclinal sedimentation. Soc Econ Soc Lond Proc 36: 18-43
Paleontol Mineral Spec Pub119: 292-303 Davies DK (1966) Sedimentary structures and subfacies of
Crowell JC (1978) Gondwanan glaciation, cyclothems, con- Mississippi River point bar. J Geol 74: 234-239
tinental positioning, and climate change. Am J Sci 278: Davies DK, Williams BPJ, Vessell RK (1991) Reservoir
1345-1372 models for meandering and straight fluvial channels:
Crowell JC, Link MH (eds) (1982) Geologic history of Ridge examples from the Travis Peak Formation, East Texas.
Basin, southern California. Pacific Section, Society of Trans Gulf Coast Assoc Geol Soc 41: 152-174
Economic Paleontologists and Mineralogists, Tulsa, Davies DK, Williams BPJ, Vessell RK (1993) Dimensions
Oklahoma and quality of reservoirs originating in low and high
Crowley KD (1983) Large-scale bed configurations (macro- sinuosity channel systems, Lower Cretaceous Travis
forms), Platte River basin, Colorado and Nebraska: pri- Peak Formation, East Texas, USA. In: North CP,
mary structures and formative processes. Geol Soc Am Prosser DJ (eds) Characterization of fluvial and
Bull 94: 117-133 aeolian reservoirs. Geol Soc Lond Spec Pub173: 95-121
Crumeyrolle P, Rubino J-L, Clauzon G (1991) Miocene Davis JL, Annan AP (1986) High resolution sounding
depositional sequences within a tectonically controlled using ground-probing radar. Geosci Can 13: 205-208
transgressive-regressive cycle. In: Macdonald DIM (ed) Davis JL, Annan AP (1989) Ground-penetrating radar for
(1991) Sedimentation, tectonics and eustasy: sea-level high-resolution mapping of soil and rock stratigraphy.
changes at active margins. Int Assoc Sedimentol Spec Geophys Prospect 37: 531-551
PubI12:373-390 Davis WM (1898a) The Triassic formations of Connecti-
Cuevas Gozalo MC, Martinius AW (1993) Outcrop data- cut. US Geol Surv Annu Rep 18 (2): 1-192
base for the geological characterization of fluvial reser- Davis WM (1898b) Physical geography. Ginn, Boston,
voirs: an example from distal fluvial fan deposits in the 432 p
Loranca Basin, Spain. In: North CP, Prosser DJ (ed) Davis WM (1899) The geographical cycle. Geogr J 14:
Characterization of fluvial and aeolian reservoirs. Geol 481-504
Soc Lond Spec Publ 73: 79-94 Davis WM (1900) The fresh-water Tertiary formations of
Cullingford RA, Davidson DA, Lewin J (1980) Timescales in the Rocky Mountains region. Proc Am Acad Arts Sci
geomorphology. Wiley, Chichester 35:345-373
Curray JR (1956) The analysis of two-dimensional orienta- Dawson MR, Bryant ID (1987) Three-dimensional facies
tion data. J Geol 64: 117-131 geometry in Pleistocene outwash sediments, Worces-
Curry WH, Curry WH III (1972) South Glenrock oil field, tershire, UK. In: Ethridge FG, Flores RM, Harvey MD
Wyoming: prediscovery thinking and postdiscovery de- (eds) Recent developments in fluvial sedimentology.
scription. In: King RE (ed) Stratigraphic oil and gas Soc Econ Paleontol Mineral Spec Pub139: 191-196
fields. Am Assoc Petrol Geol Mem 16: 415-427 De Boer PL, Smith DG (eds) (1994) Orbital forcing and
Dahlstrom CDA (1970) Structural geology in the eastern cyclic sequences. Int Assoc Sedimentol Spec Publ 19:
margin of the Canadian Rocky Mountains. Bull Can 559
Petrol Geo118: 332-406
534 References
De Boer PL, Smith DG (eds) (1994b) Orbital forcing Dickinson WR (1981a) Plate tectonics and the continental
and cyclic sequences. In: De Boer PL, Smith DG (eds) margin of California. In: Ernst WG (ed) The geo-
Orbital forcing and cyclic sequences. Int Assoc tectonic development of California. Prentice-Hall,
Sedimentol Spec Pub119: 1-14 Englewood Cliffs, New Jersey, pp 1-28
De Boer PL, Pragt JSJ, Oost AP (1991) Vertically persistent Dickinson WR (1981b) Plate tectonic evolution of the
facies boundaries along growth anticlines and climate- southern Cordillera. Arizona Geol Soc Dig 14: 113-l35
controlled sedimentation in the thrust-sheet-top South Dickinson WR, Klute MA, Hayes MJ, Janecke SU, Lundin
Pyrenean Tremp-Graus Foreland Basin. Basin Res 3: ER, McKittrick MA, Olivares MD (1988) Paleogeo-
63-78 graphic and paleotectonic setting of Laramide sedi-
DeCelles PG, Tolson RB, Graham SA, Smith GA, Ingersoll mentary basins in the central Rocky Mountain region.
RV, White J, Schmidt CJ, Rice R, Moxon I, Lemke L, Geol Soc Am Bull 100: 1023-1039
Handschy JW, Follow MF, Edwards DP, Cavazza W, Dickinson WR, Seely DR (1979) Structure and stratigraphy
Caldwell M, Bargar E (1987) Laramide thrust-generated of forearc regions. Am Assoc Petrol Geol Bull 63: 2-31
alluvial-fan sedimentation, Sphinx Conglomerate, Diemer JA, Belt ES (1991) Sedimentology and paleo-
southwestern Montana. Am Assoc Petrol Geol Bull 71: hydraulics of the meandering river systems of the Fort
l35-155 Union Formation, southeastern Montana. Sediment
DeCelles PG, Gray MB, Ridgway KD, Cole RB, Pivnik DA, Geol 75: 85-108
Pequera N, Srivastava P (1991) Controls on synoro- Diessel CFK (1992) Coal-bearing depositional systems.
genic alluvial-fan architecture, Beartooth Conglomer- Springer, Berlin Heidelberg New York
ate (Paleocene), Wyoming and Montana. Sedimen- Dixon EEL (1921) The geology of the South Wales
tology 38: 567-590 Coalfield, part XIII: the country around Tenby. Geo-
DeLuca JL, Eriksson KA (1989) Controls on synchronous logical Survey of Great Britain, Memoir
ephemeral- and perennial-river sediments in the Dobrin MB (1977) Seismic exploration for stratigraphic
middle sandstone member of the Triassic Chinle For- traps. In: Payton CE (ed) Seismic stratigraphy - appli-
mation, northeastern New Mexico. Sediment Geol 61: cations to hydrocarbon exploration. Am Assoc Petrol
155-175 Geol Mem 26: 329-351
DennyCS (1967) Fans and pediments. AmJ Sci265: 81-105 Doeglas DJ (1962) The structure of sedimentary deposits of
Derbyshire E, Owen LA (1990) Quaternary alluvial fans in braided rivers. Sedimentology 1: 167-190
the Karakoram Mountains. In: Rachocki AH, Church M Dolson 1, Muller D, Evetts MJ, Stein JA (1991) Regional
(eds) Alluvial fans: a field approach. Wiley, Chichester, paleotopographic trends and production, Muddy
pp 27-53 Sandstone (Lower Cretaceous), central and northern
Derksen SJ, McLean-Hodgson J (1988) Hydrocarbon po- Rocky Mountains. Am Assoc Petrol Geol Bull 75:
tential and structural style of continental rifts: ex- 409-435
amples from East Africa and southeast Asia. SEAPEX 8: Dorsey R1, Burns B (1994) Regional stratigraphy, sedimen-
47-62 tology, and tectonic significance of Oligocene-Miocene
Desikachar SV (1984) Exploration plays in north east sedimentary and volcanic rocks, northern Baja
Indian hydrocarbon bearing province. Petrol Asia J: Calfironia, Mexico. Sediment Geo188: 231-251
51-59 Dott RH Jr, Bourgeois J (1982) Hummocky stratification:
Desloges JR, Church M (1987) Channel and floodplain significance of its variable bedding sequences. Geol Soc
facies in a wandering gravel-bed river. In: Ethridge FG, Am Bull 93: 663-680
Flores RM, Harvey MD (eds) Recent developments in Dott RH Jr, Bourgeois J (1983) Hummocky stratification:
fluvial sedimentology. Soc Econ Paleontol Mineral Spec significance of its variable bedding sequences: reply to
Publ 39: 99-109 discussion by RG Walker et al. Geol Soc Am Bull 94:
Devine PE, Wheeler DM (1989) Correlation, interpreta- 1245-1251
tion, and exploration potential of Lower Wilcox valley- Dott RH Jr, Byers CW, Fielder GW, Stenzel SR, Winfree KE
fill sequences, Colorado and Lavaca counties, Texas. (1986) Aeolian to marine transition in Cambro-Ordovi-
Trans Gulf Coast Assoc Geol Soc 39: 57-74 cian cratonic sheet sandstones of the northern Missis-
Dewey JF (1977) Suture zone complexities: a review. sippi valley, USA. Sedimentology 33: 345-368
Tectonophysics 40: 53-67 Doyle JD, Sweet ML (1995) Three-dimensional distribu-
Dewey JF (1982) Plate tectonics and the evolution of the tion of lithofacies, bounding surfaces, porosity, and
British Isles. J Geol Soc Lond l39: 371-414 permeability in a fluvial sandstone - Gypsy sandstone
Dewey JF, Hempton MR, Kidd WSF, Saroglu F, Seng6r of northern Oklahoma. Am Assoc Petrol Geol Bull 79:
AMC (1986) Shortening of continental lithosphere: the 70-96
neotectonics of eastern Anatolia - a young collision Dranfield P, Begg SH, Carter RR (1987) wytch Farm
zone. In: Coward MP, Ries AC (eds) Collision tectonics. oilfield: reservoir characterization of the Triassisc
Geol Soc Lond Spec Pub119: 3-36 Sherwood Sandstone for input to reservoir simulation
Diaz-Molina M (1993) Geometry and lateral accretion studies. In: Brooks 1, Glennie KW (eds) Petroleum geol-
patterns in meander loops: examples from the Upper ogy of northwest Europe. Graham and Trotman, Lon-
Oligocene-Lower Miocene, Loranca Basin, Spain. In: don, pp 149-160
Marzo M, Puigdefabregas C (eds) Alluvial sedimenta- Drew F (1873) Alluvial and lacustrine deposits and glacial
tion. Int Assoc Sedimentol Spec Pub117: ll5-l31 records of the Upper-Indus Basin. QJ Geol Soc Lond 29:
Dickinson WR (1974) Plate tectonics and sedimentation. 441-471
In: Dickinson WR (ed) Tectonics and sedimentation. Dreyer T (1990) Sand body dimensions and inflll se-
Soc Econ Paleontol Mineral Spec Pub122: 1-27 quences of stable, humid-climate delta plain channels.
References 535
In: Buller AT, BergE, Hjelmeland 0, Kleppe J, Torsaeter Eckelmann WR, Dewitt RJ, Fisher WL (1975) Prediction of
0, Aasen JO (eds) North Sea oil and gas reservoirs II. fluvial-deltaic reservoir geometry, Prudhoe Bay field,
Graham and Trotman, London, pp 337-351 Alaska. In: Proceedings of the 9th world petroleum
Dreyer T (1993a) Geometry and facies of large-scale flow congress, vol 2: geology
units in fluvial-dominated fan-delta-front sequences. Eicher DL (1969) Paleobathymetry of the Cretaceous
In: Ashton M (ed) Advances in reservoir geology. Geol Greenhorn Sea in eastern Colorado. Am Assoc Petrol
Soc Lond Spec Pub169: l35-174 Geol Bull 53: 1075-1090
Dreyer T (1993b) Quantified fluvial architecture in ephem- Einsele G, Liu B, Durr S, Frisch W, Liu G, Luterbacher HP,
eral stream deposits of the Esplugafreda Formation Ratschbacher L, Ricken W, Wendt J, Wetzel A, Yu G,
(Palaeocene), Tremp-Graus Basin, northern Spain. In: Zheng H (1994) The Xigaze forearc basin: evolution and
Marzo M, Puigdefabregas C (eds) Alluvial sedimenta- facies architecture (Cretaceous, Tibet). Sediment Geol
tion. Int Assoc Sedimentol Spec Publl7: 337-362 90: 1-32
Dreyer T, Piilt L-M, H0Y T, Knarud R, Steel RJ, Cuevas J-L Einstein HA (1950) The bed-load function for sediment
(1993) Sedimentary architecture of field analogues for transportation in open channel flows. US Department
reservoir information (SAFARI): a case study of the of Agriculture, Washington DC (Soil conservation ser-
fluvial Escanilla Formation, Spanish Pyrenees. In: Flint vice technical bulletin 1026)
SS, Bryant ID (eds) The geological modelling of hydro- Eisbacher GH (1985) Pericollisional strike-slip faults and
carbon reservoirs and outcrop analogues. Int Assoc synorogenic basins, Canadian Cordillera. In: Biddle
Sedimentol Spec Publl5: 57-80 KT, Christie-Blick N (eds) Strike-slip deformation, ba-
Dubiel RF (1991) Architectural-facies analysis of non- sin formation and sedimentation. Soc Econ Paleontol
marine depositional systems in the Upper Triassic Mineral Spec Pub137: 265-280
Chinle Formation, southeastern Utah. In: Miall AD, Eisbacher GH, Carrigy MA, Campbell RB (1974)
Tyler N (eds) The three-dimensional facies architec- Paleodrainage pattern and late orogenic basins of the
ture of terrigenous clastic sediments, and its implica- Canadian Cordillera. In: Dickinson WR (ed) Tectonics
tions for hydrocarbon discovery and recovery. Soc and sedimentation. Soc Econ Paleontol Mineral Spec
Econ Paleontol Mineral Conc Sedimentol Paleontol 3: Pub122: 143-166
103-110 Ekes C (1993) Bedload-transported pedogenic mud aggre-
Dubiel RF, Parrish JT, Parrish JM, Good SC (1991) The gates in the Lower Old Red Sandstone in southwest
Pangaean megamonsoon - evidence from the upper Wales. J Geol Soc Lond 150: 469-472
Triassic Chinle Formation, Colorado Plateau. Palaios 6: Elliott L (1989) The Surat and Bowen basins. Aust Petrol
347-370 Explor Assoc J 29: 398-416
Du Boys PFD (1879) Le RhOne et Ie rivier a lit affouillable. Embry AF (1990a) A tectonic origin for third-order depo-
Ann Ponts Chaussees 18 (5): 141-195 sitional sequences in extensional basins - implications
Dueck RN, Paauwe EFW (1994) The use of borehole imag- for basin modelling. In: Cross TA (ed) Quantitative
ing techniques in the exploration for stratigraphic dynamic stratigraphy. Prentice-Hall, Englewood Cliffs,
traps: an example from the Middle Devonian Gilwood pp 491-501
channels in north-central Alberta. Bull Can Petrol Geol Embry AF (1990b) Geological and geophysical evidence in
42: l37-154 support of the hypothesis of anticlockwise rotation of
Dueholm KS, Olsen T (1993) Reservoir analog studies northern Alaska. Mar Geo193: 317-329
using multimodal photogrammetry: a new tool for the Embry AF (1991) Mesozoic history of the Arctic Islands.
petroleum industry. Am Assoc Petrol Geol Bull 77: In: Trettin HP (ed) Geology of the Innuitian Orogen
2023-2031 and Arctic Platform of Canada and Greenland. Geologi-
Dumont JF (1993) Lake patterns as related to neotectonics cal Survey of Canada. Geol Can 3: 369-433
in subsiding basins: the example of the Ucamara De- Epry C (19l3) Ripple marks. Annual report of the Smith-
pression, Peru. Tectonophysics 222: 69-78 sonian Institution, pp 307-318
Dumont JF, Fournier M (1994) Geodynamic environment Ethridge FG, Flores RM (eds) (1981) Recent and ancient
of Quaternary morphostructures of the Subandean nonmarine depositional environments: models for ex-
foreland basins of Peru and Bolivia: characteristics and ploration. Soc Econ Paleontol Mineral Spec Pub131: 349
study methods. Q Int 21: 129-142 Ethridge FG, Jackson TJ, Youngberg AD (1981) Floodbasin
Dunbar CO, Rodgers J (1957) Principles of stratigraphy. sequence of a fine-grained meander belt subsystem: the
Wiley, New York coal-bearing Lower Wasatch and Upper Fort Union
Dury GH (1964) Principles of underfit streams. US Geo- Formations, southern Powder River Basin, Wyoming.
logical Survey professional paper 452-A In: Ethridge FG, Flores RM (eds) Recent and ancient
Eaves E (1976) Citronelle oil field, Mobile county, Ala- nonmarine depositional environments, models for ex-
bama. In: Braunstein J (ed) North American oil and gas ploration. Soc Econ Paleontol Mineral Spec Publ 31:
fields. Am Assoc Petrol Geol Mem 24: 259-275 191-209
Ebanks WJ Jr, Weber JF (1982) Development of a shallow Ethridge FG, Flores RM, Harvey MD (eds) (1987) Recent
heavy-oil deposit in Missouri. Oil Gas J Sept 27: developments in fluvial sedimentology. Soc Econ
222-234 Paleontol Mineral Spec Publ 39
Eberth DA, Miall AD (1991) Stratigraphy, sedimentology Ethridge FG, Schumm SA (1978) Reconstructing paleo-
and evolution of a vertebrate-bearing, braided to anas- channel morphologic and flow characteristics: meth-
tomosed fluvial system, Cutler Formation (Permian- odology, limitations and assessment. In: Miall AD (ed)
Pennsylvanian), north-central New Mexico. Sediment Fluvial sedimentology. Can Soc Petrol Geol Mem 5:
Geol 72: 225-252 703-721
536 References
Eugster HP, Hardie LA (1975) Sedimentation in an ancient Tertiary of the Betic Cordillera, Spain. Sedimentology
playa-lake complex: the Wilkins Peak Member of the 40:879-893
Green River Formation of Wyoming. Geol Soc Am Bull Fielding CR (1984) A coal depositional model for the
86:319-334 Durham Coal Measures ofNE England. J Geol Soc Lond
Evans JE (1991) Facies relationships, alluvial architecture, 141:919-932
and paleohydrology of a Paleogene humid-tropical al- Fielding CR (1986) Fluvial channel and overbank deposits
luvial fan system: Chumstick Formation, Washington from the Westphalian of the Durham Coalfield, NE
State, USA. J Sediment Petrol 61: 732-755 England. Sedimentology 33: 119-140
Evans JE, Terry DO Jr (1994) The significance of incision Fielding CR (ed) (1993a) Current research in fluvial sedi-
and fluvial sedimentation in the Basal White River mentology. Sediment Geo185: 1-656 (special issue)
Group (Eocene-Oligocene), Badlands of South Dakota, Fielding CR (1993b) A review of recent research in fluvial
USA. Sediment Geo190: 137-152 sedimentology. Sediment Geo185: 3-14
Everest R (1832) A quantitative study of stream transpor- Fielding CR, Crane RC (1987) An application of statistical
tation. J Asiatic Soc BengalI: 238-240 modelling to the prediction of hydrocarbon recovery
Eyles N (1993) Earth's glacial record and its tectonic set- factors in fluvial reservoir sequences. In: Ethridge FG,
ting. Earth Sci Rev 35: 1-248 Flores RM, Harvey MD (eds) Recent developments in
Eyles N, Eyles CH (1992) Glacial depositional systems. In: fluvial sedimentology. Soc Econ Paleontol Mineral Spec
Walker RG, James NP (eds) Facies models, response to PubI39:321-327
sea level change. Geological Association of Canada, St. Fielding CR, Falkner AJ, Scott SG (1993) Fluvial response
John's Newfoundland, pp 73-100 to foreland basin overftlling; the Late Permian Rangal
Eyles N, Eyles CH, Miall AD (1983) Lithofacies types and coal measures in the Bowen Basin, Queensland, Austra-
vertical proftle models; an alternative approach to the lia. Sediment Geol 85: 475-497
description and environmental interpretation of glacial Figueiredo AMF (1994) Reconcavo Basin, Brazil: a prolific
diamict and diamictite sequences. Sedimentology 30: intracontinental rift basin. In: Landon SM (ed) Interior
393-410 rift basins. Am Assoc Petrol Geol Mem 59: 157-203
Fairhead JD (1986) Geophysical controls on sedimentation Finch J (1823) Geological essay on the Tertiary formations
within the African Rift System. In: Frostick LE, Renaut in America. Am J Sci 7: 31-43
RW, Reid I, Tiercelin JJ (eds) Sedimentation in the Fischer AG (1986) Climatic rhythms recorded in strata.
African rifts. Geol Soc Lond Spec Pub125: 19-27 Annu Rev Earth Planet Sci 14: 351-376
Farrell KM (1987) Sedimentology and facies architecture Fischer AG, De Boer PL, Premoli Silva I (1990) Cyclo-
of overbank deposits of the Mississippi River, False stratigraphy. In: Ginsburg RN, Beaudoin B (eds) Creta-
River region, Louisiana. In: Ethridge FG, Flores RM, ceous resources, events and rhythms: Background and
Harvey MD (eds) Recent developments in fluvial sedi- plans for research. Kluwer Academic, Dordrecht, pp
mentology. Soc Econ Paleontol Mineral Spec Publ 39: 139-172
111-120 Fisher MJ (1984) Triassic. In: Glennie KW (ed) Introduc-
Farshori MZ, Hopkins JC (1989) Sedimentology and tion to the petroleum geology of the North Sea.
petroleum geology of fluvial and shoreline deposits Blackwell Scientific, Oxford, pp 85-101
of the Lower Cretaceous Sunburst Sandstone Member, Fisher WL, McGowen JH (1967) Depositional systems in
Mannville Group, southern Alberta. Bull Can Petrol the Wilcox Group of Texas and their relationship to
Geo137: 371-388 occurrence of oil and gas. Trans Gulf Coast Assoc Geol
Fenneman NM (1906) Floodplains produced without Soc 17: 105-125
floods. Am Geogr Soc Bull 38: 89-91 Fisk HN (1944) Geological investigation of the alluvial
Ferguson RI (1993) Understanding braiding processes in valley of the lower Mississippi River. Mississippi River
gravel-bed braided rivers: progress and unsolved prob- Commission, Vicksburg, Mississippi
lems. In: Best JL, Bristow CS (eds) Braided rivers. Geol Fisk HN (1947) Fine-grained alluvial deposits and their
Soc Lond Spec Publ 75: 73-87 effect on Mississippi River activity. Mississippi River
Ferguson RI, Werritty A (1983) Bar development and Commission, Vicksburg, Mississippi
channel changes in the gravelly River Feshie, Scotland. Fisk HN (1960) Recent Mississippi River sedimentation
In: Collinson JD, Lewin J (eds) Modern and ancient and peat accumulation: Compte rendu 4th congres
fluvial systems. Int Assoc Sedimentol Spec Publ 6: l'avancement des etudes de stratigraphie et de geologie
181-193 du Carbonifere, Heerlen 1958, vol 1, pp 187-199
Ferm JC, Cavaroc VV Jr (1968) A nonmarine sedimentary Fisk HN (1952) Geological investigations of the
model for the Allegheny Rocks of West Virginia. In: Atchafalaya basin and the problem of Mississippi River
Klein G, De V (ed) Late Paleozoic and Mesozoic conti- diversion. US Army Corps of Engineers, Waterways
nental sedimentation, northeastern North America. Experimental Station, Vicksburg, Misissippi (2 vols)
Geol Soc Am Spec Paper 106: 1-20 Flach PD, Mossop GD (1985) Depositional environments
Ferm JC, Staub JR (1984) Depositional controls of mine- of Lower Cretaceous McMurray Formation, Athabasca
able coal bodies. In: Rahmani RA, Flores RM (eds) Oil Sands, Alberta. Am Assoc Petrol Geol Bull 69:
Sedimentology of coal and coal-bearing sequences. Int 1195-1207
Assoc Sedimentol Spec Publ 7: 275-289 Flemming BW (1988) Zur Klassifikation subaquatischer,
Fernandez J, Bluck BJ, Viseras C (1993) The effects of stromungstransversaler Transportkorper. Boch Geol
fluctuating base level on the structure of alluvial fan Geotechn Arb 29: 44-47
and associated fan delta deposits: an example from the
References 537
Flint S (1985) Alluvial fan and playa sedimentation in an nomic and field applications. American Association of
Andean arid closed basin: the Pacencia Group, Petroleum Geologists Field Seminar, Tulsa, pp 312-343
Antofagasta Province, Chile. J Geol Soc Lond 142: Fouch TD, Lawton TF, Nichols DJ, Cashion WB, Cobban
533-546 WA (1983) Patterns and timing of synorogenic sedi-
Flint SS, Bryant ID (eds) (1993) The geological modelling mentation in Upper Cretaceous rocks of central and
of hydrocarbon reservoirs and outcrop analogues. Int northeast Utah. In: Reynolds M, Dolly E (eds) Mesozoic
Assoc Sedimentol Spec Pub115: 269 paleogeography of west-central United States. Society
Flores RM (1981) Coal deposition in fluvial paleoenviron- of Economic Paleontologists and Mineralogists, Rocky
ments of the Paleocene Tongue River Member of the Mountain section, symposium, vol 2, pp 305-334
Fort Union Formation, Powder River area, Powder Frakes LA (1979) Climates throughout geologic time.
River basin, Wyoming and Montana. In: Ethridge FG, Elsevier, Amsterdam
Flores RM (eds) Recent and ancient nonmarine deposi- Franczyk KJ, Pitman JK (1991) Latest Cretaceous non-
tional environments: models for exploration. Soc Econ marine depositional systems in the Wasatch Plateau
Paleontol Mineral Spec Pub131: 161-190 area: reflections of foreland to intermontane basin
Flores RM (ed) (1983a) Fluvial systems, their economic transitions. In: Chidsey TC Jr (ed) Geology of east-
and field applications. American Association of Petro- central Utah. Utah Geol Assoc Pub119: 77-93
leum Geologists Field Seminar, Tulsa Franklin EH, Clifton BB (1971) Halibut field, Southeastern
Flores RM (1983b) Basin analysis of coal-rich Tertiary Australia. Am Assoc Petrol Geol Bull 55: 1262-1279
fluvial deposits, northern Powder River Basin, Mon- Fraser AJ, Gawthorpe RL (1990) Tectono-stratigraphic de-
tana and Wyoming. In: Collinson JD, Lewin J (eds) velopment and hydrocarbon habitat of the Carbonifer-
Modern and ancient fluvial systems. Int Assoc Sedi- ous in northern England. In: Hardman RFP, Brooks J
mentol Spec Publ6: 501-515 (eds) Tectonic events responsible for Britain's oil and
Flores RM (1984) Comparative analysis of coal accumula- gas reserves. Geol Soc Lond Spec Publ 55: 49-86
tion in Cretaceous alluvial deposits, southern United Fraser GS, DeCelles PG (1992) Geomorphic controls on
States Rocky Mountain basins. In: Stott DF, Glass DJ sediment accumulation and margins of foreland ba-
(eds) The Mesozoic of Middle North America. Can Soc sins. Basin Res 4: 233-252
Petrol Geol Mem 9: 373-385 Frazier DE (1967) Recent deltaic deposits of the Missis-
Flores RM, Ethridge FG, Miall AD, Galloway WE, Fouch TD sippi River: their development and chronology. Trans
(1985) Recognition of fluvial depositional systems and Gulf Coast Assoc Geol Soc 17: 287-315
their resource potential. Soc Econ Paleontol Mineral Frazier DE (1974) Depositional episodes: their relation-
Short Course 19: 290 ship to the Quaternary stratigraphic framework in the
Flores RM, Hanley JH (1984) Anastomosed and associated northwestern portion of the Gulf Basin. Bureau of Eco-
coal-bearing fluvial deposits: upper tongue river mem- nomic Geology, University of Texas, geological circular
ber, Paleocene Fort Union Formation, northern Pow- 74-1
der River Basin, Wyoming, USA. In: Rahmani RA, Friedkin JF (1945) A laboratory study of the meandering
Flores RM (eds) Sedimentology of coal and coal-bear- of alluvial rivers. Mississippi River Commission,
ing sequences. Int Assoc Sedimentol Spec Publ 7: Vicksburg
85-103 Friedman GM (1967) Dynamic processes and statistical
Focke JW, van Popta J (1989) Reservoir evaluation of the parameters compared for size frequency distribution
Permian Gharif Formation, Sultanate of Oman. Society of beach and river sands. J Sediment Petrol 37: 327-354
of Petroleum Engineers paper 17978, pp 517-528 Friedman GM (1971) Distinction between dune, beach and
Folk RL (1966) A review of grain-size parameters. Sedi- river sands from their textural characteristics. J Sedi-
mentology 6: 73-93 ment Petrol 31: 514-529
Folk RL, Ward we (1957) Brazos River bar, a study in the Friend PF (1978) Distinctive features of some ancient river
significance of grain size parameters. J Sediment Petrol systems. In: Miall AD (ed) Fluvial sedimentology. Can
27:3-26 Soc Petrol Geol Mem 5: 531-542
Forbes DL (1983) Morphology and sedimentology of a Friend PF (1983) Towards the field classification of alluvial
sinuous gravel-bed channel system, lower Babbage architecture or sequence. In: Collinson JD, Lewin J
River, Yukon coastal plain, Canada. In: Collinson JD, (eds) Modern and ancient fluvial systems. Int Assoc
Lewin J (eds) Modern and ancient fluvial systems. Int Sedimentol Spec Publ6: 345-354
Assoc Sedimentol Spec Publ6: 195-206 Friend PF (1985) Molasse basins of Europe: a tectonic
Forgotson JM, Stark PH (1972) Well-data files and the assessment. Trans R Soc Edinb 76: 451-462
computer, a case history from northern Rocky Moun- Friend PF, Alexander-Marrack PD, Nicholson J, Yeats AK
tains. Am Assoc Petrol Geol Bull 56: 1114-1127 (1976) Devonian sediments of east Greenland 1.
Fortuin AR, de Smet MEM (1991) Rates and magnitudes of Meddelelser om Gronland, vol 206, no 1
late Cenozoic vertical movements in the Indonesian Friend PF, Johnson NM, McRae LE (1989) Time-level plots
Banda Arc and the distinction of eustatic effects. In: and accumulation patterns of sediment sequences.
Macdonald DIM (ed) Sedimentation, tectonics and Geol Mag 126: 491-498
eustasy: sea-level changes at active margins. Int Assoc Friend PF, Moody-Stuart M (1972) Sedimentation of
Sedimentol Spec Pub112: 79-89 the Wood Bay Formation (Devonian) of Spitsbergen:
Fouch TD (1983) Oil- and gas-bearing Upper Cretaceous regional analysis of a late orogenic basin. Norsk
and Paleogene fluvial rocks in central and northeast Polarinstitutt Skrifter no 157
Utah. In: Flores RM (ed) Fluvial systems: their eco-
538 References
Friend PF, Sinha R (1993) Braiding and meandering pa- Galloway WE, Hobday DK (1995) Terigenous clastic depo-
rameters. In: Best JL, Bristow CS (eds) Braided rivers. sitional systems, 2nd edition. Springer, Berlin Heidel-
Geol Soc Lond Spec Pub175: 105-111 berg New York (in press)
Friend PF, Slater MJ, Williams RC (1979) Vertical and Galloway WE, Hobday DK, Magara K (1982) Frio Forma-
lateral building of river sandstone bodies, Ebro Basin, tion of the Texas Gulf Coast Basin - depositional sys-
Spain. J Geol Soc Lond 136: 39-46 tems, structural framework, and hydrocarbon origin,
Frostick LE, Reid I (1977) The origin of horizontal laminae migration, distribution, and exploration potential.
in ephemeral stream channel fill. Sedimentology 24: University of Texas at Austin, Bureau of Economic
1-10 Geology, report of investigations 122
Frostick LE, Reid I (1989) Climatic versus tectonic controls Galloway WE, Williams TA (1991) Sediment accumulation
of fan sequences: lessons from the Dead Sea, Israel. J rates in time and space: Paleogene genetic stratigraphic
Geol Soc Lond 146: 527-538 sequences of the northwestern Gulf of Mexico. Geology
Frostick LE, Renaut RW, Reid I, Tiercelin JJ (eds) (1986) 19:986-989
Sedimentation in the African rifts. Geol Soc Lond Spec Gansser A (1964) Geology of the Himalayas. Interscience,
Pub125:382 London
Fulford MM, Busby CJ (1993) Tectonic controls on non- Gardiner S, Thomas DV, Bowering ED, McMinn LS (1990)
marine sedimentation in a Cretaceous fore-arc basin, A braided fluvial reservoir, Peco field, Alberta, Canada.
Baja California, Mexico. In: Frostick LE, Steel RJ (eds) In: Barwis JH, McPherson JG, Studlick RJ (eds) Sand-
Tectonic controls and signatures in sedimentary suc- stone petroleum reservoirs: Springer, Berlin Heidel-
cessions. Int Assoc Sedimentol Spec Pub120: 301-333 berg New York, pp 31-56
Fulthorpe CS (1991) Geological controls on seismic se- Garrison RK, Chancellor R (1991) Berwick field: the geo-
quence resolution. Geology 19: 61-65 logic half of the seismic stratigraphic story in the Lower
Gabela VH (1990) Exploration and geologic framework of Tuscaloosa Mississippi. Trans Gulf Coast Assoc Geol
the Cafio Limon oil filed, Llanos Orientales de Colom- Soc 41: 299-307
bia. In: Ericksen GE, Pinochet MTC, Reinemund JA Gawthorpe RL, Colella A (1990) Tectonic controls on
(eds) Geology of the Andes and its relation to hydrocar- coarse-grained delta depositional systems in rift ba-
bon and mineral resources, Circum-Pacific Council for sins. In: Colella A, Prior DB (eds) Coarse-grained del-
Energy and Mineral Resources. Earth Sci Ser 11: tas. Int Assoc Sedimentol Spec PubllO: 113-127
363-382 Gawthorpe RL, Collier REL, Alexander J, Bridge JS, Leeder
Gagliano SM, Van Beek JL (1970) Geologic and geomor- MR (1993) Ground penetrating radar: application to
phic aspects of deltaic processes, Mississippi delta sys- sandbody geometry and heterogeneity studies. In:
tem. In: Flores RM (ed) Hydrologic and geologic North CP, Prosser DJ (eds) Characterization of fluvial
studies of coastal Louisiana. Centre for Wetland Re- and aeolian reservoirs. Geol Soc Lond Spec Publ 73:
sources, Louisiana State University, report 1 421-432
Galay VI, Kellerhals R, Bray DI (1973) Diversity of river Geddes A (1960) The alluvial morphology of the Indo-
types in Canada. In: Fluvial processes and sedimenta- Gangetic Plains. Trans Inst Br Geogr 28: 253-276
tion: proceedings of the Hydrology Symposium, Geehan GW, Lawton TF, Sakurai S, Klob H, Clifton TR,
Edmonton, National Research Council, Canada, pp Inman KF, Nitzberg KE (1986) Geologic prediction of
217-250 shale continuity, Prudhoe Bay field. In: Lake LW,
Galloway WE (1975) Process framework for describing the Carroll HB Ir (eds) Reservoir characterization. Aca-
morphologic and stratigraphic evolution of the deltaic demic, Orlando, pp 63-82
depositional systems. In: Broussard ML (ed) Deltas, Geikie A (1882) Text book of geology. Macmillan, London
models for exploration. Houston Geological Society, Geikie A (1905) The founders of Geology (republished in
Houston, pp 87-98 1962 by Dover, New York)
Galloway WE (1980) Deposition and early hydrologic evo- Genik GJ (1993) Petroleum geology of Cretaceous-Tertiary
lution of Westwater Canyon wet alluvial-fan systems. rift basins in Niger, Chad, and Central African Repub-
New Mexico Bureau Mines Miner Resourc Mem 38: lic. Am Assoc Petrol Geol Bull 77: 1405-1434
59-69 George GT, Berry JK (1993) A new lithostratigraphy and
Galloway WE (1981) Depositional architecture of Ceno- depositional model for the Upper Rotliegend of the UK
zoic Gulf Coastal plain fluvial systems. In: Ethridge FG, sector of the southern North Sea. In: North CP, Prosser
Flores RM (eds) Recent and ancient nonmarine deposi- DJ (ed) Characterization of fluvial and aeolian reser-
tional environments: models for exploration. Soc Econ voirs. Geol Soc Lond Spec Pub173: 291-319
Paleontol Mineral Spec Pub131: 127-155 Germanoski D, Schumm SA (1993) Changes in braided
Galloway WE (1989a) Genetic stratigraphic sequences in river morphology resulting from aggradation and deg-
basin analysis I: architecture and genesis of flooding- radation. J Geo1101: 451-466
surface bounded depositional units. Am Assoc Petrol Gersib GA, McCabe PJ (1981) Continental coal-bearing
Geol Bull 73: 125-142 sediments of the Port Hood Formation (Carbonifer-
Galloway WE (1989b) Genetic stratigraphic sequences in ous), Cape Linzee, Nova Scotia, Canada. In: Ethridge
basin analysis II: application to northwest Gulf of FG, Flores RM (eds) Recent and ancient nonmarine
Mexico Cenozoic basin. Am Assoc Petrol Geol Bull 73: depositional environments: moods for exploration.
143-154 Soc Econ Paleontol Mineral Spec Pub131: 95-108
Galloway WE, Hobday DK (1983) Terrigenous clastic Ghignone JI, de Andrade G (1970, General geology and
depositional systems. Springer, Berlin Heidelberg New major oil fields of Reconcavo Basin, Brazil. In:
York
References 539
Halbouty MT (ed) Geology of giant petroleum fields. Gordon I, Heller PL (1993) Evaluating major controls on
Am Assoc Petrol Geol Mem 14: 337-358 basinal stratigraphy, Pine Valley, Nevada: implications
Gibling MR, Bird DJ (1994) Late Carboniferous cyclothems for syntectonic deposition. Geol Soc Am BulI105: 47 -55
and alluvial paleovalleys in the Sydney Basin, Nova Gomez B, Naff RL, Hubbell DW (1989) Temporal varia-
Scotia. Geol Soc Am Bull 106: 105-117 tions in bedload transport rates associated with the
Gibling MR, Rust BR (1990) Ribbon sandstones in the migration of bedforms. Earth Surface Proc Landforms
Pennsylvanian Waddens Cove Formation, Sydney 14: 135-156
Basin, Atlantic Canada: the influence of siliceous Gonzalez E (1990) Hydrocarbon resources in the coastal
duricrusts on channel-body geometry. Sedimentology zone of Chile. In: Ericksen GE, Pinochet MTC,
37:45-65 Reinemund JA (eds) Geology ofthe Andes and its rela-
Gibling MR, Rust BR (1993) Alluvial ridge-and-swale tion to hydrocarbon and mineral resources, Circum-
topography: a case study from the Morien Group of Pacific Council for Energy and Mineral Resources.
Atlantic Canada. In: Marzo M, Puigdefabregas C (eds) Earth Sci Ser 11: 383-404
Alluvial sedimentation. Int Assoc Sedimentol Spec Goodwin PW, Anderson EJ (1985) Punctuated aggra-
Pub117: 133-150 dational cycles: a general hypothesis of episodic strati-
Gibling MR, Wightman WG (1994) Palaeovalleys and graphic accumulation. J Geol 93: 515-533
proterozoan assemblages in a Late Carboniferous Gould HR (1970) The Mississippi Delta complex. In: Mor-
cyclothem, Sydney Basin, Nova Scotia. Sedimentology gan JP (ed) Deltaic sedimentation: modern and ancient.
41:699-719 Soc Econ Paleontol Mineral Spec Pub115: 3-30
Gibling MR, Calder JH, Ryan R, Can de Poll HW, Yeo GW Grabau AW (1906) Types of sedimentary overlap. Geol Soc
(1992) Late Carboniferous and Early Permian drainage Am Bull 17: 567-636
patterns in Atlantic Canada. Can J Earth Sci 29: 338-352 Grabau AW (1907) Types of cross-bedding and their strati-
Gilbert GK (1880) Land, sculpture, geology of the Henry graphic significance. Science 25: 295-296
Mountains. US Geographical and Geological Survey of Grabau AW (1913a) Principles of stratigraphy. Seiler, New
the Rocky Mountain Region, 2nd edn, 160 p York, 1185 p
Gilbert GK (1884) Ripple marks. Science 3: 375-376 Grabau AW (1913b) Early Paleozoic delta deposits of
Gilbert GK (1899) Ripple marks and cross-bedding. Geol North America. Geol Soc Am Bull 24: 399-528
Soc Am Bull 10: 135-140 Grabau AW (1917) Problems of the interpretation of sedi-
Gilbert GK (1914) The transportation of debris by running mentary rocks. Geol Soc Am Bull 28: 735-744
water. US Geological Survey professional paper 86 Gradzinski R, Gagol J, Slaczka A (1979) The Tumlin Sand-
Glennie KW (1970) Desert sedimentary environments. stone (Holy Cross Mts., central Poland): Lower Triassic
Elsevier, Amsterdam (Developments in sedimentology deposits of aeolian dunes and interdune areas. Acta
14) Geol Polon 29: 151-175
Glennie KW (1972) Permian Rotliegendes of northwest Graham SA, Dickinson WR, Ingersoll RV (1975) Hima-
Europe interpreted in light of modern desert sedimen- layan-Bengal model for flysch dispersal in the Appala-
tation studies. Am Assoc Petrol Geol Bull 56: 1048-1071 chian-Ouachita system. Geol Soc Am BulI86: 273-286
Glennie KW (1983) Lower Permian Rotliegend desert sedi- Griffith WM (1927) A theory of silt and scour. Institute of
mentation in the North Sea area. In: Brookfield ME, Civil Engineers Proceedings, Tulsa, Oklahoma, pp
Ahlbrandt TS (eds) Eolian sediments and processes. 223-314
Dev Sedimentol38: 521-541 Guccione MJ (1993) Grain-size distribution of overbank
Glennie KW (1987) Desert sedimentary environments, sediment and its use to locate channel positions. In:
present and past - a summary. Sediment Geol 50: Marzo M, Puigdefabregas C (eds) Alluvial sedimenta-
135-165 tion. Int Assoc Sedimentol Spec Pub117: 185-194
Gloppen TG, Steel RJ (1981) The deposits, internal struc- Gurnis M (1990) Bounds on global dynamic topography
ture and geometry in six alluvial fan-fan delta bodies from Phanerozoic flooding of continental platforms.
(Devonian - Norway) - a study in the significance of Nature 344: 754-756
bedding sequences in conglomerates. In: Ethridge FG, Gurnis M (1992) Long-term controls on eustatic and
Flores RM (eds) Recent and ancient nonmarine deposi- epeirogenic motions by mantle convection. GSA Today
tional environments: models for exploration. Soc Econ 2: 141-157
Paleontol Mineral Spec Pub131: 49-69 Gustavson TC (1974) Sedimentation on gravel outwash
Godin P (1991) Fining-upward cycles in the sandy-braided fans, Malaspina Glacier foreland, Alaska. J Sediment
river deposits ofthe Westwater Canyon Member (Up- Petrol 44: 374-389
per Jurassic), Morrison Formation, New Mexico. Sedi- Gustavson TC (1978) Bed forms and stratification types of
ment Geo170: 61-82 modern gravel meander lobes, Nueces River, Texas.
Gohain K, Parkash B (1990) Morphology of the Kosi Sedimentology 25: 401-426
megafan. In: Rachocki AH, Church M (eds) Alluvial Hack JT (1957) Studies of longitudinal stream proflles in
fans: a field approach. Wiley, New York, pp 151-178 Virginia and Maryland. US Geological Survey profes-
Gole CV, Chitale SV (1966) Inland delta building activity of sional paper 294-B
Kosi River. Journal of the Hydraulics Division. Proc Am Hacquebard PA, Donaldson JR (1969) Carboniferous coal
Soc Civil Eng 92 (HY2): 111-126 deposition associated with flood-plain and limnic envi-
Golin V, Smyth M (1986) Depositional environments and ronments in Nova Scotia. In: Dapples EC, Hopkins ME
hydrocarbon potential of the Evergreen Formation, (eds) Environments of coal deposition. Geol Soc Am
ATP 145P, Surat Basin, Queensland. Aust Petrol Expl Spec Paper 114: 143-191
Assoc J 26: 156-171
540 References
Hahmann P (1912) Die Bildung von Sandduenen bei level research: an integrated approach. Soc Econ
gleichmaBiger Stromung. Ann Phys 637-676 Paleontol Mineral Spec Pub142: 71-108
Hallam A (1963) Major epeirogenic and eustatic changes Hardie LA, Smoot JP, Eugster HP (1978) Saline lakes and
since the Cretaceous and their possible relationship to their deposits: a sedimentological approach. In: Matter
crustal structure. Am J Sci 261: 397-423 A, Tucker ME (eds) Modern and ancient lake sedi-
Hallam A (1984) Continental humid and arid zones during ments. Int Assoc Sedimentol Spec Publ2: 7-41
the Jurassic and Cretaceous. Palaeogeogr Palaeo- Harms JC (1966) Stratigraphic traps in valley fill, western
climatol Palaeoecol47: 195-223 Nebraska. Am Assoc Petrol Geol Bull 50: 2119-2149
Hamblin AP, Rust BR (1989) Tectono-sedimentary analy- Harms JC, Fahnestock RK (1965) Stratification, bed forms,
sis of alternate-polarity half-graben basin-fill succes- and flow phenomena (with an example from the Rio
sions: Late Carboniferous Horton Group, Cape Breton Grande). In: Middleton GV (ed) Primary sedimentary
Island, Nova Scotia. Basin Res 2: 239-255 structures and their hydrodynamic interpretation. Soc
Hambrey MJ, Harland WB (1981) Criteria for the identifi- Econ Paleontol Mineral Spec Publ12: 84-115
cation of glacigenic deposits. In: Hambrey MJ, Harland Harms JC, Mackenzie DB, McCubbin DG (1963) Stratifica-
WB (eds) Earth's Pre-Pleistocene glacial record. Cam- tion in modern sands of the Red River, Louisiana. J Geol
bridge University Press, Cambridge, pp 14-27 71:566-580
Hamilton DS, Galloway WE (1989) New exploration Harms JC, Southard JB, Spearing DR, Walker RG (1975)
techniques in the analysis of diagenetically complex Depositional environments as interpreted from pri-
reservoir sandstones, Sydney Basin, NSW. Aust Petrol mary sedimentary structures and stratification se-
Explor Assoc J 29: 235-257 quences. Soc Econ Paleontol Mineral Short Course 2
Hamilton DS, Tadros NZ (1994) Utility of coal seams as Harms JC, Southard JB, Walker RG (1982) Structures and
genetic stratigraphic sequence boundaries in non- sequences in clastic rocks. Soc Econ Paleontol Mineral
marine basins: an example from the Gunnedah Basin, Short Course 9
Australia. Am Assoc Petrol Geol Bull 78: 267-286 Harris AL, Fettes DJ (eds) (1988) The Caledonian-Appala-
Hamilton W (1979) Tectonics of the Indonesian region. US chian orogen. Geological Society, London (Spec Pap 38)
Geological Survey professional paper 1078 Harris PT (1988) Large-scale bedforms as indicators of
Hamilton W (1985) Subduction, magmatic arcs, and fore- mutually evasive sand transport and the sequential
land deformation. In: Howell DG (ed) Tectono- infilling of wide-mouthed estuaries. Sediment Geo157:
stratigraphic terranes of the circum-Pacific region. 273-298
Circum-Pacific Council for Energy and Mineral Re- Hartley AJ (1993) Sedimentological response of an alluvial
sources. Earth Sci Ser 1: 259-262 system to source area tectonism: the Seilao Member of
Hamilton W (1987) Crustal extension in the Basin and the Late Cretaceous to Eocene Purilactis Formation of
Range province, southwestern United States. In: Cow- northern Chile. In: Marzo M, Puigdefabregas C (eds)
ard MP, Dewey JF, Hancock PL (eds) Continental ex- Alluvial sedimentation. Int Assoc Sedimentol Spec
tensional tectonics. Geol Soc Lond Spec Publ 28: PubI17:489-500
155-176 Harvey AM (1990) Factors influencing Quaternary alluvial
Hamilton WS, Cameron CP (1986) Facies relationships fan development in southeast Spain. In: Rachocki AH,
and depositional environments of Lower Tuscaloosa Church M (eds) Alluvial fans: a field approach. Wiley,
Formation reservoir sandstones in the McComb field Chichester, pp 247-269
area, southwest Mississippi. Trans Gulf Coast Assoc Hastings JO Jr (1990) Coarse-grained meander-belt reser-
Geol Soc 36: 141 voirs, Rocky Ridge field, North Dakota. In: Barwis JH,
Hamlin KH, Cameron CP (1987) Sandstone petrology and McPherson JG, Studlick RJ (eds) Sandstone petroleum
diagenesis of Lower Tuscaloosa Formation reservoirs reservoirs. Springer, Berlin Heidelberg New York, pp
in the McComb and Little Creek field areas, southwest 57-84
Mississippi. Trans Gulf Coast Assoc Geol Soc 37: Haszeldine RS (1983a) Fluvial bars reconstructed from a
95-104 deep, straight channel, Upper Carboniferous coalfield
Hanneman DL, Wideman CJ (1991) Sequence stratigraphy of northeast England. J Sediment Petrol 53: 1233-1248
of Cenozoic continental rocks, southwestern Montana. Haszeldine RS (1983b) Descending tabular cross-bed sets
Geol Soc Am Bull 103: 1335-1345 and bounding surfaces from a fluvial channel in the
Hanneman DL, Wideman CJ, Halvorsen JW (1994) Calcic Upper Carboniferous coalfield of north-east England.
paleosols: their use in subsurface stratigraphy. Am In: Collinson JD, Lewin J (eds) Modern and ancient
Assoc Petrol Geol Bull 78: 1360-1371 fluvial systems. Int Assoc Sedimentol Spec Publ 6:
Happ SC, Rittenhouse G, Dobson GC (1940) Some prin- 449-456
ciples of accelerated stream valley sedimentation. US Hayden HH (1821) Geological essays, or an enquiry into
Department of Agriculture, technical bulletin 695 some of the geological phenomena to be found in vari-
Harms JC (1966) Stratigraphic traps in valley fill, western ous parts of America and elsewhere. Am J Sci 3: 47-57
Nebraska. Am Assoc Petrol Geol Bull 50: 2119-2149 Hayes BJR, Christopher JE, Rosenthal L, Los G, McKercher
Haq BU, HardenbolJ, Vail PR (1987) Chronologyoffluctu- B (1994) Cretaceous Mannville Group of the Western
ating sea levels since the Triassic (250 million years ago Canada Sedimentary Basin. In: Mossop GD, Shetsen I
to present). Science 235: 1156-1167 (eds) Geological atlas of the Western Canada Sedimen-
Haq BU, Hardenbol J, Vail PR (1988) Mesozoic and tary Basin. Canadian Society of Petroleum Geologists,
Cenozoic chronostratigraphy and cycles of sea-level Calgary, Alberta, pp 317-334
change. In: Wilgus CK, Hastings BS, Kendall CGSC, Hayes JB, Harms JC, Wilson T Jr (1976) Contrasts between
Posamentier HW, Ross CA, Van Wagoner)C (~ds) Sea- braided and meandering stream deposits, Beluga and
References 541
Sterling formations (Tertiary), Cook Inlet, Alaska. In: ancient fluvial systems. Int Assoc Sedimentol Spec Publ
Miller TP (ed) Recent and ancient sedimentary envi- 6:61-83
ronments in Alaska. Alaska Geological Society, An- Hider A (1882) Appendix D, Report of assistant engineer
chorage, Alaska, pp Jl-J27 Arthur Hider upon obervations at Lake Providence,
Hayes MO, Kana TW (eds) (1976) Terrigenous clastic November, 1879 to November, 1880. Report of the Mis-
depositional environments. Coastal Research Division, sissippi River Commission, pp 80-98
Department of Geology, University of South Carolina, Higham N (1963) A very scientific gentleman: the major
technical report ll-CRD achievements of Henry Clifton Sorby. Pergamon,
Heath R (1989) Exploration in the Cooper Basin. Aust Oxford
Petrol Explor Assoc J 29: 366-378 Hirst JPp (1991) Variations in alluvial architecture across
Heckel PH (1986) Sea-level curve for Pennsylvanian the Oligo-Miocene Huesca fluvial system, Ebro basin,
eustatic marine transgressive-regressive depositional Spain. In: Miall AD, Tyler N (eds) The three-dimen-
cycles along midcontinent outcrop belt, North sional facies architecture of terrigenous clastic sedi-
America. Geology 14: 330-334 ments, and its implications for hydrocarbon discovery
Hein FJ, Walker RG (1977) Bar evolution and development and recovery. Soc Econ Paleontol Mineral Conc
of stratification in the gravelly, braided, Kicking Horse Sedimentol Paleontol3: 111-121
River, British Columbia. Can J Earth Sci 14: 562-570 Hirst JPp, Blackstock CR, Tyson S (1993) Stochastic mod-
Heller PL, Gordon I (1994) Evaluating major controls on elling of fluvial sandstone bodies. In: Flint SS, Bryant ID
basinal stratigraphy, Pine Valley, Nevada: implications (eds) The geological modelling of hydrocarbon reser-
for syntectonic deposition: reply to Discussion. Geol voirs and outcrop analogues. Int Assoc Sedimentol
Soc Am Bull 106: 156-157 Spec Pub115: 237-252
Heller PL, Paola C (1989) The paradox of Lower Cretaceous Hjulstr6m F (1935) Studies in the morphological activity of
gravels and the initiation of thrusting in the Sevier rivers as illustrated by the River Fyris. Geol Inst
orogenic belt, United States Western Interior. Geol Soc Uppsala Bull 25: 221-528
Am Bull 101: 864-875 Hobbs WH (1906) Gaudix formation of Granada, Spain.
Heller PL, Paola C (1992) The large-scale dynamics of Geol Soc Am Bull 17: 285-294
grain-size variation in alluvial basins, 2: application to Hobday DK (1978) Fluvial deposits of the Ecca and Beau-
syntectonic conglomerate. Basin Res 4: 91-102 fort Groups in the eastern Karoo Basin, South Africa.
Heller PL, Angevine CL, Winslow NS, Paola C (1988) Two- In: Miall AD (ed) Fluvial sedimentology. Canadian
phase stratigraphic model of foreland-basin sequences. Society of Petroleum Geologists Memoir 5, pp 413-429
Geology 16: 501-504 Hobday DK, Woodruff CM Jr, McBride MW (1981)
Heller PL, Angevine CL, Paola C (1989) Comment on Paleotopographic and structural controls on non-ma-
"Thrusting and gravel progradation in foreland basins: rine sedimentation of the Lower Cretaceous Antlers
a test of post-thrusting gravel dispersal". Geology 17: Formation and correlatives, North Texas and south-
959-960 eastern Oklahoma. In: Ethridge FG, Flores RM (eds)
Heller PL, Beekman F, Angevine CL, Cloetingh SAPL Recent and ancient nonmarine depositional environ-
(1993) Cause of tectonic reactivation and subtle uplifts ments: models for exploration. Soc Econ Paleontol
in the Rocky Mountain region and its effect on the Mineral Spec Pub131: 71-87
stratigraphic record. Geology 21: 1003-1006 Hoey TB (1992) Temporal variations in bedload transport
Hempton MR, Dunne LA (1984) Sedimentation in pull- rates and sediment storage in gravel-bed rivers. Prog
apart basins: active examples in eastern Turkey. J Geol Phys Geogr 16: 319-338
92: 513-530 Hoffman PF (1991) Did the breakout of Laurentia turn
Herries RD (1993) Constrasting styles of fluvial-aeolian Gondwanaland inside-out? Science 252: 1409-1412
interaction at a downwind erg margin: Jurassic Hoffman PF, Grotzinger IP (1993) Orographic precipita-
Kayenta-Navajo transition, northeastern Arizona. In: tion, erosional unloading, and tectonic style. Geology
North CP, Prosser DJ (ed) Characterization of fluvial 21: 195-198
and aeolian reservoirs. Geol Soc Lond Spec Publ 73: Hogg MD (1988) Newtonia field: a model for mid-dip
199-218 Lower Tuscaloosa retrograde deltaic sedimentation.
Hersch JB (1987) Exploration methods - Lower Tuscaloosa Trans Gulf Coast Assoc Geol Soc 38: 461-471
trend, southwest Mississippi. Trans Gulf Coast Assoc H0imyr 0, Kleppe A, Nystuen JP (1993) Effects ofhetero-
Geol Soc 37: 105-112 geneities in a braided stream channel sandsbody on the
Heward AP (1978a) Alluvial fan sequence and mega- simulation of oil recovery: a case study from the Lower
sequence models: with examples from Westphalian D - Jurassic Statfjord Formation, Snorre field, North Sea.
Stephanian B coalfields, northern Spain. In: Miall AD In: Ashton M (ed) Advances in reservoir geology. Geol
(ed) Fluvial sedimentology. Can Soc Petrol Geol Mem 5: Soc Lond Spec Pub169: 105-134
669-702 Holbrook JM, Dunbar RW (1992) Depositional history of
Heward AP (1978b) Alluvial fan and lacustrine sediments Lower Cretaceous strata in northeastern New Mexico:
from the Stephanian A and B (La Magdalena, Ciiiera - implications for regional tectonics and depositional
Matallana and Sabero) coalfields, northern Spain. Sedi- sequences. Geol Soc Am Bull 104: 802-813
mentology 25: 451-488 Holdsworth BK, Collinson JD (1988) Millstone Grit
Hickin EJ (1969) A newly identified process of point bar cyclicity revisited. In: Besly BM, Kelling G (eds) Sedi-
formation in natural streams. Am J Sci 267: 999-1010 mentation in a synorogenic basin complex: the Upper
Hickin EJ (1983) River channel changes: retrospect and Carboniferous of northwest Europe. Blackie, Glasgow,
prospect. In: Collinson JD, Lewin J (eds) Modern and pp 132-152
542 References
Holmes A (1965) Principles of physical geology, rev edn. Huggett RJ (1991) Climate, earth processes and earth his-
Nelson, London tory. Springer, Berlin Heidelberg New York
Holmes DA (1968) The recent history of the Indus. Geogr J Humphreys AA, Abbot HL (1861) Report upon the physics
134: 367-382 and hydraulics of the Mississippi. US Army Corps of
Homewood P, Allen PA, Williams GD (1986) Dynamics of Topographical Engineers, professional paper 4
the Molasse basin of western Switzerland. In: Allen PA, Hunt AR (1882) On the formation of ripple marks. R Soc
Homewood P (eds) Foreland basins. Int Assoc Sedi- Lond Proc 34: 1-19
mentol Spec Publ8: 199-217 Hunt D, Tucker ME (1992) Stranded parasequences and
Hooke JM (1986) The significance of mid-channel bars in the forced regressive wedge systems tract: deposition
an active meandering river. Sedimentology 33: 839-850 during base-level fall. Sediment Geo181: 1-9
Hooke R, Le B (1967) Processes on arid-region alluvial Hunter GM (1966) Red earth field, A and C pools. Oilfields
fans. J Geol 75: 438-460 of Alberta: supplement. Alberta Society of Petroleum
Hoover HC, Hoover LH (1950) Georgius Agricola: de re Geologists, Calgary, Alberta, pp 85-86 (Spec Publ)
metallica (translated from the first Latin Edition of Hunter RE, Richmond BM (1988) Daily cycles in coastal
1556). Dover, New York dunes. Sediment Geol 55: 43-67
Hopkins JC (1981) Sedimentology of quartzose sandstones Hunter RE, Rubin DM (1983) Interpreting cyclic cross-
of Lower Mannville and associated units, Medicine bedding, with an example from the Navajo Sandstone.
River area, central Alberta. Bull Can Petrol Geol 29: In: Brookfield ME, Ahlbrandt TS (eds) Eolian sedi-
12-41 ments and processes. Elsevier, Amsterdam. Dev
Hopkins JC (1985) Channel-fill deposits formed by aggra- Sedimentol 38: 429-454
dation in deeply scoured, superimposed distributaries Hutchison CS (1989) Geological evolution of south-east
of the Lower Kootenai Formation. J Sediment Petrol 55: Asia. Oxford Monogr Geol Geophys 13
42-52 Hutton J (1795) Theory of the earth, 2 vols (reprinted
Hopkins JC, Hermanson SW, Lawton DC (1982) Morphol- 1959). Engelmann, Wheldon and Wesley, Weinheim
ogy of channel-sand bodies in the Glauconitic Sand- Imbrie J (1985) A theoretical framework for the Pleis-
stone member (Upper Mannville), Little Bow area, tocene ice ages. J Geol Soc Lond 142: 417-432
Alberta. Bull Can Petrol Geo130: 274-285 Ingersoll RV (1988) Tectonics of sedimentary basins. Geol
Hopkins JC, Wood JM, Krause FF (1991) Waterflood re- Soc Am Bull 100: 1704-1719
sponse of reservoirs in an estuarine valley fill: Upper Inman DL (1949) Sorting of sediments in the light of fluid
Mannville G, U, and W pools, Little Bow field, Alberta. mechanics. J Sediment Petrol 19: 51-70
Am Assoc Petrol Geol Bull 75: 1064-1088 Inman DL (1952) Measures for describing size distribution
Horne JC, Ferm JC (1976) Carboniferous depositional en- of sediments. J Sediment Petrol 22: 125-145
vironments in the Pocahontas Basin, eastern Kentucky Iseya F, Ikeda H (1989) Sedimentation in coarse-grained
and southern West Virginia, Guidebook. Department sand-bedded meanders: distinctive deposition of sus-
of Geology, University of South Carolina pended sediment. In: Taira A, Masuda F (eds) Sedimen-
Horne JC, Ferm JC, Caruccio FT, Baganz BP (1978) Depo- tary facies in the active plate margin. Terra Scientific,
sitional models in coal exploration and mine planning Tokyo, pp 81-112
in Appalachian region. Am Assoc Petrol Geol Bull 62: Ito M, Masuda F (1986) Evolution of clastic piles in an arc-
2379-2411 arc collision zone. Late Cenozoic depositional history
Hossack JR (1984) The geometry oflistric growth faults in around the Tanzawa Mountains, Central Honshu, Ja-
the Devonian basins of Sunnfjord, W Norway. J Geol pan. Sediment Geo149: 223-259
Soc Lond 141: 629-638 Iwaniw E (1984) Lower Cantabrian basin margin deposits
Houbolt JJHC (1968) Recent sediments in the southern in NE Leon, Spain - a model for valley-ftll sedimenta-
Bight of the North Sea. Geologie Mijnbouw 47: 245-273 tion in a tectonically active, humid climatic setting.
Howell DG (1989) Tectonics of suspect terranes. Chapman Sedimentology 31: 91-110
and Hall, London Jackson JR (1834) Hints on the subject of geographical
Hoyt JH, Henry VJ Jr (1967) Influence of island migration arrangement and nomenclature. R Geogr Soc J 4: 72-88
on barrier-island sedimentation. Geol Soc Am Bull 78: Jackson RG II (1975) Hierarchical attributes and a unifying
77-86 model of bed forms composed of cohesionless material
Hsu KJ, Li J, Chen H, Wang Q, Sun S, Sengor AMC (1990) and produced by shearing flow. Geol Soc Am Bull 86:
Tectonics of South China: key to understanding West 1523-1533
Pacific geology. Tectonophysics 183: 9-39 Jackson RG II (1976a) Large scale ripples of the lower
Hubert JF, Filipov AJ (1989) Debris-flow deposits in allu- Wabash River. Sedimentology 23: 593-632
vial fans on the west flank of the White Mountains, Jackson RG II (1976b) Depositional model of point bars in
Owens Valley, California, USA. Sediment Geol 61: the lower Wabash River. J Sediment Petrol 46: 579-594
177-206 Jackson RG II (1976c) Sedimentological and fluid dynamic
Hubert JF, Reed AA, Carey PJ (1976) Paleogeography of the implications of the turbulent bursting phenomenon in
East Berlin Formation, Newark Group, Connecticut geophysical flows. J Fluid Mech 77: 531-560
Valley. Am J Sci 276: 1183-1207 Jackson RG II (1976d) Velocity - bed-form - texture pat-
Huggenberger P (1993) Radar facies: recognition offacies terns of meander bends in the lower Wabash River of
patterns and heterogeneities within Pleistocene Rhine Illinois and Indiana. Geol Soc Am Bull 86: 1511-1522
gravels, NE Switzerland. In: Best JL, Bristow CS (eds) Jackson RG II (1978) Preliminary evaluation oflithofacies
Braided rivers. Geol Soc Lond Spec Pub175: 163-176 models for meandering alluvial streams. In: Miall AD
References 543
(ed) Fluvial sedimentology. Can Soc Petrol Geol Mem 5: Jolley EJ, Turner P, Williams GD, Hartley AJ, Flint S (1990)
543-576 Sedimentological response of an alluvial system to
Jackson RG II (1981) Sedimentology of muddy fine- Neogene thrust tectonics, Atacama Desert, northern
grained channel deposits in meandering streams of the Chile. J Geol Soc Lond 147: 769-784
American middle west. J Sediment Petrol 51: 1169-1192 Jones BG, Rust BR (1983) Massive sandstone facies in the
Jamieson TF (1860) On the drift and rolled gravel of the Hawkesbury Sandstone, a Triassic fluvial deposit near
North of Scotland. Q J Geol Soc Lond 16: 347-371 Sydney, Australia. J Sediment Petrol 53: 1249-1260
Jeffreys H (1929) On the transport of sediment by streams. Jones CM (1977) Effects of varying discharge regimes on
Cambr Philos Soc Proc 25: 272-276 bed-form sedimentary structures in modern rivers.
Jervey MT (1988) Quantitative geological modeling of Geology 5: 567-570
siliciclastic rock sequences and their seismic expres- Jones CM, McCabe PJ (1980) Erosion surfaces within giant
sion. In: Wilgus CK, Hastings BS, Kendall CGSC, fluvial cross-beds of the Carboniferous in northern
Posamentier HW, Ross CA, Van Wagoner JC (eds) Sea England. J Sediment Petrol 50: 613-620
level research - an integrated approach. Soc Econ Jopling AV (1963) Hydraulic studies on the origin of bed-
Paleontol Mineral Spec Pub142: 47-69 ding. Sedimentology 2: 115-121
Jerzykiewicz T, Sweet AR (1988) Sedimentological and Jopling AV (1965) Hydraulic factors controlling the shape
palynological evidence of regional climate changes in of laminae in laboratory deltas. J Sediment Petrol 35:
the Campanian to Paleocene sediments of the Rocky 777-791
Mountain Foothills, Canada. Sediment Geo159: 29-76 Jopling AV (1975) Early studies on stratified drift. In:
Jin Y, Liu D, Luo C (1985) Development of Daqing oil field Jopling AV, McDonald BC (eds) Glaciofluvial and
by waterflooding. J Petrol Technol37: 269-274 glaciolacustrine sedimentation. Soc Econ Paleontol
Jirik LA (1990) Reservoir heterogeneity in Middle Frio Mineral Spec Pub123: 4-21
fluvial sandstones: case studies in Seeligson field, Jim Jopling AV, Walker RG (1968) Morphology and origin of
Wells county, Texas. Trans Gulf Coast Assoc Geol Soc ripple-drift cross-lamination, with examples from the
49:335-351 Pleistocene of Massachusetts. J Sediment Petrol 38:
Jo HR, Rhee CW, Ryang WH (1994) Evaluating major 971-984
controls on basinal stratigraphy, Pine Valley, Nevada: Jordan DW, Pryor WA (1992) Hierarchicallevels ofhetero-
implications for syntectonic deposition: Discussion. geneity in a Mississippi River meander belt and appli-
Geol Soc Am Bull 106: 155-156 cation to reservoir systems. Am Assoc Petrol Geol Bull
Johansson CE (1963) Orientation of pebbles in running 76: 1601-1624
water. A laboratory study. Geogr Ann 45: 85-112 Jordan TE (1981) Thrust loads and foreland basin evolu-
John BH, Almond CS (1987) Lithostratigraphy of the tion, Cretaceous western United States. Am Assoc
Lower Eromanga Basin sequence in south-west Petrol Geol Bull 65: 2506-2520
Queensland. Aust Petrol Explor Assoc J 27: 196-214 Jordan TE (1995) Retroarc foreland and related basins. In:
Johnson AM (1970) Physical processes in geology. Free- Busby CJ, Ingersoll RV (eds) Tectonics of sedimentary
man, San Francisco basins. Blackwell Scientific, Oxford: 331-362
Johnson D (1932) Streams and their significance. J Geo140: Jordan TE, Flemings PB (1990) From geodynamic models
481-496 to basin flll- a stratigraphic perspective. In: Cross TA
Johnson GD (1977) Paleopedology of Ramapithecus-bear- (ed) Quantitative dynamic stratigraphy. Prentice-Hall,
ing sediments, North India. Geol Rundsch 66: 192-216 Englewood Cliffs, pp 149-163
Johnson GD, Vondra CF (1972) Siwalik sediments in a Jordan TE, Flemings PB (1991) Large-scale stratigraphic
portion of the Punjab re-entrant: the sequence at architecture, eustatic variation, and unsteady tec-
Haritalyangar Disrict, Bilaspur, HP. Himal Geol 3: tonism: a theoretical evaluation. J Geophys Res 96B:
118-144 6681-6699
Johnson GD, Raynolds RGH, Burbank DW (1986) Late Jordan TE, Flemings PB, Beer JA (1988) Dating thrust-fault
Cenozoic tectonics and sedimentation in the north- activity by use of foreland-basin strata. In: Kleinspehn
western Himalayan foredeep: I. Thrust ramping and KL, Paola C (eds) New perspectives in basin analysis.
associated deformation in the Potwar region. In: Allen Springer, Berlin Heidelberg New York, pp 307-330
PA, Homewood P (eds) Foreland basins. Int Assoc Karges HE (1962) Significance of Lower Tuscaloosa sand
Sedimentol Spec Publ 8: 273-291 patterns in southwest Mississippi. Trans Gulf Coast
Johnson NM, Stix 1, Tauxe L, Cerveny PF, Tahirkheli RAK Assoc Geol Soc 12: 171-173
(1985) Paleomagnetic chronology, fluvial processes, Karlsen DA, Larter S (1989) A rapid correlation method for
and tectonic implications of the Siwalik deposits near petroleum population mapping within individual pe-
Chinji Village, Pakistan. J Geo193: 27-40 troleum reservoirs: application to petroleum reservoir
Johnson NM, Jordan TE, Johnsson PA, Naeser CW (1986) description. In: Collinson JD (ed) Correlation in hydro-
Magnetic polarity stratigraphy, age and tectonic setting carbon exploration. Graham and Trotman, London, pp
of fluvial sediments in an eastern Andean foreland 77-85
basin, San Juan Province, Argentina. In: Allen PA, Katz BJ (ed) (1990) Lacustrine basin exploration: case
Homewood P (eds) Foreland basins. Int Assoc studies and modern analogs. Am Assoc Petrol Geol
Sedimentol Spec Publ8: 63-75 Mem50
Johnsson MJ, Basu A (1993) Processes controlling the Kauffman EG (1969) Cretaceous marine cycles of the
composition of clastic sediments. Geol Soc Am Spec Western Interior. Mountain Geol6: 227-245
Paper 284
544 References
Kauffman EG (1984) Paleobiogeography and evolutionary Kleinspehn KL (1988) Sedimentary basins in the context of
response dynamic in the Cretaceous Western Interior allochthonous terranes. In: Kleinspehn KL, Paola C
Seaway of North America. In: Westermann EG (ed) (eds) New perspectives in basin analysis: Springer, Ber-
Jurassic-Cretaceous biochronology and paleogeogra- lin Heidelberg New York, pp 295-305
phy of North America. Geol Assoc Can Spec Paper 27: Klicman DP, Cameron CP, Meylan MA (1988) Petrology
273-306 and depositional environments of Lower Tuscaloosa
Keighin CW, Fouch TD (1981) Depositional environments Formation (Upper Cretaceous) sandstones in the
and diagenesis of some nonmarine Upper Cretaceous North Hustler and Thompson field areas, southwest
reservoir rocks, Uinta Basin, Utah. In: Ethridge FG, Mississippi. Trans Gulf Coast Assoc Geol Soc 38: 47-58
Flores RM (eds) Recent and ancient nonmarine deposi- Klimetz MP (1983) Speculations on the Mesozoic plate
tional environments: models for exploration. Soc Econ tectonic evolution of eastern China. Tectonics 2:
Paleontol Mineral Spec Pub131: 109-125 l39-166
Kelly SB (1993) Cyclical discharge variations recorded in Klitgord KD, Schouten H (1986) Plate kinematics of the
alluvial sediments: an example from the Devonian of central Atlantic. In: Vogt PR, Tucholke BE (eds) The
southwest Ireland. In: North CP, Prosser DJ (eds) Char- Western North Atlantic region. Geological Society of
acterization of fluvial and aeolian reservoirs. Geol Soc America, Boulder, Colorado, pp 351-378 (Geology of
Lond Spec Pub173: 157-166 North America, vol M)
Kelly SB, Olsen HO (1993) Terminal fans - a review with Klovan JE (1966) The use of factor analysis in determining
reference to Devonian examples. Sediment Geol 85: depositional environments from grain-size distribu-
339-374 tions. J Sediment Petrol 36: 115-125
Kennedy JF (1963) The mechanics of dunes and antidunes Kluth CF, Coney PJ (1981) Plate tectonics of the ancestral
in erodible-bed channels. J Fluid Mech 16: 521-544 Rocky Mountains. Geology 9: 10-15
Kennedy RG (1895) The prevention of silting in irrigation Knight SH (1929) The Fountain and Casper Formations of
canals. Inst Civil Eng 119: 281-290 the Laramie Basin. University of Wyoming Publica-
Kerr DR (1990) Reservoir heterogeneity in the Middle Frio tions in Science. Geology 1: 1-82
Formation: case studies in Stratton and Agua Dulce Knoll MD, Rea J, Knight R, Grimm K, Beckie R (1994)
fields, Nueces County, Texas. Trans Gulf Coast Assoc Architectural-element analysis of ground penetrating
Geol Soc 49: 363-372 radar data: a multidisciplinary approach to aquifer
Kerr DR, Jirik LA (1990) Fluvial architecture and reservoir characterization. Symposium on the application of
compartmentalization in the Oligocene Middle Frio geophysics to environmental and engineering prob-
Formation, south Texas. Trans Gulf Coast Assoc Geol lems, Boston
Soc 49: 374-380 Kochel RC (1990) Humid fans of the Appalachian Moun-
Kindle EM (1911) Cross-bedding and absence of fossils tains. In: Rachocki AH, Church M (eds) Alluvial fans: a
considered as criteria of continental deposits. Am J Sci field approach. Wiley, Chichester, pp 109-129
32:225-230 Kochel RC, Johnson RA (1984) Geomorphology and sedi-
Kindle EM (1917) Recent and fossil ripple marks. Geologi- mentology of humid-temperate alluvial fans, central
cal Survey of Canada, museum bulletin 25 Virginia. In: Koster EH, Steel RJ (eds) Sedimentology of
King PB (1977) The evolution of North America, revedn. gravels and conglomerates. Can Soc Petrol Geol Mem
Princeton University Press, Princeton 10: 109-122
King WSH (1916) The nature and formation of sand Kocurek G (1981) Significance of interdune deposits and
ripples and dunes. Geogr J 46: 189-209 bounding surfaces in aeolian dune sands. Sedimentol-
Kirk M (1983) Bar developments in a fluvial sandstone ogy 28: 753-780
(Westphalian "A"), Scotland. Sedimentology 30: Kocurek G (1988) First-order and super bounding surfaces
727-742 in eolian sequences - bounding surfaces revisited.
Kirk RH (1980) Statfjord field: a North Sea giant. In: Sediment Geo156: 193-206
Halbouty MT (ed) Giant oil and gas fields of the decade: Kocurek G, Hunter RE (1986) Origin of polygonal fractures
1968-1978. Am Assoc Petrol Geol Mem 30: 95-116 in sand, uppermost Navajo and Page sandstones, Page,
Kirschbaum MA, McCabe PJ (1992) Controls on the accu- Arizona. J Sediment Petrol 56: 895-904
mulation of coal and on the development of anasto- Kocurek G, Knight J, Havholm K (1991) Outcrop and semi-
mosed fluvial systems in the Cretaceous Dakota regional three-dimensional architecture and recon-
formation of southern Utah. Sedimentology 39: struction of a portion of the eolian Page Sandstone
581-598 (Jurassic). In: MiallAD, Tyler N (eds) The three-dimen-
Kirschner CE, Lyon CA (1973) Stratigraphic and tectonic sional facies architecture of terrigenous clastic sedi-
development of Cook Inlet petroleum province. In: ments, and its implications for hydrocarbon discovery
Pitcher MG (ed) Arctic geology. Am Assoc Petrol Geol and recovery. Soc of Econ Paleontol and Mineral Conc
Mem 19: 396-407 Sedimentol Paleontol 3: 25-43
Klein G deV (1987) Current aspects of basin analysis. Kolb CR, Van Lopik JR (1966) Depositional environment
Sediment Geo150: 95-118 of the Mississippi River deltaic plain - southeastern
Klein G deV, Willard DA (1989) Origin of the Pennsylva- Louisiana. In: Shirley ML (ed) Deltas in their geologic
nian coal-bearing cyclothems of North America. Geo- framework. Houston Geological Society, Houston,
logy 17: 152-155 Texas, pp 17-61
Kleinspehn KL (1985) Cretaceous sedimentation and tec- Kominz MA, Bond GC (1991) Unusually large subsidence
tonics, Tyaughton-Methow Basin, southwestern Brit- and sea-level events during middle Paleozoic time: new
ish Columbia. Can J Earth Sci 22: 154-174
References 545
evidence supporting mantle convection models for su- Langbein WB, Schumm SA (1958) Yield of sediment in
percontinent assembly. Geology 19: 56-60 relation to mean annual precipitation. Trans Am
Koster EH (1978) Transverse ribs: their characteristics, Geophys Union 39: 1076-1084
origin and paleohydraulic significance. In: Miall AD Langford RP (1989) Fluvial-aeolian interactions, part I:
(ed) Fluvial sedimentology. Can Soc Petrol Geol Mem 5: modern systems. Sedimentology 36: 1023-1035
161-186 Langford RP, Bracken B (1987) Medano Creek, Colorado, a
Koster EH, Steel RJ (eds) (1984) Sedimentology of gravels model for upper-flow-regime fluvial deposition. J Sedi-
and conglomerates. Can Soc Petrol Geol Mem 10 ment Petrol 57: 863-870
Kranzler I (1966) Origin of oil in lower member of Tyler Langford RP, Chan MA (1989) Fluvial-aeolian interac-
Formation of central Montana. Am Assoc Petrol Geol tions, part II, ancient systems. Sedimentology 36:
Bull 50: 2245-2259 1037-1051
Kraus MJ (1987) Integration of channel and floodplain Larsen V, Steel RJ (1978) The sedimentary history of a
suites, II. Vertical relations of alluvial paleosols. J Sedi- debris flow-dominated, Devonian alluvial fan - a study
ment Petrol 57: 602-612 of textural inversion. Sedimentology 25: 37-59
Kraus MJ (1992) Mesozoic and tertiary paleosols. In: Mar- Lash GG (1990) The Shochary Ridge sequence, southeast-
tini IP, Chesworth W (eds) Weathering, soils and ern Pennsylvania - a possible Ordovician piggyback
paleosols. Elsevier, Amsterdam, pp 525-542 (Develop- basin fill. Sediment Geol 68: 39-53
ments in earth surface processes, no 2) Lawrence DA, Williams BPJ (1987) Evolution of drainage
Kraus MJ, Bown TM (1988) Pedofacies analysis; a new systems in response to Acadian deformation: the Devo-
approach to reconstructing ancient fluvial sequences. nian Battery Point Formation, eastern Canada. In:
Geol Soc Am Spec Paper 216: 143-152 Ethridge FG, Flores RM, Harvey MD (eds) Recent de-
Kraus MJ, Bown TM (1993) Short-term sediment accumu- velopments in fluvial sedimentology. Soc Econ
lation rates determined from Eocene alluvial paleosols. Paleontol Mineral Spec Pub139: 287-300
Geology 21: 743-746 Lawson AC (1913) The petrographic designation of allu-
Kraus MJ, Middleton LT (1987) Dissected paleotopogra- vial fan formations. Univ Calif Publ Dep G 7: 325-334
phy and base-level changes in a Triassic fluvial se- Lawton TF (1985) Style and timing of frontal structures,
quence. Geology 15: 18-21 thrust belt, central Utah. Am Assoc Petrol Geol Bull 69:
Krumbein WC (1934) Size frequency of sediments. J Sedi- 1145-1159
ment Petrol 4: 65-77 Lawton TF (1986a) Compositional trends within a clastic
Krumbein WC (1941) Measurement and geologic signifi- wedge adjacent to a fold-thrust belt: Indianola Group,
cance of shape and roundness of sedimentary particles. central Utah, USA. In: Allen PA, Homewood P (eds)
J Sediment Petrol 11: 64-72 Foreland basins. Int Assoc Sedimentol Spec Publ 8:
Kuenen PH (1957) Longitudinal filling of oblong sedimen- 411-423
tary basins. Geol Mijnbouw 18: 189-195 Lawton TF (1986b) Fluvial systems of the Upper Creta-
Kuenzi WD, Horst OH, McGehee RV (1979) Effect ofvolca- ceous Mesaverde Group and Paleocene North Horn
nic activity on fluvial-deltaic sedimentation in a mod- Formation, central Utah: a record of transition from
ern arc-trench gap, southwestern Guatemala. Geol Soc thin-skinned to thick-skinned in the foreland region.
Am Bull 90 (1): 827-838 In: Peterson JA (ed) Paleotectonics and sedimentation
Kvale EP, Vondra CF (1993) Effects of relative sea-level in the Rocky Mountain region, United States. Am Assoc
changes and local tectonics on a Lower Cretaceous Petrol Geol Mem 41: 423-442
fluvial to transitional marine sequence. Bighorn Basin, Lawton TF, Geehan GW, Voorhees BJ (1987) Lithofacies
Wyoming, USA. In: Marzo M, Puigdefabregas C (eds) and depositional environments of the Ivishak Forma-
Alluvial sedimentation. Int Assoc Sedimentol (Spec tion, Prudhoe Bay field. In: Tailleur IL, Weimer PJ (eds)
PubI17), pp 383-399 Alaskan North Slope geology. Alaska Geol Soc Econ
Lajoie J, Stix J (1992) Volcaniclastic rocks. In: Walker RG, Paleontol Mineral Pac Sect 1: 61-76
James NP (eds) Facies models, response to sea level Leblanc Smith G, Eriksson KA (1979) A fluvioglacial and
change. Geological Association of Canada, St. John's, glaciolacustrine deltaic depositional model for Permo-
Newfoundland, pp 101-118 Carboniferous coals of the northeastern Karoo Basin,
Lane EW (1935) Stable channels in erodible materials. South Africa. Palaeogeogr Palaeoclimatol Palaeoecol
Trans Am Soc Civil Eng 63: 123-142 27:67-84
Lane EW (1955) The importance of fluvial morphology in Leckie DA (1994) Canterbury Plains, New Zealand - impli-
hydraulic engineering. Am Soc Civil Eng Proc 81 (745): cations for sequence stratigraphic models. Am Assoc
1-17 Petrol Geol Bull 78: 1240-1256
Lane EW (1957) A study of the shape of channels formed by Leckie DA, Fox C, Tarnocai C (1989) Multiple paleosols of
natural streams flowing in erodible material. US Army the late Albian Boulder Creek Formation, British Co-
Corps of Engineers, Missouri River Division, Omaha, lumbia, Canada. Sedimentology 36: 307-323
Nebraska (Sediment series 9) Leddy JO, Ashworth PI, Best JL (1993) Mechanisms of
Lang SC (1993) Evolution of Devonian alluvial systems in anabranch avulsion within gravel-bed braided rivers:
an oblique-slip mobile zone - an example from the observations from a scaled physical model. In: Best JL,
Broken River Province, northeastern Australia. Sedi- Bristow CS (eds) Braided rivers. Geol Soc Lond Spec
ment Geo185: 501-535 Publ 75: 119-127
Langbein WB, Leopold LB (1966) River meanders - theory Lee RA (1982) Petroleum geology of the Malacca Strait
of minimum variance. US Geological Survey profes- contract area (Central Sumatra Basin). Proceedings of
sional paper 422-H
546 References
the 11th annual convention of the Indonesian Petro- Leopold LB, Bull WB (1979) Base level, aggradation, and
leum Association, vol 1, pp 243-263 grade. Proc Am Philos Soc 123: 168-202
Leeder MR (1973) Fluviatile fining-upward cycles and the Leopold LB, Maddock T Jr (1953) The hydraulic geometry
magnitude of palaeo channels. Geol Mag 110: 265-276 of stream channels and some physiographic implica-
Leeder MR (1975) Pedogenic carbonates and flood sedi- tions. US Geological Survey professional paper 252
ment accumulation rates: a quantitative model for arid- Leopold LB, Wolman MG (1957) River channel patterns;
zone lithofacies. Geol Mag 112: 257-270 braided, meandering, and straight. US Geological Sur-
Leeder MR (1978) A quantitative stratigraphic model for vey professional paper 282-B
alluvium, with special reference to channel deposit Leopold LB, Wolman MG (1960) River meanders. Geol Soc
density and interconnectedness. In: Miall AD (ed) Flu- Am Bull 71: 769-794
vial sedimentology. Can Soc Petrol Geol Mem 5: Leopold LB, Wolman MG, Miller JP (1964) Fluvial pro-
587-596 cesses in geomorphology. Freeman, San Francisco
Leeder MR (1982) Upper Paleozoic basins of the British Le Roux JP (1992) Determining the channel sinuosity of
Isles - Caledonide inheritance versus Hercynian plate ancient fluvial systems from paleocurrent data. J Sedi-
margin processes. J Geol Soc Lond 139: 479-491 ment Petrol 62, p. 283-291
Leeder MR (1983) On the interactions between turbulent Le Roux JP (1994) The angular deviation of paleocurrent
flow, sediment transport and bedform mechanics in directions as applied to the calculation of channel sinu-
channelized flows. In: Collinson JD, Lewin J (eds) Mod- osities. J Sediment Res A64: 86-87
ern and ancient fluvial systems. Int Assoc Sedimentol Levey RA (1978) Bed-form distribution and internal strati-
Spec Publ6: 5-18 fication of coarse-grained point bars, Upper Congaree
Leeder MR (1988) Recent developments in Carboniferous River, SC. In: Miall AD (ed) Fluvial sedimentology. Can
geology: a critical review with implications for the Brit- Soc Petrol Geol Mem 5: 105-127
ish Isles and N.W. Europe. Proc Geol Assoc 99: 73-100 Levorsen AI (1967) Geology of petroleum, second edition.
Leeder MR (1993) Tectonic controls upon drainage basin Freeman, San Francisco
development, river channel migration and alluvial ar- Lewis GW, Lewin J (1983) Alluvial cutoffs in Wales and the
chitecture: implications for hydrocarbon reservoir de- Borderlands. In: Collinson JD, Lewin J (eds) Modern
velopment and characterization. In: North CP, Prosser and ancient fluvial systems. Int Assoc Sedimentol Spec
DJ (eds) Characterization of fluvial and aeolian reser- Publ6: 145-154
voirs. Geol Soc Lond Spec Publ 73: 7-22 Lewin J (1976) Initiation of bed forms and meanders in
Leeder MR, Alexander J (1987) The origin and tectonic coarse-grained sediment. Geol Soc Am Bull 87: 281-285
significance of asymmetric meander belts. Sedimento- Li S, Finlayson B (1993) Flood management on the lower
logy 34: 217-226 Yellow River: hydrological and geomorphological per-
Leeder MR, Gawthorpe RL (1987) Sedimentary models for spectives. Sediment Geol 85: 285-296
extensional tilt-block/half-graben basins. In: Coward Link MH (1984) Fluvial facies of the Miocene Ridge Route
MP, Dewey JF, Hancock PL (eds) Continental extension Formation, Ridge Basin, California. Sediment Geol 38:
tectonics. Geol Soc Lond Spec Pub128: 139-152 263-286
Leeder MJ, Hardman M (1990) Carboniferous geology of Link MH, Osborne RH (1978) Lacustrine facies in the
the southern North Sea basin and controls on hydro- Pliocene Ridge Basin Group: Ridge Basin, California.
carbon prospectivity. In: Hardman RFP, Brooks J (eds) In: Matter A, Tucker ME (eds) Modern and ancient lake
Tectonic events responsible for Britain's oil and gas sediments. Int Assoc Sedimentol Spec Pub12: 169-187
reserves. Geol Soc Lond Spec Pub155: 87-105 Linsley PN, Potter HC, McNab G, Racher D (1980) The
Leeder MR, Jackson JA (1993) The interaction between Beatrice field, inner Moray Firth, U.K. North Sea. In:
normal faulting and drainage in active extensional ba- Halbouty MT (ed) Giant oil and gas fields of the decade:
sins, with examples from the western United States and 1968-1978. Am Assoc Petrol Geol Mem 30: 117-129
central Greece. Basin Res 5: 79-102 Liu H (1986) Geodynamic scenario and structural styles of
Leeder MR, Seger MJ, Stark CP (1991) Sedimentation and Mesozoic and Cenozoic basins in China. Am Assoc
tectonic geomorphology adjacent to major active and Petrol Geol Bull 70: 377-395
inactive normal faults, southern Greece. J Geol Soc Long DGF (1978) Proterozoic stream deposits: some prob-
Lond 148: 331-343 lems of recognition and interpretation of ancient sandy
Lees GM (1955) Recent earth movements in the Middle fluvial systems. In: Miall AD (ed) Fluvial sedimento-
East. Geol Rundsch 43: 221-226 logy. Can Soc Petrol Geol Mem 5: 313-342
Legarreta L, Gulisano CA (1989) Analisis estratgrafico Long DGF (1981) Dextral strike-slip faults in the Canadian
sequencial de la cuenca neuquina (Triasico superior- Cordillera and depositional environments of related
Terciario inferior). Cuencas Sediment Argent 6: fresh-water intermontane coal basins. In: Miall AD (ed)
221-243 Sedimentation and tectonics in alluvial basins. Geol
Legarreta L, Uliana MA (1991) Jurassic-Cretaceous marine Assoc Can Spec Paper 23: 153-186
oscillations and geometry of back-arc basin fIll, central Lopez-Gomez J, Arche A (1993) Architecture of the
Argentine Andes. In: Macdonald DIM (ed) Sedimenta- Caiiizar fluvial sheet sandstones, Early Triassic, Iberian
tion, tectonics and eustasy. Int Assoc Sedimentol (Spec Ranges, eastern Spain. In: Marzo M, Puigdefabregas C
PubI12), pp 429-450 (eds) Alluvial sedimentation. Int Assoc Sedimentol
Leliavsky S (1955) An introduction to fluid hydraulics. Spec Pub117: 363-381
Constable, London (reprinted 1966 by Dover, New Lorenz JC, Heinze DM, Clark JA, Searls CA (1985) Deter-
York) mination of widths of meander-belt sandstone reser-
References 547
voirs from vertical downhole data, Mesaverde Group, MacKay D (1945) Ancient river beds and dead cities. An-
Piceance Creek Basin, Colorado. Am Assoc Petrol Geol tiquity 19: 135-144
Bull 69: 7l0-721 Mackin JH (1937) Erosional history of the Big Horn Basin,
Lorenz JC, Warpinski NR, Branagan PT (1991) Subsurface Wyoming. Geol Soc Am Bull 48: 813-894
characterization of Mesaverde reservoirs in Colorado: Mackin JH (1948) Concept of the graded river. Geol Soc
geophysical and reservoir-engineering checks on pre- Am Bull 59: 463-512
dictive sedimentology. In: Miall AD, Tyler N (eds) The Maizels J (1989) Sedimentology, paleoflow dynamics and
three-dimensional facies architecture of terrigenous flood history of jokulhlaup deposits: paleohydrology of
clastic sediments, and its implications for hydrocarbon Holocene sediment sequences in southern Iceland
discovery and recovery. Soc Econ Paleontol Mineral sandur deposits. J Sediment Petrol 59: 204-223
Conc Sedimentol Paleontol3: 57-79 Maizels J (1990) Long-term paleochannel evolution during
Loutit TS, Hardenbol J, Vail PR, Baum GR (1988) Con- episodic growth of an exhumed Plio-Pleistocene
densed sections: the key to age dating and correlation alluvial fan, Oman. In: Rachocki AH, Church M (eds)
of continental margin sequences. In: Wilgus CK, Alluvial fans: a field approach. Wiley, Chichester, pp
Hastings BS, Kendall CGSC, Posamentier HW, Ross 27l-304
CA, Van Wagoner JC (eds) Sea-level research: an inte- Maizels J (1993) Lithofacies variations within sandur de-
grated approach. Soc Econ Paleontol Mineral Spec Publ posits: the role of runoff regime, flow dynamics and
42: 183-213 sediment suply characteristics. Sediment Geol 85:
Lucas PT, Drexler JM (1976) Altamont-Bluebell- a major, 299-325
naturally fractured stratigraphic trap, Uinta Basin, Manspeizer W (1985) The Dead Sea Rift: impact of climate
Utah. In: Braunstein J (ed) North American oil and gas and tectonism on Pleistocene and Holocene sedimenta-
fields. Am Assoc Petrol Geol Mem 24: 121-135 tion. In: Biddle KT, Christie-Blick N (eds) Strike-slip
Lucchitta I, Suneson N (1981) Flash flood in Arizona - deformation, basin formation and sedimentation. Soc
observations and their application to the identification Econ Paleontol Mineral Spec Pub137: 143-158
of flash-flood deposits in the geologic record. Geology Marple RT, Talwani P (1993) Evidence of possible tectonic
9:414-418 upwarping along the South Carolina coastal plain from
Lundberg N, Dorsey RJ (1988) Synorogenic sedimentation an examination of river morphology and elevation
and subsidence in a Plio-Pleistocene collisional basin, data. Geology 21: 651-654
eastern Taiwan. In: Kleinspehn K, Paola C (eds) New Marriott SB, Wright VP (1993) Palaeosols as indicators of
perspectives in basin analysis: Springer, Berlin Heidel- geomorphic stability in two Old Red Sandstone alluvial
berg New York, pp 265-280 suites, South Wales. J Geol Soc Lond 150: 1109-1120
Lustig LK (1965) Clastic sedimentation in Deep Springs Martin JH (1993) A review of braided fluvial hydrocarbon
Valley, California. US Geological Survey professional reservoirs: the petroleum engineer's perspective. In:
paper 352-F Best JL, Bristow CS (eds) Braided rivers. Geol Soc Lond
Luterbacher HP, Eichenseer H, Betzler C, Van den Hurk Spec Publ 75: 333-367
AM (1991) Carbonate-siliciclastic depositional systems Martin R (1966) Paleogeomorphology and its application
in the Paleogene of the south Pyrenean foreland basin: to exploration for oil and gas (with example from west-
a sequence-stratigraphic approach. In: Macdonald ern Canada). Am Assoc Petrol Geol Bull 50: 2277-2311
DIM (ed) Sedimentation, tectonics and eustasy: sea- Martini IP, Chesworth W (eds) (1992) Weathering, soils
level changes at active margins. Int Assoc Sedimentol and paleosols. Elsevier, Amsterdam (Developments in
Spec Pub112: 391-407 earth surface processes, no 2)
Luttrell PR (1993) Basinwide sedimentation and the con- Martini IP, Kwong JK, Sadura S (1993) Sediment rafting
tinuum of paleoflow in an ancient river system: and cold climate fluvial deposits: Albany River,
Kayenta Formation (Lower Jurassic), central portion Ontario. In: Marzo M, Puigdefabregas C (eds) Alluvial
Colorado Plateau. Sediment Geo185: 411-434 sedimentation. Int Assoc Sedimentol Spec Publ 17:
Luyendyk BP, Hornafius JS (1987) Neogene crustal rota- 63-76
tions, fault slip, and basin development in southern Martinsen OJ, Martinsen RS, Steidtmann JR (1993)
California. In: Ingersoll RV, Ernst WG (eds) Cenozoic Mesaverde Group (Upper Cretaceous), southeastern
basin development of coastal California (Rubey vol 6). Wyoming: allostratigraphy versus sequence stratigra-
Prentice-Hall, Englewood Cliffs, pp 259-283 phy in a tectonically active area. Am Assoc Petrol Geol
Lyell C (1830-1833) Principles of geology, 3 vols. Murray, Bull 77: 1351-1373
London (reprinted by Johnson, New York, 1969) Marzo M, Puigdefabregas C (eds) (1993) Alluvial sedimen-
Lyons PL, Dobrin MB (1972) Seismic exploration for tation. Int Assoc Sedimentol Spec Publ17
stratigraphic traps. In: King RA (ed) Stratigraphic oil Massari F (1983) Tabular cross-bedding in Messinian flu-
and gas fields. Am Assoc Petrol Geol Mem 16: 225-243 vial channel conglomerates, southern Alps, Italy. In:
MacDonald AC, Halland EK (1993) Sedimentology and Collinson JD, Lewin J (eds) Modern and ancient fluvial
shale modeling of a sandstone-rich fluvial reservoir: systems. Int Assoc Sedimentol Spec Publ6: 287-300
Upper Statfjord Formation, Statfjord field, North Sea. Massari F, Mellere D, Doglioni C (1993) Cyclicity in non-
Am Assoc Petrol Geol Bull 77: 1016-1040 marine foreland-basin sedimentary fill: the Messinian
Mack GH, James WC (1994) Paleoclimate and the global conglomerate-bearing succession of the Venetian Alps
distribution of paleosols. J Geo1102: 360-366 (Italy). In: Marzo M, Puigdefabregas C (eds) Alluvial
Mack GH, James WC, Monger HC (1993) Classification of sedimentation. Int Assoc Sedimentol Spec Publ 17:
paleosols. Geol Soc Am Bull 105: 129-136 501-520
548 References
Mastalerz K, Wojewoda J (1993) Alluvial-fan sedimenta- McGee WJ (1897) Sheetflood erosion. Geol Soc Am Bull 8:
tion along an active strike-slip fault: Plio-Pleistocene 87-112
pre-Kaczawa fan, SW Poland. In: Marzo M, McGowen JH, Garner LE (1970) Physiographic features
Puigdefabregas C (eds) Alluvial sedimentation. Int and stratification types of coarse-grained point bars;
Assoc Sedimentol Spec Pub117: 293-304 modern and ancient examples. Sedimentology 14:
Mather AE (1993) Basin inversion: some consequences for 77-112
drainage evolution and alluvial architecture. Sedimen- McKee BA, Nittrouer CA, Demaster DJ (1983) Concepts of
tology 40: 1069-1089 sediment deposition and accumulation applied to the
Mathisen ME, Vondra CF (1983) The fluvial and pyroclas- continental shelf near the mouth of the Yangtze River.
tic deposits of the Cagayan Basin, northern Luzon, Geology 11: 631-633
Philippines - an example of non-marine volcaniclastic McKee ED (1938) Original structures in Colorado River
sedimentation in an interarc basin. Sedimentology 30: flood deposits of Grand Canyon. J Sediment Petrol 8:
369-392 77-83
Matsuzawa A (1988) Oil and gas exploration in Bohai Bay, McKee ED (I939) Some types of bedding in the Colorado
China. In: Wagner HC, Wagner LC, Wang FFH, Wong River delta. J Geo147: 64-81
FL (eds) Petroleum resources of China and related McKee ED (1957) Flume experiments on the production of
subjects. Circum Pac Council Energy Miner Res Earth stratification and cross-stratification. J Sediment
Sci Ser 10: 263-275 Petrol 27: 129-134
May SR, Ehman KD, Gray GG, Crowell JC (1993) A new McKee ED, Weir GW (1953) Terminology for stratification
angle on the tectonic evolution of the Ridge Basin, a and cross-stratification in sedimentary rocks. Geol Soc
"strike-slip" basin in southern California. Geol Soc Am Am Bull 64: 381-389
Bull 105: 1357-1372 McKee ED, Crosby EJ, Berryhill HL Jr (1967) Flood depos-
McCabe PJ (1975) The sedimentology and stratigraphy of its, Bijou Creek, Colorado, June 1965. J Sediment Petrol
the Kinderscout Grit Group (Namurian, R) between 37:829-851
Wharfedale and Longdendale. PhD Thesis, University McKenzie DP (1978) Some remarks on the development of
of Keele sedimentary basins. Earth Planet Sci Lett 40: 25-32
McCabe PJ (1977) Deep distributary channels and giant McLean JR (1977) The Cadomin Formation: stratigraphy,
bedforms in the Upper Carboniferous of the central sedimentology, and tectonic implications. Bull Can
Pennines, northern England. Sedimentology 24: Petrol Geol 25: 792-827
271-290 McLean JR, Jerzykiewicz T (1978) Cyclicity, tectonics and
McCabe PJ (1984) Depositional environments of coal and coal: some aspects of fluvial sedimentology in the
coal-bearing strata. In: Rahmani RA, Flores RM (eds) Brazeau-Paskapoo Formations, Coal Valley area,
Sedimentology of coal and coal-bearing sequences. Int Alberta, Canada. In: Miall AD (ed) Fluvial sedimentol-
Assoc Sedimentol Spec Publ 7: 13-42 ogy. Can Soc Petrol Geol Mem 5: 441-468
McCabe PJ, Jones CM (1977) Formation of reactivation McLean JR, Wall JH (1981) The Early Cretaceous
surfaces within superimposed deltas and bedforms. J Moosebar Sea in Alberta. Bull Can Petrol Geol 29:
Sediment Petrol 47: 707-715 334-377
McCabe pJ, Parrish JT (1992) Tectonic and climatic con- McMillan NJ (1973) Shelves of Labrador Sea and Baffin
trols on the distribution and quality of Cretaceous Bay, Canada. In: McCrossan RG (ed) The future petro-
coals. In: McCabe PJ, Parrish JT (eds) Controls on the leum provinces of Canada - their geology and poten-
distribution and quality of Cretaceous coals. Geol Soc tial. Can Soc Petrol Geol Mem 1: 473-517
Am Spec Paper 267: 1-15 McPherson JG, Blair TC (1993) Alluvial fans: fluvial or not?
McCarthy TS, Ellery WN, Stanistreet IG (1992) Avulsion Keynote address, 5th international conference on
mechanisms on the Okavango fan, Botswana: the con- fluvial sedimentology, Brisbane, Australia, July 1993,
trol of a fluvial system by vegetation. Sedimentology 39: Keynote addresses and abstracts, pp K33-K41
779-795 McPherson JG, Shanmugam G, Moiola RJ (1987) Fan-del-
McCaslin J (1983) Sohio to test Denver basin's Arbuckle. tas and braid deltas: varieties of coarse-grained deltas.
Oil Gas J 81: 87-88 Geol Soc Am Bull 99: 331-340
McClay KR, Norton MG, Coney P, Davis GH (1986) Col- Meadows NS, Beach A (1993) Structural and climatic con-
lapse of the Caledonian orogen and the Old Red Sand- trols on facies distribution in a mixed fluvial and
stone. Nature 323: 147-149 aeolian reservoir: the Triassic Sherwood Sandstone in
McDonald BC, Lewis CP (1973) Geomorphologic and sedi- the Irish Sea. In: North CP, Prosser DJ (eds) Character-
mentologic processes of rivers and coasts, Yukon ization of fluvial and aeolian reservoirs. Geol Soc Lond
coastal plain. Task Force on northern Oil Development, Spec Publ 73: 247-264
Information Canada Catalogue no R72-12173 Mebberson AJ (1989) The future for exploration in the
McDonald RE (1976) Big Piney-La Barge producing com- Gippsland Basin. Aust Petrol Explor Assoc J 29:
plex, Sublette and Lincoln conties, Wyoming. In: 431-439
Braunstein J (ed) North American oil and gas fields. Am Medlicott HB, Blanford WT (1879) Manual of the geology
Assoc Petrol Geol Mem 24: 91-120 of India, part II
McDonnell KL (1978) Transition matrices and the deposi- Mello MR, Maxwell JR (1990) Organic geochemical and
tional environments of a fluvial sequence. J Sediment biological marker characterization of source rocks and
Petrol 48: 43-48 oils derived from lacustrine environments in the Bra-
McDougall JW (I989) Tectonically-induced diversion zilian continental margin. In: Katz BJ (ed) Lacustrine
of the Indus River west of the Salt Range, Pakistan. basin exploration. Am Assoc Petrol Geol Mem 50:
Palaeogeogr Palaeoclimatol Palaeoecol 71: 301-307 77-97
References 549
Melton FA (1936) An empirical classification of flood- engineers and earth scientists: Pergamon, New York,
plain streams. Geogr Rev 26: 593-609 pp 168-183
Melvin J (1987) Fluvio-paludal deposits in the Lower Miall AD (1984a) Principles of sedimentary basin analysis.
Kekiktuk Formation (Mississippian), Endicott Field, Springer, Berlin Heidelberg New York
northeast Alaska. In: Ethridge FG, Flores RM, Harvey Miall AD (1984b) Variations in fluvial style in the Lower
MD (eds) Recent developments in fluvial sedimentol- Cenozoic synorogenic sediments of the Canadian Arc-
ogy. Soc Econ Paleontol Mineral Spec Pub139: 343-352 tic Islands. Sediment Geo138: 499-523
Melvin J (1993) Evolving fluvial style in the Kekiktuk For- Miall AD (1985) Architectural-element analysis: a new
mation (Mississippian), Endicott field area, Alaska: method of facies analysis applied to fluvial deposits.
base level response to contemporaneous tectonism. Earth Sci Rev 22: 261-308
AAPG Bull 77: 1723-1744 Miall AD (1986) Effects of Caledonian tectonism in Arctic
Mertz KA Jr, Hubert JF (1990) Cycles of sand-flat sand- Canada. Geology 14: 904-907
stone and playa-lacustrine mudstone in the Triassic- Miall AD (1987) Recent developments in the study of flu-
Jurassic Blomidon redbeds, Fundy rift basin, Nova vial facies models. In: Ethridge FG, Flores RM (eds)
Scotia; implications for tectonics and climatic controls. Recent developments in fluvial sedimentology. Soc
Can J Earth Sci 27: 442-451 Econ Paleontol Mineral Spec Publ 39: 1-9
Miall AD (1970a) Devonian alluvial fans, Prince of Wales Miall AD (1988a) Reservoir heterogeneities in fluvial sand-
Island, Arctic Canada. J Sediment Petrol 40: 556-571 stones: lessons from outcrop studies. Am Assoc Petrol
Miall AD (1970b) Continental-marine transition in the Geol Bull 72: 682-697
Devonian of Prince of Wales Island, Northwest Territo- Miall AD (1988b) Facies architecture in clastic sedimen-
ries. Can J Earth Sci 7: 125-144 tary basins. In: Kleinspehn K, Paola C (eds) New per-
Miall AD (1973) Markov chain analysis applied to an an- spectives in basin analysis. Springer, Berlin Heidelberg
cient alluvial plain succession. Sedimentology 20: New York, pp 67-81
347-364 Miall AD (1988c) Architectural elements and bounding
Miall AD (1974) Paleocurrent analysis of alluvial sedi- surfaces in fluvial deposits: anatomy of the Kayenta
ments: a discussion of directional variance and vector Formation (Lower Jurassic), southwest Colorado. Sedi-
magnitude. J Sediment Petrol 44: 1174-1185 ment Geol 55: 233-262
Miall AD (1976) Paleocurrent and paleohydrologic analy- Miall AD (1989) Architectural elements and bounding sur-
sis of some vertical proflles through a Cretaceous faces in channelized clastic deposits: notes on compari-
braided stream deposit, Banks Island, Arctic Canada. sons between fluvial and turbidite systems. In: Taira A,
Sedimentology 23: 459-484 Masuda F (eds) Sedimentary facies in the active plate
Miall AD (1977) A review of the braided river depositional margin. Terra Scientific, Tokyo, pp 3-15
environment. Earth Sci Rev 13: 1-62 Miall AD (1990) Principles of sedimentary basin analysis,
Miall AD (ed) (1978a) Fluvial sedimentology. Can Soc 2nd edn. Springer, Berlin Heidelberg New York
Petrol Geol Mem 5 Miall AD (1991 a) Hierarchies of architectural units in clas-
Miall AD (1978b) Fluvial sedimentology: an historical re- tic rocks, and their relationship to sedimentation rate.
view. In: Miall AD (ed) Fluvial sedimentology. Can Soc In: Miall AD, Tyler N (eds) The three-dimensional fa-
Petrol Geol Mem 5: 1-47 cies architecture of terrigenous clastic sediments, and
Miall AD (1978c) Lithofacies types and vertical proflle its implications for hydrocarbon discovery and recov-
models in braided river deposits: a summary. In: Miall ery. Soc Econ Paleontol Mineral Conc Sedimentol
AD (ed) Fluvial sedimentology. Can Soc Petrol Geol Paleontol3: 6-12
Mem 5: 597-604 Miall AD (1991b) Sedimentology of a sequence boundary
Miall AD (1978d) Tectonic setting and syndepositional within the nonmarine Torrivio Member, Gallup Sand-
deformation of molasse and other nonmarine-paralic stone (Cretaceous), San Juan Basin, New Mexico. In:
sedimentary basins. Can J Earth Sci 15: 1613-1632 Miall AD, Tyler N (eds) The three-dimensional facies
Miall AD (1979a) Tertiary fluvial sediments in the Lake architecture of terrigenous clastic sediments, and its
Hazen intermontane basin, Ellesmere Island, Arctic implications for hydrocarbon discovery and recovery.
Canada. Geological Survey of Canada paper 79-9 Soc Eco Paleontol Mineral Conc Sedimentol Paleontol
Miall AD (1979b) Mesozoic and tertiary geology of Banks 3:224-232
Island, Arctic Canada: the history of an unstable craton Miall AD (1991c) Stratigraphic sequences and their
margin. Geol Surv Can Mem 387 chronostratigraphic correlation. J Sediment Petrol 61:
Miall AD (1980) Cyclicity and the facies model concept in 497-505
geology. Bull Can Petrol Geol 28: 59-80 Miall AD (1991d) Late Cretaceous and Tertiary basin de-
Miall AD (1981a) Analysis of fluvial depositional systems. velopment and sedimentation, Arctic Islands. In:
Am Assoc Petrol Geol Educ Course Notes Ser 20 Trettin HP (ed) Geology of the Innuitian Orogen and
Miall AD (1981b) Alluvial sedimentary basins: tectonic Arctic Platform of Canada and Greenland, chap 15.
setting and basin architecture. In: Miall AD (ed) Sedi- Geol Surv Can Geol Can 3: 437-458
mentation and tectonics in alluvial basins. Geol Assoc Miall AD (1992a) Alluvial deposits. In: Walker RG, James
Can Spec Paper 23: 1-33 NP (eds) Facies models: response to sea level change.
Miall AD (1983a) Basin analysis of fluvial sediments. In: Geological Association of Canada, St. John's, New-
Collinson JD, Lewin JL (eds) Modern and ancient flu- foundland, pp 119-142
vial systems. Int Assoc Sedimentol Spec Publ6: 279-286 Miall AD (1992b) The Exxon global cycle chart: an event
Miall AD (1983b) Glaciofluvial transport and deposition. for every occasion? Geology 20: 787-790
In: Eyles N (ed) Glacial geology: an introduction for
550 References
Miall AD (1993) The architecture of fluvial-deltaic se- RJ (eds) Sandstone petroleum reservoirs. Springer,
quences in the Upper Mesaverde Group (Upper Creta- Berlin Heidelberg New York, pp 109-l30
ceous), Book Cliffs, Utah. In: Best JL, Bristow CS (eds) Miller RS, Groth JL (1990) Depositional environment and
Braided rivers. Geol Soc Lond Spec Publ 75: 305-332 reservoir properties ofthe Lower Tuscaloosa "B" Sand-
Miall AD (1994) Reconstructing fluvial macroform archi- stone Baywood field, St. Helena Parish, Louisiana.
tecture from two-dimensional outcrops: examples Trans Gulf Coast Assoc Geol Soc 40: 601-605
from the Castlegate Sandstone, Book Cliffs, Utah. J Milne G (1935) Some suggested units for classification and
Sediment Res B64: 146-158 mapping, particularly for East African soils. Soil Res
Miall AD (1995a) Description and interpretation of fluvial Berl 4: 183-198
deposits: a critical perspective: discussion. Sedimentol- Minter WEL (1978) A sedimentological synthesis of placer
ogy 42: 379-384 gold, uranium and pyrite concentrations in Proterozoic
Miall AD (199 5b) Collision-related foreland basins. In: Witwatersrand sediments. In: Miall AD (ed) Fluvial
Busby-Spera C, Ingersoll RV (eds) Tectonics of sedi- sedimentology. Can Soc Petrol Geol Mem 5: 801-829
mentary basins. Blackwell Scientific, Oxford: 393-424 Mitchell AHG, McKerrow WS (1975) Analogous evolution
Miall AD, Gibling MR (1978) The Siluro-Devonian clastic of the Burma Orogen and the Scottish Caledonides.
wedge of Somerset Island, Arctic Canada, and some Geol Soc Am Bull 86: 305-315
regional paleogeographic implications. Sediment Geol Mitchum RM Jr, Van Wagoner JC (1991) High-frequency
21:85-127 sequences and their stacking patterns: sequence-strati-
Miall AD, Smith ND (1989) Rivers and their deposits. graphic evidence of high-frequency eustatic cycles.
Society of Economic Paleontologists and Mineralo- Sediment Geo170: l31-160
gists, Tulsa, Oklahoma, slide set 4 Mitrovica JX, Beaumont C, Jarvis GT (1989) Tilting of
Miall AD, Tyler N (eds) (1991) The three-dimensional continental interiors by the dynamical effects of sub-
facies architecture of terrigenous clastic sediments, duction. Tectonics 8: 1079-1094
and its implications for hydrocarbon discovery and Mj0s R, Walderhaug 0, Prestholm E (1993) Crevasse splay
recovery. Soc Econ Paleontol Mineral Conc Mod Ser 3 sandstone geometries in the Middle Jurassic Ravenscar
Miall AD, Kerr JW, Gibling MR (1978) The Somerset Island Group of Yorkshire, UK. In: Marzo M, Puigdefabregas
Formation: an Upper Silurian to Lower Devonian inter- C (eds) Alluvial sedimentation. Int Assoc Sedimentol
tidal/supratidal succession, Boothia Uplift region, Arc- Spec Pub117: 167-184
tic Canada. Can J Earth Sci 15: 181-189 Molenaar CM, Rice DD (1988) Cretaceous rocks of the
Middleton GV (ed) (1965) Primary sedimentary structures Western Interior Basin. In: Sloss LL (ed) Sedimentary
and their hydrodynamic interpretation. Soc Econ cover - North American Craton. US Geol Soc Am Geol
Paleontol Mineral Spec Publ12 North Am D-2: 77-82
Middleton GV (1973) Johannes Walther's law of the corre- Mollard JD (1973) Airphoto interpretation of fluvial fea-
lation of facies. Geol Soc Am Bull 84: 979-988 tures. In: Fluvial processes and sedimentation. Pro-
Middleton GV (1976) Hydraulic interpretation of sand size ceedings of the Hydrology Symposium, Edmonton,
distributions. J Geol 84: 405-426 National Research Council, Canada, pp 341-380
Middleton GV (1977) Introduction - progress in hydraulic Molnar P, England P (1990) Late Cenozoic uplift of moun-
interpretation of sedimentary structures. In: tain ranges and global climatic change: chicken or egg?
Middleton GV (ed) Sedimentary process: hydraulic in- Nature 346: 29-34
terpretation of primary sedimentary structures. Soc Molnar P, Tapponnier P (1975) Cenozoic tectonics of Asia:
Econ Paleontol Mineral Repr Ser 3: 1-15 effects of a continental collision. Science 189: 419-426
Middleton GV (1978) Sedimentology - history. In: Moody-Stuart M (1966) High and low sinuosity stream
Fairbridge RW, Bourgeois J (eds) The encyclopedia of deposits, with examples from the Devonian of
sedimentology. Dowden, Hutchinson and Ross, Spitzbergen. J Sediment Petrol 36: 1102-1117
Stroudsberg, pp 707-712 Moore PD (1987) Ecological and hydrological aspects of
Middleton GV, Hampton MA (1976) Subaqueous sedi- peat formation. In: Scott AC (ed) Coal and coal-bearing
ment transport and deposition by sediment gravity strata: recent advances. Geol Soc Lond Spec Publ 32:
flows. In: Stanley DJ, Swift DJP (eds) Marine sediment 7-15
transport and environmental management. Wiley, New Moore PS, Hobday DK, Mai H, Sun ZC (1986) Comparison
York, pp 197-218 of selected non-marine petroleum-bearing basins in
Middleton GV, Southard JB (1977) Mechanics of sediment Australia and China. Aust Petrol Explor Assoc J 26:
movement. Society of Economic Paleontologists and 285-309
Mineralogists short course 3, Tulsa, Oklahoma Moore RC (1964) Paleoecological aspects of Kansas Penn-
Middleton LT, BlakeyRC (1983) Processes and controls on sylvanian and Permian cyclothems. In: Merriam DF
the intertonguing of the Kayenta and Navajo Forma- (ed) Symposium on cyclic sedimentation. Kansas Geol
tions, northern Arizona: Eolian-fluvial interactions. In: Surv Bull 169: 287-380
Brookfield ME, Ahlbrandt TS (eds) Eolian sediments Moorman BJ, Judge AS, Smith DG (1991) Examining
and processes. Elsevier, Amsterdam, pp 6l3-634 fluvial sediments using ground penetrating radar in
Mike K (1975) Utilization of the analysis of ancient river British Columbia. Geol Surv Can Paper 91 (1A): 31-36
beds for the detection of Holocene crustal movements. Morgan JP, McIntire WG (1959) Quaternary geology of the
Tectonophysics 29: 359-368 Bengal Basin, East Pakistan and India. Geol Soc Am Bull
Miller DD, McPherson JG, Covington TE (1990) Fluvio- 70:319-342
deltaic reservoir, South Belridge field, San Joaquin Val- Morgan KH (1993) Development, sedimentation and eco-
ley, California. In: Barwis JH, McPherson JG, Studlick nomic potential of palaeo river systems of the Yilgarn
References 551
Craton of Western Australia. Sediment Geol 85: (eds) Fan deltas: sedimentology and tectonic settings.
637-656 Blackie, Glasgow, pp 84-90
Morgridge DL, Smith WB (1972) Geology and discovery of Mutti E, Normark WR (1987) Comparing examples of
Prudhoe Bay field, eastern Arctic Slope, Alaska. In: modern and ancient turbidite systems: problems and
King RE (ed) Stratigraphic oil and gas fields. Am Assoc concepts. In: Leggett JK, Zuffa GG (eds) Marine clastic
Petrol Geol Mem 16: 489-501 sedimentology: concepts and case studies. Graham and
Morison SR, Hein FJ (1987) Sedimentology of the Whie Trotman, London, pp 1-38
Channel gravels, Klondike area, Yukon Territory: flu- Muwais W, Smith DG (1990) Types of channel-fllis inter-
vial deposits of a confined valley. In: Ethridge FG, preted from dipmeter logs in the MCMurray Formation,
Flores RM, Harvey MD (eds) Recent developments in northeast Alberta. Bull Can Petrol Geol 38: 53-63
fluvial sedimentology. Soc Econ Paleontol Mineral Spec Nadler CT, Schumm SA (1981) Metamorphosis of South
Pub139: 205-216 Platte and Arkansas rivers, eastern Colorado. Phys
Morley CK (1986) A classification of thrust fronts. Am Geogr 2: 95-115
Assoc Petrol Geol Bull 70: 12-35 Nadon G (1991) Architectural element analysis of a
Morton RA, Price WA (1987) Late Quaternary sea-level foreland basin clastic wedge. PhD Thesis, University of
fluctuations and sedimentary phases of the Texas Toronto
coastal plain and shelf. In: Nummedal D, Pilkey OH, Nadon G (1994) The genesis and recognition of anasto-
Howard JD (eds) Sea-level fluctuation and coastal mosed fluvial deposits: data from the St Mary River
evolution. Soc Econ Paleontol Mineral Spec Publ 41: Formation, southwestern Alberta. J Sediment Res B64:
181-198 451-463
Mosley MP (1976) An experimental study of channel Nami M, Leeder MR (1978) Changing channel morphology
confluences. J Geol 84: 535-562 and magnitude in the ScalbyFormation (M. Jurassic) of
Mosley MP, Schumm SA (1976) Stream junctions - a pro- Yorkshire, England. In: Miall AD (ed) Fluvial sedimen-
bable location for bedrock placers. Econ Geol 72: tology. Can Soc Petrol Geol Mem 5: 431-440
691-697 Nanson GC (1980) Point bar and floodplain formation of
Mossop GD, Flach PD (1983) Deep channel sedimentation the meandering Beatton River, northeastern British
in the Lower Cretaceous McMurray Formation, Columbia, Canada. Sedimentology 27: 3-30
Athabasca Oil Sands, Alberta. Sedimentology 30: Nanson GC, Croke JC (1992) A genetic classification of
493-509 floodplains. Geomorphology 4: 459-486
Mossop GD, Shetsen I (1994) Geological atlas of the West- Nanson GC, Page K (1983) Lateral accretion of fine-
ern Canada Sedimentary Basin. Canadian Society of grained concave benches on meandering streams. In:
Petroleum Geologists, Calgary, Alberta Collinson JD, Lewin J (eds) Modern and ancient fluvial
Mpodozis C, Ramos V (1990) The Andes of Chile systems. Int Assoc Sedimentol Spec Publ6: 133-144
and Argentina. In: Ericksen GE, Pinochet MTC, Nanson GC, Rust BR (1985) Channel patterns and mud
Reinemund JA (eds) Geology of the Andes and its deposits in an arid zone river: Cooper Creek, central
relation to hydrocarbon and mineral resources. Australia, 3rd international sedimentology conference,
Circum Pac Council Energy Miner Res Earth Sci Ser 11: Fort Collins, abstracts, p 31
59-90 Nanson GC, Rust BR, Taylor G (1986) Coexistent mud
Mukerji AB (1976) Terminal fans of inland streams in braids in an arid-zone river: Cooper Creek, central
Sutlej-Yamuna Plain, India. Z Geomorphol20: 190-204 Australia. Geology 14: 175-178
Mulder TJ, Burbank DW (1993) The impact of incipient Nanz RH Jr (1954) Genesis of Oligocene sandstone reser-
uplift on patterns of fluvial deposition: an example voir, Seeligson field, Jim Wells and Kleberg Counties,
from the Salt Range, northwest Himalayan foreland, Texas. Am Assoc Petrol Geol Bull 38: 96-117
Pakistan. In: Marzo M, Puigdefabregas C (eds) Alluvial Nascimento OS, Bornemann E, Jobim LDC, Carvalho MD,
sedimentation. Int Assoc Sedimentol Spec Publ 17: Pimentel AM, Bonet EJ, Rodriques EB, Sandoval JRL,
521-539 Lassandro V, Rodrigues TC, Hocott CR (1982) Aracas
Munoz A, Ramos A, Sanchez-Moya Y, Sopena A (1992) field-reservoir heterogeneities and secondary recovery
Evolving fluvial architecture during a marine trans- performance (abstract). Am Assoc Petrol Geol Bull 66:
gression: Upper Buntsandstein, Triassic, central Spain. 612
Sediment Geo175: 257-281 Neidell NS, Beard JH (1985) Seismic visibility of strati-
Munoz A, Ramos A, Sanchez-Moya Y, Sopena A (1992) graphic objectives. Society of Petroleum Engineers pa-
Evolving fluvial architecture during a marine trans- per 14175
gression: Upper Buntsandstein, Triassic, central Spain. Nemec W (1988) Coal correlations and intrabasinal sub-
Sediment Geol 75: 257-281 sidence: a new analytical perspective. In: Kleinspehn K,
Murty KN (1983) Geology and hydrocarbon prospects of Paola C (eds) New perspectives in basin analysis.
Assam Shelf - recent advances and present status: Pe- Springer, Berlin Heidelberg New York, pp 161-188
troleum Asia Journal, Nov 1983, pp 1-14 Nemec W, Muszynski A (1982) Volcaniclastic alluvial
Muto T (1987) Coastal fan processes controlled by sealevel aprons in the Tertiary of Sofia district (Bulgaria). Ann
changes: a Quaternary example from the Tenryugawa Soc Geol Polon 52: 239-303
systesm, Pacific coast of central Japan. J Geol 95: Nemec W, Postma G (1993) Quaternary alluvial fans in
716-724 southwestern Crete: sedimentation processes and geo-
Muto T (1988) Stratigraphical patterns of coastal-fan sedi- morphic evolution. In: Marzo M, Puigdefabregas C
mentation adjacent to high-gradient submarine slopes (eds) Alluvial sedimentation. Int Assoc Sedimentol
affected by sea-level changes. In: Nemec W, Steel RJ Spec Pub117: 235-276
552 References
Nemec W, Steel RJ (1984) Alluvial and coastal conglomer- (ed) Correlation in hydrocarbon exploration. Graham
ates: their significant features and some comments on and Trotman, London, pp 273-289
gravelly mass-flow deposits. In: Koster EH, Steel RJ Okolo SA (1983) Fluvial distributary channels in the
(eds) Sedimentology of gravels and conglomerates. Can Fletcher Bank Grit (Namurian R2b), at Ramsbottom,
Soc Petrol Geol Mem 10: 1-31 Lancashire, England. In: Collinson JD, Lewin J (eds)
Nemec W, Steel RJ (eds) (1988) Fan deltas: sedimentology Modern and ancient fluvial systems. Int Assoc
and tectonic settings. Blackie, Glasgow Sedimentol Spec Pub16: 421-433
Nichols GJ (1987) Structural controls on fluvial distribu- Olsen H (1988) The architecture of a sandy braided-mean-
tary systems - the Luna system, northern Spain. In: dering river system: an example from the Lower Trias-
Ethridge FG, Flores RM, Harvey MD (eds) Recent sic Solling Formation (M. Buntsandstein) in
developments in fluvial sedimentology. Soc Econ W-Germany. Geol Rundsch 77: 797-814
Paleontol Mineral Spec Pub139: 269-277 Olsen H (1989) Sandstone-body structures and ephemeral
Nijman W, Puigdefabregas C (1978) Coarse-grained point stream processes in the Dinosaur Canyon Member,
bar structure in a molasse-type fluvial system, Eocene Moenave Formation (Lower Jurassic), Utah, USA. Sedi-
Castisent sandstone Formation, south Pyrenean Basin. ment Geo161: 207-221
In: Miall AD (ed) Fluvial sedimentology. Can Soc Petrol Olsen H (1990) Astronomical forcing of meandering river
Geol Mem 5: 487-510 behaviour: Milankovitch cycles in Devonian of East
Nilsen TH (1968) The relationship of sedimentation to Greenland. Palaeogeogr Palaeoclimatol Palaeoecol 79:
tectonics in the Solund Devonian district of southwest- 99-115
ern Norway. Universitetsforlaget, Oslo (Norges geo- Olsen H (1994) Orbital forcing on continental depositional
logiske undersokelse, no 259) systems - lacustrine and fluvial cyclicity in the Devo-
Nilsen TH (ed) (1984) Fluvial sedimentation and related nian of East Greenland. In: De Boer PL, Smith DG (eds)
tectonic framework, western North America. Sediment Orbital forcing and cyclic sequences. Int Assoc Sedi-
Geo138 (special issue) mentol Spec Publl9: 429-438
Nilsen TH (1985) Modern and ancient alluvial fan depos- Olsen H, Larsen PH (1993) Structural and climatic controls
its. Van Nostrand Reinhold, New York (Hutchinson on fluvial depositional systems: Devonian, north-east
Ross Benchmark, vol 87) Greenland. In: Marzo M, Puigdefabregas C (eds) Allu-
Nilsen TH (1987) Paleogene tectonics and sedimentation vial sedimentation. Int Assoc Sedimentol Spec Publl7:
of coastal California. In: Ingersoll RV, Ernst WG (eds) 401-423
Cenozoic basin development of coastal California, Olsen PE (1990) Tectonic, climatic, and biotic modulation
Rubey, vol VI. Prentice-Hall, Englewood Cliffs, pp oflacustrine ecosystems - examples from Newark Su-
81-123 pergroup of eastern North America. In: Katz BJ (ed)
Nilsen T, MCLaughlin RJ (1985) Comparison of tectonic Lacustrine basin exploration: case studies and modern
framework and depositional patterns of the Hornelen analogs. Am Assoc Petrol Geol Mem 50: 209-224
strike-slip basin of Norway ad the Ridge and Little Olsen T (1993) Large fluvial systems: the Atane Formation,
Sulphur Creek strike-slip basins of California. In: a fluvio-deltaic example from the Upper Cretaceous of
Biddle KT, Christie-Blick N (eds) Strike-slip deforma- central West Greenland. Sediment Geo185: 457-473
tion, basin formation and sedimentation. Soc Econ Olsen T, Steel RJ, H0gseth K, Skar T, R0e S-L (1995)
Paleontol Mineral Spec Pub137: 79-103 Sequential architecture in a fluvial succession: se-
North American Commission on Stratigraphic Nomencla- quence stratigraphy in the Upper Cretaceous
ture (1983) North American Stratigraphic Code. Am Mesaverde Group, Price, Utah. J Sediment Res
Assoc Petrol Geol Bull 67: 841-875 B65:265-280
Nummedal D, Swift DJP (1987) Transgressive stratigraphy Olson JS, Potter PE (1954) Variance components of cross-
at sequence-bounding unconformities: some prin- bedding directions in some basal Pennsylvanian sand-
ciples derived from Holocene and Cretaceous ex- stones of the eastern Interior Basin: statistical methods.
amples. In: Nummedal D, Pilkey OH, Howard JD (eds) J Geol 62: 26-49
Sea-level fluctuation and coastal evolution. Soc Econ Ono Y (1990) Alluvial fans in Japan and South Korea. In:
Paleontol Mineral Spec Pub141: 241-260 Rachocki AH, Church M (eds) Alluvial fans: a field
NummedalD, PilkeyOH,HowardJD (eds) (1987) Sea-level approach. Wiley, Chichester, pp 91-107
fluctuation and coastal evolution. Soc Econ Paleontol Oomkens E (1970) Depositional sequences and sand dis-
Mineral Spec Pub141 tribution in the postglacial Rh6ne delta complex. In:
Nummedal D, Riley GW, Temple PL (1993) High-re- Morgan JP (ed) Deltaic sedimentation: modern and
solution sequence architecture: a chronostratigraphic ancient. Soc Econ Paleontol Mineral Spec Publ 15:
model based on equilibrium profIle studies. In: 198-212
Posamentier HW, Summerhayes CP, Haq BU, Allen GP Oomkens E, Terwindt JHJ (1960) Inshore estuarine sedi-
(eds) Sequence stratigraphy and facies associations. Int ments in the Haringvliet (The Netherlands). Geol
Assoc Sedimentol Spec Pub118: 55-68 Mijnbouw 39: 701-710
Nybniten G, Skolem E, 0stby K (1990) Reservoir simula- Ordonez S, Garcia del Cura MA (1983) Recent and Tertiary
tion of the Snorre field. In: Buller AT, Berg E, fluvial carbonates in central Spain. In: Collinson JD,
Hjelmeland 0, Kleppe J, Torsaeter 0, Aasen JO (eds) Lewin J (eds) Modern and ancient fluvial systems. Int
North Sea oil and gas reservoirs II. Graham and Assoc Sedimentol Spec Publ 6: 485-497
Trotman, London, pp 103-114 Ore HT (1964) Some criteria for recognition of braided
Nystuen JP, Knarud R, Jorde K, Stanley KO (1989) Correla- stream deposits. Wyoming Contrib Geo13: 1-14 .
tion of Triassic to Jurassic sequences, Snorre Field and Ori GG (1979) Barre di meandre nelle alluvioni ghiaiose del
adjacent areas, northern North Sea. In: Collinson JD fiume Reno (Bologna). Bull Soc Geol Ital 98: 35-54
References 553
Ori GG (1982) Braided to meandering channel patterns in Peale AC (1879) Report on the geology of the Green River
humid-region alluvial fan deposits, River Reno, Po District. In: Hayden FV (ed) US geological and geo-
Plain (northern Italy). Sediment Geo131: 231-248 graphical survey of the territories, 9th annual report.
Ori GG (1993) Continental depositional systems of the Washington DC
Quaternary of the Po Plain (northern Italy). Sediment Pelletier BR (1958) Pocono paleocurrents in Pennsylvania
Geo183: 1-14 and Maryland. Geol Soc Am Bull 69: 1033-1064
Ori GG, Friend PF (1984) Sedimentary basins formed and Peper T, Beekman F, Cloetingh S (1992) Consequences of
carried piggyback on active thrust sheets. Geology 12: thrusting and intraplate stress fluctuations for vertical
475-478 motions in foreland basins and peripheral areas.
Orton GJ (1988) A spectrum of Middle Ordovician fan GeophysJlntlll:l04-126
deltas and braidplain deltas, North Wales: a conse- Perlmutter MA, Matthews MD (1989) Global cyclostrati-
quence of varying fluvial clastic input. In: Nemec W, graphy - a model. In: Cross TA (ed) Quantitative dy-
Steel RJ (eds) Fan deltas: sedimentology and tectonic namic stratigraphy. Prentice Hall, Englewood Cliffs,
setting: Blackie, London, pp 23-49 New Jersey, pp 33-260
Ouchi S (1985) Response of alluvial rivers to slow active Peterson F (1984) Fluvial sedimentation on a quivering
tectonic movement. Geol Soc Am Bull 96: 504-515 craton: influence of slight crustal movements on fluvial
Ouyang J, Ma Z, He D, Tian H (1988) Sedimentary facies processes, Upper Jurassic Morrison Formation, west-
recognition in China with well-logging data. In: ern Colorado Plateau. Sediment Geo138: 21-50
Wagner HC, Wagner LC, Wang FFH, Wong FL (eds) Petterson 0, Storli A, Ljosland E, Massie I (1990) The
Petroleum resources of China and related subjects. Gullfaks field: Geology and reservoir development. In:
Circum-Pacific Council for Energy and Mineral Re- Buller AT, Berg E, Hjelmeland 0, Kleppe J, Torsaeter 0,
sources. Earth Sci Ser 10: 467-474 Aasen JO (eds) North Sea oil and gas reservoirs II.
Owens JS (1908) Experiments on the transporting power of Graham and Trotman, London, pp 67-90
sea currents. Geogr J 31: 415-420 Pettijohn FJ (1949) Sedimentary rocks. Harper and Row,
Paola C (1988) Subsidence and gravel transport in alluvial New York
basins. In: Kleinspehn K, Paola C (eds) New perspec- Pettijohn FJ (1957) Sedimentary rocks, 2nd edn. Harper,
tives in basin analysis. Springer, Berlin Heidelberg New New York
York, pp 231-243 Pettijohn FJ (1962) Paleocurrents and paleogeography.
Paola C (1990) A simple basin-filling model for coarse- Am Assoc Petrol Geol Bull 46: 1468-1493
grained alluvial systems. In: Cross TA (ed) Quantitative Pettijohn FJ, Potter PE (1964) Atlas and glossary of pri-
dynamic stratigraphy. Prentice-Hall, Englewood Cliffs, mary sedimentary structures. Springer, Berlin Heidel-
pp 363-374 berg New York
Paola C, Heller PL, Angevine CL (1992) The large-scale Pettijohn FJ, Potter PE, Siever R (1972) Sand and sand-
dynamics of grain-size variation in alluvial basins, 1: stone. Springer, Berlin Heidelberg New York
theory. Basin Res 4: 73-90 Pfiffner OA (1986) Evolution of the north Alpine foreland
Padgett V, Ehrlich R (1976) Paleohydrologic analysis of a basin in the central Alps. In: Allen PA, Homewood P
Late Carboniferous fluvial system, southern Morocco. (eds) Foreland basins. Int Assoc Sedimentol Spec Publ
Geol Soc Am Bull 87: 1101-1104 8:219-228
Parkash B, AwasthiAK, Gohain K (1983) Lithofacies of the Pienkowski G (1991) Eustatically-controlled sedimenta-
Markanda terminal fan, Kurukshetra district, Haryana, tion in the Hettangian-Sinemurian (Early Jurassic) of
India. In: Collinson JD, Lewin J (eds) Modern and an- Poland and Sweden. Sedimentology 38: 503-518
cient fluvial systems. Int Assoc Sedimentol Spec Publ6: Pierson TC (1980) Erosion and deposition by debris flows
337-344 at Mt. Thomas, New Zealand. Earth Surface Proc 5:
Parrish JT (1993) Climate of the supercontinent Pangea. J 1952-1984
Geo1101: 215-233 Pirson SJ (1970) Geologic well log analysis. Gulf, Houston
Parrish JT, Barron EJ (1986) Paleoclimates and economic Pirson SJ (1977) Geologic well log analysis, 2nd edn. Gulf,
geology. Soc Econ Paleontol Mineral Short Course 18 Houston
Parrish JT, Peterson F (1988) Wind directions predicted Pitman JK, Fouch TD, Goldhaber MB (1982) Evolution of
from global circulation models and wind directions some Tertiary unconventional reservoir rocks, Uinta
determined from eolian sandstones of the western Basin, Utah. Am Assoc Petrol Geol Bull 66: 1581-1596
United States - a comparison. Sediment Geol 56: Pitman WC III (1978) Relationship between eustacy and
261-282 stratigraphic sequences of passive margins. Geol Soc
Parsley AJ (1984) North Sea hydrocarbon plays. In: Am Bull 89: 1389-1403
Glennie KW (ed) Introduction to the petroleum geo- Pitman WC III (1986) Effects of sea level change on basin
logy of the North Sea. Blackwell Scientific, Oxford, pp stratigraphy. Am Assoc Petrol Geol Bull 70: 1762
205-230 Pitman WC III, Golovchenko X (1988) Sea-level changes
Passega R (1957) Texture as a characteristic of clastic and their effect on the stratigraphy of Atlantic-type
deposition. Am Assoc Petrol Geol Bull 41: 1952-1984 margins. In: Sheridan RE, GrowJA (eds) The geology of
Payton CE (ed) (1977) Seismic stratigraphy - applications North America, vol 1-2, The Atlantic continental mar-
to hydrocarbon exploration. Am Assoc Petrol Geol gin, United States, Geological Society of America, Boul-
Mem26 der, Colorado, pp 429-436
Pazzaglia FJ (1993) Stratigraphy, petrography, and corre- Platt NH, Keller B (1992) Distal alluvial deposits in a
lation of late Cenozoic middle Atlantic Coastal Plain foreland basin setting - the Lower Freshwater Molasse
deposits: implications for late-stage passive-margin (Lower Miocene), Switzerland: sedimentology, archi-
geologic evolution. Geol Soc Am Bull 105: 1617-1634 tecture and palaeosols. Sedimentology 39: 545-565
554 References
Platt NH, Wright VP (1992) Palustrine carbonates and the work: concepts, examples, and exploration signifi-
Florida Everglades: towards an exposure index for the cance. Am Assoc Petrol Geol Bull 76: 1687-1709
fresh-water environment. J Sediment Petrol 62: Posaner EM, Goldthorpe WH (1986) The development and
1058-1071 early performance of the North Rankin field. Aust
Playfair J (1802) Illustrations of the Huttonian theory of Petrol Explor Assoc J 26: 420-427
the Earth. Cadell and Davies, London (Reprinted by Potter PE (1955) The petrology and origin of the Lafayette
Dover, New York, 1964) Gravel, part I: mineralogy and petrology. J Geol 63:
Plint AG (1983) Sandy fluvial point-bar sediments from 1-38
the middle Eocene of Dorset, England. In: Collinson JD, Potter PE (1959) Facies models conference. Science 129:
Lewin J (eds) Modern and ancient fluvial systems. Int 1292-1294
Assoc Sedimentol Spec Publ6: 355-368 Potter PE (1978) Significance and origin of big rivers. J
Plint AG (1988) Sharp-based shoreface sequences and Geol 86: 13-33
"offshore bars" in the Cardium Formation of Alberta: Potter PE, Pettijohn FJ (1963) Paleocurrents and basin
their relationship to relative changes in sea level. In: analysis. Academic, New York (2nd edn: 1977,
Wilgus CK, Hastings BS, Kendall CGSC, Posamentier Springer, Berlin Heidelberg New York)
HW, Ross CA, Van Wagoner JC (eds) Sea-level re- Powell JW (1875) Exploration ofthe Colorado River of the
search: an integrated approach. Soc Econ Paleontol West. Washington (see also US 43rd congress, 1st ses-
Mineral Spec Pub142: 357-370 sion, H miscellaneous documents 265, 1874)
Plint AG (1990) An allostratigraphic correlation of the Pratt BR, Miall AD (1993) Anatomy of a bioclastic grain-
Muskiki and Marshybanks formations (Coniacian- stone megashoal (Middle Silurian, southern Ontario)
Santonian) in the Foothills and subsurface of the revealed by ground-penetrating radar. Geology 21:
Alberta basin. Bull Can Petrol Geol 38: 288-306 223-226
Plint AG, Eyles N, Eyles CH, Walker RG (1992) Control of Prosser S (1993) Rift-related depositional systems and
sea level change. In: Walker RG, James NP (eds) Facies their seismic expression. In: Williams GD, Dobb A
models: response to sea level change. Geological Asso- (eds) Tectonics and seismic sequence stratigraphy.
ciation of Canada, St. John's, Newfoundland, pp 15-25 Geol Soc Lond Spec Publ 71: 35-66
Plint AG, Hart BS, Donaldson WS (1993) Lithospheric Pryor WA (1960) Cretaceous sedimentation in upper Mis-
flexure as a control on stratal geometry and facies dis- sissippi embayment. Am Assoc Petrol Geol Bull 44:
tribution in Upper Cretaceous rocks of the Alberta 1473-1504
foreland basin. Basin Res 5: 69-77 Puigdefabregas C (1973) Miocene point-bar deposits in
Plint AG, Walker RG, Bergman KM (1986) Cardium For- the Ebro Basin, northern Spain. Sedimentology 20:
mation 6. Stratigraphic framework of the Cardium in 133-144
subsurface. Bull Can Petrol Geo134: 213-225 Puigdefabregas C, Van Vliet A (1978) Meandering stream
Poag CW, Sevon WD (1989) A record of Appalachian deposits from the Tertiary of the southern Pyrenees. In:
denudation in postrift Mesozoic and Cenozoic sedi- Miall AD (ed) Fluvial sedimentology. Can Soc Petrol
mentary deposits of the US Middle Atlantic continental Geol Mem 5: 469-485
margin. Geomorphology 2: 119-157 Puigdefabregas C, Munoz JA, Marzo M (1986) Thrust belt
Posamentier HW, Allen GP (1993) Siliciclastic sequence development in the eastern Pyrenees and related depo-
stratigraphic patterns in foreland ramp-type basins. sitional sequences in the southern foreland basin. In:
Geology 21: 455-458 Allen PA, Homewood P (eds) Foreland basins. Int
Posamentier HW, Allen GP (in press) Siliciclastic sequence Assoc Sedimentol Spec Publ 8: 229-246
stratigraphy Putnam PE (1982a) Fluvial channel sandstones within up-
Posamentier HW, James DP (1993) An overview of se- per Mannville (Albian) of Lloydminster area, Canada-
quence-stratigraphic concepts: uses and abuses. In: geometry, petrography, and paleogeographic implica-
Posamentier HW, Summerhayes CP, Haq BU, Allen GP tions. Am Assoc Petrol Geol Bull 66: 436-459
(eds) Sequence stratigraphy and facies associations. Int Putnam PE (1982b) Aspects of the petroleum geology of
Assoc Sedimentol Spec Pub118: 3-18 the Lloydminster heavy oil fields, Alberta and
Posamentier HW, Vail PR (1988) Eustatic controls on Saskatchewan. Bull Can Petrol Geo130: 81-111
clastic deposition II - sequence and systems tracts Putnam PE (1983) Fluvial deposits and hydrocarbon accu-
models. In: Wilgus CK, Hastings BS, Kendall CGSC, mulations: examples from the Lloydminster area,
Posamentier HW, Ross CA, Van Wagoner JC (eds) Sea- Canada. In: Collinson JD, Lewin J (eds) Modern and
level research: an integrated approach. Soc Econ ancient fluvial systems. Int Assoc Sedimentol Spec Publ
Paleontol Mineral Spec Pub142: 125-154 6:517-532
Posamentier HW, Weimer P (1993) Siliciclastic sequence Putnam PE (1993) A multidisciplinary analysis of Belly
stratigraphy and petroleum geology - where to from River-Brazeau (Campanian) fluvial channel reservoirs
here? Am Assoc Petrol Geol Bull 77: 731-742 in west-central Alberta, Canada. Bull Can Petrol Geol
Posamentier HW, Jervey MT, Vail PR (1988) Eustatic 41: 186-217
controls on clastic deposition I - conceptual frame- Putnam PE, Oliver TA (1980) Stratigraphic traps in chan-
work. In: Wilgus CK, Hastings BS, Kendall CGSC, nel sandstones in the Upper Mannville (Albian) of east-
Posamentier HW, Ross CA, Van Wagoner JC (eds) Sea- central Alberta. Bull Can Petrol Geol 28: 489-508
level research: an integrated approach. Soc Econ Qian N (1990) Fluvial processes in the lower Yellow River
Paleontol Mineral Spec Pub142: 109-124 after levee breaching at Tongwaxiang in 1855. Int J
Posamentier HW, Allen GP, James DP, Tesson M (1992) Sediment Res 5: 1-14
Forced regressions in a sequence stratigraphic frame-
References 555
Rachocki AH, Church M (eds) (1990) Alluvial fans: a field Retallack GJ (1986) Fossil soils as grounds for intepreting
approach. Wiley, New York long-term controls on ancient rivers. J Sediment Petrol
Rahmani RA, Flores RM (eds) (1984) Sedimentology of 56: 1-18
coal and coal-bearing sequences. Int Assoc Sedimentol Retallack GJ (1988) Field recognition of paleosols. In:
Spec Pub17 Reinhardt J, Sigleo WR (eds) Paleosols and weathering
Rahmani RA, Lerbekmo JF (1975) Heavy mineral analysis through geologic time: Principles and applications.
of Upper Cretaceous and Paleocene sandstones in Geol Soc Am Spec Paper 216: 1-20
Alberta and adjacent areas of Saskatchewan. In: Reynolds DW, Vincent JK (1972) Midland gas field, west-
Caldwell WGE (ed) The Cretaceous system in the West- ern Kentucky. In: King RE (ed) Stratigraphic oil and gas
ern Interior of North America. Geol Assoc Can Spec fields. Am Assoc Petrol Geol Mem 16: 585-598
Paper 13: 607-632 Rhee CW, Ryand WH, Chough SK (1993) Contrasting de-
Rahmanian VD, Moore PS, Mudge WJ, Spring DE (1990) velopment patterns of crevasse channel deposits in
Sequence stratigraphy and the habitat of hydrocar- Cretaceous alluvial successions, Korea. Sediment Geol
bons, Gippsland Basin, Australia. In: Brooks J (ed) 85:401-410
Classic petroleum provinces. Geol Soc Lond Spec Publ Riba 0 (1976) Syntectonic unconformities of the Alto
50:525-541 Cardener, Spanish Pyrenees, a genetic interpretation.
Ramos A, Sopena A (1983) Gravel bars in low-sinuosity Sediment Geoll5: 213-233
streams (Permian and Triassic, central Spain). In: Ricci Lucchi F (1986) The Oligocene to Recent foreland
Collinson JD, Lewin J (eds) Modern and ancient fluvial basins of the northern Apennines. In: Allen PA,
systems. Int Assoc Sedimentol Spec Pub16: 301-312 Homewood P (eds) Foreland basins. Int Assoc
Ramos A, Sopena A, Perez-Arlucea M (1986) Evolution of Sedimentol Spec Publ 8: 105-139
Buntsandstein fluvial sedimentation in the northwest Richards K, Chandra S, Friend PF (1993) Avulsive channel
Iberian Ranges (Central Spain). J Sediment Petrol 56: systems: characteristics and examples. In: Best JL,
862-875 Bristow CS (eds) Braided rivers. Geol Soc Lond Spec
Ramsbottom WHC (1979) Rates of transgression and re- Publ 75: 195-203
gression in the Carboniferous ofNW Europe. J Geol Soc Rider MH (1990) Gamma-ray log shape used as a facies
Lond 136: 147-153 indicator: critical analysis of an oversimplified meth-
Riisiinen M, Neller R, Salo J, Jungner H (1992) Recent and odology. In: Hurst A, Lovell MA, Morton AC (eds)
ancient fluvial depositional systems in the Amazon Geological applications of wireline logs. Geol Soc Lond
foreland basin, Peru. Geol Mag 129: 293-306 Spec Pub148: 27-37
Riisiinen M, Salo JS, Kalliola RJ (1987) Fluvial perturbance Ridgway KD, DeCelles PG (1993) Stream-dominated allu-
in the Western Amazon Basin: regulation by long-term vial fan and lacustrine depositional systems in Ceno-
sub-Andean tectonics. Science 238: 1398-1401 zoic strike-slip basins, Denali fault system, Yukon
Rascoe BR Jr, Adler FJ (1983) Permo-Carboniferous hy- Territory, Canada. Sedimentology 40: 645-666
drocarbon accumulations, Mid-Continent, USA. Am Rieu EV (1953) Homer: The Iliad. Methuen, London
Assoc Petrol Geol Bull 67: 979-1001 Rivenaes JC (1992) Application of a dual-lithology, depth
Ray RR (1982) Seismic stratigraphic interpretation of the dependent diffusion equation in stratigraphic simula-
Fort Union Formation, western Wind River Basin: ex- tion. Basin Res 4: 133-146
ample of subtle trap exploration in a nonmarine se- Roberts DG (1988) Basin evolution and hydrocarbon ex-
quence. In: Halbouty MT (ed) The deliberate search for ploration in the South China Sea. In: Wagner HC,
the subtle trap. Am Assoc Petrol Geol Mem 32: 169-180 Wagner LC, Wang FFH, Wong FL (eds) Petroleum re-
Reade TM (1884) Ripple marks in drift in Shropshire and sources of China and related subjects. Circum-Pacific
Cheshire. Q J Geol Soc Lond 40: 267-269 Council for Energy and Mineral Resources. Earth Sci
Reading HG (ed) (1986) Sedimentary environments and Ser 10: 157-177
facies, 2nd edn. Blackwell, Oxford Roberts HH, Adams RD, Cunningham RHW (1980) Evolu-
Reiche P (1938) An analysis of cross-lamination: the tion of sand-dominant subaerial phase, Atchafalaya
Coconino Sandstone. J Geol 46.905-932 Delta. Am Assoc Petrol Geol Bull 64: 264-279
Reineck HE, Wunderlich R (1968) Classification and ori- Robertson JD (1991) Reservoir management using 3-D
gin of flaser and lenticular bedding. Sedimentology 11: seismic data. In: Lake LW, Carroll HB Jr, Wesson TC
99-104 (eds) Reservoir characterization II. Academic, San Di-
Reinhardt J, Sigleo WR (eds) (1988) Paleosols and weath- ego, pp 340-354
ering through geologic time: principles and applica- Robinson JE (1981) Well spacing and the identification of
tions. Geol Soc Am Spec Paper 216 subsurface drainage systems. Bull Can Petrol Geol 29:
Reinfelds I, Nanson G (1993) Formation of braided river 250-258
floodplains, Waimakariri River, New Zealand. Sedi- Rodine JD, Johnson AM (1976) The ability of debris,
mentology 40: 1113-1127 heavily freighted with coarse clastic materials to flow
Retallack GJ (1981) Fossil soils: indicators of ancient ter- on gentle slopes. Sedimentology 23: 213-234
restrial environments. In: Niklas KJ (ed) Paleobotany, Rodolfo KS (1975) The Irrawaddy delta: tertiary setting
paleoecology and evolution, vol I. Praeger, New York, and modern offshore sedimentation. In: Broussard ML
pp 55-102 (ed) Deltas, models for exploration: Houston Geologi-
Retallack GJ (1984) Completeness of the rock and fossil cal Society, Houston, pp 339-356
record: some estimates using fossil soils. Paleobiology R0e S-L, Hermansen M (1993) Processes and products
10:59-78 of large, late Precambrian sandy rivers in northern
556 References
Norway. In: Marzo M, Puigdefabregas C (eds) Alluvial Rust BR (1984) Proximal braidplain deposits in the Middle
sedimentation. Int Assoc Sedimentol Spec Publ 17: Devonian Malbaie Formation of eastern Gaspe, Que-
151-166 bec, Canada. Sedimentology 31: 675-695
Rogers RR (1994) Nature and origin of through-going Rust BR, Gostin VA (1981) Fossil transverse ribs in Ho-
discontinuities in nonmarine foreland basin strata, locene alluvial fan deposits, Depot Creek, South Aus·
Upper Cretaceous, Montana: implications for sequence tralia. J Sediment Petrol 51: 441-444
analysis. Geology 22: 1119-1122 Rust BR, Jones BG (1987) The Hawkesbury Sandstone
Rosen MR (1994) The importance of groundwater in pla- south of Sydney, Australia: Triassic analogue for the
yas: a review of playa classifications and the sedimen- deposit of a large braided river. J Sediment Petrol 57:
tology and hydrology of playas. In: Rosen MR (ed) 222-233
Paleoclimate and basin evolution of playa systems. Rust BR, Koster EH (1984) Coarse alluvial deposits. In:
Geol Soc Am Spec Paper 289: 1-18 Walker RG (ed) Facies models, 2nd edn. Geosci Can
Rosenthal L (1988) Wave dominated shorelines and in- Reprint Ser 1: 53-69
cised channel trends: Lower Cretaceous Glauconite Rust BR, Legun AS (1983) Modern anastomosing-fluvial
Formation, west-central Alberta. In: James DP, Leckie deposits in arid central Australia, and a Carboniferous
DA (eds) 1988, Sequences, stratigraphy, sedimentol- analogue in New Brunswick, Canada. In: Collinson JD,
ogy: surface and subsurface. Can Soc Petrol Geol Mem Lewin J (eds) Modern and ancient fluvial systems. Int
15:207-230 Assoc Sedimentol Spec Publ6: 385-392
Ross WC (1990) Modeling base-level dynamics as a control Rust RB, Nanson GC (1989) Bedload transport of mud as
on basin-fill geometries and facies distribution: a con- pedogenic aggregates in modern and ancient rivers.
ceptual framework. In: Cross TA (ed) Quantitative dy- Sedimentology 36: 291-306
namic stratigraphy. Prentice-Hall, Englewood Cliffs, Rust BR, Gibling MR, Legun AS (1984) Coal deposition in
pp 387-399 an anastomosed fluvial system: the Pennsylvanian
Rouse H (1939) Experiments on the mechanics of sediment Cumberland Group south of Joggins, Nova Scotia,
suspension. 5th international congress on applied me- Canada. In: Rahmani RA, Flores RM (eds) Sedimento-
chanics, Cambridge, Massachusetts, pp 550-554 logy of coal and coal-bearing sequences. Int Assoc
Royden LH (1985) The Vienna Basin: a thin-skinned pull- Sedimentol Spec Publ 7: 105-120
apart basin. In: Biddle KT, Christie-Blick N (eds) Ryer TA (1984) Transgressive-regressive cycles and the
Strike-slip deformation, basin formation and sedimen- occurrence of coal in some Upper Cretaceous strata of
tation. Soc Econ Paleontol Mineral Spec Publ 37: Utah, USA. In: Rahmani RA, Flores RM (eds) Sedimen-
319-338 tology of coal and coal-bearing sequences. Int Assoc
Royden L, Horvath F, Rumpler J (1983) Evolution of the Sedimentol Spec Publ7: 217-227
Pannonian Basin system 1. Tectonics 2: 63-90 Ryer TA, Langer AW (1980) Thickness change involved in
Rubey WW (1938) The force required to move particles the peat-to-coal transformation for a bituminous coal
on a stream bed. US Geol Surv Profess Paper 189-E: of Cretaceous age in central Utah. J Sediment Petrol 50:
121-141 987-992
Rubey WW, Bass NW (1925) The geology of Russell Ryseth A (1989) Correlation of depositional patterns in the
County, Kansas. Kansas State Geol Surv Bull 10: 1-86 Ness Formation, Osberg area. In: Collinson JD (ed)
Ruddiman WF, Kutzbach JE (1990) Late Cenozoic plateau Correlation in hydrocarbon exploration. Graham and
uplift and climate change. R Soc Edinb Trans 81: Trotman, London, PP 313-326
301-314 Sadler PM (1981) Sedimentation rates and the complete-
Ruddiman WF, Prell WL, Raymo ME (1989) History of ness of stratigraphic sections. J Geol 89: 569-584
Late Cenozoic uplift in southern Asia and the American Sagoe K-MO, Visher GS (1977) Population breaks in grain-
west: rationale for general circulation modeling experi- size distributions of sand - a theoretical model. J Sedi-
ments. J Geophys Res 94: 18379-18391 ment Petrol 47: 285-310
Russell RJ (1954) Alluvial morphology of Anatolian rivers. Sahu BK (1964) Depositional mechanisms from the size
Ann Assoc Am Geogr 44: 363-391 analysis of clastic sediments. J Sediment Petrol 34:
Rust BR (1972) Structure and process in a braided river. 73-83
Sedimentology 18: 221-245 Salazar RB (1990) Hydrocarbon resources in the sub-
Rust BR (1975) Fabric and structure in glaciofluvial grav- Andean basins of Colombia. In: Eriksen GE, Pinochet
els. In: Jopling AV, McDonald BC (eds) Glaciofluvial MTC, Reinemund JA (eds) Geology of the Andes and its
and glaciolacustrine sedimentation. Soc Econ Paleo- relation to hydrocarbon and mineral resources.
ntol Mineral Spec PubI23.238-248 Circum-Pacific Council for Energy and Mineral Re-
Rust BR (1978a) A classification of alluvial channel sys- sources. Earth Sci Ser 11: 345-362
tems. In: Fluvial sedimentology. In: Miall AD (ed) Flu- Saleeby JS (1983) Accretionary tectonics of the North
vial sedimentology. Can Soc Petrol Geol Mem 5: American Cordillera. Annu Rev Earth Planet Sci 15:
187-198 45-73
Rust BR (1978b) Depositional models for braided allu- Salter T (1993) Fluvial scour and incision: models for their
vium. In: Miall AD (ed) Fluvial sedimentology. Can Soc influence on the development of realistic reservoir ge-
Petrol Geol Mem 5: 605-625 ometries. In: North CP, Prosser DJ (eds) Characteriza-
Rust BR (1981) Sedimentation in an arid-zone anastomos- tion of fluvial and aeolian reservoirs. Geol Soc Lond
ing fluvial system: Cooper's Creek, central Australia. J Spec Pub173: 33-51
Sediment Petrol 51: 745-755
References 557
Sanford BV, Thompson FJ, McFall FJ (1985) Plate tectonics Schumm SA, Brakenridge GR (1987) River responses. In:
- a possible controlling mechanism in the development Ruddiman WF, Wright HE Jr (eds) North America and
of hydrocarbon traps in southwestern Ontario. Bull adjacent oceans during the last deglaciation. Geol Soc
Can Petrol Geo133: 52-71 Am Geol North Am K-3: 221-240
Sanford RM (1970) Sarir oil field, Libya - desert surprise. Schumm SA, Khan HR (1972) Experimental study of chan-
In: Halbouty MT (ed) Geology of giant petroleum nel patterns. Geol Soc Am Bull 83: 1755-1770
fields. Am Assoc Petrol Geol Mem 14: 449-476 Schumm SA, Lichty RW (1963) Channel widening and
Saucier AE (1976) Tectonic influence on uraniferous flood plain construction along Cimarron River in
trends in the Late Jurassic Morrison Formation. In: southwestern Kansas. US Geol Surv Profess Paper 352-
Woodward LA, Northrop SA (eds) Tectonics and min- D: 71-88
eral resources of southwestern North America. New Schumm SA, Mosley MP, Weaver WE (1987) Experimental
Mex Geol Soc Spec Publ6: 151-157 fluvial geomorphology. Wiley, New York
Saucier RT (1974) Quaternary geology of the lower Missis- Schwartz DE (1978) Hydrology and current orientation
sippi Valley. Arkansas Archaeol Surv Res Ser 6 analysis of a braided to meandering transition; the Red
Saunderson HC, Lockett FPJ (1983) Flume experiments on River in Oklahoma and Texas, USA. In: Miall AD (ed)
bedforms and structures at the dune-plane bed transi- Fluvial sedimentology. Can Soc Petrol Geol Mem 5:
tion. In: Collinson JD, Lewin J (eds) Modern and an- 231-256
cient fluvial systems. Int Assoc Sedimentol Spec Publ6: Seeber L, Gornitz V (1983) River proflles along Himalayan
49-58 arc as indicators of active tectonics. Tectonophysics 92:
Schermer E, Howell DG, Jones DL (1984) The origin of 335-367
allochthonous terranes: perspectives on the growth and Sengor AMC (1976) Collision of irregular continental mar-
shaping of continents. Annu Rev Earth planet Sci 12: gins: implications for foreland deformation of Alpine-
107-131 type orogens. Geology 4: 779-782
Schlumberger Limited (1970) Fundamental of dipmeter Sengor AMC, Goriir D, Saroglu F (1985) Strike-slip fault-
interpretation. Schlumberger, New York ing and related basin formation in zones of tectonic
Schultz AW (1984) Subaerial debris flow deposition in the escape: Turkey as a case study. In: Biddle KT, Christie-
Upper Paleozoic Cutler Formation, western Colorado. J Blick N (eds) Strike-slip deformation, basin formation
Sediment Petrol 54: 759-772 and sedimentation. Soc Econ Paleontol Mineral Spec
Schumm SA (1963) A tentative classification of alluvial PubI37:227-264
river channels. US Geol Surv Circ 477 Seranne M, Seguret M (1987) The Devonian basins of
Schumm SA (1968a) Speculations concerning paleo- western Norway: tectonics and kinematics of an ex-
hydrologic controls of terrestrial sedimentation. Geol tel).ding crust. In: Coward MP;Dewey JF, Hancock PL
Soc Am Bull 79: 1573-1588 (eds) Continental extension tectonics. Geol Soc Lond
Schumm SA (1968b) River adjustment to altered hydro- Spec Pub128: 537-548
logic regimen - Murrumbidgee River and paleochan- Serra 0 (1989) Formation MicroScanner imager inter-
nels, Australia. US Geol Surv Profess Paper 598 pretation. Schlumberger Educational Services,
Schumm SA (1969) River metamorphosis. J Hydr Div Am Schlumberger
Soc Civil Eng 95: 255-273 Shanley KW, McCabe PJ (1991) Predicting facies architec-
Schumm SA (1972a) River morphology. Benchmark ture through sequence stratigraphy - an example from
papers in geology. Dowden, Hutchinson and Ross, the Kaiparowits Plateau, Utah. Geology 19: 742-745
Stroudsburg, Pennsylvania Shanley KW, McCabe PJ (1993) Alluvial architecture in a
Schumm SA (1972b) Fluvial paleochannels. In: Rigby JK, sequence stratigraphic framework: a case history from
Hamblin WK (eds) Recognition of ancient sedimentary the Upper Cretaceous of southern Utah, USA. In: Flint
environments. Soc Econ Paleontol Mineral Spec Publ SS, Bryant ID (eds) The geological modelling of hydro-
16:98-107 carbon reservoirs and outcrop analogues. Int Assoc
Schumm SA (1977) The fluvial system. Wiley, New York Sedimentol Spec Pub115: 21-56
Schumm SA (1979) Geomorphic thresholds: the concept Shanley KW, McCabe PJ (1994) Perspectives on the se-
and its applications. Trans Inst Br Geogr 4: 485-515 quence stratigraphy of continental strata. Am Assoc
Schumm SA (1981) Evolution and response of the fluvial Petrol Geol Bull 78: 544-568
system, sedimentological implications. In: Ethridge Shanley KW, McCabe pJ, Hettinger RD (1992) Tidal influ-
FG, Flores RM (eds) Recent and ancient nonmarine ences in Cretaceous fluvial strata from Utah, USA: a key
depositional environments: models for exploration. to sequence stratigraphic interpretation. Sedimentol-
Soc Econ Paleontol Mineral Spec Pub131: 19-29 ogy 39: 905-930
Schumm SA (1985a) Explanation and extrapolation in geo- Shannon PM, Naylor D (1989) Petroleum basin studies.
morphology: seven reasons for geologic uncertainty. Graham and Trotman, London
Trans Jpn Geomorphol Union 6: 1-18 Shantzer EV (1951) Alluvium of plains rivers in a temper-
Schumm SA (1985b) Patterns of alluvial rivers. Annu Rev ate zone and its significance for understanding the laws
Earth Planet Sci 13: 5-27 governing the structure and formation of alluvial
Schumm SA (1988) Variability of the fluvial system in suites. Trans Inst Geol Nauk Akad Nauk SSSR Geol Ser
space and time. In: Rosswall T, Woodmansee RG, 135
Risser PG (eds) Scales and global change. Wiley, New Sharp RP, Nobles LH (1953) Mudflows in 1941 at
York, pp 225-250 Wrightwood, southern California. Geol Soc Am Bull 64:
Schumm SA (1993) River response to base1evel change: 547-560
implications for sequence stratigraphy. J Geol 101: Shaui D, Qian K, Song Y, Ge R (1988) Stratigraphic-litho-
279-294 logic oil and gas pools in the Jiyang Depression, China.
558 References
In: Wagner HC, Wagner LC, Wang FFH, Wong FL (eds) Singh H, Parkash B, Gohain K (1993) Facies analysis of the
Petroleum resources of China and related subjects. Kosi megafan deposits. Sediment Geo185: 87-113
Circum-Pacific Council for Energy and Mineral Re- Singh IB, Kumar S (1974) Mega- and giant ripples in the
sources. Earth Sci Ser 10: 329-344 Ganga, Yamuna, and Son Rivers, Uttar Pradesh, India.
Shaw J, Kellerhals R (1977) Paleohydraulic interpretation Sediment Geo112: 53-66
of antidune bedforms with applications to antidunes in Sinha R, Friend PF (1994) River systems and their sedi-
gravel. J Sediment Petrol 47: 257-266 ment flux, Indo-Gangetic plains, northern Bihar, India.
Shelton JW (1967) Stratigraphic models and general crite- Sedimentology 41: 825-845
ria for recognition of alluvial, barrier-bar, and turbid- Sloss LL (1962) Stratigraphic models in exploration. Am
ity-current sand deposits. Am Assoc Petrol Geol Bull Assoc Petrol Geol Bull 46: 1050-1057
51:2441-2461 Sloss LL (1963) Sequences in the cratonic interior of North
Shelton JW, Noble RL (1974) Depositional features of America. Geol Soc Am Bull 74: 93-113
braided-meandering stream. Am Assoc Petrol Geol Sloss LL (1979) Global sea level changes: a view from the
Bull 58: 742-749 craton. In: Watkins JS, Montadert L, Dickerson PW
Shen HW (1978) Sediment transport models. In: Miall AD (eds) Geological and geophysical investigations of con-
(ed) Fluvial sedimentology. Can Soc Petrol Geol Mem 5: tinental margins. Am Assoc Petrol Geol Mem 29:
49-60 461-468
Shepherd RG (1987) Lateral accretion surfaces in ephem- Smith DG (1973) Aggradation of the Alexandra-North
eral-stream point bars, Rio Puerco, New Mexico. In: Saskatchewan River, Banff Park, Alberta. In: Morisawa
Ethridge FG, Flores RM, Harvey MD (eds) Recent de- M (ed) Fluvial geomorphology. Proceedings of the 4th
velopments in fluvial sedimentology. Soc Econ annual geomorphology symposium, publications in
Paleontol Mineral Spec Publ 39: 93-98 geomorphology. SUNY-Binghamton, New York, pp
Shields A (1936) Anwendung der Ahnlichkeitsmechanik 201-219
und der Turbulenzforschung auf die Geschiebebewe- Smith DG (1976) Effect of vegetation on lateral migration
gung. Mitteil Preuss Versuch Aust Wasserbau of anastomosed channels of a glacial meltwater river.
Schiffbau Bed 26 Geol Soc Am Bull 87: 857-860
Shlemon RJ (1975) Subaqueous delta formation - Smith DG (1983) Anastomosed fluvial deposits: modern
Atchafalaya Bay, Lousiana. In: Broussard ML (ed) Del- examples from western Canada. In: Collinson JD,
tas. Houston Geological Society, Texas, pp 209-222 Lewin J (eds) Modern and ancient fluvial systems. Int
Shultz AW (1984) Subaerial debris-flow deposition in the Assoc Sedimentol Spec Publ6: 155-168
Upper Paleozoic Cutler Formation, western Colorado. J Smith DG (1986) Anastomosing river deposits, sedimenta-
Sediment Petrol 54: 759-772 tion rates and basin subsidence, Magdalena River,
Shultz EH (1982) The chronosome and supersome: terms northwestern Colombia, South America. Sediment
proposed for low-rank chronostratigraphic units. Bull Geo146: 177-196
Can Petrol Geo130: 29-33 Smith DG (1987) Meandering river point bar lithofacies
Shurr GW (1984) Geometry of shelf-sandstone bodies in models: modern and ancient examples compared. In:
the Shannon Sandstone of southeastern Montana. In: Ethridge FG, Flores RM, Harvey MD (eds) Recent de-
Tillman RW, Siemers CT (eds) Siliciclastic shelf sedi- velopments in fluvial sedimentology. Soc Econ
ments. Soc Econ Paleontol Mineral Spec Pub134: 63-83 Paleontol Mineral Spec Publ 39: 83-91
Shuster MW, Steidtmann JR (1987) Fluvial-sandstone ar- Smith DG, Locking T (1989) Anastomosed river flood du-
chitecture and thrust-induced subsidence, northern ration and depositional rates in the upper Columbia
Green River Basin, Wyoming. In: Ethridge FG, Flores River, BC, Canada; implications for peat accumulation
RM, Harvey MD (eds) Recent developments in fluvial and interpretation of ancient coal deposits. In: 4th
sedimentology. Soc Econ Paleontol Mineral Spec Publ international conference on fluvial sedimentology,
39:279-285 Barcelona, Oct 1989, programme and abstracts, p 223
Siegenthaler C, Huggenberger P (1993) Pleistocene Rhine Smith DG, Putnam PE (1980) Anastomosed river deposits:
gravel: deposits of a braided river system with domi- modern and ancient examples in Alberta, Canada. Can
nant pool preservation. In: Best JL, Bristow CS (eds) J Earth Sci 17: 1396-1406
Braided rivers. Geol Soc Lond Spec Pub175: 147-162 Smith DG, Smith ND (1980) Sedimentation in anasto-
Simlote VN, Ebanks WJ Jr, Eslinger AV, Harpole KJ (1985) mosed river systems: examples from alluvial valleys
Synergistic evaluation of a complex conglomerate res- near Banff, Alberta. J Sediment Petrol 50: 157-164
ervoir for EOR, Barrancas Formation, Argentina. J Smith GA (1987) Sedimentology of volcanism-induced
Petrol Technol 37: 295-305 aggradation in fluvial basins: examples from the Pacific
Simons DB, Richardson EV (1961) Forms of bed roughness northwest, USA. In: Ethridge FG, Flores RM, Harvey
in alluvial channels. Am Soc Civil Eng Proc 87 (HY3): MD (eds) Recent developments in fluvial sedimentol-
87-105 ogy. Soc Econ Paleontol Mineral Spec Pub139: 217-228
Simons DB, Richardson EV, Nordin CF Jr (1965) Sedimen- Smith GA (1994) Climatic influences on continental depo-
tary structures generated by flow in alluvial channels. sition during late-stage fIlling of an extensional basin,
In: Middleton GV (ed) Primary sedimentary structures southeastern Arizona. Geol Soc Am Bull 106: 1212-1228
and their hydrodynamic interpretation. Soc Econ Smith ND (1970) The braided stream depositional envi-
Paleontol Mineral Spec Publ12: 34-52 ronment: comparison of the Platte River with some
Sinclair HD, Allen PA (1992) Vertical versus horizontal Silurian clastic rocks, north central Appalachians. Geol
motions in the Alpine orogenic wedge: stratigraphic Soc Am Bull 81: 2993-3014
response in the foreland basin. Basin Res 4: 215-232
References 559
Smith ND (1971) Transverse bars and braiding in the lower Sonnenberg SA (1987) Tectonic, sedimentary, and seismic
Platte River, Nebraska. Geol Soc Am Bull 82: 3407-3420 models for D sandstone, Zenith field area, Denver Ba-
Smith ND (1972) Some sedimentological aspects of planar sin, Colorado. Am Assoc Petrol Geol Bull 71: 1366-1377
cross-stratification in a sandy braided river. J Sediment Sorby HC (1852) On the oscillation of the currents drifting
Petrol 42: 624-634 the sandstone beds of the southeast of Northum-
Smith ND (1974) Sedimentology and bar formation in the berland, and on their general direction in the coalfield
upper Kicking Horse River, a braided outwash stream. in the neighbourhood of Edinburgh. Proc West York-
J Geo182: 205-224 shire Geol Soc 3: 232-240
Smith ND (1978) Some comments on terminology for bars Sorby HC (1859) On the structures produced by the cur-
in shallow rivers. In: Miall AD (ed) Fluvial sedimentol- rents present during the deposition of stratified rocks.
ogy. Can Soc Petrol Geol Mem 5: 85-88 Geologist 2: 137-147
Smith ND, Smith DG (1984) William River, an outstanding Sorby HC (1908) On the application of quantitative meth-
example of channel widening and braiding caused by ods to the study of the structure and history of rocks. Q
bed-load addition. Geology 12: 78-82 J Geol Soc Lond 64: 171-233
Smith ND, Cross TA, Dufficy JP, Clough SR (1989) Southard JB (1971) Representation of bed configurations
Anatomy of an avulsion. Sedimentology 36: 1-23 in depth-velocity-size diagrams. J Sediment Petrol 41:
Smith RMH (1987) Morphology and depositional history 903-915
of exhumed Permian point bars in the southwest Ka- Southard JB, Boguchwal LA, Romea RD (1980) Test of scale
roo, South Africa. J Sediment Petrol 57: 19-29 modelling of sediment transport in steady unidirec-
Smith SA (1987) Gravel counterpoint bars: examples from tional flow. Earth Surface Proc 5: 17-23
the River Tywi, South Wales. In: Ethridge FG, Flores Southard JB, Smith ND, Kuhnle RA (1984) Chutes and
RM, Harvey MD (eds) Recent developments in fluvial lobes: newly identified elements of braiding in shallow
sedimentology. Soc Econ Paleontol Mineral Spec Publ gravelly streams. In: Koster EH, Steel RJ (eds) Sedimen-
39:75-81 tology of gravels and conglomerates. Can Soc Petrol
Smith SA (1990) The sedimentology and accretionary style Geol Mem 10: 51-59
of an ancient gravel-bed stream: the Budleigh Salterton Spencer DW (1963) The interpretation of grain size distri-
Pebble Beds (Lower Triassic), southwest England. Sedi- bution curves of clastic sediments. J Sediment Petrol
ment Geo167: 199-219 33: 180-190
Smoot JP (1983) Depositional subenvironments in an arid Spieker EM (1946) Late Mesozoic and Early Cenozoic his-
closed basin; the Wilkins Peak Member of the Green tory of central Utah. US Geol Surv Profess Paper 205-D:
River Formation (Eocene), Wyoming, USA. Sedimen- 117-161
tology 30: 801-827 Spurr JE (1894a) False bedding in stratified drift deposits:
Sneed ED, Folk RL (1958) Pebbles in the lower Colorado American Geologist, v 13, p 43-47
River, Texas, a study in particle morphogenesis. J Geol Spurr JE (1894b) Oscillation and single current ripple
66: 114-150 marks. Am Geol13: 201-206
Sneh A (1983) Desert stream sequences in the Sinai Penin- Srivastava P, Parkash B, Sehgal JL, Kumar S (1994) Role of
sula. J Sediment Petrol 53: 1271-1280 neotectonics and climate in development of the Ho-
Sneider R, Massell W, Mathis R, Loren D, Wichmann P locene geomorphology and soils of the Gangetic Plains
(1991) The integration of geology, geophysics, petro- between the Ramganga and Rapti rivers. Sediment Geol
physics and petroleum engineering in reservoir deli- 94: 129-151
neation, description and management, Proceedings of Srivastava SP, Tapscott CR (1986) Plate kinematics of the
the 1st Arche conference, Oct 1990. American Associa- North Atlantic. In: Vogt PR, Tucholke BE (eds) The
tion of Petroleum Geologists, Tulsa, Oklahoma Western North Atlantic region. The Geology of North
Soegaard K (1990) Fan-delta and braid-delta systems in America vol M. Geological Society of America, Boulder,
Pennsylvanian Sandia Formation, Taos Trough, north- Colorado, pp 379-404
ern New Mexico: depositional and tectonic implica- Stabler CL (1990) Andean hydrocarbon resources - an
tions. Geol Soc Am Bull 102: 1325-1343 overview. In: Eriksen GE, Pinochet MTC, Reinemund
Soegaard K (1992) Architectural elements of fan-delta JA (eds) Geology of the Andes and its relation to hydro-
complex in Pennsylvanian Sandia Formation, Taos carbon and mineral resources. Circum-Pacific Council
Trough, Northern New Mexico. In: Miall AD, Tyler N for Energy and Mineral Resources. Earth Sci Ser 11:
(eds) The three-dimensional facies architecture of ter- 431-438
rigenous clastic sediments, and its implications for hy- Stamp LD (1925) Seasonal rhythms in the Tertiary sedi-
drocarbon discovery and recovery. Soc Econ Paleontol ments of Burma. Geol Mag 62: 515-528
Mineral Conc Models Ser, pp 217-223 Stancliffe RJ, Adams ER (1986) Lower Tuscaloosa fluvial
Soeparjadi RA, Valachi LZ, Sosromihardjo S (1986) Oil and channel styles at Liberty field, Amite County, Missis-
gas developments in Far East in 1985. Am Assoc Petrol sippi. Trans Gulf Coast Assoc Geol Soc 36: 305-313
Geol Bull 70: 1479-1565 Stanistreet IG (1993) Ancient and modern examples of
Song G (1988) Jurassic paleo-landform oil fields in the tectonic escape basins: the Archaean Witwatersrand
Ordos Basin, North China. In: Wagner HC, Wagner LC, Basin compared with the Cenozoic Maracaibo Basin.
Wang FFH, Wong FL (eds) Petroleum resources of In: Frostick LE, Steel RJ (eds) Tectonic controls and
China and related subjects. Circum-Pacific Council for signatures in sedimentary successions. Int Assoc
Energy and Mineral Resources. Earth Sci Ser 10: Sedimentol Spec Pub120: 363-376
371-386
560 References
Stanistreet IG, McCarthy TS (1993) The Okavango fan and Steel RJ, Gjelberg J, Helland-Hansen W, Kleinspehn K,
the classification of subaerial fan systems. Sediment Nottvedt A, Rye-Larsen M (1985) The Tertiary strike-
Geo185: 115-133 slip basins and orogenic belt of Spitsbergen. In: Biddle
Stanistreet IG, Cairncross B, McCarthy TS (1993) Low KT, Christie-Blick N (eds) Strike-slip deformation, ba-
sinuosity and meandering bedload rivers of the sin formation and sedimentation. Soc Econ Paleontol
Okavango fan: channel confinement by vegetated Mineral Spec Pub137: 339-359
levees without fine sediment. Sediment Geol 85: Steel RJ, Maehl S, Nilsen H, Roe SL, Spinnangr A (1977)
135-156 Coarsening-upward cycles in the alluvium of Hornelen
Stanley KO (1976) Sandstone petrofacies in the Cenozoic Basin (Devonian), Norway. Sedimentary response to
High Plains sequence, eastern Wyoming and Nebraska. tectonic events. Geol Soc Am Bull 88: 1124-1134
Geol Soc Am Bull 87: 297-309 Stephens M (1994) Architectural element analysis within
Stanley KO, Jorde K, Raestad N, Stockbridge CP (1990) the Kayenta Formation (Lower Jurassic) using ground-
Stochastic modeling of reservoir sand bodies for input probing radar and sedimentological profIling, south-
to reservoir simulation, Snorre field, northern North western Colorado. Sediment Geo190: 179-211
Sea, Norway. In: Buller AT, Berg E, Hjelmeland 0, Stewart DJ (1983) Possible suspended-load channel depos-
Kleppe J, Torsaeter 0, Aasen JO (eds) North Sea oil its from the Wealden Group (Lower Cretaceous) of
and gas reservoirs I: Graham and Trotman, London, southern England. In: Collinson JD, Lewin J (eds) Mod-
pp 91-101 ern and ancient fluvial systems. Int Assoc Sedimentol
Stanley KO, Wayne WJ (1972) Epeirogenic and climatic Spec Publ6: 369-384
controls of Early Pleistocene fluvial sediment dispersal Stewart JH (1978) Basin-range structure in western North
in Nebraska. Geol Soc Am Bull 83: 3675-3690 America: a review. In: Smith RB, Eaton GP (eds) Ceno-
Stanmore PJ, Johnstone EM (1988) The search for strati- zoic tectonics and regional geophysics of the Western
graphic traps in the southern Patchawarra Trough, Cordillera. Geol Soc Am Mem 152: 1-31
South Australia. Aust Petrol Explor Assoc J 28: 156-165 Stockmal GS, Beaumont C (1987) Geodynamic models of
Stapp RW (1967) Relationship ofLower Cretaceous depo- convergent margin tectonics: the southern Canadian
sitional environment to oil accumulation, northeastern Cordillera and the Swiss Alps. In: Beaumont C, Tankard
Powder River Basin, Wyoming. Am Assoc Petrol Geol AJ (eds) Sedimentary basins and basin-forming mecha-
Bull 51: 2044-2055 nisms. Can Soc Petrol Geol Mem 12: 393-411
Statham I (1976) Debris flows on vegetated screes in the Stockmal GS, Beaumont C, Boutilier R (1986) Geodynamic
Black Mountain, Carmarthenshire. Earth Surface Proc models of convergent margin tectonics: transition from
1: 173-180 rifted margin to overthrust belt and consequences for
Stear WM (1983) Morphological characteristics of ephem- foreland-basin development. Am Assoc Petrol Geol
eral stream channel and overbank splay sandstone bod- Bull 70: 181-190
ies in the Permian Lower Beaufort Group, Karoo Basin, Stockmal GS, Cant DJ, Bell JS (1992) Relationship of the
South Africa. In: Collinson JD, Lewin J (eds) Modern stratigraphy of the Western Canada foreland basin to
and ancient fluvial systems. Int Assoc Sedimentol Spec Cordilleran tectonics: insights from geodynamic mod-
Publ 6: 405-420 els. In: Macqueen RW, Leckie DA (eds) Foreland basin
Stear WM (1985) Comparison ofthe bedform distribution and fold belts. Am Assoc Petrol Geol Mem 55: 107-124
and dynamics of modern and ancient sandy ephemeral Stollhofen H, Stanistreet IG (1994) Interaction between
flood deposits in the southwestern Karoo region, South bimodal volcanism, fluvial sedimentation and basin
Africa. Sediment Geo145: 209-230 development in the Permo-Carboniferous Saar-Nahe
Steel RJ (1974) New Red Sandstone floodplain and pied- Basin (south-west Germany). Basin Res 6: 245-267
mont sedimentation in the Hebridian Province, Scot- Stott DF, Aitken JD (eds) (1993) Sedimentary cover of the
land. J Sediment Petrol 44: 336-357 craton in Canada. Geological Survey of Canada, Ot-
Steel RJ, Aasheim SM (1978) Alluvial sand deposition in a tawa, Canada (Geology of Canada no 5)
rapidly subsiding basin (Devonian, Norway). In: Miall Struijk AP, Green RT (1991) The Brent field, Block 211/29,
AD (ed) Fluvial sedimentology. Can Soc Petrol Geol UK North Sea. In: Abbotts IL (ed) United Kingdom oil
Mem 5: 385-412 and gas fields, 25 year Commemorative volume. Geo-
Steel RJ, Gloppen TG (1980) Late Caledonian (Devonian) logical Society, London, pp 63-72
basin formation, western Norway: signs of strike-slip Stuart lA, Cowan G (1991) The South Morecambe field,
tectonics during infIlling. In: Ballance PF, Reading HG Blocks 11012a, 11013a, 110/8a, UK East Irish Sea. In:
(eds) Sedimentation in oblique-slip mobile zones. Int Abbotts IL (ed) United Kingdom oil and gas fields, 25
Assoc Sedimentol Spec Publ4: 79-103 year Commemorative volume. Geological Society, lon-
Steel RJ, Ryseth A (1990) The Triassic - Early Jurassic don, pp 527-541
succession in the northern North Sea: megasequence Stuart WJ, Kennedy S, Thomas AD (1988) The influence of
stratigraphy and intra-Triassic tectonics. In: Hardman structural growth and other factors on the configura-
RFP, Brooks J (eds) Tectonic events responsible for tion of fluviatile sandstones, Permian Cooper Basin.
Britain's oil and gas reserves. Geol Soc Lond Spec Publ Aust Petrol Explor Assoc J 28: 255-266
55: 139-168 Summerson CH (ed) (1976) Sorby on sedimentology: a
Steel RJ, Dalland A, KalgraffK, Larsen V (1981) The Cen- collection of papers from 1851 to 1908 by Henry Clifton
tral Tertiary Basin of Spitsbergen: sedimentary devel- Sorby. Geological Milestones I. Comparative sedimen-
opment of a sheared-margin basin. In: Kerr JW, tology laboratory, University of Miami, Miami
Fergusson AJ (eds) Geology of the North Atlantic bor- Sundborg A (1956) The River Klariilven, a study of fluvial
derlands. Can Soc Petrol Geol Mem 7: 647-664 processes. Geogr Ann 38: 125-316
References 561
Surell A (1841) Etude sur les torrents des Hautes-Alpes. Collinson JD (ed) Correlation in hydrocarbon explora-
Dunod, Paris tion. Graham and Trotman, London, pp 45-51
Surell A (1870) Etude sur les torrents des Hautes-Alpes, Titheridge DG (1993) The influence of half-graben
2nd edn Dunod, Paris syndepositional tilting on thickness variation and seam
Suter JR, Berryhill HL, Penland S (1987) Late Quaternary splitting in the Brunner Coal Measures, New Zealand.
sea-level fluctuations and depositional sequences, Sediment Geo187: 195-213
southwest Louisiana continental shelf. In: Nummedal Todd SP, Went DJ (1991) Lateral migration of sand-bed
D, Pilkey OH, Howard JD (eds) Sea-level fluctuation rivers: examples from the Devonian Glashabeg Forma-
and coastal evolution. Soc Econ Paleontol Mineral Spec tion, SW Ireland and the Cambrian Alderney Sand-
Pub141: 199-219 stone Formation, Channel Islands. Sedimentology 38:
Swift DIP, Rice DD (1984) Sand bodies on muddy shelves: 997-1020
a model for sedimentation in the Western Interior Cre- Tolman CF (1909) Erosion and deposition in the southern
taceous seaway, North America. In: Tillman RW, Arizona boIs on region. J Geo117: 136-163
Siemers CT (eds) Siliciclastic shelf sediments. Soc Econ Tornqvist TE (1993) Holocene alternation of meandering
Paleontol Mineral Spec Publ 34: 43-62 and anastomosing fluvial systems in the Rhine-Mesue
Swift DJP, Hudelson PM, Brenner RL, Thompson P (1987) delta (central Netherlands) controlled by sea-level rise
Shelf construction in a foreland basin: storm beds, shelf and subsoil erodibility. J Sediment Petrol 63: 683-693
sandbodies, and shelf-slope depositional sequences in Tornqvist TE (1994) Middle and late Holocene avulsion
the Upper Cretaceous Mesaverde Group, Book Cliffs, history of the River Rhine (Rhine-Meuse delta, Nether-
Utah. Sedimentology 34: 423-457 lands). Geology 22: 711-714
Talbot MR, Holm K, Williams MAJ (1994) Sedimentation Tornqvist TE, van Ree MHM, Faessen ELJH (1993) Longi-
in low-gradient desert margin systems: a comparison of tudinal facies architectural changes of a Middle
the Late Triassic of northwest Somerset (England) and Holocene anastomosing distributary system (Rhine-
the late Quaternary of east-central Australia. In: Rosen Meuse delta, central Netherlands). Sediment Geol 85:
MR (ed) Paleoclimate and basin evolution of playa 203-219
systems. Geol Soc Am Spec Paper 289: 97-117 Trewin NH (1993a) Mixed aeolian sandsheet and fluvial
Tandon SK, Gibling MR (1994) Calcrete and coal in late deposits in the Tumblagooda Sandstone, Western Aus-
Carboniferous cyclothems of Nova Scotia, Canada: cli- tralia. In: North CP, Prosser DJ (eds) Characterization
mate and sea-level changes linked. Geology 22: 755-758 of fluvial and aeolian reservoirs. Geol Soc Lond Spec
Tang Z (1982) Tectonic features of oil and gas basins in PubI73:219-230
eastern part of China. Am Assoc Petrol Geol Bull 66: Trewin NH (1993b) Controls on fluvial deposition in
509-521 mixed fluvial and aeolian facies within the
Tankard AJ, Jackson MPA, Eriksson KA, Hobday DK, Tumblagooda Sandstone (Late Silurian) of Western
Hunter DR, Minter WEL (1982) Crustal evolution of Australia. Sediment Geo185: 387-400
Southern Africa. Springer, Berlin Heidelberg New York Trifonov VG (1978) Late Quaternary tectonic movements
Tanner WF (1955) Paleogeographic reconstructions from of western and central Asia. Geol Soc Am Bull 89:
cross-bedding studies. Am Assoc Petrol Geol Bull 39: 1059-1072
2471-2483 Trowbridge AC (1911) The terrestrial deposits of Owens
Tapponnier P, Peltzer G, Armijo R (1986) On the mechan- Valley, California. J Geo119: 706-747
ics of the collision between India and Asia. In: Coward Tunbridge IP (1981) Sandy high-energy flood sedimenta-
MP, Ries AC (eds) Collision tectonics. Geol Soc Lond tion - some criteria for recognition, with an example
Spec Pub119: 115-157 from the Devonian of SW England. Sediment Geol 28:
Taylor G, Woodyer KD (1978) Bank deposition in sus- 79-96
pended-load streams. In: Miall AD (ed) Fluvial sedi- Tunbridge IP (1984) Facies models for a sandy ephemeral
mentology. Can Soc Petrol Geol Mem 5: 257-275 stream and clay playa complex; the Middle Devonian
Teng LS (1990) Geotectonic evolution oflate Cenozoic arc- Trentishoe Formation of North Devon, UK. Sedimen-
continent collision in Taiwan. Tectonophysics 183: tology 31: 697-716
57-76 Turner BR (1978) Sedimentary patterns of uranium
Thakur GC (1991) Waterflood surveillance techniques - a mineralisation in the Beaufort Group of the southern
reservoir management approach. J Petrol Technol 43: Karoo (Gondwana) Basin, South Africa. In: Miall AD
1180-1192 (ed) Fluvial sedimentology. Can Soc Petrol Geol Mem 5:
Theriault P, Desrochers A (1993) Carboniferous calcretes 831-848
in the Canadian Arctic. Sedimentology 40: 449-466 Turner BR (1983) Braidplain deposition of the Upper Tri-
Thomas RG, Smith DG, Wood JM, Visser J, Calverley- assic Molteno Formation in the main Karoo (Gond-
Range EA, Koster EH (1987) Inclined heterolithic wana) Basin, South Africa. Sedimentology 30: 77-89
stratification - terminology, description, interpreta- Turner BR, Munro M (1987) Channel formation and mi-
tion and significance. Sediment Geo153: 123-179 gration by mass-flow processes in the Lower Carbonif-
Thorne CR, Russell APG, Alam MK (1993) Planform pat- erous fluviatile Fell Sandstone Group, northeast
tern and channel evolution of the Brahmaputra River, England. Sedimentology 34: 1107-1122
Bangladesh. In: Best JL, Bristow CS (eds) Braided riv- Turner P (1980) Continental red beds. Elsevier Scientific,
ers. Geol Soc Lond Spec Publ 75: 257-276 Amsterdam (Developments in sedimentology, vol 29)
Tipper JC (1989) Computer modelling of seismic facies: Twenhofel WH (1932) Treatise on sedimentation, 2nd edn.
implications for seismic and sequence stratigraphy. In: Williams and Wilkins, New York
562 References
Tye RS (1991) Fluvial-sandstone reservoirs of the Travis Vanoni VA (1946) Transportation of suspended sediment
Peak Formation, East Texas Basin. In: Miall AD, Tyler, by water. Am Assoc Civil Eng Trans 111: 67-102
N (eds) The three-dimensional facies architecture of Van Overmeeren RA, Staal JH (1976) Floodfan sedimenta-
terrigenous clastic sediments, and its implications for tion and gravitational anomalies in the Salar de Punta
hydrocarbon discovery and recovery. Soc Econ Negra, northern Chile. Geol Rundsch 65: 195-211
Paleontol Mineral Conc Models Ser 3: 172-188 Van Straaten LMJU (1954) Sedimentology of Recent tidal
Tyler N, Ethridge FG (1983) Fluvial architecture ofJurassic flat deposits and the Psammites du Condroz (Devo-
uranium-bearing sandstones, Colorado Plateau, west- nian). Geol Mijnbouw 16: 25-47
ern United States. In: Collinson JD, Lewin J (eds) Mod- Van Wagoner JC, Mitchum RM Jr, Posamentier HW, Vail
ern and ancient fluvial systems. Int Assoc Sedimentol PR (1987) Seismic stratigraphy interpretation using
Spec Publ6: 533-547 sequence stratigraphy, Part 2: key definitions of se-
Tyler N, Galloway WE, Garrett CM Jr, Ewing TE (1984) Oil quence stratigraphy. In: Bally AW (ed) Atlas of seismic
accumulation, production characteristics, and targets stratigraphy. Am Assoc Petrol Geol Stud Geol 27 (1):
for additional recovery in major oil reservoirs of Texas. 11-14
University of Texas, Bureau of Economic Geology Van Wagoner JC, Mitchum RM, Campion KM, Rahmanian
(Geological circular 84-2) VD (1990) Siliciclastic sequence stratigraphy in well
Udden JA (1914) Mechanical composition of clastic sedi- logs, cores, and outcrops. Am Assoc Petrol Geol Meth-
ments. Geol Soc Am Bull 25: 655-744 ods Explor Ser 7
Vail PR (1987) Seismic stratigraphy interpretation using Van Wagoner JC, Nummedal D, Jones CR, Taylor DR,
sequence stratigraphy, part 1: seismic stratigraphy in- Jennette DC, Riley GW (1991) Sequence stratigraphy
terpretation procedure. In: Bally AW (ed) Atlas of seis- applications to shelf sandstone reservoirs. American
mic stratigraphy. Am Assoc Petrol Geol Stud Geol 27 Association of Petroleum Geologists Field Conference
(1): 1-10 Guidebook, Tulsa
Vail PR, Todd RG (1981) Northern North Sea Jurassic Veevers JJ (ed) (1984) Phanerozoic earth history of Austra-
unconformities, chronostratigraphy and sea-level lia. Oxford University Press, Oxford
changes from seismic stratigraphy. In: Illing LV, Verdier AC, Oki T, Suardy A (1980) Geology of the Handil
Hobson GD (eds) Petroleum geology of the continental field (East Kalimantan, Indonesia). In: Halbouty MT
shelf of northwest Europe. Institute of Petroleum, Lon- (ed) Giant oil and gas fields of the decade: 1968-1978.
don, pp 216-235 Am Assoc Petrol Geol Mem 30: 399-421
Vail PR, Mitchum RM Jr, Todd RG, Widmier JM, Thomp- Vessell RK, Davies DK (1981) Nonmarine sedimentation
son S III, Sangree JB, Bubb IN, Hatlelid WG (1977) in an active fore arc basin. In: Ethridge FG, Flores RM
Seismic stratigraphy and global changes of sea-level. (eds) Recent and ancient nonmarine depositional envi-
In: Payton CE (ed) Seismic stratigraphy - applications ronments: models for exploration. Soc Econ Paleontol
to hydrocarbon exploration. Am Assoc Petrol Geol Mineral Spec Pub131: 31-45
Mem 26: 49-212 Vincent P, Gartner JE, Attali G (1977) Geodip: an approach
Vail PR, Hardenbol J, Todd RG (1984) Jurassic uncon- to detailed dip determination using correlation by pat-
formities, chronostratigraphy and sea-level changes tern recognition. 6th formation evaluation symposium,
from seismic stratigraphy and biostratigraphy. In: Canadian Well Logging Society, paper L
Schlee JS (ed) Interregional unconformities and hydro- Vincent P, Gartner JE, Attali G (1979) Geodip: an approach
carbon exploration. Am Assoc Petrol Geol Mem 36: to detailed dip determination using correlation by pat-
129-144 tern recognition. J Petrol Technol31: 232-240
Vail PR, Audemard F, Bowman SA, Eisner PN, Perez-Crus Vis eras C, Fernandez J (1994) Channel migration patterns
C (1991) The stratigraphic signatures of tectonics, and related sequences in some alluvial fan valleys. Sedi-
eustasy and sedimentology - an overview. In: Einsele G, ment Geo188: 201-217
Ricken W, Seilacher A (eds) Cycles and events in Visher GS (1965a) Fluvial processes as interpreted from
stratigraphy. Springer, Berlin Heidelberg New York, pp ancient and recent fluvial deposits. In: Middleton GV
617-659 (ed) Primary sedimentary structures and their hydro-
Vandenberghe J (1993) Changing fluvial processes under dynamic interpretation. Soc Econ Paleontol Mineral
changing periglacial conditions. Z Geomorph NF 88: Spec Pub112: 84-115
17-28 Visher GS (1965b) Use of vertical profrle in environmental
Vandenberghe J, Kasse C, Bohnke S, Kozarski S (1994) reconstruction. Am Assoc Petrol Geol Bull 49: 41-61
Climate-related river activity at the Weichselian-Ho- Visher GS (1969) Grain size distributions and depositional
locene transition: a comparative study of the Wart a and processes. J Sediment Petrol 39: 1074-1106
Maas rivers. Terra Nova 6: 476-485 Visher GS, Saitta S, Phares RS (1971) Pennsylvanian delta
van der Lingen GJ (1982) Development ofthe North Island patterns and petroleum occurrences in eastern Okla-
subduction system, New Zealand. In: Leggett JK (ed) homa. Am Assoc Petrol Geol Bull 55: 1206-1230
Trench-forearc geology. Geol Soc Lond Spec Publ 10: Visser JNJ, Dukas BA (1979) Upward-fining fluviatile
259-272 megacycles in the Beaufort Group, north of Graaff-
Van Houten FB (1973) Origin of red beds: a review - Reinet, Cape Province. Trans Geol Soc South Afr 82:
1961-1972. Annu Rev Earth Planet Sci 1: 39-61 149-154
Van Houten FB (1981) The Odyssey of Molasse. In: Miall Visser MJ (1980) Neap-spring cycles reflected in Holocene
AD (ed) Sedimentation and tectonics in alluvial basins. subtidal large-scale bedform deposits: a preliminary
Geol Assoc Can Spec Paper 23: 35-48 note. Geology 8: 543-546
References 563
von Hoff KEA (1822-1824) Geschichte der durch Uber- Wasson RJ (1977a) Catchment processes and the evolution
lieferung nachgewiesenen natnrlichen Veranderungen of alluvial fans in the lower Derwent valley, Tasmania. Z
der Erdoberflache, 2 vols. Perthes, Gotha GeomorphoI21:'147-168
von Zittel KA (1901) History of geology and paleontology Wasson RJ (1977b) Last-glacial alluvial fan sedimentation
to the end of the nineteenth century (translated by MM in the Lower Derwent valley, Tasmania. Sedimentology
Ogilvie-Gordon). Scott, London 24: 781-799
Vos RG, Tankard AJ (1981) Braided fluvial sedimentation Waters KH, Rice JW (1975) Some statistical and probabi-
in the Lower Paleozoic Cape Basin, South Africa. Sedi- listic techniques to optimize the search for stratigra-
ment Geo129: 171-193 phic traps using seismic data. Ninth World Petroleum
Wadman DH, Lamprecht DE, Mrosovsky I (1979) Joint Congress Proceedings, Tokyo (Panel Discussion 9, pa-
geologic/engineering analysis of the Sadlerochit reser- per 1)
voir, Prudhoe Bay field. J Petrol Technol31: 933-940 Watts AB (1981) The US Atlantic margin: subsidence his-
Walker RG (1976) Facies models 3: sandy fluvial systems. tory, crustal structure and thermal evolution. Ameri-
Geosci Can 3: 101-109 can Association of Petroleum Geologists (Education
Walker RG (ed) (1979) Facies models. Geological Associa- Course Notes Series 19)
tion of Canada, St. John's, Newfoundland (Geosci Can Watts KJ (1987) The Hutton Sandstone-Birkhead Forma-
Reprint Ser 1) tion transition, ATP 269P(1), Eromanga Basin. Aust
Walker RG (ed) (1984) Facies models, 2nd edn. Geological Petrol Explor Assoc J 27: 215-228
Association of Canada, St. John'S, Newfoundland Weber KJ (1993) The use of3-D seismic in reservoir geo-
(Geosci Can Reprint Ser 1) logical modelling. In: Flint SS, Bryant ID (eds) The
Walker RG (1990) Perspective - facies modelling and se- geological modelling of hydrocarbon reservoirs and
quence stratigraphy. J Sediment Petrol 60: 777-786 outcrop analogues. Int Assoc Sedimentol Spec Pub115:
Walker RG, Cant DJ (1979) Sandy fluvial systems. In: 181-188
Walker RG (ed) Facies models. Geosci Can Reprint Ser Weber KJ, Van Geuns LC (1990) Framework for construct-
1:23-31 ing clastic reservoir simulation models. J Petrol
Walker RG, Cant DJ (1984) Sandy fluvial systems. In: Technol42: 1248-1253, 1296-1297
Walker, R. G (ed) Facies models, 2nd edn. Geosci Can Wecker HRB (1989) The Eromanga Basin. Aust Petrol
Reprint Ser 1: 71-89 Explor Assoc J 29: 379-397
Walker RG, James NP (eds) (1992) Facies models: response Weimer P, Posamentier HW (1993) Recent developments
to sea level change. Geological Association of Canada, and applications in siliciclastic sequence stratigraphy.
St. John's, Newfoundland In: Weimer P, Posamentier HW (eds) Siliciclastic se-
Walker TR (1967) Formation of red beds in modern and quence stratigraphy. Am Assoc Petrol Geol Mem 58:
ancient deserts. Geol Soc Am Bull 78: 353-368 3-12
Walling DE, Webb BW (1983) Patterns of sediment yield. Weimer RJ (1960) Upper Cretaceous stratigraphy, Rocky
In: Gregory KJ (ed) Background to paleohydrology. Mountain area. Am Assoc Petrol Geol Bull 44: 1-20
Wiley, Chichester, pp 69-100 Weimer RJ (1970) Rates of deltaic sedimentation and
Walther J (189311894) Einleitung in die Geologie als intrabasin deformation, Upper Cretaceous of Rocky
historische Wissenschaft. Fischer, Jena, 3 vols Mountain region. In: Morgan JP (ed) Deltaic sedimen-
Walton AW, Bouquet DJ, Evenson RA, Rofheart DH, tation, modern and ancient. Soc Econ Paleontol Min-
Woody MD (1986) Characterization of sandstone res- eral Spec Pub115: 270-292
ervoirs in the Cherokee Group (Pennsylvanian, Weimer RJ (1986) Relationship of unconformities, tecton-
Desmoinesian) of southeastern Kansas. In: Lake LW, ics, and sea level change in the Cretaceous of the West-
Carroll HB Jr (eds) Reservoir characterization. Aca- ern Interior, United States. In: Peterson JA (ed)
demic, Orlando, pp 39-62 Paleotectonics and sedimentation in the Rocky Moun-
Wang QM, Coward MP (1993) The Jiuxi Basin, Hexi corri- tain region, United States. Am Assoc Petrol Geol Mem
dor, NW China: foreland structural features and hydro- 41:397-422
carbon potential. J Petrol Geo116: 169-182 Weirich TE (1929) Cushing oil and gas field, Creek County,
Wanless HR, Weller JM (1932) Correlation and extent of Oklahoma. In: Structure of typical American oil fields,
Pennsylvanian cyclothems. Geol Soc Am Bull vol 2. American Association of Petroleum Geologists,
43: 1003-1016 Tulsa, pp 396-406
Wanless HR, Tyrrell KM, Tedesco LP, Dravis JJ (1988) Wells NA (1983) Transient streams in sand-poor red beds:
Tidal-flat sedimentation from Hurricane Kate, Caicos Early Middle Eocene Kuldana Formation of northern
platform, British West Indies. J Sediment Petrol 58: Pakistan. In: Collinson JD, Lewin J (eds) Modern and
724-738 ancient fluvial systems. Int Assoc Sedimentol Spec Publ
Warris BJ (1988) The geology of the Mount Horner oilfield, 6:393-403
Perth Basin, Western Australia. Aust Petrol Explor Wells NA, Dorr JA Jr (1987) Shifting of the Kosi River,
Assoc J 28: 88-99 northern India. Geology 15: 204-207
Warwick PD, Flores RM (1987) Evolution of fluvial styles Wells SG, Harvey AM (1987) Sedimentologic and geomor-
in the Eocene Wasatch Formation, Powder River Basin, phic variations in storm-generated alluvial fans,
Wyoming. In: Ethridge FG, Flores RM (eds) Recent and Howgill fells, northwest England. Geol Soc Am Bull 98:
ancient nonmarine depositional environments: models 182-198
for exploration. Soc Econ Paleontol Mineral Spec Publ Wentworth CK (1922) A scale of grade and class terms for
31: 303-310 clastic sediments. J Geo130: 377-392
564 References
Werren EG, Shew RD, Adams ER, Stancliffe RJ (1990) Willis BJ, Behrensmeyer AK (1994) Architecture of Mi-
Meander-belt reservoir geology, mid-dip Tuscaloosa, ocene overbank deposits in northern Pakistan. J Sedi-
Little Creek field, Mississippi. In: Barwis JH, ment Res B64: 60-67
McPherson JG, Studlick RJ (eds) Sandstone petroleum Wilson JL (1975) Carbonate facies in geologic history.
reservoirs. Springer, Berlin Heidelberg New York, pp Springer, Berlin Heidelberg New York
85-107 Wilson LG (1972) Charles Lyell, the years to 1841: the
Wescott WA (1990) The Yallahs fan delta: a coastal fan in a revolution in geology. Yale University Press, New Ha-
humid tropical climate. In: Rachocki AH, Church M ven, Connecticut
(eds) Alluvial fans: a field approach. Wiley, Chichester, Wilson RCL, Williams CA (1979) Oceanic transform struc-
pp 213-225 tures and the development of Atlantic continental mar-
Wescott WA (1993) Geomorphic thresholds and complex gin sedimentary basins - a review. J Geol Soc Lond 136:
response of fluvial systems - some implications for 311-320
sequence stratigraphy. Am Assoc Petrol Geol Bull 77: Winder CG (1965) Alluvial cone construction by alpine
1208-1218 mudflow in a humid temperate region. Can J Earth Sci
Weston PJ, Alexander J (1993) Computer modelling of flow 2:270-277
lines over deformed surfaces: the implications for pre- Wing SL (1984) Relation of paleovegetation to geometry
diction of alluvial facies distribution. In: Marzo M, and cyclicity of some fluvial carbonaceous deposits. J
Puigdefabregas C (eds) Alluvial sedimentation. Int Sediment Petrol 54: 52-66
Assoc Sedimentol Spec Pub117: 211-217 Winker CD (1982) Cenozoic shelf margins, northwest Gulf
White DA (1980) Assessing oil and gas plays in facies-cycle of Mexico basin. Trans Gulf Coast Assoc Geol Soc 32:
wedges. Am Assoc Petrol Geol Bull 64: 1158-1178 427-448
White DA (1988) Oil and gas play maps in exploration and Winkley BR (1977) Manmade cutoffs on the lower Missis-
assessment. Am Assoc Petrol Geol Bull 72: 944-949 sippi River; conception, construction and river re-
Wightman DM, Tilley BJ, Last BM (1981) Stratigraphic sponse. US Army Corps of Engineers, Vicksburg,
traps in channels sandstones in the upper Mannville Mississippi
(Albian) of east-central Alberta: discussion. Bull Can Winn RD Jr, Steinmetz JC, Kerekgyarto WL (1993) Strati-
Petrol Geo129: 622-625 graphy and rifting history of the Mesozoic-Cenozoic
Wightman DM, Pemberton SG, Singh C (1987) Deposi- Anza Rift, Kenya. Am Assoc Petrol Geol Bull 77:
tional modelling of the Upper Manville (Lower Creta- 1989-2005
ceous), east central Alberta: implications for the Winsemann J, Seyfried H (1991) Response of deep-water
recognition of brackish water deposits. In: Tillman RW, fore-arc systems to sea-level changes, tectonic activity
Weber KJ (eds) Reservoir sedimentology. Soc Econ and volcaniclastic input in Central America. In:
Paleontol Mineral Spec Pub140: 189-220 Macdonald DIM (ed) Sedimentation, tectonics and
Wilgus CK, Hastings BS, Kendall CGSC, Posamentier HW, eustasy: sea-level changes at active margins. Int Assoc
Ross CA, Van Wagoner JC (eds) (1988) Sea-level Sedimentol Spec Pub112: 273-292
. changes: an integrated approach. Soc Econ Paleontol Wise DU, BeltES, Lyons PC (1991) Clastic diversion by fold
Mineral Spec Publ 42 salients and blind thrust ridges in coal-swamp develop-
Williams GE (1971) Flood deposits of the sandbed ephem- ment. Geology 19: 514-517
eral streams of central Australia. Sedimentology 17: Withrow PC (1968) Depositional environments of Penn-
1-40 sylvanian Red Fork Sandstone in northeastern
Williams H, Soek HF (1993) Predicting reservoir sandbody Anadarko Basin, Oklahoma. Am Assoc Petrol Geol Bull
orientation from dipmeter data: the use of sedimentary 52: 1638-1654
dip profiles from outcrop studies. In: Flint SS, Bryant Wiygul GJ, Young A (1987) A subsurface study of the
ID (eds) The geological modelling of hydrocarbon res- Lower Tuscaloosa Formation at Olive field, Pike and
ervoirs and outcrop analogues. Int Assoc Sedimentol Amite Counties, Mississippi. Trans Gulf Coast Assoc
Spec Pub115: 143-156 Geol Soc 37: 295-302
Williams PF, Rust BR (1969) The sedimentology of a Wizevich MC (1992a) Photomosaics of outcrops: useful
braided river. J Sediment Petrol 39: 649-679 photographic techniques. In: Miall AD, Tyler N (eds)
Willis BJ (1989) Palaeochannel reconstruction from point The three-dimensional facies architecture of terrig-
bar deposits: a three-dimensional perspective. Sedi- enous clastic sediments, and its implications for hydro-
mentology 36: 757-766 carbon discovery and recovery. Soc Econ Paleontol
Willis BJ (1993a) Interpretation of bedding geometry Mineral Conc Sedimentol Paleontol3: 22-24
within ancient point-bar deposits. In: Marzo M, Wizevich MC (1992b) Sedimentology of Pennsylvanian
Puigdefabregas C (eds) Alluvial sedimentation. Int quartzose sandstones of the Lee Formation, central
Assoc Sedimentol Spec Pub117: 101-114 Appalachian Basin: fluvial interpretation based on lat-
Willis BJ (1993b) Ancient river systems in the Himalayan eral profile analysis. Sediment Geol 78: 1-47
foredeep, Chinji Village area, northern Pakistan. Sedi- Wizevich MC (1993) Depositional controls in a bedload-
ment Geo188: 1-76 dominated fluvial system: internal architecture of the
Willis BJ (1993c) Evolution of Miocene fluvial systems in Lee Formation, Kentucky. Sediment Geo185: 537-556
the Himalayan foredeep through a two-kilometer-thick Woidneck K, Behrman P, Soule C, Wu J (1987) Reservoir
succession in northern Pakistan. Sediment Geol 88: description of the Endicott field, North Slope, Alaska.
77-121 In: Tailleur I, Weimer P (eds) Alaska North Slope geo-
References 565
logy. Society of Economic Paleontologists and Miner- Wright VP (1992) Paleopedology, stratigraphic relation-
alogists Pacific Section, Los Angeles, California, and ships and empirical models. In: Martini IP, Chesworth
Alaska Geological Society, Anchorage, Alaska, pp W (eds) Weathering, soils and paleosols: developments
43-59 in earth surface processes, no 2. Elsevier, Amsterdam,
Wolman MG, Leopold LB (1957) River flood plains: some pp 475-499
observations on their formation. US Geological Survey Wright VP, Marriott SB (1993) The sequence stratigraphy
professional paper 282-C of fluvial depositional systems: the role of floodplain
Woncik J (1972) Recluse field, Campbell County, Wyo- sediment storage. Sediment Geo186: 203-210
ming. In: King RA (ed) Stratigraphic oil and gas fields. Yang C-S, Nio S-D (1989) An ebb-tide delta depositional
Am Assoc Petrol Geol Mem 16: 376-382 model - a comparison between the modern Eastern
Wood JM (1989) Alluvial architecture of the Upper Creta- ScheIdt tidal basin (southwest Netherlands) and the
ceous Judith River Formation, Dinosaur Provincial Lower Eocene Roda Sandstone in the southern
Park, Alberta, Canada. Bull Can Petrol Geo137: 169-181 Pyrenees (Spain). Sediment Geo164: 175-196
Wood JM, Hopkins JC (1989) Reservoir sandstone bodies Yang C-S, Nio S-D (1993) Application of high-resolution
in estuarine valley fill: Lower Cretaceous Glauconitic sequence stratigraphy to the Upper Rotliegend in the
Member, Little Bow Field, Alberta, Canada. Am Assoc Netherlands offshore. In: Weimer P, Posamentier HW
Petrol Geol Bull 73: 1361-1382 (eds) Siliciclastic sequence stratigraphy. Am Assoc
Wood JM, Hopkins JC (1992) Traps associated with paleo- Petrol Geol Mem 58: 285-316
valleys and interfluves in an unconformity bounded Yang W (1985) Daqing oil field, People's Republic of
sequence: Lower Cretaceous Glauconitic Member, China: a giant field with oil of nonmarine origin. Am
southern Alberta, Canada. Am Assoc Petrol Geol Bull Assoc Petrol Geol Bull 69: 1101-1111
76:904-926 Yinan Q, Peihua X, Jingsiu X (1987) Fluvial sandstone
Wood LJ, Ethridge.FG, Schumm SA (1993) An experimen- bodies as hydrocarbon reservoirs in lake basins. In:
tal study of the influence of subaqueous shelf angles Ethridge FG, Flores RM, Harvey MD (eds) Recent de-
on coastal plain and shelf deposits. In: Weimer P, velopments in fluvial sedimentology. Soc Econ
Posamentier HW (eds) Siliciclastic sequence stratigra- Paleontol Mineral Spec Pub139: 329-342
phy. Am Assoc Petrol Geol Mem 58: 381-391 Yoshida S, Willis A, Miall AD (in press) Tectonic control of
Woodland AW, Evans WB (1964) The geology of South nested sequence architectures in the Castlegate Sand-
Wales Coalfield, part IV: the country around stone (Upper Cretaceous), Book Cliffs, Utah. J Sedi-
Pontypridd and Maesteg, 3rd edn. Geological Survey of ment Res
Great Britain Memoir, London Zaitlin BA, Dalrymple RW, Boyd R (1994) The strati-
Wopfner H, Callen R, Harris WK (1974) The Lower Ter- graphic organization of incised-valley systems associ-
tiary Eyre Formation of the southwestern Great Arte- ated with relative sea-level changes. In: Dalrymple RW,
sian Basin. J Geol Soc Aust 21: 17-51 Boyd R, Zaitlin BA (eds) Incised-valley systems: origin
Worrall DM, Snelson S (1989) Evolution of the northern and sedimentary sequences. Soc Econ Paleontol Min-
Gulf of Mexico, with emphasis on Cenozoic growth eral Spec Pub151: 45-60
faulting and the role of salt. In: Bally AW, Palmer AR Zhao X, Zhang F, Yang H (1988) Seismic-reflection studies
(eds) The geology ofN orth America - an overview. Geol in the Huanghua Depression, Bohai Bay Basin, China.
Soc Am Geol North Am A: 97-138 In: Wagner HC, Wagner LC, Wang FFH, Wong FL (eds)
Worsley TW, Nance D, Moody JB (1984) Global tectonics Petroleum resources of China and related subjects.
and eustasy for the past 2 billion years. Mar Geol 58: Circum-Pacific Council for Energy and Mineral Re-
373-400 sources. Earth Sci Ser 10: 277-296
Worsley TW, Nance D, Moody JB (1986) Tectonic cycles Ziegler AM, Raymond AL, Gierlowski TC, Horrell MA,
and the history of the earth's biogeochemical and Rowley DB, Lottes AL (1987) Coal, climate and terres-
paleoceanographic record. Paleoceanography 1: trial productivity: the present and early Cretaceous
233-263 compared. In: Scott AC (ed) Coal and coal-bearing
Wright MD (1959) The formation of cross-bedding by a strata: recent advances. Geol Soc Lond Spec Publ 32:
meandering or braided stream. J Sediment Petrol 29: 25-49
610-615 Ziegler PA (1988) Evolution of the Arctic-North Atlantic
Wright VP (ed) (1986) Paleosols: their recognition and and the western Tethys. Am Assoc Petrol Geol Mem 43
interpretation. Blackwell Scientific, Oxford
Wright VP (1990) Estimating rates of calcrete formation
and sediment accretion in ancient alluvial deposits.
Geol Mag 127: 273-276
Author Index
Aasheim, SM 35,40,42,311 Bachman,SB 386 Bird, DJ 454,459,460,463,466,467,

