Integrating Petrophysics and Allostratigraphy To Find Sweet Spots

Western University
Scholarship@Western
Electronic Thesis and Dissertation Repository
6-4-2018 10:00 AM
Integrating Petrophysics and Allostratigraphy to Find Sweet Spots

in the Upper Cretaceous Belle Fourche and Second White Specks
Alloformations, West-Central Alberta, Canada
Kienan P. Marion
The University of Western Ontario
Supervisor
Cheadle, Burns A.
Graduate Program in Geology

A thesis submitted in partial fulfillment of the requirements for the degree in Master of Science
© Kienan P. Marion 2018
Follow this and additional works at: https://ir.lib.uwo.ca/etd
Part of the Geology Commons, and the Stratigraphy Commons
Recommended Citation
Marion, Kienan P., "Integrating Petrophysics and Allostratigraphy to Find Sweet Spots in the Upper
Cretaceous Belle Fourche and Second White Specks Alloformations, West-Central Alberta, Canada"
(2018). Electronic Thesis and Dissertation Repository. 5440.
https://ir.lib.uwo.ca/etd/5440
This Dissertation/Thesis is brought to you for free and open access by Scholarship@Western. It has been accepted
for inclusion in Electronic Thesis and Dissertation Repository by an authorized administrator of
Scholarship@Western. For more information, please contact [email protected].
Solving the Second White Specks
Abstract
The Upper Cretaceous Second White Specks Formation – an organic-rich, calcareous

mudrock succession within the lower Colorado Group – is a prolific self-sourcing tight oil
reservoir in Alberta. Due to the low porosity and permeability of this interval, localized
natural fracture networks have previously provided the only means for oil to flow at
economic rates. This study, focused in west-central Alberta, used allostratigraphic
methods to subdivide the Second White Specks Formation into allomembers that define
hydraulic flow units. The petrophysical properties (porosity, organic content, clay
volume, and brittleness) of each allomember were modelled using a basic suite of
geophysical wireline logs and sparse core data. When overlain with historic oil
production results, modelled petrophysical parameters delineated previously
unrecognized “sweet spots” that likely have increased oil potential. Applying the
innovative petrophysical workflow developed in this study may increase the likelihood of
drilling successful wells in similar exploration scenarios with limited datasets.
Keywords: allostratigraphy, brittleness, Cenomanian-Turonian boundary, formation

evaluation, lower Colorado Group, petroleum geology, petrophysics, reservoir
characterization, Second White Specks, Western Canada Sedimentary Basin
ii
“It's a dangerous business, Frodo, going out of your door,” he

used to say. “You step into the Road, and if you don't keep your
feet, there is no knowing where you might be swept off to.
Do you realize that this is the very path that goes through
Mirkwood, and that if you let it, it might take you to the Lonely
Mountain or even further and to worse places?”
– J.R.R. Tolkien, The Fellowship of the Ring (1954)
iii
Acknowledgments
I thank you, gentle reader, for having the fortitude to pick up this thesis – it had become a
veritable tome by the time I ran out of things to say. There are lots of pictures, I promise!
I would like to thank Husky Energy, Devon Canada Corporation, and Tuzo Energy for
funding this research as part of the Second White Specks Liquids-Rich Research
Consortium at Western University. I am grateful to Schlumberger Canada, geoLOGIC
systems, Divestco, and IHS Canada for their generous donations of software and data to
this project. Additional research funding for this thesis was provided by a Society of
Petrophysicists and Well Log Analysts (SPWLA) Foundation Grant and an American
Association of Petroleum Geologists (AAPG) Grants-in-Aid award in 2016.
This thesis would not have existed without the unrelenting support of Dr. Burns Cheadle,
my indefatigable supervisor. Without his mentorship, which began in 2012 during my
undergraduate study at Western, I would not be where I am today. His patience and
encouragement allowed me to power through when I thought I could go no further.
I am further indebted to my thesis examination committee – Dr. Mahi Singh, Dr.

Catherine Neish, and Dr. Phil McCausland – for the donation of their time and invaluable
scientific expertise.
I extend my profound gratitude to Dr. Guy Plint, whose passion for sedimentary geology
kindled my own back in 2010. His good-humored skepticism and deep curiosity for all
things related to the natural world helped to show me that there is always more to learn. I
would also like to express my appreciation to his wife Annemarie and his daughter Tessa
for being incredibly kind and welcoming to me during my time in London.
I was blessed with wonderful geoscience mentors during my time at NAL Resources,
Imperial Oil, and Shell Canada. I am eternally grateful for the excellent training and
professional guidance that was provided to me by Kevin Lounsbury, Bob Leibel,
Ndubuisi Obi, Sean McDonald, Dave Llewelyn, and Jenna Lapointe in the fields of
exploration geology, production geology, and petrophysics. Not only has that training
iv
helped me further my career, but it also came in extremely handy over the course of my
thesis research.
I doubt I could give my parents, Karen and Don, the thanks they deserve without
continuing on for several pages. Without them, I could never have reached this point.
Their enduring love has kept me going through many hardships, and they have always
helped me keep things in perspective.
To my brother, Riley: always stay lit.
My colleagues (now friends) in the Petroleum Geology and Basin Analysis Laboratories
at Western provided invaluable input on this work. Mailyng Aviles and Jessica Flynn
both deserve special recognition: they were always open-minded and served as great
sounding boards for some of my more outlandish ideas. I also acknowledge the support of
Matea Drljepan, Beth Hooper, Katharin Pavan, Karen Grey, and Negar Nazari in this
regard.
Over the past eight years I have spent at Western, I often felt as though I was stuck
halfway between two worlds: my body was in London, but my heart was in my
hometown of Calgary. Marianna Chibotar-Rutkevich, my piano teacher, became my first
true friend at Western in 2010. Her compassion, warmth, and artistry have inspired me to
become a better person.
To my friends: I owe you a drink or five. May we meet again.
v
Table of Contents
Abstract........................................................................................................................... ii
Acknowledgments .......................................................................................................... iv
Table of Contents ........................................................................................................... vi
List of Figures ................................................................................................................ xi
List of Tables ................................................................................................................ xv
Variables and Abbreviations......................................................................................... xvi
PART I: INTRODUCTION & LITERATURE REVIEW
Chapter 1 ........................................................................................................................ 1
1 Introduction ................................................................................................................ 1
Summary ............................................................................................................. 1
Problem ............................................................................................................... 1
1.2.1 Influence of brittle behaviour...................................................................... 2
1.2.2 Influence of porosity................................................................................... 4
1.2.3 Stratal architecture of the Second White Specks ......................................... 5
Aims & Objectives .............................................................................................. 6
1.3.1 Research aim .............................................................................................. 6
1.3.2 Objectives .................................................................................................. 6
Scope................................................................................................................... 7
Significance ......................................................................................................... 8
Overview of thesis structure............................................................................... 10
Chapter 2 ...................................................................................................................... 11
2 Stratigraphy, sedimentology, and play analysis of the Second White Specks and
Belle Fourche Formations ......................................................................................... 11
Overview ........................................................................................................... 11
Geologic setting of Lower Colorado Group Shales ............................................ 11
2.2.1 Tectonic setting ........................................................................................ 13
2.2.2 Climate and physiography ........................................................................ 17
Establishing nomenclature for Upper Cretaceous lower Colorado Group rocks
in west-central Alberta ....................................................................................... 20
2.3.1 Fish Scales Formation .............................................................................. 23
2.3.2 Belle Fourche Formation .......................................................................... 25
2.3.3 Second White Specks Formation .............................................................. 26
Petroleum system analysis of the lower Colorado Group in the Willesden
Green – Gilby area............................................................................................. 28
2.4.1 Source rock quality, maturation, and migration......................................... 28
2.4.2 Storage capacity ....................................................................................... 30
2.4.3 Flow capacity ........................................................................................... 32
Highlights .......................................................................................................... 35
Chapter 3 ...................................................................................................................... 37
3 Using wellbore petrophysical data for quantitative geological interpretation ............. 37
Overview ........................................................................................................... 37
Role of petrophysics in exploration geology ...................................................... 37
Connecting petrophysics with stratigraphy ......................................................... 38
Overview of well logging .................................................................................. 39
3.4.1 Borehole petrophysics .............................................................................. 40
vi
3.4.2 Drilling mud ............................................................................................. 41

3.4.3 Caliper log................................................................................................ 43
Petrophysical modelling..................................................................................... 43
3.5.1 Modelling Archie-type reservoirs ............................................................. 45
3.5.2 Modelling shaley sand reservoirs .............................................................. 48
Interpreting geophysical well log data from unconventional reservoirs .............. 50
3.6.1 Petrophysical tops guidance for mudstone reservoirs ................................ 50
3.6.2 Well log responses as proxies for mudstone reservoir composition ........... 51
3.6.2.1 Radioactivity logs .............................................................................. 53
3.6.2.1.1 Gamma ray logs ....................................................................... 54
3.6.2.1.2 Spectral gamma ray logs .......................................................... 56
3.6.2.1.3 Bulk density log ....................................................................... 60
3.6.2.1.4 Neutron log .............................................................................. 64
3.6.2.2 Acoustic logs ..................................................................................... 65
3.6.2.2.1 Tool specifications ................................................................... 66
3.6.2.2.2 Environmental effects on sonic logs ......................................... 66
3.6.2.3 Electric logs....................................................................................... 67
3.6.2.3.1 Tool specifications ................................................................... 69
3.6.2.3.2 Environmental effects .............................................................. 69
Highlights .......................................................................................................... 70
PART II: RESEARCH METHODOLOGY
Chapter 4 ...................................................................................................................... 71
4 Stratigraphic method................................................................................................. 71
Overview ........................................................................................................... 71
Stratigraphic nomenclature and bounding surfaces ............................................. 71
Controls on sedimentation ................................................................................. 72
4.3.1 Eustasy ..................................................................................................... 72
4.3.2 Milankovitch orbital forcing ..................................................................... 73
4.3.3 Sediment supply ....................................................................................... 74
4.3.4 dA/dS ratio ............................................................................................... 74
Seismic stratigraphy .......................................................................................... 75
Sequence stratigraphy ........................................................................................ 76
Allostratigraphy ................................................................................................. 77
Highlights .......................................................................................................... 80
Chapter 5 ...................................................................................................................... 81
5 Petrophysical method ............................................................................................... 81
Overview ........................................................................................................... 81
Using the petrophysical model to calculate brittleness........................................ 81
Database development ....................................................................................... 81
Core-to-log correlation ...................................................................................... 85
5.4.1 Total organic carbon (TOC) determination ............................................... 85
5.4.2 X-ray fluorescence (XRF) analyses .......................................................... 87
5.4.3 Normalized mineralogy ............................................................................ 88
Allostratigraphic correlation .............................................................................. 89
Petrophysical modelling workflow..................................................................... 90
5.6.1 Environmental corrections ........................................................................ 94
5.6.1.1 Gamma ray corrections ...................................................................... 94
vii
5.6.1.2 Density log corrections ...................................................................... 95

5.6.1.3 Resistivity corrections ....................................................................... 95
5.6.2 Volume modelling .................................................................................... 95
5.6.2.1 Kerogen volume from logs ................................................................ 98
5.6.2.1.1 DlogR or Exxon method ........................................................... 98
5.6.2.1.2 Issler et al. method ................................................................. 101
5.6.2.1.3 Application of TOC modelling to the lower Colorado Group.. 101
5.6.2.2 Clay volume from logs .................................................................... 104
5.6.2.2.1 Application of clay volume to the lower Colorado Group ....... 105
5.6.2.3 Porosity from logs ........................................................................... 105
5.6.2.3.1 Total porosity ......................................................................... 106
5.6.2.3.2 Total porosity for the lower Colorado Group .......................... 110
5.6.2.3.3 Effective porosity ................................................................... 112
5.6.2.3.4 Effective porosity for the lower Colorado Group .................... 112
5.6.2.4 QFM and carbonate mineral volumes .............................................. 114
5.6.3 Brittleness .............................................................................................. 116
5.6.3.1 Lithological brittleness indices ........................................................ 118
5.6.3.2 Application of brittleness indices to the lower Colorado Group........ 121
Mapping .......................................................................................................... 121
5.7.1 Spatial interpolation ............................................................................... 121
5.7.2 Surface-to-surface operations ................................................................. 124
Analysis........................................................................................................... 125
Highlights ........................................................................................................ 126
PART III: RESULTS
Chapter 6 .................................................................................................................... 128
6 Allostratigraphic correlation of the Lower Second White Specks and Upper Belle
Fourche alloformations in the Willesden Green – Gilby region, Alberta, Canada .... 128
Overview ......................................................................................................... 128
Allostratigraphic framework ............................................................................ 128
Lower Colorado Group facies .......................................................................... 131
6.3.1 Facies observed in core........................................................................... 135
6.3.1.1 Facies 1: bentonite beds ................................................................... 135
6.3.1.2 Facies 2: Very dark grey, planar laminated mudstone ...................... 136
6.3.1.3 Facies 3: Weakly bioturbated, medium-grey laminated mudstone,
interbedded siltstone, and very fine-grained sandstone ..................... 138
6.3.1.4 Facies 4: Wavy laminated calcareous mudstone, siltstone, and fine-
grained sandstone ............................................................................ 139
6.3.1.5 Facies 5: Wavy to undulating laminated calcareous siltstone and
fine-grained sandstone ..................................................................... 140
6.3.2 Facies association ................................................................................... 141
6.3.2.1 Offshore to mid-shelf shallow muddy ramp setting .......................... 142
Fine-grained sediment transport in ancient epicontinental seas ......................... 142
Core-to-log correlation .................................................................................... 146
6.5.1 Electrofacies ........................................................................................... 147
6.5.1.1 Facies 1 ........................................................................................... 147
6.5.1.2 Facies 2 ........................................................................................... 147
6.5.1.3 Facies 3 ........................................................................................... 147
viii
6.5.1.4 Facies 4 and 5 .................................................................................. 147

6.5.1.5 Electrofacies summary..................................................................... 148
Allostratigraphic correlation ............................................................................ 148
6.6.1 Upper Belle Fourche (BF2 and BF3) ...................................................... 151
6.6.2 Second White Specks (Allomember VII) ................................................ 160
6.6.3 Overall interpretation.............................................................................. 161
Highlights ........................................................................................................ 161
Chapter 7 .................................................................................................................... 162
7 Depositional trends of the Second White Specks and Belle Fourche alloformations
in the Willesden Green – Gilby region, Alberta, Canada ......................................... 162
Overview ......................................................................................................... 162
Structure mapping ........................................................................................... 162
Isochore mapping ............................................................................................ 165
7.3.1 Allomember BF1 .................................................................................... 166
7.3.2 Allomember BF2 .................................................................................... 166
7.3.3 Allomember BF3 .................................................................................... 171
7.3.4 Allomember VII ..................................................................................... 171
Isolith mapping ................................................................................................ 172
7.4.1 Allomember BF1 .................................................................................... 173
7.4.2 Allomember BF2 .................................................................................... 173
7.4.3 Allomember BF3 .................................................................................... 178
7.4.4 Allomember VII ..................................................................................... 179
Highlights ........................................................................................................ 179
Chapter 8 .................................................................................................................... 180
8 Petrophysical property mapping of the Lower Second White Specks and Upper
Belle Fourche alloformations in the Willesden Green – Gilby region, Alberta,
Canada ................................................................................................................... 180
Overview ......................................................................................................... 180
TOC ................................................................................................................ 180
Porosity ........................................................................................................... 183
8.3.1 Total porosity ......................................................................................... 183
8.3.2 Effective porosity ................................................................................... 188
8.3.3 Porosity-thickness .................................................................................. 188
Brittleness........................................................................................................ 191
Highlights ........................................................................................................ 191
Chapter 9 .................................................................................................................... 196
9 Sweet spot mapping of the Second White Specks and Belle Fourche alloformations
in the Willesden Green – Gilby region, Alberta, Canada ......................................... 196
Overview ......................................................................................................... 196
Sweet spot maps .............................................................................................. 196
9.2.1 Location of producing wells ................................................................... 196
9.2.2 Overlay maps ......................................................................................... 196
9.2.2.1 Allomember BF3 ............................................................................. 197
9.2.2.2 Allomember VII .............................................................................. 197
Highlights ........................................................................................................ 200
PART IV: DISCUSSION & CONCLUSIONS
Chapter 10................................................................................................................... 201
ix
10 Discussion .............................................................................................................. 201

Defining hydraulic flow units .......................................................................... 201
10.1.1 Significance of allostratigraphic bounding surfaces ................................ 201
10.1.2 Quantifying lateral facies variation ......................................................... 203
Petrophysical property modelling..................................................................... 205
10.2.1 TOC ....................................................................................................... 206
10.2.2 Porosity .................................................................................................. 207
10.2.3 Brittleness .............................................................................................. 210
Sweet spot mapping ......................................................................................... 212
Recommendations for future Second White Specks exploration ....................... 213
Chapter 11................................................................................................................... 215
11 Conclusions ............................................................................................................ 215
Future work ..................................................................................................... 218
Curriculum Vitae......................................................................................................... 220
References .................................................................................................................. 222
x
List of Figures
Fig. 1.1: Two adjacent (~1 km well spacing) Second White Specks-targeted wells, drilled
by the same operator days apart, in the Willesden Green field of Alberta. 102/14-31
produced economic volumes of light oil (122,551 Bbl), whereas 102/8-36 was
unsuccessful. 3
Fig. 1.2: Base map of the Willesden Green – Gilby study area in Alberta, illustrating the
spread of available data across the entire area. The scale bar is in metres (m). 9
Fig. 2.1: A composite map of the Second White Specks lithostratigraphic interval in
southern Alberta. 12
Fig. 2.2: Schematic of a strike-oriented cross-section through a foreland basin. Modified

from DeCelles and Giles (1996), and Allen and Allen (2013). 14
Fig. 2.3: Paleogeographic map of the WCSB during the mid- to Late Cenomanian. 15
Fig. 2.4: Paleogeographic map of the WCSB during the early Turonian. 16
Fig. 2.5: Modified from Blakey (2014). Early Cenomanian through Early Turonian (~99
Ma to ~93.2 Ma) paleogeography of North America, with approximate shoreline
locations. 18
Fig. 2.6: Comprehensive table of formations for the study area, combined with WIS sea
level changes from Schröder-Adams (2014) and Haq (2014). 22
Fig. 2.7: 1-dimensional burial history diagram (top) of 16-18-52-5W5, modified from
Roberts et al. (2005). The inset map is the same as Fig. 2.1, with the study area for this
thesis outlined in orange. The 16-18 well is indicated on the inset map as a red dot. 29
Fig. 2.8: A comparison of different oil and gas reservoirs in terms of reservoir quality;
modified from CSUR (Canadian Society of Unconventional Resources). The Second
White Specks in the Willesden Green area has characteristics of both unconventional and
conventional reservoirs and is characterized as a “hybrid shale”. 31
Fig. 2.9: IP4 (initial average oil rate for the first 4 months) map for Second White Specks
wells in the study area. 33
Fig. 2.10: Cumulative production map for Second White Specks Formation wells in the
study area. Green indicates oil production, red indicates gas production, yellow indicates
condensate production, and blue indicates water production. Most Second White Specks
wells in this area are primarily oil producers and produce negligible water and
condensate, with minor gas production. 34
Fig. 3.1: Schematic of a centered caliper-type tool attached to an instrument assembly,

with idealized responses to beds that could potentially be encountered in the lower
Colorado Group. 42
xi
Fig. 3.2: Comparison of general petrophysical models for Archie-type, shaley sandstone,
and unconventional reservoirs in terms of complexity. The level of petrophysical
complexity increases with the number of components, which makes them more difficult
to resolve. 47
Fig. 3.3: A representative well log (7-19-45-06W5) for the Second White Specks interval.
52
Fig. 3.4: Schematic of a bulk density logging tool in a conventional reservoir. 61
Fig. 5.1: A process flow diagram illustrating the integrated petrophysical and geological
workflows used in this thesis. 82
Fig. 5.2: A screenshot of the Techlog software used in this project, with individual wells
in the left-hand column, the properties of any selected parameters in the second column,
and a cross-section visualized across the main section of the screen. 84
Fig. 5.3: Modified from Ambrose et al. (2010) – this is the petrophysical model used for
the Second White Specks in this study. 91
Fig. 5.4: Modified from Asquith (1990) – this represents the previous petrophysical
model (Fig. 5.3) within a very simplified geological context. 92
Fig. 5.5: Issler model TOC vs. Rock-Eval TOC analyses from the core study interval in
the 100/07-19-045-06W5 well. The regression shows a strong linear relationship with a
moderately high coefficient of determination (r2 = 0.811). 102
Fig. 5.6: A log display of the 7-19 well, showing the sonic and resistivity curves that were
used as input for the TOC model as well as the porosity model. 103
Fig. 5.7: A multi-well crossplot with 1946 bulk density and neutron porosity data points
from 5 wells (102/14-31-41-06W5, purple; 100/15-36-40-6W5, pink; 100/16-28-37-
05W5, brown; 100/14-18-37-7W5, yellow; and 100/10-16-042-06W5, light green)
throughout the Second White Specks and Belle Fourche intervals. The y-intercept of the
regression is an estimate of the matrix density (100% rock volume) of the Second White
Specks and Belle Fourche Formations. 111
Fig. 5.8: Histograms of uncorrected (above) and kerogen-corrected (below) modelled

total density porosities from a selection of wells in the study area within the Second
White Specks and Belle Fourche intervals, presented logarithmically. 113
Fig. 5.9: This is a thorium vs. potassium % chart with data plotted in green from well
100/02-14-38-06W5. This data plots largely in the mixed-layer clay field, providing a
basis for using a mixed-layer clay density for effective porosity. 115
Fig. 5.10: The above graph displays a linear regression of carbonate volume computed
from Jiang et al. (2017) in addition to carbonate XRD data from Furmann et al. 2014,
xii
versus modelled kerogen for wells 100/02-14-038-05W5, 100/07-19-045-06W5/00,

102/08-15-036-05W5/00, and 102/08-36-041-07W5/00. . 117
Fig. 5.11: A screenshot from Surfer of the grid geometry parameters used for mapping in
this study. Using identical grid geometries for each contour map ensured they could be
accurately overlain and compared to each other. 122
Fig. 6.1: Correlation of the allostratigraphic framework used in this thesis (left) to the
framework used by Tyagi et al. (2007), using the K1 disconformity as a datum. Dashed
lines indicate the extension of the framework used in this thesis into Tyagi et al. (2007),
and vice versa. 129
Fig. 6.2: An approximately south-to-north oriented cross section displaying the five
Second White Specks and Belle Fourche cores logged by this author. Core photos are
included for reference to facies. 132
Fig. 6.3: An approximately south-to-north oriented cross section displaying the available
LAS data for cores in the study area. 133
Fig. 6.4: Cartoons summarizing the paleogeography and paleobathymetry of the

Kaskapau delta and its basinal equivalent Second White Specks and Belle Fourche
alloformations – vertical scale is extremely exaggerated. This approximates a shallow
marine muddy ramp setting with a very low gradient. The approximate location of facies
2 through 5 are shown as circled numbers. 143
Fig. 6.5: A composite cross-section, using the “red” bentonite as a datum, of the Lower
Second White Specks and Belle Fourche alloformations across the study area. Net
reservoir intervals are shaded in purple (Belle Fourche) and green (Second White Specks)
150
Fig. 6.6: Reference map for summary allostratigraphic cross-sections, with core locations
for reference. 152
Fig. 6.7: Allostratigraphic cross-section B to B’, which runs north-south. 153
Fig. 6.8: Allostratigraphic cross-section C to C’, which runs north-south. 154
Fig. 6.9: Allostratigraphic cross-section D to D’, which runs north-south. 155
Fig. 6.10: Allostratigraphic cross-section E to E’, which runs east-west. 156
Fig. 6.11: Allostratigraphic cross-section F to F’, which runs east-west. 157
Fig. 6.12: Allostratigraphic cross-section G to G’, which runs east-west. 158
Fig. 6.13: Allostratigraphic cross-section H to H’, which runs east-west. 159
Fig. 7.1: 2D structure map of the top of allomember VII, with the Mesozoic deformation
front for reference. 163
xiii
Fig. 7.2: 3D structure map of the top of allomember VII. 164
Fig. 7.3: Isochore map of allomember BF1. 167
Fig. 7.6: Isochore map of allomember VII. 170
Fig. 7.7: Isolith (isochore thickness overlaid on clay volume) of allomember BF1. 174
Fig. 7.10: Isolith (isochore thickness overlaid on clay volume) of allomember VII. 177
Fig. 8.1: Comparison of corrected GR (green and yellow colour fill, first track on the
left), TOC (blue and red colour fill, second from the right), total porosity (black line with
white fill, far right) and effective porosity (far right, blue fill) for 102/14-31-041-06W5,
102/08-36-041-07W5, and 100/07-19-045-06W5. 182
Fig. 8.2: TOC of allomember BF3. 184
Fig. 8.3: TOC of allomember VII. 185
Fig. 8.4: Total porosity of allomember BF3. 186
Fig. 8.5: Total porosity of allomember VII. 187
Fig. 8.6: Effective porosity of allomember BF3. 189
Fig. 8.7: Effective porosity of allomember VII. 190
Fig. 8.8: Porosity thickness for allomember BF3. 192
Fig. 8.9: Porosity thickness for allomember VII. 193
Fig. 8.10: Brittleness of allomember BF3. 194
Fig. 8.11: Brittleness of allomember VII. 195
Fig. 9.1: Sweet spot map for allomember BF3. 198
Fig. 9.2: Sweet spot map for allomember VII. 199
xiv
List of Tables
Table 3.1: Table of geophysical well log responses to commonly encountered minerals
and compounds in the Lower Colorado Group. .............................................................. 57
Table 5.1: Table of equations and parameters used for petrophysical modelling in this
project. All variables in red font are measurements taken directly from the well logs..... 93
Table 5.2: Table of TOC equations, with remarks related to their physical basis and
limitations. .................................................................................................................... 99
Table 5.3: Table of total and effective porosity equations, with remarks related to their
usefulness for unconventional reservoirs. .....................................................................109
Table 5.4: Table of mineralogical brittleness indices in chronological order, with their
associated innovations and limitations. Compiled from Jarvie et al. (2007), Wang and
Gale (2009), Jin et al. (2014a); Jin et al. (2014b), Katz et al. (2016); Mathia et al. (2016)
and Rybacki et al. (2016). ............................................................................................119
Table 6.1: The facies codes used in this thesis, along with their descriptions and
associated depositional environments. Facies colours reflect the colours used in later
cross-sections to indicate those facies. Vertical shifts in facies indicate lateral shift in
depositional environment, which are interpreted to be driven by changes in relative
advection energy. .........................................................................................................134
Table 6.2: Petrophysical tops guidance for the Second White Specks and Belle Fourche
alloformations in the Willesden Green – Gilby area of west-central Alberta. This includes
distinguishing log characteristics in addition to problems the interpreter may encounter
while picking tops. .......................................................................................................149
xv
Variables and Abbreviations

dA = rate of change in accommodation
dS = rate of change in sediment supply
g = gamma ray intensity (API)
gAPI = sum of gamma ray energies for spectral gamma ray (API)
gB = gamma ray intensity of an organic free shale (API)
fcl = clay porosity (v/v)
fD = density porosity (v/v)
fDE = effective density porosity (v/v)
fE = effective porosity (v/v)
fS = sonic porosity (v/v)
fT = total porosity (v/v)
f = porosity (v/v)
rb = bulk density (g/cm3)
rdryclay = dry clay density (g/cm3)
re = electron density (g/cm3)
rfl = fluid density (g/cm3)
rker = kerogen density (g/cm3)
rma = matrix density (g/cm3)
rwetclay = wet clay density (g/cm3)
%RO = vitrinite reflectance
∆logR = Passey method for calculating total organic carbon (wt %)
∆"baseline = interval transit time baseline used for Passey method (µs/m)
∆"fl = interval transit time through fluid (µs/m)
∆"ma = interval transit time through matrix (µs/m)
2WS (Second White Specks)
a = constant related to sandstone type (unitless)
A = locally determined calibration component (unitless)
API (American Petroleum Institute units)
xvi
Bbl (barrels of oil equivalent)

BI = brittleness index (v/v or wt %)
BIA (Bow Island Arch)
BF (Belle Fourche)
BFSM (Base of Fish Scales marker)
BOE (barrels of oil equivalent)
C or Cp = compaction component (unitless)
CEC (cation exchange capacity)
CGR = uranium-free computed gamma ray (API)
CGRlog = uranium-free gamma ray value in API from log at a given depth
CGRmax = uranium-free computed gamma ray value in API or 95th percentile “shale”
cutoff
CGRmin = uranium-free computed gamma ray value in API for 5th percentile “sand” cutoff
DSSI (dipole sonic shear imager)
DT (sonic transit time)
FSU (Fish Scales Upper)
Ga (billion years ago)
GR (gamma ray)
GRlog = gamma ray value in API from log at a given depth
GRmax = gamma ray value in API for 95th percentile “shale” cutoff
GRmin = gamma ray value in API for 5th percentile “sand” cutoff
HCS (hummocky cross-stratification)
K = kerogen conversion factor (unitless)
keV (kilo electronvolts)
LAS (Log Ascii Standard)
LIP (Large Igneous Province)
LLD (laterolog)
LOM (level of organic maturity)
m = cementation exponent (unitless)
Ma (million years ago)
MeV (mega electronvolts)
xvii
mD (millidarcies)
n = saturation exponent (unitless)
nD (nanodarcies)
NDIR (nondispersive infrared)
NEU = neutron porosity log reading (v/v)
∆" = interval transit time (µs/m)
OAE (Ocean Anoxic Event)
OOIP (original oil in place)
PDMSL (present-day mean sea level)
PEF (photoelectric effect)
PSF (polished slip faces)
Rbaseline = baseline resistivity used for Passey method (ohm.m)
RT = deep resistivity of partially brine saturated formation (ohm.m)
RW = resistivity of saturating brine (ohm.m)
RR (rig-release date)
SGA (Sweetgrass Arch)
SGR (spectral gamma ray)
SO = oil saturation (v/v)
SW = water saturation (v/v)
SWB (storm wave base)
T = tortuosity (unitless)
TOC = total organic carbon (wt %)
TVD (true vertical depth, in metres)
v/v (volume %)
Vbulk = bulk volume (v/v)
Vcal = calcite volume (v/v)
Vcarb = carbonate volume (v/v)
Vcl = clay volume (v/v)
Vdol = dolomite volume (v/v)
Vker = kerogen volume (v/v)
Vmatrix = matrix volume (v/v)
xviii
Vqtz = quartz volume (v/v)

VQFM = quartz-feldspar-mica volume (v/v)
Vsh = shale volume (v/v)
WCSB (Western Canada Sedimentary Basin)
WEGSF (wave-enhanced sedimentary gravity flow)
wker = kerogen wt %
WQFM = quartz-feldspar-mica wt %
wt% (weight %)
XRD (x-ray diffraction)
XRF (x-ray fluorescence)
Z:A = ratio of average atomic number to atomic weight
xix
Solving the Second White Specks 1. Introduction
Chapter 1
1 Introduction
Summary
This thesis is about the petrophysical analysis of an unconventional oil reservoir. The
study developed an innovative method for identifying “sweet spots” in a fracture-
controlled unconventional reservoir with limited available core and public geophysical
wireline log data. The method involves overlaying maps of petrophysically-derived
properties related to effective porosity and mineralogical brittleness. The stratigraphic
zones represented by each map are defined using high-resolution allostratigraphic
correlations. Using this workflow may increase the likelihood of drilling successful wells
in similar exploration scenarios.
Problem
The calcareous, organic-rich mudstones of the lower Cretaceous Colorado Group
represent a major unexploited hydrocarbon resource in the Western Canada Sedimentary
Basin (WCSB). In particular, the Second White Specks Formation in Alberta has
produced economic volumes of light oil from certain wells (>100,000 barrels of oil
equivalent or BOE; exceptionally >1,000,000 BOE), and has an estimated OOIP (original
oil in place) of 450 billion barrels (Osadetz et al. 2010). The Second White Specks
Formation is a demonstrably effective source rock – a Colorado Group geochemical
fingerprint can be recognized clearly in all oils found in Viking, Belly River, and
Cardium reservoirs in Alberta (Creaney et al. 1994).
Generating consistent and repeatable results from Second White Specks Formation-
targeted wells has proven difficult, despite containing significant reserves (Canadian
Discovery 2011). Due to the extremely low permeability of the Second White Specks
Formation (<1 mD), fractures are required for oil to flow through the rock matrix.
Historically, successful wells targeted small thrust structures where the Second White
Specks was naturally fractured – unfortunately, those pools are areally constrained and
most have been exploited already (Clarkson & Pedersen 2011). Second White Specks
1
Formation wells have also been consistently plagued by unreliable production results and
severe wellbore instability. The non-brittle and clay-rich nature of the Second White
Specks makes it very difficult to yield commercial production rates through hydraulic
fracture stimulation methods. These issues have prevented widespread pursuit of the
Second White Specks as a viable resource play in Canada (Letourneau et al. 1996;
Clarkson & Pedersen 2011; MacKay 2014; Fox & Garcia 2015).
As an example, two seemingly identical adjacent wells, drilled by the same operator days
apart (Fig. 1.1), resulted in a success (102/014-31-041-06W5) and an abject failure
(102/08-36-041-07W5). Can we reliably predict what geographic areas and
stratigraphic zones are more likely to produce consistently well from the Second
White Specks Formation, and why? This thesis builds on current knowledge of the
stratigraphic architecture of the Second White Specks Formation in west-central Alberta
to answer this research question (Tyagi et al. 2007; Varban & Plint 2008a; Tyagi 2009;
Plint et al. 2012b; Zajac 2016). Most significantly, this thesis expands on new
understandings of two factors governing oil production in west-central Alberta:
1) brittle behaviour (Wang & Gale 2009; Clarkson & Pedersen 2011; Cho & Perez 2014;
Furmann et al. 2016; Mathia et al. 2016), and;
2) enhanced porosity (Schieber 2010; Jiang & Cheadle 2013; Furmann et al. 2016).
1.2.1 Influence of brittle behaviour

Historically, unpredictable inflow performance from the Second White Specks Formation
has been tied solely to natural fractures in stiff, brittle rocks. Open fractures can locally
improve permeability (Bloch et al. 1999; Gale et al. 2007; MacKay 2014). Brittle
behaviour and fracability is enhanced in mudstone reservoirs with significant silica,
feldspar, or carbonate content, because calcareous and siliceous minerals (except for clay
minerals) impart their brittle characteristics to their host rocks (Cho & Perez 2014). The
Second White Specks Formation has a significant carbonate lithological component
derived from dispersed coccoliths, coccolith-rich fecal pellets, as well as other bioclastic
grains and carbonate cements (Bloch et al. 1999). The location of natural fractures in this
interval may be tied to increased brittle mineral content, as well as structural elements
2
Fig. 1.1: Two adjacent (~1 km well spacing) Second White Specks-targeted wells, drilled by the same operator days apart, in the Willesden Green field of
Alberta. 102/14-31 produced economic volumes of light oil (122,551 Bbl), whereas 102/8-36 was unsuccessful.
Petrophysically, these wells appear to be identical (the logs are, from left to right: caliper (mm), gamma ray (API), sonic (us/m), and resistivity (ohm.m). The
102/014-31 well (right, starred) was drilled and perforated first (indicated by vertical blue dots), and successfully produced from the Second White Specks.
The 102/08-36 well, drilled 10 days later, was perforated in the same interval and produced insignificant oil volumes. Tops are from Zajac, 2016.
3
(Letourneau et al. 1996; Clarkson & Pedersen 2011).
Previous studies have not attempted to continuously model variations in petrophysical

attributes (e.g., brittleness or porosity) throughout the Second White Specks Formation.
Instead, the focus has been on historical production results (Clarkson & Pedersen 2011),
or indirect reservoir quality indicators like resistivity maps (Zajac 2016). Other studies
have relied solely on individual core analysis points to calculate brittleness and porosity
(Furmann et al. 2014; Furmann et al. 2016) – this means that vertical and spatial
variations in reservoir quality were not adequately captured. Instead, a continuous model
of brittleness for the entirety of the Second White Specks Formation can be derived from
wireline logs if calibration to core is adequately established (Sondergeld et al. 2010).
Accurate models of mineralogy and brittleness may be the keys to success in this play.
Clarkson and Pedersen (2011) recognized that advances in horizontal drilling, hydraulic
multi-stage hydraulic fracture stimulation, and improvements to brittleness models have
facilitated production of unexpelled hydrocarbons from other fracture-influenced
unconventional reservoirs (e.g., the Duvernay Formation in Alberta). The economic
success of the Duvernay has helped fund the development of a comprehensive database of
lithological and petrophysical information, including cores. The mineralogy and
brittleness of the Duvernay Formation, therefore, are well established. A long history of
inconsistent production and geomechanical drilling problems has discouraged core
recovery from the Second White Specks Formation (Fox & Garcia 2015), making
calibration of any petrophysical model dependent on indirect measures of lithology based
on well logs.
1.2.2 Influence of porosity

Changes in primary porosity may be an important factor in successful Second White
Specks Formation wells. New revelations from mudstone microstructure studies have
demonstrated that porous depositional fabrics can be preserved even after deep burial
(Schieber 2010; Jiang & Cheadle 2013). Because depositional modes control sediment
fabric variation across a basin, a representative paleogeographic interpretation is critical
to understand reservoir quality variations in the Second White Specks petroleum system.
4
To date, petrophysically-derived reservoir quality indicators have not been correlated to

inflow performance from Second White Specks Formation-targeted wells, nor has such a
connection been sufficiently tested. Establishing the relationship would require robust
knowledge of lithological heterogeneity in three dimensions, but this is limited by two
key issues with most Second White Specks datasets:
1) Data scarcity caused by a lack of calibration data (i.e., core data, including
mineralogical analyses and porosity), and;
2) Widespread data quality problems like borehole environmental effects (e.g., hole
washout, thin-bed effects, organic matter effects on gamma ray measurements).
Recent studies have failed to address these challenges, limiting accurate characterization
of lithological and petrophysical heterogeneity. Conventional core analyses are especially
rare in this interval because it has not been a primary exploration target. This limits
petrophysical analyses of the Second White Specks Formation because of the paucity of
framework and clay mineralogical data, as well as porosity measurements – these are
required for calibration. Despite the aforementioned data issues, heterogeneity can be
resolved if adequate environmental corrections are applied. Environmental effects (e.g.,
borehole washout and high organic content) can be mitigated via robust data conditioning
of wireline log measurements, thereby improving the accuracy of acquired data from
affected intervals. Borehole corrections like these can help elucidate ambiguous
lithologies in logged zones where core data is limited or nonexistent, as is the case for the
Second White Specks Formation. In combination with existing core data, properly
conditioned borehole measurements may significantly reduce the uncertainty surrounding
brittleness and porosity variations in the Second White Specks Formation.
1.2.3 Stratal architecture of the Second White Specks

If increased brittleness and enhanced porosity do correlate with production from Second
White Specks wells, completion method notwithstanding, it is essential to determine if
and how brittleness and porosity vary across the study area. Once brittleness and porosity
parameters have been modelled, they must be described in an allostratigraphic framework
in order to analyze hydraulic flow units in their depositional context. Lithostratigraphic
methods are used to subdivide rock units on the basis of lithic characteristics;
5
allostratigraphic methods, in comparison, are used to subdivide rock units via correlation
of bounding surfaces with temporal significance (NACSN 1983, 2005). Prior to the work
of Tyagi et al. (2007), the Second White Specks was only described within a
lithostratigraphic framework (Stott 1963; Wall & Rosene 1977; Leckie et al. 1994; Bloch
et al. 1999). These studies failed to resolve the vertical and lateral distribution of coeval
strata (allomembers), bounded by time-stratigraphic discontinuities, that constitute the
high-resolution record of Second White Specks time. Reconciling the strata with the
absolute geologic time scale makes it possible to relate the Second White Specks
allomembers to higher-order allogenic drivers such as the tectonic and eustatic history of
the WCSB.
Aims & Objectives
1.3.1 Research aim