Abbot, HL 10,11 Badgley, C 73 472,474
Abdullatif, OM 249 Bailey, EB 7-9, 13 Blackbourn, GA 479
Abrahams, MJ 449 Baker, VR 33,36,47,96,98,178,192, Blackwelder, E 18,31
Adams, ER 485 197,431,433 Blair, TC 59,99, 145, 199,206,208,
Adler, FJ 486,506 Balducchi, A 484, 492, 497 245,246,249,252,271,311,313,
Ager, DV 10, 57, 476 Ballance,PF 178,346,370,415 343,345,346,375,424,427-432,
Agricola, G 6, 7 Bally,AW 376,404 450
Ahmad, R 424 Baltzer, F 346,348,399,400,404,419 Blakey, RC 45, 136,241,387,442
Aitken, JD 68 Banks, NL 152,436 Blanche, JB 487
Aitken, JF 412,454,463,466 Bardaji, T 361 Blanford, WT 12,21
Albright, W A 497 Barrell, J 23, 24, 343 Blissenbach, E 31,99,428
Alexander,J 160,199,293,346-349, Barron, EJ 421 Blixt, JE 501
379 Barry, RG 422 Bloch, S 492
Algeo, TJ 445 Bartow, JA 386, 390 Blodgett, RH 50, 148, 149,200,234,
Allen, GP 359,453,461-463,482 Bass, NW 21 437
Allen, JR 372,376 Basu, A 432 Bloomer, RR 490,496
Allen, JRL 3,5, 10, 19,20-22, Bates, RL 184,245 Bluck,BJ 30,50,99,103,106,141,
25-28,30,33-39,42-46,49,52, Beach, A 284, 487 159,161,199,214,217,317,409,
58,62,65,69,70,81,83,85,87, Beard, JH 286 411
99, 102-104, 115, 127, 135, 146, Beaty, CB 424, 428 Blum, MD 353, 448, 458, 464, 465
148, 151-153, 155, 159-161, 172, Beaumont,C 398,404 Bond, GC 371
199,200,219,235,252,271-273, Beck, RA 365, 398 Boothroyd, JC 3,31,99, 199,208,
308,309,322,335,337,338,436, Beer,IA 71,74 210,424,430,431
453,454,471 Beerbower, JR 30,311 Boss, RF 489, 496
Allen, MB 409 Begin, ZB 353 Botvinkina, LN 25
Allen, PA 46,372,376,397-399,403, Behrensmeyer, AK 45,47,73, 127, Bourgeois, J 40,62,63,69,70,201
444 176,177,179,252-254,256,293, Bowen, DW 487,491,496
Almond, CS 492 296,320,321 Bowen, JM 497
Alonso Zarza, MA 443 Belt,ES 160,199,352 Bown, TM 45,73,127,169,172-175,
Ambrose, WA 61 Ben-Avraham,Z 392 184-188,254,440
Anad6n,P 346,347,403 Bendix, I 435 Bracken,B 112,200,245
Anderson, B 318 Benkhelil, I 395 Brakenridge, GR 58
Anderson, EJ 68 Bentham, PA 74,171,202,205,322, Brayshaw, AC 167
Annan, AP 292 409,485 Brenner, RL 65
Anstey, NA 285, 286, 289 Berg, RR 39, 484, 485, 492, 493, 496 Brice, JC 195
Arche, A 159,199,214,219,253,308, Berger, A 444 Bridge, IS 3, 36, 43, 44, 54, 73,
465 Bernard, HA 5,21,28,29,33,49,65, 74,78,81,85,87,89,91,143,149,
Aristotle 6 66,72,159,199,215,217 153,155,156,160,161,172,192,
Armentrout, JM 389 Berne,S 64,70,71,74 194,195,199,200,233,240,241,
Arnott, RWC 465,471,475 BerrY,IK 241 252,296,335,337,339-342,347,
Ashley, GM 20,31,38,60,78,99, Bersier, A 26,27,55,458 436
109-111,113,114,146,198,199, Besly, BM 387,432,439,442, 443 Brierley, GJ 92,95,169,172,196,
208,210,235,431,432,436 Best, JL 3,81,131,166,167,198,322 197,199,202,211,216,403,424,
Ashmore, PE 99,141-143,167,192, Beston, NB 487 430,432
195,214,317 Beutner, EC 159, 160, 199 Brinkman, R 21
Ashton,M 2,3,492 Bhattacharyya, DP 148,264,308 Bristow, CS 3,51,52,54,72,81,
Atkinson, CD 190,242,347,484, Biddle, KT 346, 389, 390 131-133,151,152,154,156,158,
492,496,507,509 Billi, P 199, 212 160,167,198,200,232,238,322,
Aubry, WM 454, 467 Bilodeau, WL 59,252,271,311,343, 333,435,436
Autin, WJ .252,307,308,458 345,346,375 Broadhurst, FM 438
568 Author Index
Bromley, MH 200, 244, 428, 434, Church, M 18, 135, 196, 197, 199, Damanti, JF 376,409
442,465 210,211,249,311,424,431 Dana, JD 11, 12, 14
Brookfield, ME 42, 43, 62, 70, 7l, Clark, ABS 493 Darby, DA 424,431,432
370,400,402 Clark, SL 289, 290 Darwin, GH 10, 11, 19, 20
Brown, AR 74,88,291,292,492 Clement, W A 289 Davies, DK 130, 145, 160, 199,276,
Brown, D 492 Clemmensen, LB 442 333,479,484,485
Brown, LF Jr 284-286 Cleveland, MN 486 Davis, JL 292
Brown, S 384,512 Clifford, HJ 484,487,492,497,509, Davis, WM 12-15,59
Brush, LM Jr 22 511 Dawson, MR 199,365
Bryant, ID 2,3, 199,429,438,492 Clifton, BB 497 de Andrade, G 497
Bucher, WH 20,21 Cloetingh, S 68, 370-372, 412 De Boer, PL 424,444-447
Buck, SG 146, 148 Cluzel, D 402 De Saussure 7
Bull, WB 18,31, 146, 199,207,311, Cole, MR 389 de Smet, MEM 463
344,352,358,424,428,449 Cole, RD 481 DeCelles, PG 75,81-84,87,88,95,
Buller, AT 342,492 Colella, A 361 249,252,258,259,307,311-314,
Burbank, DW 293, 346, 350, 352, Coleman, JM 22,30,51,52,54,63, 346,400,424,427,430
365,375,398,403,404,406,408, 114,150,152,166,172,173,232, DeLuca, JL 434, 443
461 333,436 Denny,CS 311-313,428
Burchfiel, BC 387 Collinson, JD 2,3,22,33,35,37, Derbyshire, E 424, 431
Biirgisser, HM 145 39,42,46, 81, 83, 99, 114, 116, Derksen, SJ 487
Burke,K 387,400,486,515 146, 149, 151, 153,201, 2Q7, 248, Desikachar, SV 485, 486
Burnett, AW 46, 349, 350 252,264,267,323,331,334,429, Desloges, JR 199,211
Burns, B 370,458 432 Desmarest 7
Busby,CJ 376,386 Conaghan,PJ 114,150 Desrochers, A 190
Busch, DA 282, 284, 484, 493, 496, Coney, PJ 401 Devine, PE 491
513 Conybeare, CEB 479,491,496,501 Dewey, JF 400,401,403
Butcher,SW 353,355,356,463 Conybeare, WD 8,21 Diaz-Molina, M 160,162
Cook, BC 485 Dickinson, WR 376, 386, 387, 389,
Cairncross, B 52 Cornish, V 19 390,396,458,519
Cameron, CP 485 Cosgrove, JL 489 Diemer, JA 43,44,85,87, 160, 199
Cameron, GIF 299 Costa, JE 438 Diessel, CFK 172,179-181
Campbell, CV 34, 35, 43, 45, 54, Costello, WR 40, 433 Dixon, EEL 24
135,136,165,322-324 Cotter, E 35,36, 121, 150, 193,200, Dobrin, MB 288, 289, 491, 496, 504,
Campbell, JE 160, 199,215, 217, 245 505
223,224 Covey, M 397 Doeglas, DJ 30
Cant, DJ 31,33,35,36,39,40,46, Cowan, EJ 45,54,84,92,121,163, Dolson, J 484
50,51,120,121,148,151,153,167, 165,167,200,238,245,252,322, Donaldson, JR 180, 183, 184
195,200,235,238,398,404,434, 323,435,437,484,487 Dorr, JA Jr 246,315-317,351
436,437,479 Cowan,G 241,365,484,487 Dorsey, RJ 370, 386, 458
Carey, WC 22 Coward, MP 386,387,401,486 Dott, RH Jr 40,62, 63, 69, 70, 201,
Carling, PA 99,103,105 Crane, RC 333-337 412
Carlston, CW 331 Crews, SG 424, 430 Doyle, JD 85, 282
Carson, MA 46, 135, 141, 158, 161, Croke,JC 48,54,55,169,200,232, Dranfield, P 485,487,497
192, 194 422 Drew, F 11,14,246
Cartwright, JA 286, 473 Crook, KAW 21,346,403 Drexler, JM 496
Casshyap, SM 432 Cross, TA 3,387,389 Dreyer, T 202, 322, 334, 335, 435,
Castle, JW 484, 520 Crostella, A 487 437,441
Cavaroc, VV Jr 35 Crowell, JC 263, 346, 393, 394, 463, Du Boys, PFD 18
Cavazza, W 159 472 Dubiel, RF 253,421,438,442,451
Cerveny, PF 399 Crowley, KD 50,103,135,148,149, Dueck, RN 302
Chadwick, OH 449 153,154,158,162,197,199,200, Dueholm, KS 76
Chamberlin, TC 14,15,317 215,220,233-235,239,242 Dukas, BA 252, 408
Chan, MA 130,435, 441 Crumeyrolle,P 253,399,404 Dumont, JF 63,387,409
Chancellor,R 485,491,493,496 Cuevas Gozalo, MC 334 Dunbar, CO 27,454
Chapin, MA 484, 493 Cullingford, RA 58 Dunbar, RW 467
Charles, HH 493, 496 Curray, JR 21 Dunne, LA 400
Chatfield, J 484,496,519-521 Curry, WH 491,496 Dury, GH 198
Chawner, WD 18,428 Curry, WH III 491,496
Chesworth, W 183 Eaves,E 485,493,496
Chit ale, SV 31,246,315,316 da Vinci, L 6 Ebanks, WJJr 485,493
Chorley, RJ 422 Dahlstrom, CDA 402,408 Eberth, DA 53,78,92,93, 127, 135,
Chow, N 68 Dalrymple, RW 62, 63, 69, 70, 72, 149,172,174,200,229,232,236,
Christie-Blick, N 346 466 253,401
Author Index 569
Eckelmann, WR 496, 509 Franczyk,KJ 473 Gostin, VA 105