The goal of this thesis is to establish the relative contributions of mappable petrophysical
properties (i.e., brittleness and porosity) to fluid inflow performance in the Second White
Specks Formation within a 110-township area, using pre-existing wireline log data and
core data collected by the author. It is hypothesized that areas with higher relative
brittleness (increased silica and carbonate content) and increased porosity will co-
locate with improved oil production from the Second White Specks Formation.
1.3.2 Objectives
This research seeks to answer three key questions related to the research aim.
1) Can the Second White Specks and Belle Fourche Formations in the study area be
subdivided into allomembers that provide a framework for mapping coeval
geologic units, which may act as hydraulic flow units? This question will be
answered through the generation of three key outputs:
• High-resolution (<20 m unit thickness) allostratigraphic cross-sections to
establish reservoir geometry in the Willesden Green and Gilby areas;
• Structure, isochore, and petrophysical property maps for mapped
allomembers.
6
2) Can the petrophysical properties (e.g., porosity, organic content, clay content,
and brittleness) of the Second White Specks and Belle Fourche Formations be
reliably modelled, using limited geophysical well data and sparse core control for
calibration? This question will be answered through the generation of two key
outputs:
• A petrophysical tops guidance for the Second White Specks and Belle Fourche
Formations in the study area, including;
• A list of generated tops (the elevations of stratigraphic horizons)
• A workflow for consistent top selection for future workers
• Correlation of sedimentary facies in core with petrophysical facies
from logs
• Continuous, static petrophysical models for porosity, organic content,
mineralogy, and brittleness, with average values for modelled petrophysical
properties, separated by allomember, for each well.
3) Is there a spatial association (co-location) between reservoir quality indicators
(e.g., brittleness and porosity thickness) and oil production? This question will be
answered by building a series of “sweet spot” maps that compare reservoir quality
indicators (e.g., brittleness and porosity-thickness) and oil production.
By answering these three questions, recommendations will be made for future

hydrocarbon exploration in the study area.
Scope
The study area is constrained to a 110-township region (~11,000 km2) in west-central
Alberta extending from Township 35 to 45, and Range 1W5 to 10W5. This region of
Alberta contains the Willesden Green and Gilby areas, as well as the town of Rocky
Mountain House. In this area, over 2200 wells penetrate the Base of Fish Scales;
however, only 118 wells were completed in the Second White Specks zone, of which only
49 produced any significant hydrocarbon volumes (Fig. 1.2). Because these wells appear
to be confined to specific geographic areas, stratigraphic correlation on a development
well program basis (³ one well per section) was focused on the surrounding area.
7
Data for this study area within the zone of interest (the Second White Specks) is limited,
which reduces the number of available wells for quantitative petrophysical analysis.
Limiting factors include:
• Seven cores available in the zone of interest, of which only three have any core
analysis publicly available;
• Limited digital log data (LAS) available; only 227 out of 444 wells used for
stratigraphic correlation had LAS data (raster logs cannot be used to create static
petrophysical models);
• No dipole sonic (DSSI) logs available (these are normally used for Young’s modulus
and Poisson’s ratio calculations, which are quantitative measures of brittleness);
• Not enough photoelectric effect (PEF) and spectral gamma ray (SGR) logs available
for field-wide quantitative lithology and mineralogy calculations.
Allostratigraphic mapping was limited to horizons overlying and underlying the perceived
“pay zone” within the Second White Specks and Belle Fourche Formations in the
Willesden Green – Gilby area. Because the goal of this thesis is to tie allomembers into
petrophysical property models, exhaustive high-resolution allostratigraphic mapping of
the entirety of the Lower Colorado Group is out of scope for this project.
Significance
Due to its anisotropic nature and fracture dependency, the Second White Specks tight oil
play lends itself well to an integrated petrophysical and stratigraphic study. This study
provides a predictive framework for reliable economic exploration, development,
completion and production of the Second White Specks Formation by revealing how
brittleness and porosity vary in the Willesden Green – Gilby areas. The framework
developed here can be extended to other areas where the Second White Specks Formation
may be economic, as well as to analogous heterolithic mudstone reservoir units in the
Western Canada Sedimentary Basin.
8
Fig. 1.2: Base map of the Willesden Green – Gilby study area in Alberta, illustrating the spread of
available data across the entire area. The scale bar is in metres (m).
Although ~2200 wells penetrate the Second White Specks in this region, only about 10% (227
wells, indicated in black) had LAS data available in the interval of interest.
9
Overview of thesis structure

This thesis consists of ten further chapters, within four main parts. In PART I (Chapters 1
through 3), the research problem is introduced, and the study is situated in relation to its
associated literature. Chapter 1 deals with the introduction and structure of the thesis.
Chapter 2, the first part of the literature review, discusses the stratigraphic context of the
Second White Specks Formation. This guides high-resolution allostratigraphy, which
provides zone boundaries for petrophysical analysis. Chapter 3, the second half of the
literature review, investigates best practices in petrophysical analysis of unconventional
reservoirs using conventional well logs.
In PART II (Chapters 4 & 5) the research methodology is established. Chapter 4 reviews

the allostratigraphic method, which is used to establish the framework for petrophysical
modelling. Chapter 5 deals with the methodological issues and research design of this
study, providing the procedural description and theoretical basis for the petrophysical
approaches used to collect, present, and analyze data.
In PART III (Chapters 6 through 9), the results of data analysis of the Second White
Specks Formation in the study area from allostratigraphic correlation, isopach and isolith
mapping, petrophysical property mapping, and sweet spot maps, respectively, are
presented.
PART IV (Chapters 10 & 11) contains the discussion, limitations of, conclusions, and
recommendations for future work related to this study. Based on this work, inferences and
recommendations are drawn to inform and improve the current practice of oil exploration
and potential development of Second White Specks-targeted wells.
10
Solving the Second White Specks 2. Stratigraphy, setting, and petroleum geology
Chapter 2
2 Stratigraphy, sedimentology, and play analysis of the
Second White Specks and Belle Fourche Formations
Overview
In this chapter, the Second White Specks and Belle Fourche Formations are placed in
their proper geologic, spatial, and temporal context in order to guide the high-resolution
allostratigraphy used later (Chapters 4 and 6), and to pose key research questions. This
chapter identifies several pressing gaps in the literature through a critical review of
previous studies of the Second White Specks and Belle Fourche Formations with regards
to their tectonic setting, environment of deposition, age, stratigraphic position, bounding
discontinuities, lithology, and petroleum system elements.
Geologic setting of Lower Colorado Group Shales

Thorough studies of conventional sandstone reservoirs within the Upper Cretaceous
Colorado Group of Alberta (e.g., the Viking and Cardium Formations) have revealed
much about their reservoir properties and sedimentary architecture; these reservoirs
contain 14% of the total hydrocarbon reserves in Western Canada (Leckie et al. 1994). In
comparison, far less is known about lower Colorado Group mudstones (Bloch et al. 1999;
Rokosh et al. 2009) – this includes the Belle Fourche and Second White Specks
Formations.
The Fish Scales, Belle Fourche, and Second White Specks Formations (lower Colorado
Group mudstones; from oldest to youngest, respectively) are the focus of this study. They
were deposited over a period of ~7.0 Ma (92.1 to 99 Ma) from the early Cenomanian to
the middle Turonian (Schröder-Adams et al. 1996). These mudstones are thickest in the
Alberta Deep Basin, where they can exceed 600 m thick (Fig. 2.1). Lower Colorado
Group mudstones appear to be largely unaffected by the subsiding Williston Basin to the
southeast, unlike the rocks that comprise the upper Colorado Group (Leckie et al. 1994;
Rokosh et al. 2009). Lower Colorado Group mudstones gradually thin eastward to about
120 m in west-central Alberta, where the study area is located, to < 50 m at the eastern
11
Fig. 2.1: A composite map of the Second White Specks lithostratigraphic interval in southern Alberta.
The study area is situated around a cluster of localized Second White Specks oil pools in the Willesden
Green field. Modified after Creaney and Allan (1990); Letourneau et al. (1996); Rokosh et al. (2009) and
Canadian Discovery (2012).
12
margin of Alberta (Leckie et al., 1994, Fig. 2.1).
2.2.1 Tectonic setting

The WCSB developed into a retroarc foreland basin (Fig. 2.2) during the Jurassic to Early
Cretaceous Columbian Orogeny, due to collision and accretion of exotic terranes on the
western margin of the North American plate (DeCelles & Giles 1996; Evenchick et al.
2007; Allen & Allen 2013). Accommodation in a retroarc foreland basin is created on the
overriding tectonic plate as a result of lithospheric flexure; this is initiated by tectonic
loading in a collisional setting (DeCelles & Gilles 1996). The Late Cretaceous to
Paleogene Laramide Orogeny created a NNW- to SSE-oriented zone of maximum
subsidence and highest sedimentation rate along the western margin of the WCSB.
Deposition of the lower Colorado Group coincided with regional tectonic downflexing of
the North American craton (Lambeck et al. 1987). An asymmetric foreland basin
architecture is revealed by an Upper Cretaceous sedimentary wedge, 700 m thick in the
west, which thins to < 50 m from SW to NE along an extremely low-gradient foredeep
ramp that onlaps onto the forebulge (Varban & Plint 2005; Varban & Plint 2008b).
The location and geometry of the forebulge in the WCSB during the Late Cenomanian
(Fig. 2.3) and Early Turonian (Fig. 2.4) is controversial (Varban & Plint 2005; Varban &
Plint 2008a; Yang & Miall 2008, 2009; Plint et al. 2012a; Percy & Pedersen 2017). The
axis and location of the forebulge in the WCSB may have moved throughout the Late
Cretaceous (Plint et al. 2012b) –as new terranes progressively accreted onto the western
margin, the thrust load would have shifted (Allen & Allen 2013). Utilizing the
palaeomagnetic observations of McCausland et al. (2006), Plint et al. (2012b) attributed
lower Colorado Group forebulge migration to progressive dextral (south and south-east)
movement of the Stikine and Yukon-Tanana terranes, which was initiated in the mid-
Cretaceous and continued into the Eocene.
The NE-SW oriented Sweetgrass Arch (SGA) and contiguous Bow Island Arch (BIA) in
Alberta and Saskatchewan may have acted as a peripheral forebulge during the Late
Cenomanian due to crustal heterogeneities that made it susceptible to flexure
13
Fig. 2.2: Schematic of a strike-oriented cross-section through a foreland basin. Modified from DeCelles and Giles (1996), and Allen and Allen
(2013).
14
Fig. 2.3: Paleogeographic map of the WCSB during the mid- to Late Cenomanian.
Adapted from Plint et al. (2012b), Schröder-Adams et al. (2012), Jiang and Cheadle (2013), and Blakey
(2017). The study area is shown as a dashed red square; it falls within the foredeep and potentially the
forebulge region of Late Cenomanian (Belle Fourche) strata. The present-day Cordilleran deformation
front is indicated as a dark brown line.
15
Fig. 2.4: Paleogeographic map of the WCSB during the early Turonian.
Adapted from Plint et al. (2012b), Schröder-Adams et al. (2012), Jiang and Cheadle (2013), and Blakey
(2014). The study area is demarcated as a dashed red square; it falls within the foredeep of early Turonian
(Second White Specks) strata. The present-day Cordilleran deformation front is indicated as a dark brown
line.
16
(Lorenz 1982). The SGA and BIA are oriented similarly and are geographically close to
the axis of the Late Cenomanian forebulge proposed by Plint et al. (2012b). These
structural features measurably impact upper Colorado Group strata in Alberta (Nielsen et
al. 2008), but their influence – if any – on lower Colorado Group strata has not been
quantified.
2.2.2 Climate and physiography

The Cretaceous period was globally characterized by an atypically warm, “greenhouse”
climate with ephemeral polar ice caps, high eustasy, and elevated atmospheric CO2
(Hallam 1985; Haq et al. 1987; Wilson & Norris 2001; Haq 2014). Peak tectonoeustatic
conditions allowed the northern Boreal Sea and southern Tethyan Sea to transgress
(flood) across the North American continent and join in the Cenomanian, creating a
shallow, epicontinental sea known as the Western Interior Seaway, or WIS (Kauffman
1969, 1977; Caldwell et al. 1993; Hay et al. 1993; Kauffman & Caldwell 1993;
Kauffman et al. 1993; Leckie et al. 1994; Bloch et al. 1999; Schröder-Adams 2014).
Comingling of warm waters from the Tethyan Sea with the denser mid-latitude waters of
the Boreal Sea in a counter-clockwise gyre, driven by Coriolis forces and wind stresses in
the Late Cenomanian, likely initiated substantial plankton blooms via nutrient upwelling
(Fig. 2.5). This produced an oxygen-poor intermediate zone that fostered organic matter
preservation (Schlanger & Jenkyns 1976; Eriksen & Slingerland 1990; Pedersen &
Calvert 1990; Hay et al. 1993; Arthur & Sageman 2004; Hay & Floegel 2012; Schröder-
Adams 2014). Initiation of global ocean oxidation event OAE-II, an oceanographic crisis
and extinction event with biological and sedimentological consequences (Philip &
Airaud-Crumiere 1991), encouraged ocean upwelling of nutrients and enhanced organic
carbon burial by increasing sea-surface fertility and productivity (Arthur & Sageman
2004). OAE-II peaked near the Cenomanian-Turonian boundary (Arthur et al. 1987).
Lower Colorado Group shales were deposited during the Greenhorn Cycle – this was a
sea-level highstand initiated during the latest Albian and coincident with OAE-II (Arthur
et al. 1987; Kauffman & Caldwell 1993; Schröder-Adams et al. 1996). From the Late
Cenomanian to Middle Turonian, sea-level rose to a maximum (~250 m above present
17
Fig. 2.5: Modified from Blakey (2014). Early

Cenomanian through Early Turonian (~99 Ma to
~93.2 Ma) paleogeography of North America,
with approximate shoreline locations.
A) Early Cenomanian (~99 – 98.5 Ma): During

the Mowry cycle, cool Arctic waters (blue
arrows) incurred into western Canada,
facilitating deposition of the Fish Scales
Formation.
B) Middle Cenomanian (~96 Ma): Eustatic rise
connected the Boreal and Tethys Seas for the
first time during the peak of the Mowry
Cycle. The Dunvegan delta complex
prograded from the northwest as the Mowry
Cycle waned, facilitating mud deposition of
the Belle Fourche Formation (outlined in
grey) in the study area.
C) Early Turonian (~93.2 Ma): At the peak of
the Greenhorn Cycle sea-level reached a
maximum within the WIS, and normal marine
conditions dominated. Mixing of cold Boreal
and warm Tethyan waters (red arrows)
promoted phytoplankton blooms (shown in
green), which helped preserve organic matter
in the Second White Specks Formation.
18
day mean sea-level, PDMSL) just after the Cenomanian-Turonian boundary within the
WIS and the Second White Specks Formation was deposited (Kauffman 1977; Haq
2014). This time of relative sea-level rise was interrupted by brief periods of relative sea-
level fall when the oceanographic connection to the Tethys Ocean was lost (Kauffman
1977; Leckie et al. 1994; Bloch et al. 1999; Schröder-Adams 2014). Following the
termination of OAE-II, global sea-level had fallen by 60 m (~190 m above PDMSL) at
the end of the Turonian (Haq 2014).
Early workers characterized the lower Colorado Group as the record of deposition of fine-
grained sediment in a deep (> 200 m water depth) and quiescent marine basin dominated
by pelagic rainout of “marine snow” (Hay et al. 1993; Schröder-Adams et al. 1996; Bloch
et al. 1999). Stratal patterns of the lower Colorado Group suggest, instead, that sediments
were deposited in an epicontinental, shallow marine muddy ramp setting (< 50 m water
depth) with low relief – even greater than 250 km offshore – via wave-enhanced sediment
gravity flows, or WESGFs (Tyagi et al. 2007; Varban & Plint 2008a; Plint 2014); these
will be discussed further in Chapter 5. This places lower Colorado Group mudstones
within storm wave base for mud, which is less than 70 m (Plint et al. 2012b). The
influence of storms on the Late Cretaceous seabed in Alberta is clearly recorded even the
most distal muddy sediments of the Second White Specks-equivalent Kaskapau
Formation; these rocks contain oscillatory wave and combined-flow ripples, as well as
gutter casts (Varban & Plint 2008a).
High eustasy that led to the flooding of the WIS during the Late Cretaceous has been
linked globally to increases in mid-ocean ridge volcanism, as well as the ascent of a
mantle plume under the Pacific Ocean (Kominz 1984; Harrison 1990; Larsen 1991).
Flare-ups in continental arc magmatism during the Late Cretaceous may have enhanced
the flux of nutrients into the ocean via windblown volcanic ash, further increasing
biological productivity in the WIS and accelerating source rock deposition (Cao et al.
2017; Lee et al. 2018). Arthur et al. (1987) hypothesized that OAE-II may have been
initiated by an as-yet unrecognized volcanogenic event. Although the cause of the
Cenomanian-Turonian boundary event remains unknown, later workers connected OAE-
II to sub-oceanic eruptions that formed Large Igneous Provinces (LIPs) in the Caribbean
19
and Madagascar (Arthur & Sageman 2004; Snow et al. 2005; Kuroda et al. 2007).
Organic carbon burial in lower Colorado Group mudstones was therefore enhanced by
volcanic ash as well as by climatic changes that induced biotic crises and anoxia.
Tyagi et al. (2007) designated a prominent bentonite ash bed, the Bighorn River “red”
bentonite, as the base of the Second White Specks Formation. Prokoph et al. (2013)
determined that the eruption(s) associated with the “red” bentonite and the d13Corg-
excursion associated with OAE-II were approximately coeval – further constraining the
eruption responsible for the “red” bentonite to 94.29 ± 0.13 Ma (Barker et al. 2011). This
intimates that the eruption that initiated OAE-II predated the currently accepted age of the
Cenomanian-Turonian boundary (93.9 Ma) by 260 to 530 Kyr. The timing of OAE-II,
coupled with the age of the bounding discontinuity between the Second White Specks and
Belle Fourche Formations (which approximates the Cenomanian-Turonian boundary),
suggests that OAE-II and the eruption responsible for the “red” bentonite could have
triggered physicochemical changes in the basin (e.g., climatic, ocean chemistry,
biological activity, relative sea-level).
Establishing nomenclature for Upper Cretaceous lower

Colorado Group rocks in west-central Alberta
Within industry circles, the Belle Fourche and Second White Specks Formations are
colloquially and collectively referred to as the “Second White Speckled Shale”, or simply
the “Second White Specks” for convenience. It was first identified by Fraser et al. (1935)
in the subsurface during a survey of southern Saskatchewan, where they described a dark,
white-speckled (coccolith-rich) shale that could be traced across the Western Canada
Sedimentary Basin (WCSB) into Alberta. Following this, Stott (1963) formalized a
stratigraphic framework across the central Alberta plains that included the Second White
Specks Formation.
Lower Colorado Group stratigraphic nomenclature varies widely, both regionally and
between authors. This can be problematic to navigate when conducting larger-scale
stratigraphic studies. The discrepancy between stratigraphic subdivisions is rooted in the
application of different methods (i.e., lithostratigraphic, allostratigraphic, biostratigraphic,
20
chemostratigraphic) in independent studies, as well as the periodic revision of

terminology over time. The highly diachronous and interfingering relationship between
lower Colorado Group mudstones has complicated efforts to regionally correlate coeval
strata of the lower Colorado Group (Singh 1983).
At least eight distinct sets of lithostratigraphic terminology exist for lower Colorado
Group rocks (Caldwell et al. 1978; Stott 1982; Glass 1990; Bloch et al. 1993; Leckie et
al. 1994; Bloch et al. 1999; Roca et al. 2008); three of these are relevant in the Willesden
Green – Gilby area. Within industry, lithostratigraphic nomenclature from the Alberta
central plains (e.g., Bloch et al. 1993, 1999; Schröder-Adams et al. 1996) is favoured.
Those authors divided the lower Colorado Group from bottom to top into four
lithostratigraphically-defined shaley units: the Westgate, Fish Scales, Belle Fourche, and
Second White Specks Formations. These units were differentiated using their geophysical
wireline log signatures and bulk geochemistry. Stott (1967, 1984) subdivided the lower
Colorado Group from the bottom up into the Shaftesbury, Fish Scales, Dunvegan, and
Kaskapau Formations; this terminology is used in the northwestern plains and Foothills of
Alberta. Lastly, in the southern Foothills of Alberta the lower Colorado Group is
subdivided into the Sunkay (Westgate, Fish Scales, and Belle Fourche) and Vimy
(Second White Specks) Members of the Blackstone Formation (Simons et al. 2003).
Genetically-related, coeval strata should be tied into a stratigraphic framework that

emphasizes the time-equivalence between rock packages, rather than their lithological
similarities. For this reason, an allostratigraphic approach to correlation is preferred to the
lithostratigraphic method (see Chapter 4). The time-equivalency of the lower Colorado
Group units discussed above has been confirmed by the correlation of foraminiferal
biozones between regions (Schröder-Adams et al. 1996; Schröder-Adams 2014). These
units are constrained further by radiometric dating of bentonite marker beds present
throughout the Lower Colorado Group (Cadrin 1992; Obradovich 1993; Barker et al.
2011), and regional allostratigraphic correlation (Bhattacharya 1989; Bhattacharya &
Walker 1991; Plint 1996, 2000; Kreitner 2002; Varban 2004; Plint & Kreitner 2005;
Varban & Plint 2005; Kreitner & Plint 2006; Roca 2007; Tyagi et al. 2007; Roca et al.
2008; Varban & Plint 2008b, a; Tyagi 2009; Plint et al. 2012a; Plint 2014). Figure 2.6
21
Fig. 2.6: Comprehensive table of formations for the study area, combined with WIS sea level changes from Schröder-Adams (2014) and Haq (2014).
This study uses the forebulge nomenclature from Tyagi et al. (2007) and Roca et al. (2008), even though the study area lies within the foredeep depozone .
Allostratigraphic terminology revises Bloch et al. (1993)’s placement of the Second White Specks-Belle Fourche boundary from the “forebulge unconformity”
upwards to the Bighorn River “red” bentonite, which approximates the Cenomanian-Turonian boundary; bentonite dates are from Barker et al. (2011).
22
summarizes the relationship between the lithostratigraphic and allostratigraphic

terminology discussed above.
The nomenclature used herein was adapted from the allostratigraphic framework
developed by Tyagi et al. (2007) for the distal foredeep and forebulge regions of the
lower Colorado Group. This study substitutes some names for allostratigraphic units with
the names of more familiar lithostratigraphic equivalents (i.e., substituting Second White
Specks alloformation for Kaskapau Formation). Altering the nomenclature in this way
should ensure that allostratigraphic units are assigned to consistent stratigraphic positions
between individual studies, and also preserves familiar industry terminology. This study
focuses on the lower Colorado Group only, as the lower Colorado Group contains the
main reservoir interval within the Willesden Green and Gilby areas.
2.3.1 Fish Scales Formation

The Fish Scales Formation is a distinct, mappable unit observable in outcrop and wireline
logs that is frequently used as a stratigraphic datum across the WCSB (Bhattacharya &
Walker 1991; Leckie et al. 1994). The Fish Scales Formation varies in thickness across
the Canadian Prairies (Bloch et al. 1999; Schröder-Adams et al. 1999). It is typically 1.5
to 21 m thick, but in southwestern Alberta it may be eroded entirely by the overlying
Belle Fourche (Leckie et al. 1994; Bloch et al. 1999).
The Fish Scales Formation is a siliceous, coarse-grained phosphatic unit that contains
bentonites < 1 to 30 cm thick, abundant fish skeletal remains, as well as phosphate and
chert nodules (Leckie et al. 1994; Schröder-Adams et al. 1996; Bloch et al. 1999;
Schröder-Adams et al. 1999; Schröder-Adams et al. 2001). Elevated radioactivity is
observed in this zone due to an increased abundance of fish debris relative to vertically
adjacent formations (Bloch et al. 1993; Roca et al. 2008). Although algal cysts are
abundant in this formation, it is otherwise barren of foraminifera – the Fish Scale
Formation has a reduced dinoflagellate assemblage relative to the rest of the lower
Colorado Group. It contains a mixture of Type II and III organic matter, and has total
organic carbon (TOC) up to 8 wt%, averaging 3.2% regionally (Schröder-Adams et al.
1996; Bloch et al. 1999).
23
Radiometric dating of bentonite horizons within the Fish Scales Formation (Obradovich
1991; Cadrin 1992; Schröder-Adams et al. 1996), as well as studies of its dinoflagellate
assemblage and relative stratigraphic position (Singh 1983; Schröder-Adams et al. 1996;
Bloch et al. 1999) firmly establish the formation in the Early Cenomanian (97.2 to 95.8
Ma; Schröder-Adams et al. 1996). Deposition of this unit corresponds with the basin-
wide disconformity and global oxidation event OAE-1d (Schröder-Adams 2014).
The base of the Fish Scales Formation (the “Base of Fish Scales Marker” or BFSM in
industry) is a diachronous surface that approximates the Albian-Cenomanian boundary in
Alberta (Stelck 1962; Stott 1982; Leckie et al. 2000; Ridgley et al. 2001; Schröder-
Adams et al. 2012), which is currently accepted to be 100.5 Ma (Cohen et al. 2013). Roca
et al. (2008) noted that the positive radioactive excursion demarcating the BFSM
becomes progressively diffuse towards northwestern Alberta, suggesting that the BFSM
in that area is a non-erosive condensed surface and disconformity. This could indicate the
onset of anoxia at the Albian-Cenomanian boundary (Schröder-Adams et al. 1996). South
and east of Roca et al. (2008)’s study area, in west-central Alberta, the BFSM is a
composite erosional and geochemical unconformity mantled by a thin bed of chert
pebbles (Tyagi et al. 2007); this may indicate the location of the Late Cenomanian
forebulge (Roca et al. 2008).
Although the top of the Fish Scales Formation (“Fish Scale Upper” or FSU) is well-
acknowledged as the base of the Belle Fourche Formation and the time-equivalent
Dunvegan Formation (Bhattacharya & Walker 1991), its chronostratigraphic significance
has been amended. Once thought to conformably underlie the Belle Fourche Formation
(Bloch et al. 1993; Schröder-Adams et al. 1996; Bloch et al. 1999; Schröder-Adams et al.
1999), the FSU is now recognized as a regional non-erosional, condensed downlap
surface for successive allomembers of the Lower Belle Fourche equivalent Dunvegan
Formation (Bhattacharya & Walker 1991; Plint 2000; Tyagi et al. 2007; Roca et al.
2008).
24
2.3.2 Belle Fourche Formation

The Belle Fourche Formation is a coarsening-upward, noncalcareous to slightly
calcareous mudstone and siltstone that varies in thickness from 20 m in eastern
Saskatchewan to approximately 150 m in the Foothills of Alberta (Bloch et al. 1999;
Schröder-Adams et al. 1999; Yang & Miall 2009). It is bounded above and below by
unconformities: it unconformably overlies the Fish Scales Formation and unconformably
underlies the Second White Specks Formation (Tyagi et al. 2007).
Although Planolites and Chondrites traces (horizontal and vertical burrows) are
ubiquitous throughout the Belle Fourche regionally, Gordia, Helminthopsis, and
Teichichnus traces (curved burrows) are only found in the western Interior Plains and
coincide with sideritic concretionary horizons (Bloch et al. 1999). The Fish Scales and
Belle Fourche Formations can be distinguished by a slight increase in both calcareous
sediment and bioturbation (Schröder-Adams et al. 1999), as well as by the presence of
inoceramids (Bloch et al. 1999). In some places, however, no compositional or textural
changes are observable over this zone (Bloch et al. 1999). TOC in the Belle Fourche
Formation is 1.7 wt% on average, and comprises a mixture of Type II and III organic
matter (Schröder-Adams et al. 1996).
The Belle Fourche Formation is Middle to Upper Cenomanian in age and was deposited
from 95.8 to 93.3 Ma (Schröder-Adams et al., 1996). This age is based on the presence of
Verneuilinoides perplexus in the Belle Fourche of Alberta and its equivalent in
Saskatchewan (Schröder-Adams et al., 1996; 1999; Bloch et al., 1999). The Belle
Fourche is subdivided into Upper and Lower portions based on Late and Middle
Cenomanian ages, respectively. These ages are derived from foraminiferal and ammonite
zoning (Ridgley et al. 2001; Tyagi et al. 2007) as well as through allostratigraphic
correlation of a correlative conformity between the AX-3 and X flooding surfaces of
Tyagi et al. (2007). Obradovich (1993) dated the X-bentonite within the Upper Belle
Fourche at 94.93 ± 0.53 Ma (Late Cenomanian); Tyagi et al. (2007) traced this bentonite
in >1000 wells through the Belle Fourche from Tp. 30 to 65, confirming that the X-
bentonite is equivalent to the A bentonite of Gilboy (1988) and lies within unit “B” of
Ridgley et al. (2001).
25
Ridgley et al. (2001) and Gilboy (1988) independently noted differential loss of Upper
Belle Fourche strata beneath the Second White Specks Formation in Montana and
Saskatchewan. Loss of strata and “wedging” appeared to be controlled by major structural
lineaments, in addition to additional minor northwest-southeast-trending and northeast-
southwest-trending lineaments – potentially defining fault-bounded basement blocks.
Differential movement and rotation of these blocks may have influenced tectonic thinning
(syndepositional and post-depositional erosion; potentially ravinement) of the Upper
Belle Fourche in Saskatchewan and Montana, creating a regional erosional unconformity
between the Belle Fourche and Second White Specks (Ridgley et al. 2001; Tyagi 2009;
Hicks 2010). Although the existence of this unconformity was refuted by Yang and Miall
(2008), it has been extensively documented by other geologists (Caldwell et al. 1978;
Bloch et al. 1993; Schröder-Adams et al. 1996; Nielsen et al. 2008; Tyagi 2009).
Reassignment of the base of the Second White Specks from the prominent erosion surface
noted by Bloch et al. (1999) to the stratigraphically higher Bighorn River “red” bentonite
by Tyagi et al. (2007) paints the existence of the unconformity noted by Ridgley et al.
(2001) and Gilboy (1988) in an uncertain light. It is not clear if Unit D of Ridgley et al.
(2001) – their Belle Fourche – would be assigned to the allostratigraphically-defined
Upper Belle Fourche of Yang and Miall (2009) or the Second White Specks of Tyagi et
al. (2007). Regardless, the nature and mechanism of the contact between the Second
White Specks and the Upper Belle Fourche, whether it is unconformable or not, has not
been adequately investigated.
2.3.3 Second White Specks Formation

The Second White Specks Formation, so-called for the visible “white specks”
encountered by drillers within the shale (Leckie et al. 1994), is a calcareous claystone to
siltstone found in eastern to southern Alberta. It grades into a calcareous siltstone in
northwestern Alberta (Bloch et al. 1999). Like the Fish Scales Formation, the Second
White Specks Formation contains horizons of bioclastic conglomerate (Schröder-Adams
et al. 1996). The formation thins to the east, where it varies from 90 m near the Peace
River Arch in northwestern Alberta to 25 m at the Saskatchewan-Manitoba border.
26
The Second White Specks Formation is the only Lower Colorado Group formation that
contains coccolith assemblages (Schröder-Adams et al. 1996). It contains a large number
of calcareous nanofossils that have aggregated into white-coloured, fine to very fine-
grained fecal pellets (the so-called “white specks”) which are dispersed throughout
parallel-laminated and unidirectional rippled beds 0.25 to 2 cm thick (Leckie et al. 1994;
Schröder-Adams et al. 1996). Starved ripples and mm-thick laminae are frequently found;
some upper surfaces are potentially wave-reworked (Schröder-Adams et al., 1996). TOC
in the Second White Specks Formation is 5.1 wt% on average, and the organic matter
found within the formation is largely type II (marine, oil-prone; Bloch et al., 1999).
The Second White Specks Formation was deposited between 93.3 and 91.2 Ma
(Schröder-Adams et al. 1996). Schröder-Adams et al. (1996) recovered one ammonite
specimen of Collignoniceras woollgari at 14-29-11-28W4 in the Second White Specks
Formation, which was assigned a Middle Turonian age. An unconformity is interpreted
between the W. aprica Subzone and the V. perplexus Zone as the Late Cenomanian to
Early Turonian age C. simplex Subzone is missing (McNeil and Caldwell, 1981). A
transitional zone that includes the benthic calcareous taxon Neobulimina albertensis is
also absent (Schröder-Adams et al., 1996).
Defining the stratigraphic base of the Second White Specks Formation has proven
problematic due to the highly diachronous nature of the unit. The base should
approximate the Cenomanian-Turonian boundary, which spans OAE-II. Bloch et al.
(1993) and Bloch et al. (1999) originally defined the base of the formation using the first
appearance of abundant coccoliths in the 6-34-30-8W4 type well. Many subsequent
studies used this definition (Schröder-Adams et al. 1996; Schröder-Adams et al. 1999;
Ridgley & Gilboy 2001; Ridgley et al. 2001; Simons et al. 2003; Fraser 2005). Using
allostratigraphic correlation from outcrop and subsurface data, in combination with
bentonite dating (Obradovich 1993), Tyagi et al. (2007) contended the Bighorn River
“red” bentonite best approximated the Cenomanian-Turonian boundary, and was the most
practical means of tracing it in the subsurface. Furmann et al. (2014) placed the base of
the Second White Specks Formation several metres below the “red” bentonite of Tyagi et
al. (2007) based on mineralogy and organic petrography, but provided no clear
27
stratigraphic reasoning for this change. Zajac (2016) also used the Furmann et al. (2014)
definition, on the basis of changes in depositional patterns from “sheet” (Belle Fourche)
to “wedge” (Second White Specks) geometries; placing the boundary between his
allomembers V and VI. This fails, however, to reconcile with the allostratigraphic
observations made by Tyagi et al. (2007).
Petroleum system analysis of the lower Colorado Group in the

Willesden Green – Gilby area
Although a basin-scale analysis of the entire Lower Colorado Group petroleum system in
the WCSB is well beyond the scope of this thesis, some regional context is necessary to
adequately investigate the geologic factors controlling the location of Second White
Specks oil pools in the Willesden Green study area. The fundamental elements of a
successful petroleum system are an effective source rock, porous and permeable
reservoir, adequate seal rock, traps formed prior to peak generation, migration, and
accumulation of petroleum, and sufficient overburden (Magoon & Dow 1994) – these
elements are known with varying levels of certainty for the Second White Specks and
Belle Fourche in west-central Alberta.
2.4.1 Source rock quality, maturation, and migration

The thick mudstone successions of the Lower Colorado Group are prolific and
demonstrably effective source rocks. The Second White Specks and Belle Fourche
Formations are enriched in marine Type II (oil-prone) organic matter (Bloch et al. 1999)
and contain 2-5 wt% TOC on average, and exceptionally up to 12 wt% (Bloch et al. 1999;
Rokosh et al. 2009). Thermal maturation of all WCSB source rocks and their generated
petroleum resources started in the west, near the Rocky Mountains, and progressed up-dip
towards the east – in some cases, migrated and trapped petroleum travelled up to 600 km
(Higley et al. 2009).
The maturity pattern of the Second White Specks Formation – which increases towards
the orogenic front – can be clearly seen in Figs. 2.1 and 2.7. Figure 2.7 is a 1-dimensional
burial history model illustrating the burial history of the Lower Colorado Group in the
proximal foredeep, north of the study area (100/16-18-52-5W5). The major subsidence
28
Fig. 2.7: 1-dimensional burial history diagram (top) of 16-18-52-5W5, modified from Roberts et al.
(2005). The inset map is the same as Fig. 2.1, with the study area for this thesis outlined in orange. The
16-18 well is indicated on the inset map as a red dot.
Vitrinite reflectance (bottom left, red dots) using Sweeney and Burnham (1990) and temperature from
DSTs (bottom right, green dots) were used by Roberts et al. (2005) to calibrate the burial history model.
29
period in the WCSB occurred during the Laramide Orogeny (Late Cretaceous to
Paleogene); the Second White Specks alloformation reached a maximum burial depth of
2.6 km in the Eocene in that particular well (Roberts et al. 2005); burial depths for the
lower Colorado Group in the Willesden Green area commonly exceed 3 km. Vitrinite
reflectance (%Ro, a measure of thermal maturity) in the Second White Specks
alloformation reached ~0.5% Ro in 16-18-52-5W5, suggesting that the Second White
Specks source rock did not reach sufficient thermal maturity to generate hydrocarbons in
that particular well.
Within the Willesden Green field, thermal maturity is not a limiting factor – those pools
lie within a NW-SE trending oil fairway that extends from the Peace River Arch in
northwestern Alberta to the US border (Letourneau et al. 1996). This is further
constrained eastward by Davies et al. (2013) to a NNE-trending fairway approximated by
R3-4W5. Davies et al. (2013) noted that the majority of Second White Specks cores they
encountered with recorded polished slip faces (PSF – post-depositional shear interfaces
that are approximately parallel to bedding, where frictional heating and shear have altered
kerogen in the host rock to a black, highly polished surface) were within the Tmax oil
window, and that PSF frequency increased westward toward the margin of the Mesozoic
Deformation Belt. PSF occurrence may reflect increased burial depth, increased
horizontal shear, and consequent increased maturation and generation of hydrocarbons.
2.4.2 Storage capacity

Reserves estimates of lower Colorado Group mudstones suggest that large volumes of oil
and gas hosted in low porosity rocks remain unexploited. Osadetz et al. (2010) estimated
the volume of oil generated by Lower Colorado Group source rocks at ~350 billion m3, of
which 95% was expelled (leaving ~17.5 billion m3 unexpelled). They list the following
reserve estimates for potential Lower Colorado Group mudstone-sourced resources:
• Potentially commercial tight rocks (5% < f < 6%) contain ~8 million m3;
• Probably commercial very tight rocks (0.03 < f < 0.04) contain ~680 million m3;
• Currently non-commercial extremely tight rocks (f < 0.02) contain ~176 billion m3.
30
Fig. 2.8: A comparison of different oil and gas reservoirs in terms of reservoir quality; modified
from CSUR (Canadian Society of Unconventional Resources). The Second White Specks in the
Willesden Green area has characteristics of both unconventional and conventional reservoirs and is
characterized as a “hybrid shale”.
31
Lower Colorado Group mudstones fall into the “hybrid shale” category of shale plays
defined by Jarvie (2012a, 2012b) because they contain interbedded organic-rich and
organic-lean intervals (Jiang 2013). Other examples of hybrid shales include the Niobrara
in Colorado, the Eagle Ford in Texas, and the Bakken in Montana (Jarvie 2012b). The
lower Colorado Group falls into the “shale oil” classification of Clarkson and Pedersen
(2011) – it is primarily self-sourced (the source rock is the reservoir) with very low
matrix permeability and high organic content (Fig. 2.8). Hybrid shales can have increased
productivity from conventional lithofacies, due to increased storage capacity from
carbonate diagenesis and primary matrix porosity that can survive deep burial (Schieber
2010; Jarvie 2012a; Jiang & Cheadle 2013).
2.4.3 Flow capacity