Ehrlich,R 36 Franklin, EH 497 Gould, HR 65, 72
Eicher, DL 36 Fraser, AJ 387,391 Grabau,AW 17,20,22-24
Einsele, G 386 Fraser, GS 311,312 Gradzinski, R 42, 64
Einstein, HA 18 Frazier, DE 65,67,68,72,318,457 Graham, JR 121,200,245
Eisbacher, GH 400,403,404,414 Friberg, JF 481 Graham, SA 403
Ekes, C 241,439 Friedkin, JF 15 Green, RT 278, 485
Elliott, L 493 Friedman, GM 18 Griffith, WM 15
Embry, AF 358, 365, 371, 412, 419, Friend, PF 30,31,35,36,39,45,48, Groth, JL 485
506 49,52,54,73,74,94,133,192,195, Grotzinger, JP 444
Emery, KO 392 229,246,248,254,290,331,337, Gubitosa, R 45, 136, 442
England, P 444 341,364,401,409,484 Guccione, MJ 254
Epry, C 20 Frostick, LE 380,424,429,437 Guettard, JE 7
Eriksson, KA 52,424,431,434, Fulford, MM 386 Gulisano, CA 454
443 Fulthorpe, CS 286 Gurnis, M 412
Ethridge, FG 2,17,37,54-56,148, Gustavson, TC 22,49, 140, 141,
149,160,192,199,280-282,329, Gabela, VH 486 159, 199
424,430 Gagliano, SM 65
Eugster, HP 481 Galay, VJ 15,32 Hack, JT 36
Evans,JE 424,431,432,463 Galloway, WE 3, 15,48, 50, 95, 200, Hacquebard, PA 180,183,184
Evans, WB 273 252,254-256,261,274,276-278, Hahmann, P 20
Everest, RIO 280,323,358,362,367-369,381, Hallam, A 450, 477
Eyles, CH 130 382,407,413,427,459,461,467, Halland, EK 275,492,512,513
Eyles, N 106, 130, 444 474,484,485 Hamblin, AP 401
Gansser, A 398 Hambrey, MJ 421
Fahnestock, RK 5,22,148 Garcia del Cura, MA 127, 129 Hamilton, DS 274,276-278,280,
Fairhead, JD 380, 395 Gardiner, S 484,496 474
Farrell, KM 169,172,173,176-178, Garner, LE 30,40,49, 148, 160, 199, Hamilton, W 385, 387, 392, 393,
199,219 215 415
Farshori, MZ 496 Garrison, RK 485,491, 493, 496 Hamilton, WS 485
Fenneman, NM 23, 27 Gawthorpe, RL 160,199,238,293, Hamlin, KH 485
Ferguson,RI 141,143,199,211 344,345,361,378,379,386,387, Hampton, MA 18
Ferm, JC 35, 148, 182 391 Hanley, JH 52, 172, 200, 254
Fernandez, J 317,361 Geehan,GW 492,509 Hanneman, DL 275,286,454,463
Fettes, DJ 401 Geikie, A 11, 12, 14 Happ,SC 23-25,28,34
Fielding, CR 3, 18,38,48,55,89, Genik, GJ 380, 381, 395, 487 Haq, BU 369,404,406
92,172,173,333-337,442 George, GT 241 Hardie, LA 150, 190,481
Figueiredo, AMF 487 Germanoski, D 143, 153, 192,361 Hardman, M 387, 391
Filipov,AJ 99,424,427,428,440 Gersib, GA 55,130,149,172 Harland, WB 421
Finch, J 8 Ghignone, JI 497 Harms, JC 5,22,77,111,112,146,
Finlayson, B 322 Gibling, MR 39, 135, 146, 149, 150, 148,466,467,490,491,496,502,
Fischer, AG 68,444-446 152,161,199,200,209,240,296, 503
Fisher,MJ 362,363,369,383,512 401,412,415,434,437,447,454, Harris, AL 401
Fisher, WL 284-286,381,417,419 459,460,463,466,467,469,472, Harris, PT 70, 74
Fisk, HN 5,15,21-26,28,179,219, 474 Hartley, AJ 424, 429, 438
318,319,469,483 Gilbert, GK 10,11,14,16-18,20,26 Harvey, AM 424, 429, 430
Flach, PD 54, 159, 160, 199,223 Gilbert, R 18,210 Hastings, JO Jr 485,496
Flemings, PB 404,461,477 Glaister, MS 103 Haszeldine, RS 43,78,81,151-153,
Flemming, BW 114 Glennie, KW 241,437,438,441 200,238,436
Flint, S 424, 429, 438, 442 Gloppen, TG 252,258,314,393 Hayden, HH 8
Flint, SS 2, 3, 392, 454, 463 Godin, P 39, 136, 146,323,325-327, Hayes, BJR 495
Flores, RM 2,3,52,55,170,172, 435,437 Hayes, JB 486
200,254,320,321 Gohain, K 315 Hayes, MO 65, 161
Focke, JW 489 Goldthorpe, WH 487 Heath, R 489
Folk, RL 18, 19 Gole, CV 31,246,315,316 Heckel, PH 68,74,252,264,476
Forbes, DL 49, 103, 141, 159, 199, Golin, V 276 Hein, FJ 22,99, 101, 102, 106, 140,
214 Golovchenko, X 356, 357 141, 143, 199
Forgotson, JM 496 Gomez, B 139 Heinberg, C 40
Fortuin, AR 463 Gonzalez, E 386 Heller, PL 3,343,352,371,372,
Fouch, TD 389,406,467,478,519, Goodwin, PV 68 374-376,403,412
521 Gordon, EA 43 Hempton, MR 400
Fournier, M 63, 387 Gordon, I 375, 376 Hendry, HE 160,199,215,217,223,
Frakes, LA 421 Gornitz, V 349, 398, 404 224
570 Author Index
Henry, VJ Jr 65,66,71 Jackson,JA 184,245,345,379,386 Kleinspehn, KL 409,411,412