Most carbonaceous mudstone reservoirs rely on permeability and porosity created by pre-
existing natural fractures in stiff, brittle rocks (Gale et al. 2007). When placed under
enough stress, rocks that break without deformation exhibit brittle behaviour (Cho &
Perez 2014). Brittle behaviour is enhanced in mudstone reservoirs with increased silica or
carbonate content, because siliceous and calcareous minerals impart their brittle
characteristics to their host rocks (Gale et al. 2014). Successful Second White Specks
vertical wells in the Willesden Green area targeted small, localized pools (Fig. 2.9 and
2.10) where Laramide faulting created sufficient fracture permeability via natural
fractures for conventional oil to flow from otherwise uneconomic siltstones and
mudstones (Podruski et al. 1988; Skuce et al. 1992; Letourneau et al. 1996; Clarkson &
Pedersen 2011).
There has been recent debate over whether commercial production rates via horizontal
drilling could be achieved in areas outside the known fault-controlled plays in Willesden
Green, where brittleness and reservoir quality (porosity and permeability) are sufficiently
high (Clarkson & Pedersen 2011; MacKay 2014). Insofar as brittleness is concerned, no
model for brittleness currently exists for lower Colorado Group mudstones. Porosity has
only been adequately examined on a well-by-well basis (Furmann et al. 2014; Furmann et
al. 2016). Furthermore, new research suggests that enhanced matrix porosity can
positively contribute to production rates, even in fracture-controlled plays
32
Fig. 2.9: IP4 (initial average oil rate for the first 4 months) map for Second White Specks wells in the study
area.
Black dots indicated all the 2WS penetrations, whereas grey dots indicate all the wells drilled in the area.
Yellow polygons delineate known 2WS pool boundaries.
33
Fig. 2.10: Cumulative production map for Second White Specks Formation wells in the study area. Green
indicates oil production, red indicates gas production, yellow indicates condensate production, and blue
indicates water production. Most Second White Specks wells in this area are primarily oil producers and
produce negligible water and condensate, with minor gas production.
34
(Schieber 2010; Clarkson & Pedersen 2011). In order to investigate this, brittleness and
reservoir quality variations across the Willesden Green field need to be understood.
Unlike most hybrid plays, which utilize horizontal drilling and multi-stage hydraulic
stimulation to optimize access to matrix porosity, the Second White Specks has been
primarily developed using vertical wells in the Willesden Green – Gilby area (Clarkson &
Pedersen 2011). Operators have historically treated the Second White Specks as a “bail-
out zone” rather than a drilling target. This is because of drilling problems such as severe
borehole instability, lost circulation, and high-pressure kicks (Letourneau et al. 1996;
Clarkson & Pedersen 2011; Fox & Garcia 2015) – these problems can make horizontal
wells more challenging (i.e., more expensive).
Although there is no historical consensus on which zones should be completed, the

coarser-grained units overlying the “red” bentonite (assigned to the Second White Specks
alloformation) are the most commonly perforated in the study area (Canadian Discovery
2014). Frequently, wells that produce from the Second White Specks Formation are
commingled (coproduced) with other producing zones (e.g., Cardium, Viking) in order to
flow enough liquid hydrocarbons to be economically viable (O’Connell 2003). So far,
horizontal wells drilled in the study area have experienced “less than spectacular”
production histories (Canadian Discovery 2011). This could be due to several factors: the
style of completion and hydraulic fracture treatment selected, completion of non-
productive zones, bypassed pay, or insufficient reservoir quality.
Highlights
• Lower Colorado Group rocks were deposited in a retroarc foreland basin that
developed into a marine muddy ramp environment where water depths likely ranged
from <40 to 70 m. This depositional setting facilitated organic matter deposition and
preservation.
• Nomenclature for lower Colorado Group rocks varies regionally and between authors.
The Fish Scales, Belle Fourche, and Second White Specks Formations each contain
unconformity- or disconformity-bounded coeval units that can be regionally
correlated across the WIS.
35
• Lower Colorado Group rocks have considerable source rock potential. They reached a
sufficient depth of burial in the study area to become thermally mature and produce
liquid hydrocarbons.
• Despite considerable hydrocarbon reserves, it has historically proven difficult to
produce oil from lower Colorado Group reservoirs. Localized natural fracture
networks provide the means for oil to flow out of the reservoir, which has low
permeability and porosity. These fracture networks are likely tied to variations in
brittleness, which are not currently understood.
36
Solving the Second White Specks 3. Using petrophysics to model geological properties
Chapter 3
3 Using wellbore petrophysical data for quantitative
geological interpretation
Overview
The theoretical underpinnings of wireline petrophysical methods are reviewed, including:
the intersection of petrophysics with stratigraphy and exploration geology, the importance
of good petrophysical modelling, and the responses of basic geophysical well logs to
shaley lithologies. This sets the stage for a critical examination of stratigraphic methods
that rely on wireline log data in Chapter 4, and a review of petrophysical modelling
methods for unconventional reservoirs in Chapter 5.
Role of petrophysics in exploration geology

Petrophysical evaluation is an essential part of petroleum exploration. The principal goals
of a petrophysicist are to:
1) ascertain if fluid hydrocarbons are present in the formation of interest, and if so, of
what phase (gas or oil);
2) determine the commercial viability of the reservoir by calculating the pore (e.g.
porosity, permeability) and fluid phase (i.e. saturation) distribution of the reservoir;
3) evaluate the physical makeup of the reservoir (e.g., mineralogy, organic richness,
brittleness), and;
4) develop a representative model of reservoir properties that can be extrapolated into
mappable units across the area of interest.
Fundamentally, petrophysics involves transforming the “wiggle traces” acquired by

geophysical well logging tools into quantitative geological information. Wireline well
logs are graphical portrayals of various recorded drilling conditions or measured
subsurface features that relate to the evaluation of an individual well (Tiab & Donaldson
2012). No remote, cost-effective methods for directly measuring mineralogy and porosity
currently exist, so interpreting this data in a geological context can be challenging.
Instead, petrophysicists principally use indirect measurements (i.e., geophysical well
37
logs) to quantify the composition and reservoir quality of reservoirs rocks in the
subsurface, often without calibration data taken directly from that particular well. This is
effectively geophysical inverse modelling. Sources of data used for petrophysical analysis
and model development consist of (Ellis & Singer 2008):
• Mudlog data, drill stem tests, and drill cuttings (direct) – not included in this study
• Core and sidewall samples (direct) – extremely limited in study area
• Logs acquired via wireline logging tools (indirect) – main source of data
• Borehole seismic and 3D seismic (indirect) – not included in this study
During the exploration phase of a petroleum play, well data is extremely limited, and
knowledge of inter-well variation is even more uncertain. Extensive core extraction and
core analysis is prohibitively expensive and is only performed when deemed absolutely
necessary. Pioneers in the field of petrophysics, a term coined by Archie (1950), have
continually sought to enhance sensor capabilities, increase computing speed, and improve
measurement resolution so as to improve measurement quality with minimal incremental
cost (Bateman 2009); innovation in this field continues to this day. Modern petrophysical
analysis utilizes a strong foundation of electrical engineering combined with rock physics
principles to characterize hydrocarbon reservoirs (Doveton 2014).
Connecting petrophysics with stratigraphy

One major function of the petrophysicist is to assist geologists with extrapolation of data
away from the wellbore, i.e. developing a scaled-up model of subsurface properties across
a field or pool. Such models are crucial to the assessment of storage and flow capacity, as
well as reservoir compartmentalization. However, these models require meaningful
constraints, and geological context, to be representative of stratigraphic architecture.
These constraints are hydraulic flow units, which are defined by Ebanks (1987) as:
“[…] a volume of the total reservoir rock within which

geological and petrophysical properties that affect fluid flow
are internally consistent and predictably different from
properties of other rock volumes (i.e., [other] flow units).
[Hydraulic flow units are] defined by geological properties,
such as texture, mineralogy, sedimentary structures, bedding
contacts, and the nature of permeability barriers, combined
38
with quantitative petrophysical properties, such as porosity,

permeability, capillarity, and fluid saturations.”
Ebanks (1987) opines further that “[…] flow units do not always coincide with geologic
lithofacies”. Lithostratigraphic methods, which do not consider significant time breaks in
the sedimentary section (e.g., unconformities, hiatuses, ravinements, and flooding
surfaces), often cannot provide adequate geological and geometrical constraints for
petrophysical analysis. These breaks in the sedimentary section can be significant
impediments to the continuous flow of hydrocarbons. Flow units, therefore, are best
defined by bounding surfaces such as permeability barriers (i.e., thick mud drapes or
unconformities). Since allomembers (time-stratigraphic units) define genetically-related,
lithologically heterogeneous packages of contemporaneous strata (Bhattacharya &
Posamentier 1994), they can provide more representative bounding surfaces (or zones)
needed for petrophysical modelling within the aforementioned hydraulic flow units.
These bounding surfaces may define barriers to the flow of hydrocarbons within a zone,
depending on the vertical stacking patterns of lithofacies within separate allomembers.
Previous studies of the lower Colorado Group have used a variety of stratigraphic
divisions. Any historically reported petrophysical or geochemical data for this zone needs
to be carefully scrutinized to determine its allostratigraphic position (e.g., Belle Fourche
or Second White Specks Formation). Moreover, the absence of consistent zoning means
that previous petrophysical analyses of the Second White Specks interval have not been
placed into the appropriate flow units.
Overview of well logging

After a vertical well is drilled, and prior to setting casing, the formations of interest are
exposed in the open borehole. At this point, cylindrical geophysical measurement tools
called sondes are lowered into the wellbore to collect in situ measurements of rock
properties. These sondes are connected in series into “tool strings” that can exceed 30 m
in total length, allowing for simultaneous measurement of different properties. An
armored cable called a wireline provides electrical power for the sondes, as well as a
means of transmitting digital data continuously out of the wellbore for geologists to view,
in well log format. The tool string is raised from the bottom of the hole at a constant
39
speed, allowing for the sondes to record data at a constant sampling rate (Glover 2000;
Ellis & Singer 2008; Andersen 2011).
Data transmission and computer processing capabilities for well logging evolved from
entirely analog in the late 1960s to fully digital by the 1990s (Bateman 2009). Prior to
digital logs, if different logging parameters (e.g., a different matrix density for porosity
logs) were desired, the entire well had to be re-logged; this was rarely done (Hill 2012).
Digital library tape formats, developed first by Schlumberger in 1968, allowed for in-
house log processing of individual data tracks post-logging (Hill 2012). Library ASCII
Standard (LAS) was adopted as an industry standard for digital log formats in 1989 and is
still used today (Struyk et al. 1989; Hill 2012). In a LAS file, the data from each sonde is
stored as a separate column called a track (Struyk et al. 1989). This data can be
manipulated by the petrophysicist in a worksheet, or in sophisticated formation evaluation
programs like Schlumberger’s Techlog™ software.
Recent advancement in well logging has been driven by unconventional reservoir

exploration. Reservoirs with “unconventional” components (e.g., clay, kerogen, and
carbonate) are more difficult to characterize with petrophysical methods. Consequently,
more types of well log data must be acquired in order to adequately describe the
formation of interest. Over 50 types of measurement sondes have been developed to
measure a diverse set of subsurface properties and monitor drilling progress (Ellis &
Singer 2008). It is essential, however, to understand and account for the uncertainties and
limitations associated with well log data. Virtually all types of well log data are proxies
for direct measurements of petrophysical properties used in modelling. If their limitations
are not recognized, well log data acquired in unconventional reservoirs can be
misinterpreted (Moore et al. 2011).
3.4.1 Borehole petrophysics

Petrophysical studies are limited by the most prevalent logging suite available. This
logging suite is usually a combination of geophysical well logs that are the most
economical for drilling operators to run in the borehole (Moore et al. 2011). The
Willesden Green field typifies this limitation: due to the vintage of the available log suite
40
and the status of the Second White Specks Formation as a “bail-out zone” in the area,
only common measurements (caliper, gamma ray, spectral gamma ray, bulk density,
compressional sonic, and resistivity) will be discussed further. More sophisticated logs
(DSSI, PEF; pulsed neutron mineralogy; nuclear magnetic resonance) were not widely
available. Section 3.6 will cover open-hole logs from vertical wells in detail.
Petrophysical values (e.g., porosity and clay volume) are imperfect approximations of
reservoir properties. This is because petrophysical calculations are made using a limited
number of wireline log measurements that have their own sources of uncertainty. The
main characteristics that affect log quality are the depth of investigation, vertical
resolution, and logging speed. The following section summarizes the specifications and
limitations of wireline log measurements used in this thesis (Glover 2000).
3.4.2 Drilling mud

Most petroleum reservoirs are located kilometres below the surface and are at high
lithostatic pressures. Drilling the wells with dense drilling mud (1.05 to 1.9 g/cm3)
balances the formation fluid pressure and allows safe access to hydrocarbons without
blowouts (Glover 2000). Drilling mud, or drilling fluid, is a suspension of mud particles
in a water- or oil-based slurry that can contain heavy minerals like barite. When the
pressure exerted by drilling fluid exceeds the lithostatic pressure of a formation and
permeability is sufficiently high, drilling mud infiltrates or “invades” into the formation.
Formation fluids are displaced by mud filtrate (i.e., the fluid component of the drilling
mud), and a crust of “mud cake” (i.e., the solid residue of the drilling mud) forms inside
the borehole (Fig. 3.1). This can cause permeability to be locally reduced, potentially
damaging the reservoir permanently. In highly invaded zones, most logging tools will
respond to mud filtrate in the reservoir, rather than formation fluids (Ellis & Singer 2008).
Because the Second White Specks is not a primary target in the study area, drillers
sometimes “mud up”, or increase drilling mud density, while drilling through the lower
Colorado Group – this conditions the hole for optimal drilling progress and improves the
rate of penetration (i.e., drilling speed). As a result, the Second White Specks formation
may have lower reservoir quality (formation damage) in intervals where significant
41
Fig. 3.1: Schematic of a centered caliper-type tool attached to an instrument assembly, with idealized
responses to beds that could potentially be encountered in the lower Colorado Group.
Modified from Glover (2000) and Krygowski (2004). As the sonde and instrument assembly are slowly
raised through the borehole, the caliper arms move if the hole size changes in any way. Breakouts in brittle
shale are of particular interest from a correlative and geomechanical perspective in this study.
42
invasion has occurred. In contrast, excessive borehole breakout was encountered when
drilling operators used drilling mud with insufficient density (Fox & Garcia 2015) –
evidently, proper mud weight selection is essential to avoid drilling problems in the
Second White Specks.
3.4.3 Caliper log

Caliper logs provide a constant measurement of hole shape and diameter during logging
runs by sensing the movement of measurement pads (calipers) that extend from the sonde
(Fig. 3.1). Caliper logs are indicators of the borehole conditions - if the caliper shows
that the hole is rough, measurements that require direct sensor contact with the borehole
wall will be unreliable. Density, neutron, and resistivity logs are impacted by hole size;
caliper logs offer a way to quantify the magnitude of mud cake accumulation or hole
washout that is occurring (Krygowski 2004).
Petrophysical modelling
A petrophysical model continuously describes the static properties (i.e., mineralogy and
porosity) of a formation, given a set of geophysical well logs and core data for calibration
(Doveton 2014). A good petrophysical model allows the geologist to predict variation in
reservoir properties within hydraulic flow units across large areas. This is essential for
sweet-spotting in unconventional reservoirs – especially when no seismic data is available
to verify field-scale anisotropy, and production data is scarce or nonexistent. Sweet spots
are loosely defined areas in an unconventional reservoir that produce better than others
(Glaser et al. 2013), which may be defined by a series of overlapping reservoir quality
parameters (e.g., porosity, TOC, brittleness). Identification of these regions, a process
termed “sweet spotting”, enables operators to select optimal well locations (Wang & Gale
2009; Glaser et al. 2013; Ter Heege et al. 2015). Ideally, these reservoir quality
parameters are directly correlated to hydrocarbon production data – but these data are
often difficult to obtain in North American basins due to their proprietary nature (Ter
Heege et al. 2015).
A search of publicly available geochemical and geophysical studies failed to reveal static
petrophysical models of the Second White Specks and Belle Fourche Formations at any
43
scale of observation – previously, core analyses were used to describe regional variations
in reservoir quality (Furmann et al. 2014; Zajac 2016). This reveals an opportunity to
develop a petrophysical model for an economically important mudstone reservoir, which
could potentially be used to detect areas with superior reservoir quality.
Methods for quantifying mineralogy from well logs originated in the mid-1970s – they
used log cross-plots such as density-sonic and neutron density to calculate porosity. These
methods could adequately characterize formations with two or fewer mineral components
(Clavier & Rust 1976; Schlumberger 2009). Modern commercial petrophysical modelling
software, developed in the 1980s, uses inverse linear and nonlinear modelling to solve for
mineral composition, porosity, and fluid saturation continuously (Mitchell & Nelson
1988; Doveton 1994b; Ellis & Singer 2008). These software models require a large
training database of core data and can fail in formations that are petrophysically complex
(Heidari 2011). “Petrophysically complex” reservoirs, like the organic-rich mudstones of
the Second White Specks and Belle Fourche Formations in west-central Alberta, have the
following characteristics (Bloch et al. 1999; Ellis & Singer 2008; Clarkson & Pedersen
2011; Heidari 2011):
1) pervasive thin beds;
2) multimodal mineralogy (3 or more components), particularly clay or kerogen; and
3) multimodal porosity, especially low porosity (<3%).
A lack of core data, combined with enhanced petrophysical complexity, means that
applying an inverse model to lower Colorado Group rocks would introduce more
uncertainty into any resulting petrophysical model than it would resolve (Yin 2011). This
is because inverse linear models should be overdetermined (see section 5.6.2). Instead,
lower Colorado Group rocks present an underdetermined problem where unknowns
outnumber the available equations. Alternatively, a deterministic petrophysical model can
be built, wherein each property is independently (rather than simultaneously) derived
from log data and then calibrated to core data (Sondergeld et al. 2010).
A deterministic petrophysical model is restricted by the type, quality, and amount of data
available. Although improvements in logging technology (e.g., lithodensity logging,
formation micro-imaging, nuclear magnetic resonance imaging, and dipole sonic logs)
44
have significantly improved the quality and types of data that can be gathered from a
single wellbore, these types of logs are costly and often inaccessible to the public. Similar
issues arise with core data, which is required for calibration of log data to rock properties.
Cores are only drilled occasionally, so representative vertical sampling of any formation
(particularly mudstones, which were historically not drilling targets) is typically poor.
Furthermore, many studies are limited by the generation of the wells drilled in the area –
advanced logging methods may not have existed when those wells were drilled. Digital
log files (LAS) may not be available for many older wells, which limits the interpreter to
raster log data. As such, despite the existence of sophisticated methods for mudstone
petrophysical analysis, this thesis will only discuss those methods that can be completed
with basic, conventional well logs (resistivity, porosity, gamma ray) and limited core
data.
Creating a good petrophysical model for an unconventional reservoir requires a solid

grounding in formation evaluation methods. The Second White Specks and Belle Fourche
Formations both have very low porosities, and contain clay, kerogen, and carbonate –
thus creating an intriguing petrophysical problem. Basic geophysical logging tools were
designed to measure Archie-type reservoirs that are petrophysically “simple”: these are
rocks with no clay, kerogen, or carbonate content, and unimodal conventional porosities.
Because petrophysical techniques were originally developed to analyze conventional
hydrocarbon targets, it is not appropriate to apply the same methods to petrophysically
complex mudstone reservoirs. Unconventional reservoirs require “unconventional”
methods (Sondergeld et al. 2010; Labani & Rezaee 2015). The methods used for
petrophysical modelling methods in this thesis are discussed and reviewed in Chapter 5.
3.5.1 Modelling Archie-type reservoirs

Wireline logging techniques were developed by the Schlumberger brothers in 1927 for
subsurface stratigraphic correlation and detection of an oil-saturated target zone (Hill
2012; Hall 2017). Archie (1942) was the first to establish an empirical relationship
between conductivity, porosity, and brine saturation in “clean” (clay-free) sandstones,
which is still the basis of most conventional log evaluation today (Doveton 1994a):
45
, & × ("
!" = % +f) × ( [1]
*
where !- is fractional water saturation, n is the saturation exponent, a is a constant

related to sandstone type (Winsauer et al. 1952), RW is the resistivity of the saturating
brine, f is the porosity of the sandstone, m is the cementation exponent, n is the saturation
exponent, and RT is the resistivity of a partially brine-saturated rock. This is known as the
Archie equation. Equation 1 offered the first opportunity for geologists to obtain
quantitative porosity information directly from resistivity measurements (Doveton
1994a). To do this routinely, however, required more computing power than was
available at the time – calculations had to be made by hand, point by point (Hill 2012).
Due to its utility, the Archie equation is a mainstay of conventional petrophysical
modelling, despite its origins as an empirical relationship between rock properties in a
simplistic reservoir model (Fig. 3.2).
The usefulness, physical significance, and accuracy of the Archie equation has been
thoroughly criticized by Luffel and Guidry (1992), Doveton (2001), and Sondergeld et al.
(2010), among others. Critically, the Archie equation does not account for the electrical
conductivity of clay minerals present in unconventional “shaley sand” or clay-rich
reservoirs, causing gross overestimation of !- (and therefore underestimation of oil
saturation, or !. ). Clay minerals affect resistivity readings, but they also affect reservoir
quality by degrading porosity and permeability; this is called the “shaley sand problem”
(Asquith 1990). Other rock properties can cause the Archie (1942) equation to fail – these
include thin beds, heterogeneous mineralogy, anisotropy, heavy minerals, multimodal
porosity, microfractures, variable water salinity, variable wettability, and when m and n
do not equal 2 (Bust et al. 2013). Application of the Archie (1942) equation is
inappropriate in clay-rich reservoirs, and the problems are compounded in organic-rich
mudstones where low-density kerogen complicates porosity determination.
46
Fig. 3.2: Comparison of general petrophysical models for Archie-type, shaley sandstone, and unconventional reservoirs in terms of
complexity. The level of petrophysical complexity increases with the number of components, which makes them more difficult to resolve.
Compiled and adapted from: Passey et al. (1990); Ellis and Singer (2008); Ambrose et al. (2010); Passey et al. (2010); Ramirez et al. (2011);
Bust et al. (2013); Crain and Holgate (2014); Doveton (2014).
47
3.5.2 Modelling shaley sand reservoirs

To remedy the limitations of the Archie (1942) equation in clay-rich reservoirs, two types
of shaley sandstone equations for calculating porosity and water saturation in clay-rich
reservoirs have been developed (Worthington 1985):
1) Resistivity-based !"# or !$% based equations – e.g., Simandoux (1963), where shales
are homogenously conductive media, regardless of clay mineralogy; key inputs are
!"# and shale resistivity.
2) Ionic double-layer or cation exchange capacity (CEC) based equations – e.g.,
Waxman and Smits (1968); Juhasz (1981); Clavier et al. (1984), where electrical
conductivity increases with increased surface area, and therefore altered by clay
mineralogy (high surface area, varies with type); this requires measurement of, or
proxies for, cation exchange capacity.
Both equation types are of the general form:
-
1) = +,
'( .) / × ', + 3 [2]
f
where 3 is the electrical conductivity contribution of the shaley portion of the reservoir.
Equation 2 is a simplified Archie equation with a shale correction added. Although ionic
double-layer water saturation equations better capture the heterogeneity of electrical
conductivity in shaley sand reservoirs, their use is limited and inefficient. This is due to a
necessary calibration to CEC – a parameter that is currently impossible to measure
directly using geophysical well logs (Doveton 2001). Doveton (2014) declared that
“clearly, the ideal shaley sandstone model has not been resolved nor ever will be.”
Despite this, the usefulness of shaley sandstone equations in optimizing water saturation
calculations is undeniable, provided that their limitations are understood.
Nuclear technology, developed in the 1940s for the Manhattan Project, was repurposed
for petrophysical logging in 1950 when a need for reliable wireline porosity measurement
arose (Hill 2012). To that end, modern gamma ray and neutron-density porosity logs
respond to the absorption or attenuation of radiation, respectively (Glover 2000). The
48
gamma ray (GR) log is particularly useful for calculating shale volume (!"# ), because it
primarily detects increases in radioactivity that can be attributed to potassium (40K) in
clay minerals (Asquith 1990). Shale volume using total GR, , !"# | 5', is calculated as
follows (Asquith 1990; Krygowski 2004):
5'%67 − 5'/9:
!"# | 5' = [3]
5'/;< − 5'/9:
where 5'%67 is the GR reading at the interval of interest, 5'/9: is the GR value in a
nearby “clean” or shale-free zone, and 5'/;< is the gamma ray value in an adjacent
shale. It has been long recognized that this linear interpolation can be an overestimate of
!"# because GR tools do not respond linearly to shaley lithologies (Asquith 1990; Ellis &
Singer 2008; Doveton 2014). Clavier et al. (1971) converted !"# to !$% (clay volume)
using the following non-linear correction, which is a pessimistic non-linear corrections of
shale volume:
!$% = 1.7 − ?3.38 − (!"# + 0.7)- [4]
!$% is an important parameter as it allows for the calculation of porosities calculated in

clay-rich intervals. Heslop (1974) first noted the need for a distinction between the oft-
confused shale volume and clay volume. Shale volume refers to the volume proportion of
all fine-grained components regardless of mineralogy, whereas clay volume refers only to
the volume proportion of clay minerals. It is generally accepted that using a non-linear
correction (e.g., Equation 4) converts !"# to !$% ; however, it is always prudent to ground-
truth and calibrate this assumption with clay mineralogy from XRD or XRF (Quirein et
al. 2010; Doveton 2014).
Fig. 3.2 illustrates the differences in petrophysical complexity between different reservoir
types. Compared to conventional Archie-type reservoirs and shaley sand reservoirs,
unconventional reservoirs have more components and complicated pore geometry;
creating an underdetermined problem. Many petrophysical models for unconventional
reservoirs have been developed (Passey et al. 1990; Ambrose et al. 2010; Passey et al.
49
2010; Ramirez et al. 2011; Bust et al. 2013; Crain & Holgate 2014). The development of
shaley sand petrophysics helped pave the way for calculations of the following
parameters using conventional well logs:
1) total organic carbon: in v/v, !EFG ; or, in weight %, TOC;
2) clay volume (!$% );
3) porosity: total (H ( ) and effective (HI ); and
4) matrix volume (!$;GJ + !KLM ).
These factors, ultimately, function as measures of reservoir quality – they can be used to
estimate brittleness, which is the end goal of this thesis.
Interpreting geophysical well log data from unconventional

reservoirs
The following section reviews how geophysical well log data is collected and should be
interpreted in unconventional reservoirs like the Second White Specks Formation, by first
considering expected well log responses. Although the advent of shale gas and tight oil
exploration has greatly progressed the capacity to characterize unconventional reservoirs,
many plays suffer from a scarcity of data. For this reason, the following section will focus
on the basic wireline log suite used in this study.
3.6.1 Petrophysical tops guidance for mudstone reservoirs

In unconventional mudstone reservoirs with high organic content, the following log
responses are expected (Fertl & Chilingar 1988; Passey et al. 1990; Labani & Rezaee
2015; Zee Ma et al. 2016):
• Elevated gamma ray, due to uranium in organic matter and potassium in clays;
• Low density response, as organic matter has very low density (1.1-1.4 g/cm3) relative
to quartzofeldspathic matrix (2.6-2.8 g/cm3) – the density response may be increased
if pyrite (FeS2) is present;
• Slow sonic log response, due to low-density organic matter, gas, and clay-bound
water reducing compressional wave velocities;
50
• High neutron-porosity response in clay-rich zones from to water bound to clays; this
can be suppressed when hydrocarbons are present (hydrogen response will be reduced
by gas and organic matter);
• Increased resistivity, due to presence of non-conductive fluid hydrocarbons and
organic matter in the rock matrix; resistivity is also increased relative to the level of
organic maturity, and conversion of kerogen to hydrocarbons (relative to organic-poor
mudrocks that have decreased resistivity due to clay conductivity).
3.6.2 Well log responses as proxies for mudstone reservoir

composition
Figure 3.3 is a well log through a portion of the Second White Specks Formation,
illustrating the vertical changes in lithology throughout the stratigraphic interval. This
well (7-19-45-6W5) vertically intersected and was perforated in the same productive
interval illustrated in Fig. 1.1 but was ultimately abandoned as the unit produced
negligible hydrocarbons. Relative to a conventional reservoir (low organic content and
clay-free), the wireline log response of the entire stratigraphic interval appears to be
unconventional, with high clay and organic content throughout. From this data alone, it is
difficult to ascertain why the so-called productive interval proved unproductive in this
particular well.
Petrophysical modelling exploits the relationship between well log responses and specific
lithologies, like those listed in section 3.6.1. Petrophysicists attempt to estimate rock
properties indirectly using measurements of different physical properties (wireline logs)
that are proxies for lithology. At each measurement point, logging tools record a signal
from the reservoir derived from mineralogy, pore geometry, fluid composition, and
borehole conditions. A table of expected petrophysical responses for common lithologies
is available from Baker Hughes (https://www.spec2000.net/freepubs/LogResponses.pdf),
although these responses are idealized and represent end member cases. The following
section considers the potential applications and drawbacks of different conventional well
logs to petrophysical modelling and geological interpretation, based on the physics of
their respective measurements (radioactivity, acoustic, and electric). Using this
51
Fig. 3.3: A representative well log (7-19-45-06W5) for the Second White Specks interval.
From left to right, the tracks include caliper (CALI, in mm) shown in red, gamma ray (GR_K) in API,
volume % mineralogy, TOC (in weight percent), bulk density (DEN_K) in kg/m3, density correction
(DENC), density porosity (PHIT_D) which is overlain by core total porosity, as well as medium and deep
resistivity (RES_DEP and RES_MED, both in ohmm).
52
information, subtle changes in wireline log responses – like as those illustrated in Figs.
1.1 and 3.3 – can be better understood in a geologic context (e.g., change in porosity).
3.6.2.1 Radioactivity logs

Radioactive decay, discovered by Becquerel (1896) and Curie et al. (1898), is a stochastic
(random) property of unstable atomic nuclei wherein the nucleus loses energy by emitting
radiation (i.e., alpha [a], beta [b], and gamma [g] rays) at an exponentially decreasing
rate, thus decaying into a more stable daughter nucleus. The emission of a gamma ray –
electromagnetic radiation with shortest wavelengths (10 picometers) and highest energies
(up to 10 MeV) – occurs after a and b decay. Gamma or g decay occurs when the
daughter nucleus is produced in an excited state and then decays to a lower energy state
by emitting gamma radiation. Gamma radiation interacts with matter through several
mechanisms. Three of these mechanisms are important for petrophysical radioactivity
logging applications, in order of increasing energy: photoelectric absorption, Compton
scattering, and pair production (Ellis & Singer 2008). The probability of a particular
photon-matter interaction occurring is dependent on the energy of the incident photons
and the atomic number of the material of interest (Ellis & Singer 2008).
The photoelectric effect is a phenomenon that occurs when a low energy photon (<0.1
MeV) interacts with a material of sufficiently high atomic number. The incident photon is
absorbed and transfers its energy to a bound electron within the material; that electron is
subsequently ejected with an energy proportional to the atomic number of the material
(Einstein 1905; Millikan 1913). Because different rock matrices have different average
atomic numbers (e.g., sandstone, limestone, dolomite), low-energy photons (in this case,
gamma rays) interact predictably with changes in lithology. The photoelectric effect is the
physical basis for PEF measurements (Ellis & Singer 2008), which will not be discussed
further here.
A gamma ray with intermediate energy (0.1 to 1 MeV) experiences Compton scattering
by colliding and subsequently transferring a portion of its energy to an orbital electron
(Compton 1923). Incident gamma rays experience a reduction in energy (attenuation) that
is directly proportional to the electron density of the material that is being traversed;
53
increased density causes more energy loss and scattering (Baker 1957). The energy of the
orbital electron is proportional to the lithology of the rock type the incident gamma ray
passed through. This electron, therefore, produces a measurable signal in petrophysical
gamma ray detectors of all types (Ellis & Singer 2008).
Similar to the photoelectric effect, pair production is a process where a gamma ray is
absorbed. In this circumstance if the gamma ray exceeds a threshold value of 1.022 MeV,
the gamma ray is replaced by an electron-positron (positively charged electron) pair
(Blackett & Occhialini 1933; Hubbell 2006). The positron is subsequently annihilated,
thereby emitting two gamma rays with reduced energies of 511 keV each (Ellis & Singer
2008). This process contributes only marginal energy to the overall signal of radioactivity
tools, so it is only consequential in very dense lithologies that are not likely to be
encountered in sedimentary rocks (Glover 2000).
For typical rock formations (no heavy minerals), the average atomic number ranges from
13 to 20. At photon energies less than 0.5 MeV, photoelectric absorption prevails over
other interactions. Compton scattering predominates gamma ray attenuation over the
energy range from 0.1 MeV to 10 MeV, but most commonly from 0.5 to 3.0 MeV
(Glover 2000). The pair production process only becomes significant for lithologies with
high atomic numbers and gamma rays surpassing a threshold of 1.022 MeV, and most
commonly at energies exceeding 3 MeV (Wilson et al. 1979; Glover 2000).
3.6.2.1.1 Gamma ray logs

Modern total gamma ray (GR) logging tools record the radioactivity of the borehole
environment. GR detectors consist of two essential parts: a detector (commonly a solid-
state sodium iodide [NaI] scintillation crystal) that absorbs incoming gamma rays, and an
amplifying-counting system called a photomultiplier. The incoming gamma rays have
experienced varying degrees of Compton scattering en route to the tool, depending on the
electron density of the formation. When a gamma ray is absorbed by the detector crystal,
the crystal emits a flash of light whose intensity is proportional to the energy of the
absorbed gamma ray. That flash is detected by the photomultiplier, which transforms it
into an electrical pulse that is recorded by the tool. Unlike all other logging tools, the GR
54
tool is passive – it records data, instead of actively emitting radiation to induce a response
measurable by the tool (Adams & Gasparini 1970; Ellis & Singer 2008).
GR curves can be used for correlation, lithological interpretation, and shale volume (!"# )
calculation. For correlation purposes, GR curves are scanned by the interpreter for
similarities in the shape of curve deflections – matching patterns between different wells
signifies that the same unit is present in both boreholes (Krygowski 2004). In traditional
clastic sedimentology, the shape and patterns of the GR curve are used to infer
depositional processes based on vertical trends in the relative proportions of sand and clay
in vertical successions of strata. Natural formation radioactivity detected by the GR tool is
dominated by the radioactive decay of potassium (40K) in clay minerals (Rider 1990).
However, because the lower Colorado Group contains organic-rich mudstones with non-
clay components that contribute to total radioactivity, interpreting GR curves requires
caveats to avoid serious analytical pitfalls.
Natural radioactivity in sedimentary rocks comes primarily from the radioactive decay of
40
K in potassium-rich and, more rarely, from 232Th- and 238Ur-bearing zones. Radioactive
isotopes are more concentrated in mudstones than in other sedimentary rocks because
clays are derived from the weathering and decomposition of igneous rocks – igneous
rocks contain volumetrically significant proportions of radioactive elements from micas
and feldspars (Glover 2000). The high CEC of clay minerals allows them to retain trace
amounts of radioactive elements from their parent rocks (Ellis & Singer 2008). Quartz,
the principal component of most sandstones, does not contain any radioactive isotopes
(Andersen 2011). For this reason, GR readings are often used as a proxy for grain size:
high GR readings are attributed to high clay1 content, whereas low GR readings are
attributed to low clay content (high sandstone or carbonate content). The intensity of a
GR log is often treated as a proxy for the sand-to-clay ratio or grain size of a zone – this is
1
“Clay” or “mud” is used as both a textural and a mineralogical term, leading to some ambiguity when
discussing sedimentary rocks. In a textural context, “clay” refers to particles with grain sizes < 4 µm; this
may include a fraction of quartz or feldspar grains in addition to clay minerals. In this thesis, “clay” is used
only in a mineralogical sense, whereas “mudstones” will refer to fine-grained lithologies.
55
a bad assumption because grain size does not always correlate with radioactivity (Rider
1990; Ellis & Singer 2008).
Mudstones have a range of possible GR responses, due to a variety of possible mudstone

compositions. They can contain a mixture of radioactive (clay minerals, micas, feldspars,
and organic matter) and non-radioactive (quartz, feldspars, carbonates, sulfides and
amorphous carbon) constituents (Aplin & Macquaker 2011). Table 3.1 displays the range
of minerals that can potentially be found in mudstones and their associated well log
responses. Clay minerals can have an array of possible compositions – for example, illite
clays (K0.6-0.85(Al,Mg)2(Si,Al)4O10(OH)2) may have up to 8% K, whereas chlorite
((Mg,Fe)3(Si,Al)4O10(OH)2*(Mg,Fe)3(OH)6) contains negligible K. Thorium is associated
with certain clay minerals and accessory zircons (Krygowski 2004). In addition, a range
of radioactive element concentrations are possible for micas and feldspars (Hurst 1990).
Mudstones may also contain a significant portion of non-radioactive quartz silt – this
could result in a lower GR response than a typical mudstone that has a higher proportion
of clay (Rider 1990). Finally, “black” mudstones with elevated organic content can
contain high concentrations of 238U salts precipitated through reduction by organic matter
in low-oxygen to intermittently anoxic marine environments (Aplin & Macquaker 2011).
Mudstones may also contain fine-grained carbonate particles or cements that have a lower
GR response (Rider 1990; Ellis & Singer 2008). Formation compositional effects on GR
readings must be considered in order to use GR measurements for petrophysical
calculations – this avoids propagating systematic errors. When they are correctly
calibrated, GR logs can be used to accurately distinguish mudstone from non-mudstones.
These examples illustrate that a GR response is not necessarily solely indicative of high
clay content (Equation 3).
3.6.2.1.2 Spectral gamma ray logs

GR tools are unable to distinguish between different sources of natural radioactivity – this
is a significant obstacle for lithological interpretation. Spectral GR (SGR) detectors
rectify this by quantifying the individual contributions of separate radioisotopes (40K,
232
Th and 238U) present in a formation; this is particularly useful for quantifying
mineralogy (Glover 2000; Ellis & Singer 2008; Hill 2012). This is done by partitioning
56
Table 3.1: Table of geophysical well log responses to commonly encountered minerals and compounds in
the Lower Colorado Group.
Compiled from Hurst (1990), Glover (2000), Ellis and Singer (2008), and Schlumberger (2009). Any single
well log response (e.g., gamma ray) is a result of measuring a complicated mixture of minerals with
different compositions. Travel time is in µs/ft.
57
gamma rays into energy bins according to their unique pulse size or spectra (Gadeken et
al. 1991). The sum of the total GR energies from these sources is equal to the total GR
response, resulting in the following relationship to total GR in shales where 232Th is in
ppm, 238U is in ppm, and 40K is in % (Ellis & Singer 2008):
NOPQ = 4Sℎ + 8U + 16W [5]
Anomalous gamma ray responses associated with uranium in organic matter can make it
difficult to quantify clay components of organic-rich formations like the lower Colorado
Group. This can be deconvolved from the total GR by utilizing the uranium-free curve
(computed gamma ray, CGR):
X5' = NOPQ − 8U = 4Sℎ + 16W [6]
CGR removes radioactive anomalies associated with uranium. The Th/K ratio, in
particular, can be used to quantify radioactive mineralogy in shales, as well as
identification of mineralogical differences between zones (Ellis & Singer 2008;
Schlumberger 2009). Changes in Th/K can indicate a sudden change in depositional
environment or an unconformity (Glover 2000). Charts of Th/K are used to determine the
types of clay minerals in a mudstone from concentrations measured in the Th- and K-
energy windows with a SGR. The intersection point of Th/K for each measurement
determines the type of radioactive minerals that the rock contains. An illite-rich “shaley”
sandstone would plot between 2.0<Th/K<3.5, with clay-rich zones plotting closer to 70%
illite (pink line) and clay-poor units located closer to the origin (Schlumberger 2009).
CGR can be used to calculate clay volumes (!$% ) that are more robust than those from
total GR, because the random contribution of uranium is eliminated (Glover 2000):
X5'%67 − X5'/9: [7]

!"# | X5' =
X5'/;< − X5'/9:
where X5' = NOPQ − 8U. Clearly, SGR data is superior to total GR data in all cases for
unconventional reservoir assessment because of their utility and flexibility.
Unfortunately, few wells have SGR logs available due to cost and time associated with
58
their acquisition. When available, they can and should be used to calibrate clay volume as
well as establish radioactive mineral assemblages.
3.6.2.1.2.1 Tool specifications

GR intensity is given in American Petroleum Institute units (API) and typically ranges
between 0 and 150 API in clastic reservoirs. If the formation of interest is particularly
radioactive, readings may exceed 300 API. This intensity is calibrated to an artificially
radioactive formation located at the University of Houston, which has approximately
twice the radioactivity of a typical shale (200 API; typical shale response is ~100 API).
The response of a GR log is a sphere with a depth of investigation of ~10 cm, and the tool
has a vertical resolution of 0.4 to 0.9 m; this depth of investigation decreases in denser
formations. The precision of the GR tool is ±4 API units (Krygowski 2004; Ellis & Singer
2008).
The GR tool measures the amount of gamma rays resulting from radioactive decay; since
radioactivity is a stochastic, spontaneous, and non-continuous process, GR logs contain a
significant amount of statistical noise. The only way to reduce these statistical
fluctuations is to increase the number of samples measured, either by increasing counting
times per sample or increasing radioactive source strength; the former increases time
spent logging the well and is expensive, while the latter presents potential safety risks
(Ellis & Singer 2008). Statistical fluctuations contribute ± 2 API units to GR readings in
mudstones, and ± 5 API units in non-mudstones (Krygowski 2004).
The sphere of influence of a spectral GR (SGR) log has a depth of investigation identical
to total GR tools but can have a higher vertical resolution of ~0.1 m. This is because the
tool is run more slowly than normal GR logs. SGR tool precision is slightly lower than
total GR tools, at ±5 API units (Krygowski 2004; Ellis & Singer 2008). Count rates are
also reduced by up to one-third compared to total GR tools. This increases the amount of
drilling rig time required to log the hole, significantly increasing the SGR tool’s
associated cost (Krygowski 2004). Consequently, SGR tools are not utilized as often as
total GR tools (Ellis & Singer 2008).
59
3.6.2.1.2.2 Environmental effects