Hermansen, M 200 Jackson, JR 14 Klicman, DP 485
Herodotus 6 Jackson, RG II 3,22,30,33,37,38, Klimetz, MP 387,515
Herries, RD 241 42,43,46,49,50,63,69,70,72, Klitgord, KD 371
Hersch, JB 485, 493 131,148,155,159-161,164,199, Klovan,JE 19,419
Heward, AP 3,31,40,42,249,252, 212,215,217,221 Kluth, CF 401
258,259,311,313-315,343-345, James, DP 354 Knight, SH 21
423,424,431,457 James, NP 68,77,308,317 Knoll, MD 293
Hickin, EJ 58,95, 141, 196, 197, James, WC 190,422,441 Kochel, RC 424, 430, 432
199,202,211,216 Jamieson, TF 10 Kocurek, G 42,59,62,63,68-71,
Hider, A 20 Jarvis, J 160, 161, 199 76,438
Higham, N 10 Jeffreys, H 17 Kolb, CR 65, 68, 71
Hirst, JPp 248,252,254,256,257, Jervey,MT 354,456,461 Kominz, MA 371
296,315,342,435 Jerzykiewicz, T 35, 123, 125, 127, Kooi, H 371
Hjulstrom, F 18 209,252,408,429,451 Koster, EH 2, 3, 22, 105, 208
Hobbs, WH 19,22,23 Jin, Y 278,485,493,497,515-517 Kranzler, I 496
Hobday, DK 3, 160, 199,416,419, Jirik, LA 485 Kraus,MJ 45,46,73,105,127,165,
427,443 Jo, HR 375 172-175,184-188,252,254,257,
Hoey, TB 139 Johansson, CE 22 262,440,465
Hoffman,PF 444,476 John, BH 492 Krumbein, WC 18, 19
Hogg, MD 485,491 Johnson, AM 18,99 Kuenen, PH 364
H0imyr, (2) 513 Johnson, D 13 Kuenzi, WD 178,181,370,386
Holbrook, JM 454, 467 Johnson, GD 127,346,389,390,398, Kumar, N 71
Holdsworth, BK 252, 264, 267 409 Kumar, S 71, 150
Holmes, A 246,315, "118, 322, 364, Johnson, NM 74,252,253,293,297, Kutzbach, JE 444
398,404 298,389,390,409 Kvale, EP 466
Holmes, DA 318,351 Johnson, RA 424,430,432
Homer 6 Johnsson, MJ 432 Lajoie, J 130
Homewood, P 399, 409 Johnstone, EM 484,493 Lane,EW 16,192,353,354
Hooke, JM 141, 159, 199 Jolley,EJ 409,410,424,429 Lang,SC 92,376
Hooke, R 18, 145, 199,206,207, Jones, BG 114,139 Langbein, WB 16,423
344,424,428 Jones,CM 42,43,54,150,422,434, Langer, AW 180
Hoover, HC 6, 7 436,437 Langford, RP 112, 130, 200, 435,
Hoover, LH 6,7 Jones, JG 114, 150 441
Hopkins, JC 46, 135, 139, 291, 304, Jopling, AV 5,22,115 Larsen, PH 347
305,465-467,490,495,496,498, Jordan,DW 85,199,219 Larsen, V 99
499 Jordan, TE 343,346,375,387,389, Larter, S 305
Hornafius, JS 393 390,403,404,461,477 Lash, GG 409
Horne, JC 35,55, 148,416 Lawrence, DA 147,148,199,200,
Hossack, JR 393 Kana, TW 65 222,435
Houbolt, JJHC 66, 70 Karges, HE 485 Lawson, AC 31
Howard, RH 489-491, 496 Karlsen, DA 305 Lawton, TF 278,296,346,365,366,
Howell, DG 409 Katz, BJ 421,439 406,407,484,485,519
Hoyt, JH 65, 66, 71 Kauffman,EG 68,406,407 Le Roux, JP 296
Hsu, KJ 402 Keighin, CW 521 Le, B 145
Hubert, JF 99, 200, 245, 415, 424, Keller, B 170-171 Leblanc Smith, G 52,78,424,431
427,428,435,438,440,441 Keller, MD 22 Leckie, DA 353,358,359,457,463,
Huggenberger, P 143,166,292,293 Kellerhals, R 22, 36 471
Huggett, RJ 422 Kelling, G 387 Leddy,JO 143,192,317
Humphreys, AA 10, 11 Kelly,SB 240,246,249,428,435 Lee, RA 487
Hunt, AR 19 Kennedy, JF 22 Leeder,MJ 387,391
Hunt, D 459,475 Kennedy, RG 16 Leeder,MR 17,21,30,34-39,57,
Hunter, GM 492,496 Kerr, DR 485 61,68,73,74, 110, 127, 158, 160,
Hunter, RE 62, 63, 70, 438 Khan, HR 198 199,202,252,267,318,329-331,
Hutchison, CS 385, 386, 392, 393 Kindle, EM 20,21,23 335-337,339-342,344-349,378,
Hutton, J 7,8 King, PB 381 379,386,387,453
King, WSH 20 Lees, GM 348
Ikeda, H 63 Kirk, M 151-153,200 Legarreta, L 454, 466
Imbrie, J 445 Kirk, RH 46,492,497,510,511 Legun,AS 52,200,438
Ingersoll, RV 376,390,393 Kirschbaum, MA 54, 200, 229, Leliavsky, S 18
Inman, DL 18 237 Leopold, LB 15, 16,36, 141, 142,
Iseya, F 63 Kirschner, CE 386,390,391,486, 191,192,296,330,331,353,
Ito, M 397 496 356-358
Iwaniw, E 424,431,432 Klein, G deY 252,376 Lerbekmo, JF 408
Author Index 571
Levey, RA 148, 160, 161, 199,215, 253,351,352,437,439,440,453, Minter, WEL 199