GR tools have a limited depth of investigation and are sensitive to the borehole
environment. GR counts and GR log character in open-hole wells are impacted by 4
major borehole environmental factors (Glover 2000; Krygowski 2004):
1) Increasing the hole size (via hole collapse or cave-ins) decreases count rates and
mutes the signal of the reservoir, because gamma rays have to travel farther to the
detector and experience more Compton scattering;
2) Invasion can increase or decrease count rates depending on drilling mud composition
because drilling fluid displaces formation fluids that have different densities;
3) Drilling mud composition: adding barite decreases count rates, because barite is an
efficient absorber of gamma ray radiation (r = 4.20 g/cm3); in comparison, adding
KCl to drilling mud increases count rates, because 40K is radioactive, and;
4) Increasing logging speed decreases gamma ray count rates and reduces measurement
resolution, because the detector does not have as much time to measure the true GR
activity of the formation.
SGR logs, like normal GR logs, are impacted by borehole environmental effects that
influence count rates, like changes in hole size and logging speed (Ellis & Singer 2008).
Although barite muds still impact count rates of SGR logs, the impact of KCl drilling
mud is only experienced by the 40K and total GR curves – 232Th and 238U are not affected
(Glover 2000).
3.6.2.1.3 Bulk density log

Figure 3.4 illustrates a bulk density logging set-up. Gamma rays are used to measure bulk
density because their scattering and transmission through media is strongly dependent on
this property. Unlike GR logs, density logs work via active measuring methods – high
energy gamma rays are emitted from a collimated 137Cs or 60Co radioactive source, which
then interact with the electrons of elements in the formation (Krygowski 2004; Hill 2012).
The 137Cs or 60Co source is chosen so that emitted gamma ray energies are sufficiently
low to interact via Compton scattering rather than pair production while exceeding the
60
Fig. 3.4: Schematic of a bulk density logging tool in a conventional reservoir.
Adapted from Glover (2000) and Ellis and Singer (2008). Assuming a matrix density of 2.55 g/cm3
(weighted average of mudcake and sandstone), the bulk density in the above reservoir would be
measured at 2.41 g/cm3. Adding clay or kerogen into the reservoir (like in the above brittle shale)
decreases porosity and decreases the matrix density, resulting in a lower bulk density reading.
61
threshold for photoelectric absorption (Ellis & Singer 2008). The emitted gamma rays
induce a responding gamma ray emission from the formation. The returning gamma rays
undergo Compton scattering as they travel back through rock and collide with electrons.
These slower gamma rays are counted by heavily shielded short- and long-range detectors
that measure gamma radiation. The attenuation the returning particles undergo is directly
related to the electron density of the formation, such that induced gamma rays undergo
more attenuation in high bulk density formations due to high electron density, therefore a
lower gamma ray count is recorded – and vice versa (Glover 2000; Krygowski 2004).
The measured bulk density of a rock is directly related to its mineralogy, porosity, and the
density of fluids filling the available porous space (Davis 1954; Baker 1957; Edwards
1959; Pickell & Heacock 1960). The bulk density of the formation (YJ ) is a function of
matrix density (Y/; ), density porosity (HZ ), and the density of the pore filling fluid (Y[% ;
Davis 1954):
YJ = (1 − HZ )Y/; + HZ ∗ Y[% [8]
Using Equation 8, density porosity can be calculated continuously alongside bulk

density, provided that matrix density and fluid density are invariant. Ideally, these
variables reflect the true lithology of the reservoir (i.e., Y/; has been derived from core).
Bulk density tools are calibrated to blocks of pure limestone (Y/; = 2.71 g/cm3) saturated
with freshwater (Pickell & Heacock 1960); these are invalid assumptions for most
reservoirs (Ellis & Singer 2008).
For a rock with average atomic number Z, atomic weight A, and density r, the electron
density re is equal to (Pickell & Heacock 1960):
^ [9]
YF = 2 YJ
O
For most light elements likely to be encountered in a formation logging scenario, Z:A is a
constant that approaches 0.5. A constant Z:A ratio is the physical basis for bulk density
logging – if Z:A deviates from 0.5, the calculated bulk density is no longer valid. This
62
relationship breaks down when hydrogen in hydrous clay minerals, formation fluids, or
organic matter are introduced (Glover 2000) – Z:A for hydrogen is closer to 1 (Gaymard
& Poupon 1968). By Equation 9, higher Z:A causes bulk density to appear low. When
substituted into Equation 8, the resulting density porosity from a high Z:A is too high. For
this reason, high porosities in low bulk density zones are invalid (Sondergeld et al. 2010).
3.6.2.1.3.1 Tool specifications

Bulk density is measured in g/cm3 or kg/m3, and usually ranges from 1.95 to 2.95 g/cm3
because this the natural range for most rocks. The tool automatically compensates for the
presence of dense mud cake, which is included as a quality control curve. If this
correction exceeds ± 0.15 g/cm3, the bulk density curve is not reliable (Glover 2000).
When run at normal logging speed (396 m/hr), the bulk density tool has a vertical
resolution of 0.26 m but can resolve beds <0.15 m thick if run slowly. Due to increased
gamma ray attenuation with density, depth of investigation decreases with bulk density of
matrix. It has a relatively shallow depth of investigation of 10 cm in most reservoir rocks,
and a precision of ±0.01 g/cm3 (Krygowski 2004); this translates to an error of 0.5%
(Glover 2000). Due to invasion of drilling fluid, it is normally assumed that the zone of
investigation is saturated with mud filtrate (Krygowski 2004).
3.6.2.1.3.2 Environmental effects

Similar to GR tools, bulk density tools have a limited depth of investigation, and are
therefore sensitive to the borehole environment. Additionally, the eccentered-mounted
tool is pressed against the walls of the borehole, which means readings are especially
sensitive to rough hole conditions and caving. Bulk density responses in open-hole wells
are impacted by the following borehole environmental factors (Glover 2000; Krygowski
2004):
1) Increasing the hole size (via hole collapse or cave-ins) decreases bulk density readings
relative to the true bulk density, because the eccentered-mounted tool is no longer in
direct contact with the formation, and the detector is therefore reading mud cake and
drilling fluid;
63
2) Choice of drilling mud composition changes the bulk density with additives like barite
increases bulk density, because barite is an efficient absorber of gamma ray radiation
(r = 4.20 g/cm3); sylvite mud (KCl) has an anomalous Z:A from chlorine;
3) Hydrogen in light hydrocarbons and potentially organic matter decreases the
measured bulk density, making density porosity erroneously high.
Other environmental effects are caused by transforming bulk density to density porosity.
Because density porosity is calibrated to specific zones, if one of the parameters (e.g.,
lithology or fluid density) changes, the calculated porosity will be incorrect. This is
particularly common in shaley units, because of the varied nature of clay mineral
densities. Additionally, if the bulk density is incorrectly calibrated to matrix density or
fluid density, resulting density porosities will also be wrong; for this reason, it is always
critical to calibrate bulk density logs to core data if possible (Glover 2000).
In highly heterogeneous reservoirs (e.g., organic-rich unconventional reservoirs), matrix

density and bulk density (Table 3.1) may vary significantly on a centimeter scale:
organic-rich mudstones have highly variable compositions depending on matrix (2.4 -
2.85 g/cm3) and organic content (<1.5 gm/cm3; Glover 2000). However, if clay volume
and kerogen volume are continuously modelled across a reservoir interval, their
contribution to density porosity can be compensated for. Otherwise, due to the hydrogen
content of kerogen, the calculated density porosity will be erroneously high (Sondergeld
et al. 2010).
3.6.2.1.4 Neutron log

Like the bulk density log, the neutron log is sensitive to the number of hydrogen atoms in
a formation – it offers a third way to calculate porosity indirectly by using logs (Glover
2000). The neutron tool bombards the formation with high energy neutrons from a
radioactive source that undergoes Compton scattering and produce gamma rays. Detected
gamma ray count rates are proportional to hydrogen atom concentration because
hydrogen is an efficient absorber of neutrons (Glover 2000), although it is important to
note that neutron tools respond to all sources of hydrogen, not just those contained in the
pore spaces (Ellis & Singer 2008).
64
The neutron porosity responds primarily to hydrogen (Ellis & Singer 2008; Gonzalez et
al. 2013). Low neutron count rates are detected from lithologies with a large number of
hydroxyls in clay matrices, pore fluids (hydrocarbons or water), or solid hydrocarbons
(bitumen and kerogen). Hydroxyls are found in clay-rich, kerogen-rich, or porous rocks
with hydrocarbon-filled porosity In comparison, high count rates are detected from low
porosity or clay-free rocks; these lithologies have minimal hydrogen present and
experience less neutron absorption (Ellis & Singer 2008).
Because the neutron log is sensitive to hydrogen, the neutron porosity log is used as a
proxy for clay volume in concert with the density log. This clay volume is considered
superior to GR-derived clay volume, but because it requires 0% clay for calibration it can
be difficult to apply this method in unconventional reservoirs (Glover 2000).
3.6.2.2 Acoustic logs

Sonic logs measure the interval travel time (∆t), or “slowness”, of a sound wave through a
formation; they can be used to derive wave velocity as well as porosity (Wyllie et al.
1956; Raymer et al. 1980; Glover 2000; Krygowski 2004). First, high amplitude and high
frequency pulses (20 to 40 kHz, 10 to 60 per second) are discharged from a tool-mounted
transmitter. In the intervening time, the pulse travels through the rock of the formation
where it attenuates and disperses, before it is detected shortly thereafter by two or more
receivers on the tool (Glover 2000). Sound energy arrives at the receivers in different
times, depending on the type of wave that arrives (in order of arrival time: P, S, Rayleigh,
and Stoneley). More complex tools measure P- and S-waves, and some (full waveform
logs) can record the entirety of the wave train (Glover 2000). The speed that sound travels
through the formation is largely dependent on the reservoir’s mineral composition and
porosity (Andersen 2011). The capacity for rocks to transmit elastic waves varies with
lithology and texture, markedly decreasing with increasing effective porosity (Ellis &
Singer 2008). Finally, the slowness of a P-wave is inversely proportional to the strength
of the material, and directly proportional to the density of the material (Glover 2000):
Y [10]
D_ ∝
a_bcde_ℎ
65
where strength is related to bulk modulus (incompressibility and stiffness) and shear
modulus (rigidity; Ellis & Singer 2008). At slow ∆t or wave velocity, the P-wave
undergoes more attenuation and dispersion, thus arriving at the receiver later; conversely,
at fast ∆t or high wave velocity, the P-wave undergoes less attenuation and dispersion,
thus arriving at the receiver earlier.
Along with being a lithology indicator, wave velocity can be used to derive porosity from
the sonic log using Wyllie et al. (1956), who related slowness (∆_) to porosity (Hg ), fluid
transit time (∆_[% ) and matrix transit time (∆_/; ) by the following equation:
∆_ = h1 − H+ i∆_/; + H+ ∗ ∆_[% [11]
although porosities derived from density logs are considered superior to porosity from
sonic (Glover 2000). Using this equation can be tricky in petrophysically complex
reservoirs because ∆_/; needs to be calibrated to the formation of interest and is
especially variable in shales (Ellis & Singer 2008).
3.6.2.2.1 Tool specifications

Lithology and porosity affect ∆t, making sonic logs useful for many petrophysical
applications (Varhaug 2016). ∆t is measured in µs/ft or µs/m, and commonly ranges from
130 to 820 µs/m in most formations. The depth of investigation depends on the frequency
(f) of the pulse emitted from the acoustic source and the slowness of the formation, and
commonly ranges from 0.025 m to 0.25 m. As wavelength is a function of velocity over
frequency, formations with reduced slowness (higher velocity) can be penetrated more
deeply by the acoustic log (Glover 2000; Krygowski 2004). Denser, more consolidated
formations typically have a faster ∆t than unconsolidated formations, or those with fluid-
filled porosity (Varhaug 2016). Finally, the maximum vertical resolution of a standard
acoustic log is equal to the detector spacing, which is commonly ~0.60 m (Glover 2000).
3.6.2.2.2 Environmental effects on sonic logs

“Cycle skipping” occurs when the first P-wave to reach the detector fails to exceed a
threshold amplitude value on the first arrival, causing that peak to be missed by one of the
66
detectors. As a result, that first P-wave is detected in the following cycle, 40 µs later, and
∆t is perceived to be higher. This effect creates spikes in the sonic log on the order of 130
µs/m, and may indicate the presence of fractures or gas-filled porosity (Glover 2000; Ellis
& Singer 2008). Cycle skipping, however, can also be caused by enlarged borehole
conditions or improper tool centralization (Krygowski 2004).
3.6.2.3 Electric logs

Resistivity logs attempt to measure the resistivity (inverse of conductivity) of the
formation of interest, both adjacent to and surrounding the borehole (Krygowski 2004).
Laterologs, a type of resistivity log used primarily in saline water-based muds, work by
creating an electric circuit in the formation through the pore network and formation fluid
(Andersen 2011). This process works via a low frequency current that flows from source
electrodes in the tool. Arrays on either side of the source electrode force the measuring
current into a thin disc-shape perpendicular to the borehole that penetrates deep into the
formation (Krygowski 2004). The amount of current returning to the tool is detected by a
measuring electrode, and the amount of current that is lost represents the virgin formation
resistivity, which can be used to calculate water saturation and fluid content (Ellis &
Singer 2008).
The pioneering work of Archie (1942), followed by Winsauer et al. (1952), established an
empirical relationship between porosity and electrical resistivity. The electrical resistivity
of brine-saturated rocks was later observed to increase with pore throat constriction
(reduced permeability), reduced porosity, and elevated tortuosity – the latter is the
effective distance available for electrical current flow through a reservoir (Owen 1952;
Towle 1962; Helander & Campbell 1966). The relationship between formation resistivity,
porosity, and tortuosity can be expressed as (Helander & Campbell 1966):
'( (S)- [12]

=
', H
where T is tortuosity and ', is the resistivity of the saturating brine. By Equation 12,
porosity is inversely related to '( and directly proportional to the square of tortuosity.
67
Although tortuosity cannot be directly measured directly in the lab, it can be estimated by
the following relationship (Pérez-Rosales 1982):
H [13]
S =
H/
where m is the cementation exponent from the Archie (1942) equation (Equation 1).
Formation resistivity, '( , can be used to calculate water saturation by taking advantage of
Equation 1 (Archie 1942; Glover 2000). As discussed previously in section 3.5.2, the
Archie (1942) equation fundamentally overlooks the elevated electrical conductivity of
certain lithologies (especially clay and organic matter). Clay minerals reduce formation
resistivity compared to equivalent clay-free zones due to enhanced conductivity from
bound water in the clay structure (Krygowski 2004). This causes Archie water saturations
to appear erroneously high (Worthington 1985; Asquith 1990). Resistivity measurements
in unconventional reservoirs, therefore, should be interpreted with caution (Bust et al.
2013). '( is affected by porosity, fluid saturation, and mineralogy (Ellis & Singer 2008):
• Low R / high conductivity: formation brines and water-based mud filtrates have low
electrical resistivity, therefore rocks with high brine saturations (high porosity) are
highly conductive;
• High R / low conductivity: matrix material (especially carbonates), hydrocarbons, and
oil-based mud filtrate all have high electrical resistivity; additionally, organic-rich
mudstones have high resistivity due to low porosity and water saturation.
Representative brine resistivity, ', , and '( values are essential for calculating water
saturation. Small changes can alter water saturation calculations by ±10% (Glover 2000).
Ideally, these parameters would be determined in a lab using water samples retrieved
from the Second White Specks Formation in the study area, but this information is not
available. There is significant potential risk of propagating error in water saturation
calculations for the Second White Specks without calibration to core data. It is more
responsible to use resistivity logs as a secondary indicator of matrix composition.
68
3.6.2.3.1 Tool specifications

Resistivity is measured in ohm×m, and ranges from 0.2 (highly conductive) to greater than
20000 ohm×m (highly resistive). The readings are presented graphically in a logarithmic
manner to enhance small changes in resistivity over a short vertical distance. Laterologs
are run centered in the borehole with water-based drilling mud, and must make electrical
contact with the formation in order to get a resistivity reading (Krygowski 2004). At
shallow depths of investigation (<0.5 m) the measured resistivity is not normally equal to
the virgin formation resistivity, because conductive mud in the borehole and mud filtrate
invasion into the formation distort the signal from the formation (Cosmo et al. 1991). To
combat this, laterologs are used with different electrode spacings to get resistivity at
multiple depths of investigation (i.e., deep, shallow, and microlog). Reducing the
electrode spacing of a laterolog improves vertical resolution at the cost of reduced
penetration depth (Glover 2000).
The deep laterolog (LLD) has a vertical resolution of 0.6 m, and a 1.5 to 2.1 m depth of
investigation. It is precise to ±0.2 ohm×m. In comparison, the shallow laterolog (LLS) has
comparable vertical resolution and precision but shallow depth of investigation (0.6 to 0.9
m). The microlog (RXO), which has closely spaced electrodes for maximum vertical
resolution (3 to 8 cm), has a significantly reduced radius of investigation (2.5 to 10 cm),
and is precise to ±0.2 ohm×m (Krygowski 2004).
3.6.2.3.2 Environmental effects

Historically, resistivity logging in thin beds has proven difficult (De Witte 1954; Cosmo
et al. 1991). Laterolog tools are calibrated by computing the current flow and potential
distribution patterns in a standard homogenous medium, where current flow spreads out
over a very large volume after leaving the tool (Woodhouse 1978); in reality, distribution
patterns are distorted by any resistivity anisotropy present (Threadgold 1972). In
heterogeneous thinly-layered formations, electrical current tends to flow preferentially
into more conductive beds; under these conditions, anisotropy is particularly extreme.
This phenomenon, called the “shoulder-bed effect”, causes laterolog readings to be
distorted at bed boundaries (Ellis & Singer 2008). Differential invasion into thin sand-
69
shale (permeable vs. non-permeable) beds can further exacerbate this problem (Minette
1990). Even if micrologs are run over the formation of interest, they do not provide
reliable formation resistivities (Glover 2000; Krygowski 2004). With current technology,
beds < 1 m thick are not easily resolved without NMR (nuclear magnetic resonance) logs,
which are expensive and not commonly run in the Second White Specks interval.
Invasion of mud filtrate into permeable zones causes virgin formation fluids to be
displaced. Although the LLD normally measures deep enough to extract the true
formation resistivity (required for the calculation of water saturation), if “formation
damage” (invasion) is extreme this value may be incorrect (Glover 2000). Resistivity
curves at different depths of investigation should be identical in impermeable beds, where
invasion does not occur. In comparison, in invaded zones, an “invasion profile” is visible:
if the mud filtrate resistivity exceeds the formation resistivity then the RXO and LLS
should read higher resistivities than the LLD, and vice versa (Krygowski 2004).
Highlights
• Petrophysicists assess hydrocarbon reservoirs by using indirect measurements
(geophysical wireline logs) collected from the subsurface to model rock properties.
Storage and flow capacity can be evaluated remotely using this data.
• Petrophysical models require meaningful constraints in order to accurately represent
natural variations in rock properties. Stratigraphic architecture controls
compartmentalization of the reservoir into hydraulic flow units.
• Wireline log measurements are used as proxies for reservoir quality indicators like
mineralogy, porosity, and water saturation. These tools were originally designed to
assess “conventional” reservoirs that did not contain clay or organic matter. As focus
has now shifted to low porosity “unconventional” reservoirs that contain large
amounts clay and kerogen, the tool responses to unconventional reservoirs need to be
well constrained. Elevated gamma ray, low density, slow sonic travel time, high
neutron porosity, and increased resistivity responses are expected in reservoirs with
high organic and clay content.
70
Solving the Second White Specks 4. Allostratigraphic method
Chapter 4
4 Stratigraphic method
Overview
Allostratigraphic correlations form the basis for zonal analyses of petrophysical properties
presented in Chapters 6 through 8. The following section reviews the allostratigraphic
method used in this study to define reservoir compartments called hydraulic flow units.
Following the allostratigraphic method validates the stratigraphic picks herein.
Stratigraphic nomenclature and bounding surfaces