217,221 454,456-460,464,466,471,472, Mitchell, AHG 386
Levorsen, AI 485 474,475 Mitchum, RM Jr 458
Lewin, J 2,3,37,99, 141, 159, 161, McCarthy, TS 208,245,248,315, Mitrovica, JX 407
162,199,317 427,437 Mj0s, R 334
Lewis, CP 218, 356 McCaslin, J 502 Molenaar, CM 253, 482
Lewis, GW 317 McClay, KR 401 Molina, J 486
Li, S 322 McDonald, BC 218,356 Mollard, JD 15, 32
Lichty, RW 193,433 McDonald, RE 485, 496 Molnar, P 370,444
Link, MH 160, 199,346,393 McDonnell, KL 325 Moody-Stuart, M 28, 30, 32, 36,
Linsley, PN 497 McDougall, JW 350,351 296,419
Liu, H 402,515,516 McGee, WJ 11 Moore, PD 180
Lockett, FPJ 38,111,115,437 McGowen, JH 30,40,49,148,160, Moore,PS 467,484-486,490,492
Locking, T 53 199,215,362,363,369,381,419 Moore, RC 253,472
Long, DGF 35,150,194,400,450 McIntire, WG 351 Moorman, BJ 292, 293
Lopez-Gomez, J 253, 308, 465 McKee, BA 72, 73 Morgan, JP 351
Lorenz,JC 148,303,304,331,333 McKee, ED 21,22,24,30,33,42,43, Morgan, KH 412
Loutit, TS 474 54,63,69,72,81,163,200,243, Morgridge, DL 487,492,496,506,
Lucas, PT 496 435,437 509
Lucchitta, I 54 McKenzie, DP 477 Morison, SR 199
Lundberg, N 386 McKerrow, WS 386 Morley, CK 349
Lustig, LK 18, 428 McLaughlin, RJ 394, 395 Morton, RA 483
Luterbacher, HP 404 McLean,JR 35,123,125,127,252, Mosley, MP 167
Luttrell, PR 200,387,392,434,437, 408,419,501 Mossop, GD 54,159,160,199,224,
442,443 McLean-Hodgson, J 487 234,365
Luyendyk,BP 393 McMillan, NJ 367 Mpodozis, C 486
Lyell, C 8,9,10,13,19,317 McPherson, JG 99,145,199, Mukerji, AB 31,246,248
Lyon,CA 386,390,391,486 206-208,245,246,249,311,313, Mulder, TJ 293,351,352
Lyons,PL 288,491,496,504,505 415,424,427-432,450 Mufi6z, A 160, 199,219,226,240,
Meadows, NS 284, 487 244,428,429,435
MacDonald, AC 275,492,512,513 Mebberson, AJ 491 Munro, M 139
Mack, GH 189, 190,422,441 Medlicott, HB 12,22 Murty, KN 485, 486
MacKay, D 318 Mello, MR 481 Muszynski, A 145,146,199,200,
Mackey, SD 252,337,341 Melton, FA 15,23,25,27 207
Mackin, JH 16, 25-28 Melvin, J 275, 278, 484, 485, 492, Mutch, TA 419
Maddock, T Jr 16,36 496 Muto, T 361, 362
Maizels, J 99, 100,424,429,431 Mertz, KA Jr 200,435,438,441 Mutti, E 61,62,68
Major, CJ 28,49, 159, 199,215,217 Miall, AD 2,3,5, 13, 15,22-24, Muwais, W 298,300,301
Makino, Y 62, 63 30-33,35-42,44,46,49-51,53,54,
Manspeizer, W 443 56,58,61,62,64,68-71,73-75,77, Nadler, CT 192,433
Marple, RT 349,351 78,82,84,85,88,89,91-93,95,97, Nadon,G 53,54,229,232
Marriott, SB 256, 435, 454, 459, 99,106,108,111-113,120,121, Nami, M 30,35,160,199
460,475 123,125,127,130-135,143,146, Nanson, G 171,199,211,325,327,
Martin, JH 278, 303, 305, 306, 492, 148-152, 154, 158, 159, 161-164, 328
497,509,513 171, 172, 174, 192, 193, 195, 198- Nanson, GC 48,50-52,54,55, 141,
Martin, R 491,496 201,205,207-210,212,213,215, 155, 159, 161, 163, 169, 199-201,
Martini, IP 183, 429, 432, 438 222,229,231,232,234-236,240, 224,232,233,240,241,422,438,
Martinius, AW 334 243-245,248,249,251-253,259, 439
Martinsen, OJ 309, 454 264,265,272,276,282,283,293, Nanz, RH Jr 5,28,39,273
Marzo, M 3 294,296-300,306-308,323,325, Nascimento, as 497
Massari, F 103, 105, 141, 199,346 343-346,353,355-357,364,371, Naylor, D 395
Mastalerz, K 346 376,378,387,393,395,401,403, Neidell, NS 286
Masuda, F 397 404,412-418,428,429,431,432, Nemec, W 145,146,199,207,246,
Mather, AE 351 434,435,437,438,441,443,447, 252,273,361,415,424,431
Mathisen, ME 387 453,454,457,458,462,467,471, Nichols, GJ 248, 296
Matsuzawa, A 486 476,477,484,492 Nijrnan, W 30,33,40,148,160,199,
Matter, A 46 Middleton, GV 5, 9, 13, 19,22, 38, 217,224
Matthews, MD 422 99,372,465 Nilsen, TH 2,3,31,249, 3ll, 386,
Maxwell, JR 481 Middleton, LT 18,105,241,387 394,395
May, SR 393 Mike, K 349 Nio, S-D 70,72,447
Mayer, DF 484,493 Miletus, T 6 Noble, RL 160
McCabe,PJ 3,4,42,43,53-55,127, Miller, DD 485,487,496 Nobles, LH 18,428
135,149,150,162,165,172,179, Miller, RS 485 Nordin, CF 22
180,182,184,200,229,233,237, Milne, G 184 Norrnark, WR 61,62,68
572 Author Index
Nummedal, D 3,4,31,68, 199,208, 461-463,471,475,482 Rodine, JD 18

357,424,430,431,456 Po saner, EM 487 Rodolfo, KS 386
Nur, A 292 Postma, G 246, 424, 431 R0e, S-L 200
NybnHen, G 513 Potter, PE 5,21,22,26,197,362, Rogers, RR 465
Nystuen, JP 252, 253, 265, 268 364,367,381,414 Rood, K 135
Powell, JW 13, 14 Rosen, MR 379
Okolo, SA 46, 135, 139, 162,200, Pratt, BR 293 Rosenthal, L 481
233 Price, DM 353,448,449 Ross, WC 453
Oliver, TA 53,303 Price, W A 483 Rouse, H 18
Olsen, H 68, 166, 200, 233, 261, Prior, DB 361 Royden,L 387
445-447 Prosser, S 189,344,345 Royden, LH 393
Olsen, HO 240, 246, 249, 428, 435 Pryor, WA 21,199,219 Rubey, WW 18,21
Olsen, PE 252, 380,415 Puigdefabregas, C 3, 30, 33, 40, 49, Rubin, DM 63, 70
Olsen, T 76,347,454,462,465,471, 148,159,160,199,217,224,346, Ruddiman, WF 444
475 404,405 Russell, RJ 6,318
Olson, JS 21 Purser, BH 346,348,399,400,419 Rust, BR 3, 15, 22, 30-32, 43, 52, 87,
Ono, Y 424, 430 Putnam,PE 52-54,135,279,280, 99,105,106,113,114,120,121,
Oomkens, E 28,67,72,73 303,484 123, 125, 132, 133, 135, 139, 140,
Ord6nez,S 127,129 143, 146, 152, 161, 172, 192, 195,
Ore, HT 30 Qian, N 358 196,198-200,208,210,213,232,
Ori, GG 49, 159, 199,214,346,399, 233,240,241,401,424,431,438,
400,409,419,424 Rachocki, AH 3,249,311 439
Orton, GJ 387 Rahmani, RA 55, 172, 408 Ryder, JM 424,431
Osborne, RH 393 Rahmanian, VD 275,385,487,491, Ryer, TA 180,472
Ouchi, S 349 497 Ryseth, A 253,267-270,275,
Owen, LA 424, 431 Ramos, A 3, 46, 49, 52, 53, 58, 105, 383-385,465
Owens, JS 20 144,159,199,200,214,243,427
Ramos, V 486 Sadler, PM 57,71,445
Paauwe, EFW 302 Ramsbottom, WHC 68,74,252, Sagoe, K-MO 19
Padgett, V 36 264-266,476 Sahu, BK 18
Page, K 50, 141, 161, 163 Rasanen,M 348,349,409 Salazar, RB 387,486
Paola, C 3,372-375 Rascoe, BR Jr 486, 506 Saleeby, JS 409,411
Parkash, B 54, 150,200,240,246, Ray, RR 286,287,475 Salisbury, RD 14,15,317
249,315,398 Raynolds, RGH 293, 346, 398, 404, Salter, T 167
Parrish, JT 421,435,439,440 408,461 Sanders, JE 72
Parsley, AJ 487,492,510,512 Reade, TM 10 Sanford, BV 412
Passega, R 18 Reading, HG 77,201,346 Sanford, RM 497
Payton, CE 55 Reiche, P 21 Saucier, AE 278
Pazzaglia, FJ 461 Reid, I 424,429,437 Saucier, RT 318
Peale, AC 14 Reineck, HE 471 Saunderson, HC 38,111,115,437
Pelletier, BR 21 Reinfelds, I 171, 199,211,325,327, Schermer, E 409, 411
Penck 12 328 Schouten, H 372
Peper, T 371, 477 Reinhardt, J 183 Schultz, AW 45,99,106,146,199,
Perlmutter, MA 422 Retallack, GJ 68, 127, 183, 185-187, 207,424
Peterson, F 46,352,421,439 421,439-441 Schumm, SA 3-5,8,15,17,32,33,
Petterson, 0 485 Reynolds, DW 18,491 35-37,46,48,56,58-60,84,136,
Pettijohn, FJ 5, 10, 19,21,22,421 Rhee, CW 173 143,150,153,167,191-198,252,
pfiffner, OA 409 Riba,O 346 253,311-313,315,316-319,329,
Phillips, W 8 Ricci Lucchi, F 399, 402, 409 348-350,352-354,358-361,423,
Pienkowski, G 467,469 Rice, DD 69, 253, 482 429,433,438,450,454,456,484
Pierson, TC 424, 430 Rice, JW 289 Schwartz, DE 22, 159, 199
Pirson, SJ 30, 39 Richards, K 174,317,322 Seeber,L 349,398,404
Pitman, JK 473,496,518,521 Richardson, EV 5, 22 Seely, DR 386
Pitman, WC III 355-357,477 Richmond, BM 62, 70 Seguret, M 393
Plato 5 Rider, MH 273,274 Sengor, AMC 400, 403
Platt, NH 170-172, 442 Ridgway, KD 400, 424 Sengor,C 387,400,486,515
Playfair, J 8, 9, 343 Rieu, EV 6 Seranne, M 393
Plint, AG 65,95, 148, 160, 199,219, Rivenaes, JC 358, 359 Serra,O 298
453,461,465,475,477 Roberts, DG 487 Sevon, WD 371, 372
Poag,CW 371,372 Roberts, HH 63, 65 Seyfried, H 370
Pommier, G 484,492,497 Robertson, JD 305 Shanley, KW 3,4, 130, 253, 352,
Posamentier, HW 55, 95, 353-355, Robinson, JE 280 453,454,456-460,464,466,470,
358-360,404,453,454,456, Rodgers, J 27 471,473-475
Author Index 573
Shannon,P~ 395 Stabler, CL 387,389,486 Tipper, JC 288

Shantzer, EV 25 Stamp, LD 24 Tirsgaard, H 442
Sharp, RP 18,428 Stancliffe, RJ 485 Titheridge, DG 351
Shaui, D 486,491 Stanistreet,IG 192,246-248,427, Todd, RG 355,459
Shaw, J 22 433,437 Todd,SP 199,347
Shelton,JW 160,496,499-501 Stanley, KO 50, 148, 149,200,278, Tolman, CF 31
Shen, HW 18 432,437,444,497,513 Tornqvist, TE 326,448,449,469,
Shepherd, RG 199,222,428,434, Stanmore, PJ 484, 493 471
435,442 Stapp, RW 496 Trewin, NH 241,451
Shetsen, I 365 Stark, PH 496 Trifonov, VG 348
Shields, A 18 Statham, I 18 Trowbridge, AC 23, 31
Shlemon, RJ 318,319 Staub, JR 182 Trujillo, AP 105
Shultz, AW 146 Stear, W~ 54, 148, 160, 163, 199, Tucker,~E 459,475
Shurr, GW 68 222,228,321,333,428,429,435, Tunbridge, IP 54, 163, 200, 428,
Shuster, ~W 54, 200 437,438 429,435,437,438
Siegenthaler, C 143, 166 Steel, RJ 2,35,40,42,99, 199,252, Turner, BR 139,252,267,408,435
Sigleo, WR 183 253,258,268-270,275,311,314, . Turner, P 3, 439
Simlote, VN 486, 497 361,383-385,393,394,415,419, Twenhofel, WH 21,24
Simons, DB 5, 22 423,465 Tye, RS 484
Sinclair, HD 397-399,403,444 Steidtmann, JR 54, 200 Tyler, N 3, 148,280-282,487,492
Singh, H 31,246 Steno, N 6
Singh, IE 150 Stephens, ~ 240, 293, 295 Udden, JA 18
Sinha, R 192, 195,364 Stewart, DJ 50, 161, 199,224,232 Uliana, ~A 466
Sloss, LL 253,412,453,454,476 Stewart, JH 387
Smith, DG 32,33, 35, 50-54, 103, Stix, J 130 Vail, PR 55,68,74,251-253,265,
135,148-150,158,159,161,162, Stockmal, GS 398, 404 268,353-355,358,359,365,453,
172,174,192,197,199,200,222, Stollhofen, H 192 457,459,460,471,473,474,476
224,229,234,236,279,298,300, Stott, DF 412 Van Beek, JL 65
301,325,326,351,365,433,437, Struijk, AP 278, 485 van der Lingen, GJ 386
444,445 Stuart,IA 484,487 Van Geuns, LC 489,493
Smith, GA 207, 344, 376, 449, 475 Stuart, WJ 484 Van Houten, FB 364, 365, 398, 399,
Smith, ND 14,22,30,32,38,50-54, Summerson, CH 10 403,419,421,439
56,99, Ill, 112, 148, 172-175, 177, Sundborg,A 18,28,160,199 Van Lopik, JR 65, 68, 72
192,197,200,229,234,237,325, Suneson, N 54 Van Overmeeren, RA 414
327,328,351,437 Surell, A 35, 246 van Popta, J 489
Smith, ~H 333 Suter, JR 68,74,252,260,359,466, Van Straaten, L~JU 27
Smith, SA 75,81,87, 105, 141, 143, 467 Van Vliet, A 30,160,199
145,159,199,211,215,424,427, Sweet, AR 451 Van Wagoner, JC 4,69,95,252,
429,442 Sweet, ~L 85, 282, 429 264,406,407,454,456,458,459,
Smith, WB 487,492,496,506,509 Svnfi,DJP 69,407,454,455 467,477
Smoot, JP 435,438,441 Vandenberghe,J 448,449
Smyth,~276 Tadros, NZ 472,474 Vanoni, VA 18
Sneed, ED 19 Talbot, ~R 241,439 Varenius, B 6
Sneh,A 54,163,200,437 Talwani, P 349,351 Veevers, JJ 385,412
Sneider, R 306 Tandon,SK 447,463 Verdier, AC 162,386
Snelson, S 376, 381 Tang,Z 497,515 Vessell, RK 130, 145
Soegaard, K 75,81,87 Tankard,AJ 199,395,396 Vincent, JK 491
Soek, HF 299, 303, 304 Tanner, WE 21 Vincent, P 298
Soeparjadi, RA 515 Tapponnier, P 370,400,401,487 Viseras, C 317
Song, G 467, 490 Tapscott, CR 372 Visher, GS .18, 19,30, 38, 275, 493,
Sonnenberg, SA 291,496,501-503, Tauxe,L 45,73,252-254,293,296, 496
505 320 Visser, JNJ 252, 408
Sopena, A 3,46, 99, 105, 144, 159, Taylor, G 33, 161, 199 Visser, ~J 63, 69
199,214,427 Teng,LS 397 Von Brunn, V 419
Sorby, HC 10,11,17,19-21 Terry, DO Jr 463 von Hoff, KEA 8, 13
Southard, JB 18,22,61,71,99, 105, Terwindt, JHJ 28 von Zittel, KA 8, 486
146,372 Tewari, RC 432 Vondra, CF 346,387,398,466
Spencer, DW 18 Thakur, GC 303 Vos, RG 199
Spieker, E~ 346 Theriault, P 190
Spurr, JE 19 Thomas, B~ 493 Wadman, DH 278,491,492,509
Srivastava, P 263,349,351 Thomas, RG 50,51,130,224,471 Walker, RG 3,22,31,33,34,36,
Srivastava, SP 372 Thompson, DB 99,146 38-40,46,50,51,77, 101, 102, 105,
Staal, JH 414 Thorne, CR 232 115, 120, 121, 140, 141, 143, 148,
574 Author Index
151,153,195,200-202,235,238, Wescott, WA 4, 193,348,360,424, Wiygul, GJ 485

308,317,433,434,436,437 430,454 Wizevich, MC 76, 139, 153, 156,
Walker, TR 421,439 Weston, PJ 348 162,163,200,233,240,323,428
Wall, JH 419,501 Wheeler, DM 491 W oidneck, K 492, 496
Walling, DE 424 Whitaker, ST 489-491, 496 Wojewoda, J 346
Walther, J 13,24,38 White, DA 479 Wolman, MG 15, 16,36, 141, 142,
Walton, AW 284 Wideman, CJ 454, 463 191, 192,331
Wang, QM 486 Wightman, DM 135, 275, 279, 280 Woncik, J 491
Wanless, HR 63,476 Wightman, WG 466, 469 Wood,JM 161,199,224,229,291,
Ward, WC 18 Wilgus, CK 4 466,467,490,495,496,498,499
Warren, GK 12 Wilkinson, BH 445 Wood, LJ 357
Warris, BJ 487 Willard, DA 252 Woodland, AW 273
Warwick, PD 254 Williams, BPJ 45, 127, 135, 147, Woodyer, KD 33,161,199
Wasson, RJ 145,199,207,424,427, 148,199,200,222,252,271-273, Wopfner, H 412
431 308,309,322,435 Worrall, DM 381
Waters, KH 289 Williams, CA 395 Worsley, TW 476
Watts, KJ 486, 492 Williams, GE 22, 33, 150, 163, 200, Wright, MD 27, 28
Wayne, WJ 432,444 240,435,437 Wright, VP 68,73,183-185,190,
Webb, BW 424 Williams, H 299, 303, 304 256,421,435,440-442,454,459,
Weber, JF 485,493 Williams, PF 22,30,31,43,87,99, 460,475
Weber,KJ 291,489,492,493 132, 14~ 199,210,213 Wunderlich, R 471
Wecker, HRB 484,492,497 Williams, TA 381,382
Weimer, P 354,404,454 Willis, A 459 Yang,C-S 70,72,447
Weimer, RJ 72,389,406,407,466, Willis, BJ 155,159-162,164,176, Yang, W 497,515
469,483,484,490,502-504 177,179,200,238,256,321,341 Yinan, Q 278,280,485,493,516,
Weir, GW 21,42,43,69,81 Wilson, JL 419,472 517
Weirich, TE 496 Wilson,LG 8,9,13 Yoshida,S 264,265,271,406,407,
Weller, JM 476 Wilson, RCL 394 454,459,462,467,477
Wells, NA 246,315-317,351,432 Winder, CG 424 Young, A 485
Wells, SG 424, 430 Wing, SL 55
Went, DJ 199,347 Winker, CD 367,368,382 Zaitlin, BA 3, 130, 224, 469, 471,
Wentworth, CK 18 Winkley, BR 17 482
Werner, AG 7 Winn, RD Jr 380, 487 Zhao, X 486
Werren, EG 484, 485, 493, 496, Winsemann, J 370 Ziegler, AM 440,450
513-515 Wise, DU 352 Ziegler, PA 380,383,511,512
Werritty, A 141,199,211 Withrow, PC 491,496
Subject Index
10-year flood 62,63,70 antitectonic model 375,403 493,496

100-year flood 59,62,63,70 apex avulsion 143 base level
l4C dating 58, 73 apex incision 312 change,effectof 352-362,456-
Aracas field 487 458
abandoned channels 139, 177-178, architectural element definition of 352
180,493,514 and climate 434-435 Basin and Range 379,386,387,388
accommodation 456-459,475 classification of 89-94, 98 basin classification
actualistic facies model, history of definition of 91 paleogeographic 413
26,31 scale of 82, 282, 283 tectonic 376,377
adiabatic rainfall, see orographic arid climate 425,426,427,428-430, basin equilibrium time 374
rainfall 442 Battery Point Formation 39, 40, 50,
Adriatic foreland 397,403 aridity, facies indications of 222, 434
aerial photographs 15, 58 424-429,432-435,438,442 bayou 14, 177
aggradation index 146, 152 Arkansas River 193,433 Beartooth Conglomerate 88, 258,
Agua Grande field 487 Arkoma Basin 479,486,487,496, 314
Alberta Basin 365, 395, 403, 408, 506 Beatrice field 487
419,479,488,495,496 armoring 103 Beaufort Group 227,228,408,416,
algal encrustations 127 Arno River 6 435
Algoa Basin 395 arroyo 435 bedload 15,17,18,50
allocyclic, see allogenic Arve River 9 bed roughness 17,18
allogenic process, definition of term ash (in coal) 179 bed-load stream 50, 196
30,31 Atchafalaya River/delta 63, 65, 72, bedform classification 38, 60, 78
allostratigraphy, use of 95, 306- 318,319 Bellshill Lake field 491,496
309,460 Athabasca Oil Sands 54, 224 Belly River Formation 496
alluvial fan (see also terminal fan) Athabasca River 208, 209 bench (point-bar) 51, 163,222
classification 207,245-249 aulacogen 382-383 Beni Basin 387
climatic setting 424,428-431 auto cyclic, see autogenic Benin Formation 395
evolution model 84, 252, 258- autogenic process, definition of term Benue Trough 383, 394, 418
259,311-317,344-346,361-362 30, 31, 3ll Bermejo Valley 389, 390
history of research 11, 23, 24 Auvergne 8 Berwick field 493, 496
sequence 252,259 avulsion 252,253,312,317-327, big rivers 54, 150,362,364,414
tectonic setting 344-346, 378 341,471 Big Wall field 486
alluvial ridge 321 axial (longitudinal) transport 364- Bijou Creek model 33, 54, 163,200,
alluvium, deformation of 399 365,375,379,398,399,403,404, 243
Alpine fault 393 413 Bijou Creek, Colorado 24, 33, 72
Altamont-Bluebell field 486 billabong 177
alternate bar 162, 165,233,238-240 Babbage River 218, 356, 357 bioturbation, floodplain 177
Amazon River 348,362,364,367, backarc basins 386 Blackhawk Formation 407,462
417 and petroleum 479,486,515 Blackwell field 490, 496
Amite River 40,215,307,308 backfilling, fan 84,311-315,362 blind structures 349
Anadarko Basin 487,506 Badlands National Park 440 boghead coal 181
analysis of variance 21 Bahia Supergroup 482, 497 Bohai Bay Basin 482,486,491
anastomosed rivers bajada 31 Book Cliffs 265,407,482
avulsion in 325-327 bank (channel) stability 135, 192 Boothia Uplift 412,415
causes of 197,325,469,471 Bantry field 486 boreal climate 425,426,431,432
fluvial style 229, 232, 234, 237 bar, definition of llO Boulder Creek Formation 463
anastomosing: history of term 14, bar-top assemblage 149 boundary layer 17
15,32-33,52 Barail Sands 485 bounding surfaces
Ancestral Rockies 392,401 Barrancas Formation 497 and allostratigraphy 305-306
antecedent river 398 Barrow Arch 506, 507 annotation of 96,98, 151
antidune 19,20,22,36, ll2, 245 Bartlesville Sandstone 275, 485, classification of 69-71,81-89
576 Subject Index
bounding surfaces (Contd.) Central African Rift System 394 compound bar 46, 235
drawing of 76-77 Central Lowlands 417 concave bench complex 161
shape of 131, 151 Central Tertiary Basin 394 condensed section 474
box canyon 139, 435 channel Congaree River 215,217,221
Brahmaputra River 30,51,54,63, architectural element 131-139, Congo River 364
72,114,436 256 connectedness, sandbody, see
bar-top model 156, 160 classification of 15, 94-95, 195- interconnectedness
crevasse channels 173 198 constriction avulsion 143
flute scours 166 classification, history of 32, 45, constructivist facies models 202
levees 172 49 contact matrix 39
scale 151,238,333 clustering 461 convergent-margin basins 386-392
tectonic setting 364,404,414 hierarchy 43, 133, 135 Cook Formation 496
braid delta 246,415,430,450,491, in seismic data 289-292 Cook Inlet 386,390,391,417,486,
509 scales 334 487,496
braided: history of term 14, 15 channel-fIll cycles 136, 148,323, Cooper Basin 484, 489, 493
braiding 326 Cormorant Formation 512
causes of 192, 194,361 Cherokee Group 416, 479, 480, 485, cornstone 183
index 15, 195 486,493,506 coset 81,436
parameter 195, 198 Cheyenne Valley field 491, 496 countercurrent ripple 116
Brazeau-Paskapoo formations 408 Chinle Formation 442,443,451, counterpoint bar 141
Brazos River 29,215 465 Coyote Creek field 492, 493, 496
break-out 370 choking avulsion 143 cratonic basins 412
Breathitt Group 463 chronosome 45 crevasse
Brent field 485,497,510 chute channel 132, 149,219,226,229,
Brent Group 384,480,497,512 bar, gravel 105,217 319,437
bright-spot 305 channel 132,141,217 splay 148,149,229,237,437
Bronson-Xenia field 284 cutoff 317,360 splay, evolution of 174-175,177
Brownstones 43, 44 Cimarron River 193,433 cross-channel bar 153, 195
Buck Tongue 471 Citronelle field 485, 493, 496 Cumberland Marshes 171,173,
Budleigh Salterton Pebble Beds clast-supported gravel 106 237,327,328
143,145,215,430,442 clastic dyke 183 cumulate paleosol 256
bundles, cycle 445, 446 clastic wedge 479,480-482 Cushing field 486
Buntsandstein 53, 144,226,244 clay ball 432 Cut Bank field 490,496,499-501
burst-sweep cycle 21,37-38,61,62 clayeyness 274 cutan 183
Bush City field 486 Clearwater Sea 481 cutbank slope 133
Clifford-Siaana field/trend 486, 487 Cutler Formation/Group 93, 135,
Cagayan Basin 387 climatic criteria 425-426 235,236,401
caliche, as lithofacies 127, 128 climbing ripple 72 cutoff, meander 8, 11, 14,361
Candeias field 288 clotted mud 240 Cuyo Basin 389, 486, 497
cannel coal 180 cluster analysis 40 cycle of erosion 12, 13, 14,59
cannibalism 403 coal cyclothem
Canterbury Plains 192,358,457, and tectonics 351-352 causes of 477
47l as architectural element 170, duration 252, 476
Caiiizar Sanstone 308 179-184 history of concept 27,28
Caiio Limon field 486 as climatic indicator 438, 439- Illinois-type 263, 264
carbonaceous mudstone 127 440,442 Late Paleozoic 74,263,446,463
carbonate, freshwater 127, 129 as lithofacies 127, 129 major 68,264,476
Cardium Formation 68, 465 as marker bed 273 minor 68,264,476,477
Carrot River 327 in stratigraphic sequences 472, tectonic 59, 27l, 404
Caseyville Formation 496 474
Castisent Sandstone 40, 224 origin of 52, 55 "D" Sandstone 484,491,496,501-
Castlegate Sandstone coarsening-upward cycle 40-41, 506
autogenic processes 323, 325 50,256,258,259 Dakota Sandstone 407,502
macro form model 151,158,299 collision-related basins 395-402 Danube River 419
paleogeography 366 and petroleum 486-487 Daqing field 479,486,493,497,
sequences 264,265,268,462, colluvial deposits 247 515-517
463,467,468,471,472 color 439,451 De Re Metallica 6, 7
Catahoula Formation 262, 382 Colorado River (Gulf of California) Dead Sea 416,424,429,443
catastrophism 10 417,439 Death Valley 189,190,206,207,
catena 184,252,440 Colorado River (Texas) 464,475 248,415
Catskill Formation 24 Columbia River 364 debris flow (sediment-gravity flow)
CB-type field 292, 488, 489, 490, Colville Trough 506, 507 and alluvial fans 245-248,313
492-493,494,496,497,513-521 composite paleosol 256 and climate 425,427,429,430
Subject Index 577
and fluvial style 206-208 enhanced recovery 304-306,509, in subsurface 29, 30

as architectural element 146 513,515 Fire Island 72
early observations of 18 Enon Conglomerate 145,396 first-order cycle 476
frequency of 63 entrainment 9, 11 Firth River 32, 356, 357
lithofacies 100, 106, 107 entrenchment, fan 84,311-315, fixed channel 45, 94
process 105 345,362 flexuralloading 371
deductive reasoning 96 eolian lithofacies flexural moat 375,404,409
deflation surface 429 and climate 427,428,429,438, flood (re. steam/water injection) see
deglaciation 58 439,441-442 enhanced recovery
delta, history of research 23 fluvial associations 130, 241 floodplain classification 201
Deluge, The 8 epeirogeny 412 floodplain evolution 320-321, 325,
Denkman Formation 481,487 ephemeral rivers 328
Denver Basin 488, 496, 502, 503, and climate 422, 423, 428-429, floods
504 435,437 overbank 177
desert varnish 313 fluvial style of 54,219,223,240- power of 8, 11
desiccation cracks 125, 429, 435, 245 flow regime 20, 22
438 epsilon crossbedding 26, 28, 36, 49, flow unit 278
diachronous collision 402 50,83,155 fluid flow 100, 492
diagonal gravel sheet 102 equilibrium point 354,461 flume experiments/models 15,
diamict(ite) 106, 146,432 equilibrium profile, see graded 17,20,21,22,38,100,141-143,
diffuse gravel sheet 102, 140 profile 361
diluvial gravel 8 Eromanga Basin 481,484,486,489, fluvial sedimentology, origins of 5
dimensions of sandbodies, see 492,497 fluvial symposia, see symposia
sandbody erosion cycle, see cycle of erosion fluvioglacial deposits, see outwash
dinosaur footprints 125 Escanilla Formation 171,202,322 flux-driven model 373
dip meter 298-302, 303, 304 escape tectonics 378,387,400,401, forced -regressive-wedge systems
dispersion (in fan cycle) 84 486,515 tract 459
distribution-driven model 374 Esplugafreda fluvial system 335 forearc basins 386
diurnal processes 61,62,69 estuarine deposits, facies and petroleum 486
Djunggar Basin 486 bedforms 70 forebulge 403
Donjek model 31, 199,219 fluvial associations 130, 224 foreland basins 387-392,400,402-
Donjek River 32,43,210,212,325 in incised valleys 466, 482, 490, 409,455
Donovan/Sligo Formation 485, 496 499 and petroleum 479,486, 491
double crest stream 15 in lowstand systems tract 466, zone A/B subdivision 462, 463
double crested ripple 119 470,471 forest moor 181
down-climbing bedforms 52, 152 Euphrates River 318,364,400,419, form roughness 18
downlap, sequence 474,475 423 Fort Union Formation 286, 287
drainage diversion 348 eustasy 68 fourth-order cycle 252, 477
dreikanter 429, 441 evaporite 190, 222, 229, 245, 429, Fraser Valley 443
drift bedding 19 438,441 Fresnel zone 292
dune evapotranspiration 448, 449 Frio Formation 368, 369, 479, 482,
classification and origin 20, 38, extensional basins 377-386,455, 485,487
60,62 461,477 FSST 459
gravel 102, 103, 104 and petroleum 479,492,515
humpback 38 tectonics 344,477 Gallup Sandstone 64
Dunvegan Formation 264 Eyre Formation 412 Galveston Island 65, 66, 72
dynamic event 42, 63, 71 gamma-ray log 273
dynamic superimposition (bedform) facies model, principle of 39 Gamtoos Basin 396
69, 110, 146 factor analysis 19,39 Ganges River 10,351,364,404
dynamic topography 412 failed rifts 382-386 ganister 183
Fall River 496 gas/oil ratio 305
East African Rift System 380,415, falling stage systems tract 459,475 genetic stratigraphic sequence 474
416 false bedding 9, 24 geomorphic cycle 59
East Coast Depression, NZ 386 fan delta 245,413,415 geomorphic threshold 58, 59, 73,
East Java Basin 392 fan segmentation 344, 345 252,429
East Shetland Basin 383,481,485, fan, see alluvial fan Gharif Formation 489
487,510 fan-head trench 11,312,313,345 giant crossbedding 43, 150
East Texas Basin 487 fanglomerate 31, 246, 346 Gilbertian delta 178,370
Eastburn field 493 festoon crossbedding 21, 114 Gippsland Basin 383,385,386,481,
Eclipse Trough 443 fifth-order cycle 252, 477 487,489,491,493,497
Egypt 6 fining-upward cycle Gironde Estuary 499
Endicott field 484,485,487,492, history of concept 24, 26-30, glacial depositional systems 260
496,506 33-34,36,40-41 glacial lithofacies 130
578 Subject Index
glacioeustasy 68,447-449,466,476, 432,442 502,503

477 hummocky cross-stratification 62, Jackson field 487
glaebule 183 63,69,70,121,245 Jarama River 219
Glauconite Formation 481 humpback dune 102,104,111,115, Jiangsu Basin 482, 486
Glenrock field 491,496 120,146,437 jigsaw-puzzle reservoirs 489, 493
GPR, see ground-penetrating radar Huntsman Shale 501-504 Jiuquan Basin 402
graded (equilibrium) profile 11, 14, Huon Gulf 403 Jiuxi Basin 486
16,353,456,457,461 hurricanes 63, 70 Jiyang Depression 491
grain flow 100, 106 Hutton Sandstone 484, 486, 492, jokulhlaup 431
grain-size analysis 18-19,56 497 Jordan Valley 416,417
granite wash 274 hydraulic geometry 16-17
Granite Wash (formation) 492,496 hydraulic mine 17 Karoo Basin 267,408,476,487
grass, effect on sediment yield 193 hydrography, river 422, 434 Kayenta Formation 137,244,295,
gravel-bed rivers, description of hyperconcentrated flow 100, 246, 387,392,434,443,465
206-217 247,425,429 Kekiktuk Formation 480, 484, 492,
gravity drainage (oil) 304 496
Great Artesian Basin 412 ice scour 429, 433 KenaiGroup 390,391,486,496
Great Valley Basin 386,388 ice-wedge cast 429, 438 kettle hollow 429
Green River Basin 482, 485, 488, ideal proflles 205 Kinderscout-Crawshaw delta 391
496 IHS, see inclined heterolithic Klaralven River 18
Green River Formation 496 sedimentation Kootenai Formation 496, 500, 501
grid-block model 282 Illinois Basin 496 Kosi River/fan
grit 264,265,266 imbrication 10,22, 101, 102, 105 and alluvial architecture 34, 322,
ground-penetrating radar 292-293, in-plane stress, see intraplate stress 323
295 incised valleys and alluvial fan classification 31,
ground-truth 293 and base level change 358,461, 245-248,427
group classification (of lithosomes) 464-465 and autogenic processes 315,316,
60-68,251,252-253,308,476 and climate change 464-465 317,321,322,323
growth increment 61,62,63 and petroleum 479,481,482-484, and foreland basins 404
growth structures 348, 403 495-506 Kutei Basin 383
Guadalupe River 357 and sequence stratigraphy 461,
Gueydan system 254, 255, 256, 262 466,467,469,474 labyrinth reservoirs 489
Gullfaks field 485,510 as architectural element 136 lacustrine facies 130
Gypsy Sandstone 282 in "group" classification 252 ladderback ripple 115, 119
in seismic data 289, 498, 503, 505 lag (bedform) 22, 115,436
half-graben basin 378,379 inclined heterolithic sedimentation Lake Eyre 163,232
Halibut field 487 (IHS) 50,51,224,229,471,472 Lake Eyre basin
Hassi Messaoud field 492, 497 inductive reasoning 96 anastomosed channels 232-233,
helical flow 155,161,194,217 Indus River 350,351,356,357,364, 438
heterogeneity, reservoir 61,292, 365 cratonic sedimentation 412, 418,
492,509,513 Indus-Ganges Basin, see Himalayan 419
hierarchy of foredeep Lake Missoula 36, 431
architectural units 62 inflltration 103, 433 laminar flow 99
bedforms 22,23,42,60 inner accretionary bank 161 Lane field 491,496
channels and bars 43 instantaneous sedimentation rate lateral accretion
cycles of sedimentation 59 62,71 element 155-163
time scales 57-60 interconnectedness, sandbody 232, history of research 15, 23, 25-28,
highstand systems tract 459,474- 302,305,334-337,475,516,517 49
475 interfluve Latrobe Group 385, 386, 491, 497
Himalayan collision 370,400,401 sedimentation 429,437,441,463, layer cake reservoirs 489
Himalayan foredeep 11, 22, 350, 474 levee 138,169,171,172-173,433,
351,364,365,370,375,397,398, vegetation 193 Liebre Fault Zone 394
404,408,486 intracratonic basins and petroleum linguoid bar 51,436
hinterland basins 395, 400-402 487 linguoid dune 146
history matching 306 intraformational unconformity linguoid ripple 20
Hjortestegvatnet fan 258 346,347,403 Lisburne Group 507
Hoback Formation 496 intraplate stress 68,352,370-372, lithofacies
hollow (architectural element) 163, 412,477 classification 77-78
165-167,238,240,435,437 Irrawaddy River 364,419 codes 77-78,79
Hornelen Basin 40, 393, 394, 395 isopach map 278, 282, 493 scale 77
HST 459 Ivishak Member 507,509 Little Bow field 490, 495, 496, 498-
Hughenden field 491 499
humid climate 425,426,430-431, "J" Sandstone 467,484,491, 496, Little Creek field 284, 493, 496,
Subject Index 579
513-515 migration 155,159, 161, 164,229, sediment load 11

Little Sulphur Creek Basin 394, 395 317,347 slope of 357
Llanos Basin 387 wavelength 330, 331 tectonic setting 364,419
Llanos Orientales Basin 486 Meander River 6, 318 Mississippi-Missouri system, sedi-
lobe, fan 312,313,315,491 meanderbelt width 333 mentsources 367,369,382
lobe, gravel 105 meandering rivers mixed-load stream 15,50, 196
loess 9 fluvial style 219-229 mobile channel 48,94,341
longitudinal gravel sheet 102, 103 history of term 14 model
longitudinal transport, see axial Medicine River field 486 actualistic facies, history of 26,
transport Mediterranean climate 450 31
losimean fan 247,248 Meervlakte Basin 387 alluvial architecture 327-342,
Loup River 436 megacycle 252,253,259,261,270, 492
Lower Muddy Sandstone 496, 502 311,384 Bijou Creek-type river 33, 54,
lowstand systems tract 459, 466- megacyclothem 68,476 163
469 megaevent 58,60 Brahmaputra River, bar-top 156,
LST 459 megaripple 62 160
Lunde Formation 268 megasequence, see megacycle Castlegate Sandstone, macroform
Luzon Central Valley 386 Mekong River 364, 370 151, 158
LWST 459 Melville Island Group 419 channel 49
Mesa Rica Sandstone 467 climatic 421,422
Mackenzie River 364 Mesa Verde Group 365, 366, 406, constructivist facies 202
macroform 407,462,480 Donjek-type river 31
cycle 323, 326 mesoevent 60 facies, principle of 39
definition of 42 mesoform, definition of 42 fan evolution 84, 252, 258-259,
macro scale set 81 Mesopotamia 399, 400 311-317,344-346
Madison River 348 mesoscale set 81 floodplain evolution
320-321,
Madura Basin 392 mesothem 68,252,265,266,267, 325,328
Magdalena River 171,172,173,232, 476 flume 100,141-143,361
236 Messla field 492,497,508,509 flume, gravel-bed braided river
Magellanes Basin 389, 390 Meuse River 326,448,469,471 141-143
magnetostratigraphy 73, 74, 293, MFS 459 grid-block 282
296,297,298 micro event 60 oil and gas field classification
Mahakam River 386 microform, definition of 42 488,489
Main Consolidated field 489,491, microscale set 81 paleogeographic 412-419
496 MicroScanner, Schlumberger 85, pedofacies 45, 46, 184, 187
major (erosion) surface 44,87 298 Platte-type river 200, 234-235,
Malcolm River 356, 357 Middle Ground field 486 434
Mallalieu field 493, 496 Midland Basin 496 production 303, 492, 493, 509,
Mancos Shale 407,462,480,482 Midland field 491 513
Mannville Sandstone Midland Valley 386,411 sand-belt 252,254
and anastomosed models 53, Milankovitch cycle Scott-type river 199,208
279-280 and bounding surfaces 62, 82 sediment supply and transfer
and petroleum 479,484,490,491, and climate change 421,444-447 372-376
495,496,498 and glacio eustasy 260 sequence 338,453-478
enhanced production of 304 and sequence stratigraphy 476, South Saskatchewan-type river
incised valleys 136,465,467,484, 477 31,35,41,50,200,235,238,434
490,491,495,496,498 time scale 62,68,82,251,252, molasse 9, 12,27,477
log character 275 261,476,477 Molasse (Alpine) 364,365,397,
Maracaibo Basin 401 Milk River 158 399,404,419
Maraii6n Basin 387 Miller field 493 Molteno Formation 267,408
Markham Valley 403 minor (erosion) surface 44,85 monsoon 423
Markov chain analysis 31,39, 121, Mississippi delta 65,67,68 moraine 130, 432
325 sedimentation rate 72,73,74 Moray Firth basin 497
mass balance 372 Mississippi River Morrison Formation 43, 54, 134,
massive sandstone 123, 139 avulsion of channels 318-319, 165,238,240,281,322,435
master bedding surface 66, 70 347 Morrow Sandstone 289,290,487,
matrix-supported gravel 106 floodplain deposits 169,172,173, 491,496
maturity, pedogenic 45, 68, 73 176,177,178,321 mosaic, facies 288
maximum flooding surface 459, hierarchy of units in 85 Mowry Shale 503
467,474 historical investigations 8, 9, 11, mud braid 233, 240, 438
McDonald Draw field 486 25,27 mudflow, early observations 18,31
meander incised valley 358, 469, 483 Murray River 356,357,364
evolution of 8, 11, 63 meander patterns 14, 169 Murrumbidgee River 36,318
580 Subject Index
Narrabeen group 276 and sedimentation rate 73 play, petroleum 479

neck cutoff 8,317,361 as clasts 432 playa 190,378,417,426,435
neotectonics 348,349,351 as climatic indicators 429, 440- plutonism 7
neptunism 7 441 Po River 6,364,399,400,403,417,
nesting of cycles 39 as lithofacies 127 419,423
net-to-gross ratio 469,472,475 as marker beds 272, 273 point bars (see also lateral accretion
New Red Sandstone 476 as overbank elements 183-190 element)
Newark basins/Supergroup 379, as sequence boundaries 463,474 accretionary dip 158,201,217,
380,476 classification 186 222,329,331
Nice 8 in subsurface 286 as reservoirs 484,492,493,514,
nickpoint 355, 356, 463 paleostress 371 515,519
Niger River 364,419 paleovalley, see incised valley facies 155-163
Nile River 6,356,357,358,362,419 palustrine limestone 130, 442 history of research 26, 27, 28, 30,
Ninian field 487 Pangea 421,476 36
nonevent 58, 60 Pannonian Basin 387 pond deposits 177,229
North China Basin 486 paraglacial climate 425,426,431, porosity and fluid flow 302-306,
North Sea Basin 383, 384, 479, 487 432 492
North Slope 414,479,480,487,491, parting lineation 120,437 Potwar Plateau 351, 352
492,496,506 Pattani trough 487 poudingues 7
North Viking Graben 480,487,489, peat:coal ratio 180 Powder River Basin 491,496
492,497 Peco field 484, 496 pre-Devonian rivers 35, 150, 193,
Nubian Sandstone 451 ped 183,438 194,450,451
pedofacies preservation potential 41-42
Oakville Formation 262, 369 and climate 440 pressure tests 302, 303, 492
oil and gas fields, classification and floodplain evolution 321 pressure-depth plot 303, 304
488,489 and sedimentation rate 73 primary current lineation 19
Okanogen Valley 443 model 45,46, 184, 187 production model 303, 492, 493,
Okavango fan 245,247,248,315, sequence 252,254,257 509,513
427,433 pedogenic beds and architecture, see profIle, outcrop 95-98
Old Channel 327 paleosols prospecting, subsurface 331
Old Red Continent 400 pedogenic carbonate 127 Prudhoe Bay oil field 479,484, 485,
Old Red Sandstone 27,36,252,271, penetration probability 334-337 487,491,492,496,506-509
337,386 Peo:.;-ia field 493 pseudoplastic flow 105, 106, 108
oncolite, freshwater 127, 129 peripheral basins 396 pull-apart basin 393
onlap (sequence) 355,473 and petroleum 479 pulse test 304
oolite, freshwater 127, 129 permeability trends 491,509 punctuated aggradational cycle 68
open moor 181 Persian Gulf 397,400,403 Putumayo Basin 387,486
open-work gravel 103, 105 Petrified National Forest 465 PV -type field 292, 488, 490-491,
operational unit 274 petroleum fields 495-506
orbital forcing 444-447,475, 477 classification 488, 489
Ordos Basin 466,481,482,484,485, reservoir geometry 484-485 Qaidam Basin 402, 486
486,491 tectonic setting 479,486-489
orographic rainfall 423, 443 petrophysical zone 278 radar facies 292-293
Osorno Basin 386 phi scale 18 rafting, clast 105
Oued Mya Basin 484,487,489,497 Phitsanulok Basin 487 Ramu Basin 387
outwash (fluvioglacial) 31,40-41, photomosaics 75-77,95-98 Rangitata River 210,213
208,424,431,432 piggyback basin 351,409 Ravenscar Group 334
overfIlled basin 403, 404 Pine Valley 375 ravinement surface 68
overpassing, gravel 103 placer deposits 6 reaction group 40
oxbow 177 plain 15 reactivation surface
plane bed 38 in bedforms 42,43, 69, 116
Pacencia Group 442 Platte River in macro forms 63, 64, 66, 70, 83,
paleocurrent analysis and climate 422, 433, 436, 437 162
and architectural analysis 98 architectural elements 148 Recluse field 491,496
and macroform reconstruction bedforms 148,436,437 Reconcavo Basin 482,487,489,497
151 changes in fluvial style 193,235, recovery, oil 304
as mapping tool 293-297 433 recurrence interval 60,71-74
methods of 78,80-81 early studies of 14, 50 red beds 421,439,442
origins of 10 fluvial styles 215, 220, 233, 239, Red Earth field 486, 492
paleogeographic models 412-419 242 Red Fork Sandstone 496, 504, 505
paleohydraulics 35-37 Platte-type Red Wash field 484,496,519-521
paleosols macroform 153, 154, 233, 235 redox conditions 423, 439, 440
and alluvial architecture 34, 45 river 50,51,200,234-235,434 reed moor 181
Subject Index 581
reentrant 409,414 sand-bed rivers 217-245 shape (clast) 19

relaxation, see lag sand-drift surface 442 sheet sandstone
research trends 3, 523-524 sandbody dimensions 328-334 as architectural element 147,149,
retroarc basins 387-392,395,400 Sandy Neck Spit 65 163,235,240,245
and petroleum 479, 486 Sao Francisco delta 417 as petroleum reservoir 479, 484,
return period 252 Sapelo Island 65, 66, 72 506-513
Reynolds number 18 sapropel 180 definition of 45, 48
Rhine River 9, 326, 364, 448, 469, Sarir field 487 sheet-scale cycle 323, 326
471 Sarir Sandstone 484,497,509,511 Sherwood Sandstone 484, 497
Rhone River, delta 9,67,68,72,73, Saskatchewan River 232, 327 Shields criterion 18
364,419,423 satellite basin 351,409,410 Shoal field 486
rib (stone) 105 savannah 450 Shoshone River 26
ribbon sandstone sawtooth curve 354, 355 sieve deposit 18
and fluvial style 182,205 Scott-type river 199,208 sigmoidal crossbedding 38
as petroleum reservoir 479, scour hollow 92,123,163,165-167, simple bar 148
484-485 435 single-crest stream 15
definition of 45, 48 scroll bar 195,222,224 sinuosity
in anastomosed systems 52, 229, seasonal process 62,63,69,177, controls of 192,359-361
235,236,237 222,423,426,438,442,451,463 estimation 36
in coal swamp 182 second-order cycle 253,476-477 Sirte Basin 487,489,497,509
in ephemeral systems 222, 321, secondary streams, fan 312 Siwalik Formation/Group
435 sediment gravity flow (see also and basin subsidence 375, 398
in incised valley 466, 499 debris flow, mudflow) architecture of 47, 254
in subsurface 279-280 gravel element 145-146 avulsion frequency 252, 253
mapping of 279-280 processes 105 early study of 12, 13,22
Ridge Basin 393, 394, 395 sediment transfer 372-376 magnetostratigraphy of 296, 297
ridge push 370 sediment yield 193,367-370,475 overbank deposits 179,321
ridge-and-swale 70, 161 sedimentary structures, sedimentation rate 74,375
rift basin 379-381,479 classification of 21 slab pull 370
rigid plug 146 sedimentation rates 71-74 Slims River 32
Rio Grande River 356,357,364,367 sedimentology, revolution in slope
Rio Puerco 435, 442 seismic of fans 246
ripple data 74 of rivers 357
classification 115, 118 facies 283, 284, 285 slug, gravel 101, 140, 142
form 20 resolution 286, 473 Snorre field 487
ripple-drift crossbedding 10, 19 seismic cross-sections (examples) Songliao Basin 387,479,482,485,
river classification, see channel basin-fill 269 486,489,497,515
classification channels 289-292 sorting 18
Rockdale delta 363 incised valleys 288, 289, 498, 503, South Belridge field 485,487, 496
Rocky Ridge field 486 505 South Ceres field 491,496,506
root bed 127,463 prograding fluvial system 285, South Morecambe field 484, 487
rootlets 125, 126, 127 287 South Pass 65
Rosario Embayment 386 seismogram 286, 289 South Pyrenean Basin 404, 405
roundness 19 separation eddy 10, ll, 19,38,50 South Saskatchewan River 153,217,
Rufiji River 318 Sepik Basin 387 223,224,437
sequence South Saskatchewan-type river 31,
sabkha, inland 219 boundary 459-466 35,41,50,200,235,238,434
Sadlerochit Group 479,480,484, concepts, in mapping 275 sphericity (clast) 19
491,492,496,507,509 definition of 453 spheroidal weathering 432
Salar de Punta Negra 414,415 model 338, 453-478 Spirit River Channel 501
Salt Wash Mtlmber 281,282 rank-ordering 404, 475-477 split (coal) 180, 182
saltation 17, 109 self-similarity 454 Squamish River 216
Salton Trough 393 tectonic control 407,458,476, Squirrel Formation 496
Salween River 370 477 stability, of bedforms 110
San Gabriel fault 393, 394 sequence stratigraphy Statfjord field 497,510,511-513
San Joaquin Valley 390, 442, 496 and climate change 447-449,463 Statfjord Formation 268, 383, 479,
San Jorge Basin 389,486 history of 34, 55, 453-454 480,485,492,497
San Luis Valley 442 Sergi Formation 497 storey 82
sand set boundary 21,69 straight rivers 197
flat 46,51,153,235,436 Sevier Basin 387,389 stratigraphic base level 352
shoal 52 Sevier orogen 480 streaming lineation 120
wave 20,22,52,60,63,70,71,72, SH-type field 292,488,489,491- strike-slip fault basins 393-395,
114 492,494,496,497,506-513 400
582 Subject Index
strike-slip fault basins (Contd.) traction currents 99-105,109 vertical accretion, history of research
and petroleum 479,487,515 traction load 19 15,23,25
structural trap 492, 495 Transcontinental Arch 504 vertical profIles
subsidence-driven model 373 transgressive systems tract 471- fluvial 203-205
success rate 336,337 473 history of interpretation 24,
successor basin 411 transitional zone (meander bend) 27-28
Sumatra basins 392,415 164,217 problems of interpretation
summary sequence 34, 39 transrotational basins 393 38-42
Sundaland Basin 489 transverse bar 155,217 Vienna Basin 393
super surface 43, 59, 68, 71 transverse bedform, gravel 104, Viking Graben 383,384,385,510,
Surat Basin 481,486,493 105 512
surveillance geology 302-306,492, transverse drainage 364-365, 379, Violin Breccia 394
509 398, 399, 404, 413 viscous fingering 304
Susitna River 417 transverse rib 105 volcaniclastic lithofacies l30
suspended load 15,17,19,50 Travis Peak Formation 479,481, volcaniclastic sedimentation 178,
suspended-load stream 50, 196 484,487 192,370
Sverdrup Basin 365,371,412 trend-surface analysis 21
swale 155 Trollheim fan 206,207,429,432 VVabashRiver 49,63,72,215,221
swamp deposits 177,180-181 tropical climate 426,430-431,432 wadi l33, l39, 140
Swanson River field 486 Tsangpo River 370 VV aimakariri River 325
Sweetgrass Arch 501 TST 459 walking-out correlation 272
Sydney Basin 365 tufa 127 VValther's Law 38
symposia on fluvial sedimentology tuff, as marker beds 273 wandering rivers
2-3, l3, 37 Tulare Formation 482, 485, 496 definition of 196
syndepositional tectonics 344-352 turbulent flow 99 description of 211-212,216
syntectonic model 374,375,403 Tuscaloosa Sandstone 282,368,479, wash-out (coal) 180
481,485,487,491,493,496 waterflood, see enhanced recovery
Table Mountain Sandstone 419 two-dimensional dune 109, 110, weathering, clast 432
Tana River 433,436 111,113,148,436,437 well spacing 282, 335, 336
Tarim Basin 402,443,486 Tyaughton-Methow Basin 411 VV est African Rift System 380, 394
tectonic Tyler Formation 484, 496 VVest Java Basin 392
classification, basins 376,378 types-I, 2 unconformity 355,459, VV estern Approaches Trough 497
cyclothem, see cyclothem, tectonic 460,463 VV estern Interior Seaway 406
geomorphology 345 types-A, B ripple 115, 118 VV estern Trough 386
systems tract 345 VV estwater Canyon Member 43,
temperate climate 425, 426, 430 U-shaped fIlls 93,174 134,165,238,240,277,322-324,
terminal fan 31,54,245,246,248, Ucayali River 63 435
249,311,433,435,450 Uinta Basin 481, 482, 484, 488, 496, width/depth ratio 45, l33, 329-334
Termit Basin 380 518,519 VVilcox Group 363, 368, 369, 482,
terrace 464, 465, 475 unconformity classification 355, 487,491
terrane analysis, and basins 409, 459,460,463 VVilliam River 192,437
411-412 underfilled basin 403, 404 VVilliston Basin 496
texture 18 underfit stream 197 VVillwood Formation 257,262
thalweg, meandering 141, 194, 197 uniformitarianism 5, 7,8-10 wing 45,256
Thames River 6 unsteady sedimentation 73 wireline log
third-order cycle 252, 477 upstream accretion 152 facies 275,276, 277
three-dimensional dune 109, 110, upward-fining, see fining upward interpretation 29, 273-276
112, 113, 148 Ure River 166 VVitwatersrand Basin 401
three-dimensional seismic data 88, VVombarra operational unit 277,
291,292,305,492 Vacas Muertas field 487 278
Tibet, effect on climate 444 variance, paleocurrent 80 VVoodbine Formation 481,487
tidal deposits, facies 61,62,63,69, vector methods 21 VVytch Farm field 484,487,497
72,130,224,471-473,499 vegetation
Tigris River 364,400, 419, 423 and climate 423, 448-449, 450 Yangtze River 72
till ball 431,432 effect on channels 35, 192,433, Yellow River 322,358,364
Tjeribon Basin 392 448-449
Torrivio Member 64 vertebrate fossils 173 Zenith field 496,501-506

Geologia Depositos

Uploaded by

Copyright:

Available Formats

Geologia Depositos

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Geologia Depositos

Uploaded by

Copyright:

Available Formats

Andrew D.

ISBN 978-3-642-08211-5 ISBN 978-3-662-03237-4 (eBook)

SPIN: 11751502 Printed on acid-free paper 30/3141/as 543

2 Historical Background ......................................... . 5

2.1 Introduction .................................................. . 5

4 Methods of Architectural-Element Analysis ........................ 75

6 Architectural Elements Formed Within Channels ................... 131

7 Architectural Elements of the Overbank Environment ............... 169

7.2.3 Crevasse-Splay Deposits (Element CS) ............................ . 174

8 Fluvial Styles and Facies Models ................................. . 191

9 The Stratigraphic Architecture of Fluvial Depositional Systems ...... . 251

10 Fluvial Depositional Systems and Autogenic Sedimentary Controls .. . 311

10.3.1 The Development of Meander Belts ............................... . 318

11 Tectonic Control of Fluvial Sedimentation ........................ . 343

12 What Does Fluvial Lithofacies Reveal About Climate? .............. . 421

12.5.1 Mass-flow Versus Traction-current Processes ....................... 425

13 Sequence Stratigraphy .......................................... . 453

14 Stratigraphic and Tectonic Controls on the Distribution

14.2.1.1 Clastic Wedges ................................................. 480

16 Future Research Trends ........................................ . 523

References ............................. ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525

Author Index 567

Subject Index 575

2.1 Introduction Jopling (l975) discussed early research on fluvio-

A-·RIVER. B-WEIR. C-GATE. D-AREA. E-MEADOW. F-FENCE. G-DITCH.

Fig. 2.3. Cross-stratified sand (C) overlain by

Fig. 2.4. Development of an asymmetric

In succeeding sections, an attempt will be made to

flow velocities (or shallower depths) than any of the

va . YlRTICAl ACCRETION DEPOSITS

.:!l • SPlAYS AlONe; 'IU.£O CHANNEL

Fig. 2.12. Formation of the Shoshone valley floor. A

Fisk's main sedimentological contribution was to 2.3.6.2 Meandering River Deposits:

Fig. 2.13. A suite of typical fining-upward fluvial cyclo-

Fig.2.1S. Point-bar model, based on obser-

Piedmont formed of alluvial fans

Fig.2.1B. The four basic fluvial chan-

o bar surfaces covered

Slims Firth Donjek Fig. 2.19. Three typical gravel-bed

1::::::::1 high speed region

Fig. 2.23. Bedform phase diagram showing how

Fig. 2.24. Comparison of four vertical profIles,

(a) (b) (c) (d)

Fig. 2.25. A Progradation and vertical

channel-avulsion events (an allogenic process) or

cross-bedding, and related them to changes in

~ Mudston,-clast Itringers (GI) ~ ·Troullh cross-bedded sandstone (S3)

r:~\ij Masslv. intraformational canlliomerate (G 2 )

~ Cross-bedded landstone (S21 5- Position in depositional sequence

Fig. 2.31. Development of compound pedo-

. . ' ... ...............

iiil Field de5CrJ>tion hollow UI

;,. ) Genetic ardysis fA hollows

~. '. :-'.< . ' .;c::::;

Fig. 2.33. Channel definition and recognition. (Friend 1983)

I ) Wos flow chonnelised?

SEDIMENT Fig. 2.35. Model of distinctive types of alluvial

D Coorse _Fine sediment

"Oft' l .,olt~ . 1ICI