The sedimentary rock record can be subdivided and correlated several different ways.
Two methods are relevant to this study: lithostratigraphy, the subdivision of rock units
based on lithic characteristics; and allostratigraphy, the subdivision of rock units via
correlation of bounding surfaces with temporal significance (NACSN 1983, 2005). This
thesis uses the allostratigraphic method for subdivision of the lower Colorado Group,
extending the work of Tyagi et al. (2007). Bounding surfaces with allostratigraphic
significance include unconformities and conformities.
Unconformities are stratigraphic surfaces that separate younger from older strata and
represent gaps in sedimentation (Cross & Lessenger 1988). These include but are not
limited to:
1) subaerial unconformities, sensu Sloss et al. (1949): these are unconformities that form
under subaerial conditions as a result of fluvial erosion or sedimentary bypass, among
other processes;
2) marine flooding surfaces: a surface that indicates an abrupt increase in water depth;
3) regressive surfaces of marine erosion, sensu Plint (1988): an erosional surface that
forms via wave scouring during forced regression in wave-dominated shallow water
settings due to relative sea-level fall.
A conformity is a surface that separates younger from older strata but does not indicate
erosion or significant non-deposition – this includes surfaces that experienced very slow
rates of deposition such as condensed sections (Loutit et al. 1988).
71
Controls on sedimentation
Sedimentary rocks are inherently cyclical – they exhibit specific and recognizable stratal
stacking patterns that reflect allogenic forcing (Helland-Hansen & Gjelberg 1994;
Dalrymple 2010a). These patterns, called “sequences”, are sedimentary successions
deposited during an entire cycle of change in accommodation and supply (Posamentier et
al. 1988; Galloway 1989; Catuneanu et al. 2009). Accommodation defines the available
space that sediment can fill (Jervey 1988). The major controls on accommodation are
changes in relative sea level and changes in the rate of sediment supply to the basin
(Jervey 1988; Posamentier et al. 1988; Galloway 1989; Plint et al. 1992; Schlager 1993;
Muto & Steel 1997, 2000; Catuneanu 2002, 2006; Catuneanu et al. 2011).
4.3.1 Eustasy
Eustasy is the change in sea-level relative to a fixed datum, like the centre of the earth
(Posamentier et al. 1988). Basin tectonics locally affect the elevation of the sea floor
which serves as a local datum, and the position of the eustatically determined sea surface
relative to this local datum is known as the relative sea level (Posamentier et al. 1988).
Many cyclical sedimentary patterns are observed globally and appear to be driven by
eustatic variation (Haq et al. 1987; Haq 2014). Eustatic change is primarily controlled by
fluctuations in oceanic water volumes (Donovan & Jones 1979; Miller et al. 2005) – this
flux comprises the creation or removal of accommodation within a particular basin.
Eustatic processes, once classified on the basis of their magnitude (Vail et al. 1977), are
now recognized as episodic, hierarchical phenomena that operate on a variety of
timescales. These processes are listed below, in order from low to high frequency (Vail et
al. 1977; Donovan & Jones 1979; Plint et al. 1992; Miall 2010):
1) First order (400 to 500 Myr periodicity) cycles are related to global cycles of
supercontinent formation and breakup;
2) Second order (10 to 100 Myr periodicity) cycles are linked to volume flux of mid-
ocean ridge spreading centres;
3) Third order (0.1 to 10 Myr periodicity) cycles are correlated with regional basement
tectonic plate kinematics, and;
72
4) Fourth and fifth order (10 Kyr to 500 Kyr periodicity) cycles are caused by
Milankovitch orbital forcing (Milankovitch 1941).
Only Milankovitch processes will be discussed further. Although higher order cyclicity
may have played a role in sedimentation during the Late Cretaceous, the time scales on
which higher order cycles operate is likely beyond the vertical resolution of stratigraphic
section and time considered in this thesis.
4.3.2 Milankovitch orbital forcing

Milankovitch cycles are attributed to the climatic imprint of Earth’s periodic orbital
variations. These cycles include orbital eccentricity, obliquity, and precession
(Milankovitch 1941). Milankovitch cycles change the amount of solar energy at the
earth’s surface, in addition to its latitudinal and seasonal distribution – thus globally
influencing climate (Arthur & Philip 1986). These climate cycles regulate fluctuations in
the global volume of continental ice sheets (glacioeustasy), which is the primary
mechanism for high frequency eustatic variation (Plint et al. 1992; Miall 2010). High
frequency sedimentary cycles attributed to Milankovitch cyclicity have been recognized
in Cretaceous rocks across the WIS (Plint 1991; Varban & Plint 2008b) although their
linkage to glacioeustasy is contentious. The existence of ice caps during Cretaceous time
has been hotly debated (Huber et al. 2002), due to the “greenhouse” climate of the
Cretaceous period that should have prohibited continental ice sheet growth. d18O isotope
data from Turonian sediments recovered in the tropical Atlantic record exhibit positive
excursions that are synchronous with the eccentricity orbital cycle, suggesting that
continental ice sheets may have been extant, at least for periods of time, during the Late
Cretaceous (Bornemann et al. 2008).
Milankovitch cyclicity has been documented in rocks from the WIS that span the
Cenomanian-Turonian boundary. For example, Sageman et al. (1997) correlated
limestone/marlstone bedding couplets of the Second White Specks-equivalent Greenhorn
Formation in Colorado with Milankovitch obliquity cycles (41 Kyr). Prokoph et al.
(2001) used core from Bloch et al. (1999)’s Second White Specks Formation type well
(6-34-30-8W4) to determine that Milankovitch cyclicity was a significant control on
73
sedimentation during OAE-II, but they were unable to link overall sea-level rise during
the Late Cenomanian and OAE-II to a specific mechanism. These observations suggest
that rocks of the lower Colorado Group in west-central Alberta contain regionally
correlative cycles that are related to eustasy and changes in accommodation.
4.3.3 Sediment supply

Clastic detritus is transported to basins at variable rates, depending on climatic factors and
relative sea-level change (Miall 2010). Climate controls the sedimentation rate by
prohibiting or encouraging erosion (Nichols 2009). Highest sedimentation rates are
encountered in temperate climates that experience seasonal variability and increased
precipitation (Miall 2010), because erosion and weathering processes (both chemical and
physical) are enhanced (Nichols 2009).
Rivers adjust their profiles as relative sea-level drops, re-equilibrating to a lower “base
level” (Nummedal et al. 1993). Shanley and McCabe (1994) defined base level as:
“[…] the potential energy surface that describes the direction

in which a stratigraphic system is likely to move, toward
sedimentation and stratigraphic preservation or sediment
bypass and erosion.”
Lowering base level triggers the erosion of topographic highs by rivers, which then
transport sediment to the ocean. Raising base level, conversely, favors sedimentation and
stratigraphic preservation. Base level is, fundamentally, the equilibrium surface to which
rivers erode – it approximates relative sea-level in the marine realm (Wheeler 1964).
4.3.4 dA/dS ratio

Variations in the ratio of change in accommodation (dA) to the rate of change sediment
supply (dS) is frequently invoked as the primary mechanism for shoreline migration. The
interaction between these rates (dA/dS) determines shoreline trajectories (Jordan &
Flemings 1991; Swift & Thorne 1991; Posamentier et al. 1992; Schlager 1993; Muto &
Steel 1997; Martinsen et al. 1999; Muto & Steel 2000). If accommodation rates are
approximately equal to sedimentation rates (dA = dS), shorelines remain stationary and
produce aggradational successions. If accommodation rates and sediment supply are not
74
balanced, depositional systems will shift in space to reach equilibrium (Catuneanu 2006).
When supply exceeds accommodation (dS > dA), a regression occurs: shorelines and
facies shift seaward, resulting in progradational (lateral outbuilding) stacking patterns.
When accommodation exceeds supply (dA > dS), the shoreline experiences a landward
transgression, which also constitutes a landward facies shift – this results in
retrogradational stacking patterns (Catuneanu 2002).
Seismic stratigraphy
Based on the observations of Vail et al. (1977), Mitchum et al. (1977) introduced the
concept of “seismic stratigraphy” from their analysis of two dimensional seismic
reflection data. Surfaces of contrasting acoustic impedance (i.e., surfaces that separate
different lithologies) generate seismic reflections (Neidell 1979). Seismic stratigraphy
associates seismic reflections with unconformities and stratal surfaces – therefore
assigning time-stratigraphic significance to strong seismic reflectors (Vail et al. 1977;
Cross & Lessenger 1988). Seismic data allows for the observation of stratal terminations,
stacking patterns, and 3D visualization of stratigraphic surfaces (Posamentier 2000).
Seismic stratigraphy is restricted by the limitations of seismic data. The bandwidth of

seismic exploration data (20 to 40 Hz, ~30 m vertical resolution) may be insufficient to
characterize high frequency stratigraphic cycles (Catuneanu et al. 2009). In some cases,
surfaces of contrasting acoustic impedance occur at sedimentary facies boundaries that
are not time-stratigraphic surfaces (i.e., lithostratigraphic boundaries) – thus, seismic
reflections may cross time lines (Cross & Lessenger 1988). Impedance contrasts over
diffuse stratigraphic boundaries with minimal facies contrast (e.g., correlative
conformities) generate little to no seismic reflection (Neidell 1979), so their time-
stratigraphic significance may be overlooked (Cross & Lessenger 1988).
Regardless of its limitations, seismic stratigraphy afforded the delineation of cyclical

stratigraphic packages bounded by regional unconformities induced by relative sea-level
fall (Vail 1987). It is now generally accepted that these cycles, or genetic sequences,
result from the complex interplay between changes in accommodation, sedimentation,
and erosion (Posamentier & Vail 1988). Integration of seismic stratigraphic methods with
75
outcrop and well log data resulted in the creation of sequence stratigraphic concepts
(Posamentier & Vail 1988; Van Wagoner et al. 1988).
Sequence stratigraphy
Van Wagoner et al. (1988) defined sequence stratigraphy as:
“[…] the study of rock relationships within a

chronostratigraphic framework of repetitive, genetically
related strata bounded by surfaces of erosion or nondeposition,
or their correlative unconformities.”
Catuneanu et al. (2009), alternatively, defined sequence stratigraphy as the study of

stratal stacking patterns and their variations within a temporally-constrained framework
of bounding surfaces. These patterns, or sequences, constitute stratal successions
“deposited during a full cycle of change in accommodation or sediment supply” and are
bound by sequence stratigraphic surfaces called sequence boundaries (Catuneanu et al.
2009).
Sequence boundaries delineate systems tracts of regionally mappable and correlative

depositional systems (Brown & Fisher 1977; Posamentier et al. 1988): these are the
lowstand, transgressive, highstand, and falling-stage system tracts (Mitchum 1977;
Posamentier & Vail 1988; Van Wagoner et al. 1988; Plint & Nummedal 2000). Systems
tracts are coeval depositional systems defined by their stratal geometry, stratigraphic
position, and parasequence stacking patterns (Posamentier et al. 1988). Parasequences
are smaller elements (metre scale) within an individual sequence (Van Wagoner 1985):
they are short-term, higher frequency flooding events (Haq et al. 1987) that may originate
from orbital forcing on Milankovitch (<500 Kyr) scales (Read & Goldhammer 1988).
Sequence stratigraphic methods can be used to interpret stratigraphic hierarchy on the

basis of interactions with allogenic factors – this is because stratigraphic architecture has
diagnostic geometry in different depositional settings (Catuneanu 2006). Systems tracts
define the four possible relationships between dA/dS and are representative of different
depositional environments and sedimentary architecture, because shifts in dA/dS can
create or destroy sequence-stratigraphic surfaces (Catuneanu 2006). Re-equilibrating to
76
base level after a shift in dA/dS takes time to occur – as a result, sequence stratigraphic
surfaces are inherently diachronous and may cross time lines (Cross & Lessenger 1988).
The applications of sequence stratigraphy to petroleum geology are clear – many

significant reservoirs contained within the WCSB are associated with unconformities
inferred as a result of relative sea level fall, and transgressive disconformities produced
during sea level rise (Bhattacharya & Posamentier 1994; Leckie et al. 1994). The
Cenomanian period is widely recognized as a highstand that reached eustatic maximum in
the early Turonian, coincident with OAE-II (Haq et al. 1987; Sageman et al. 1997; Haq
2014). The Bighorn River “red” bentonite is a marker bed that approximates the
Cenomanian-Turonian boundary in west-central Alberta (Tyagi et al. 2007; Tyagi 2009).
This horizon is associated with the Cenomanian eustatic maximum and may have some
sequence stratigraphic significance if it was reworked by waves.
Stratal patterns observed in passive continental margin settings – which have distinct
slope-shelf breaks – formed the original basis for sequence stratigraphic models (Vail et
al. 1977; Jervey 1988; Posamentier et al. 1988; Posamentier & Vail 1988). Subsidence
patterns in foreland basins fundamentally differ from passive continental margins. In a
foreland basin setting, flexural loading by thrust sheets causes subsidence rates on the
tectonically-active side of the basin to decrease progressively with distance from the
orogenic belt (Beaumont 1981; DeCelles & Giles 1996; DeCelles 2012). Passive margin
settings, in comparison, experience increased subsidence in the seaward direction.
Depositional sequences in foreland basins will exhibit landward-thickening, wedge-like
geometries (Posamentier & Allen 1993); this is the reverse of passive margin settings.
This evidence suggests that traditional sequence stratigraphy may have limited
application to foreland basin settings.
Allostratigraphy
Allostratigraphic units, or allomembers, are defined by laterally traceable bounding
discontinuities that constrain coeval rock units (NACSN 1983, 2005). Contrary to
lithostratigraphic methods, which define stratal units on the basis of similar lithologies,
allostratigraphy is predicated on the correlation of time-equivalent units regardless of
77
lateral variability – thereby assigning critical chronological and spatial context to facies
and thickness changes. Similar to seismic stratigraphy, it is assumed that allostratigraphic
surfaces correspond to physically continuous surfaces and approximate timelines
(Bhattacharya & Posamentier 1994). By constraining geological zones in this manner,
allostratigraphic maps illustrate coeval depositional systems (i.e., paleogeography) in
such a way that the influence of tectonism, changes in A/S, and erosion may be inferred
(Martinsen et al. 1993; Shank & Plint 2013). This establishes a predictive framework for
rock properties, such as clay content and thickness, based on their associations with
modern depositional environments.
Correlating allostratigraphic bounding surfaces establishes an allostratigraphic

framework, which distinguishes the stratal geometry of individual allomembers; this has
some parallels to sequence stratigraphic methods. Allomembers, when mapped across a
region, provide well-constrained “time slices” of rock properties at a vertical scale that
closely matches that of the hydraulic flow units described by Ebanks (1987). Each
allomember is expected to have a distinct set of petrophysical properties indicative of the
environmental conditions where it was deposited. This is fundamentally similar to slicing
and flattening a 3-D seismic volume along a stratigraphic surface – to a certain extent, the
resulting surface reflects the conditions it was deposited under. Typical allomember
thicknesses are in the range of 3 to 15 metres.
Allostratigraphy conceptually differs from sequence stratigraphy on how bounding

surfaces are defined. Subaerial unconformities and erosional surfaces of marine
ravinement (i.e., sequence boundaries) can be highly diachronous, limiting their
usefulness as stratigraphic timelines (Plint & Nummedal 2000; Bhattacharya 2001). In
addition, subaerial unconformities are either not preserved or rarely developed in strata
from the Kaskapau alloformation, so it is difficult to apply sequence stratigraphic criteria
and define sequences (Plint et al. 2012a). Marine transgressive surfaces – which are also
sequence boundaries – better approximate time-stratigraphic surfaces for allostratigraphy
(Vail et al. 1977; Cross & Lessenger 1988; Plint et al. 2012b) despite their moderately
diachronous nature (Roca et al. 2008). Flooding surfaces are regionally extensive and
78
commonly preserved within sedimentary strata in the WCSB, and so are more easily
correlated throughout the basin.
Smectite-rich volcanic ash beds (bentonites) are common in strata of Late Cretaceous age
(Kauffman 1984). Bentonites contain igneous minerals, such as zircons, which can be
isotopically dated (e.g., Barker et al. 2011). Volcanic ash, produced by explosive volcanic
events, is deposited quickly (days to weeks) and results in the deposition of bentonite
beds that are considered coeval (Obradovich & Cobban 1975). Bentonite ash beds are
mainly distributed via high altitude (10 to 15 km) winds, which can achieve ash
dispersion over large geographic regions (Slaughter & Earley 1965). Bentonite
preservation is enhanced when sedimentation rates are low, as ashfalls will condense into
composite ash beds that are undiluted by background sediments (Ver Straeten 2004).
Volcanic ash, however, may also be redistributed by current transport if it is deposited
above storm wave base – if this occurs, the ash bed will be physically reworked and non-
uniform in thickness (Knechtel & Patterson 1956; Ver Straeten 2004).
Of the volcanic events indicated by bentonite beds in the WIS, approximately 85% have
been correlated to major transgressive episodes (Kauffman & Caldwell 1993). Bentonites
contained within the Blackstone Formation were spatially associated with major flooding
surfaces, wave-rippled silty beds, and were typically non-bioturbated (Tyagi 2009). These
observations suggest that the bentonites observed by Tyagi (2009) were likely deposited
during marine transgressions above storm wave base, and bentonite preservation may
have been assisted by a lack of bioturbation (Tyagi 2009).
Flooding surfaces and bentonites constitute examples of basin-wide, coeval depositional

surfaces. This is supported by the relative parallelism of bentonites and flooding surfaces
(Asquith 1970; Hattin 1971; Tyagi et al. 2007; Varban & Plint 2008a). Flooding surfaces,
which are recognizable in outcrop or core and distinguishable in wireline logs (Varban &
Plint 2005), are particularly useful because their geophysical responses reflect an abrupt
upward shift to clay-rich or shaley lithologies (Rider 1990; Catuneanu 2006; Ellis &
Singer 2008).
79
Highlights
• Sedimentary cycles, or sequences, reflect environmental responses to cyclical changes
in accommodation and sediment supply to the basin. These sequences can be
stratigraphically correlated and mapped.
• Allostratigraphic zones, which are defined by coeval depositional surfaces, best
approximate hydraulic flow units. Coeval depositional surfaces include flooding
surfaces, bentonite horizons, unconformities, and correlative conformities.
80
Solving the Second White Specks 5. Petrophysical method
Chapter 5
5 Petrophysical method
Overview
Geological data (wireline logs and core data) was transformed into a 3-dimensional
petrophysical model for the Second White Specks and Belle Fourche alloformations that
can be used for first-pass sweet spotting, even when data is sparse. The process outlined
in this chapter integrates the allostratigraphic method outlined in Chapter 4 with the
petrophysical best practices examined in Chapter 3, resulting in a repeatable workflow
that reduces uncertainty.
Using the petrophysical model to calculate brittleness

The primary goal of this thesis is to assess the degree of spatial association between high
relative brittleness, increased porosity, and improved oil production from the Second
White Specks Formation. The project workflow (Fig. 5.1) has been reverse-engineered
with that goal in mind. In common practice, geological and petrophysical studies are
performed separately and without reference to each other – this can limit their usefulness
and general applicability. The approach used in this study integrates stratigraphic and
petrophysical methods to improve the predictive capacity of the resulting maps.
Database development
The LAS files, core analysis, production data, and raster logs used in this project were
imported into industry petrophysical modelling software and geological mapping
programs (i.e., Schlumberger’s Techlog™ and geoLOGIC’s geoSCOUT™ program) in
the Petroleum Geoscience Laboratory at the University of Western Ontario and used to
build a project database of stratigraphic tops and petrophysical calculations. This database
was continuously updated and modified as correlations and parameters were refined. The
final user database is available as an attachment to this thesis.
81
Fig. 5.1: A process flow diagram illustrating the integrated petrophysical and geological workflows
used in this thesis.
Green boxes indicate raw data inputs – the green boxes on the top row are petrophysical inputs,
whereas the left-side green boxes indicate geological inputs. Purple boxes indicate volume models
that were used for the brittleness index (BI), whereas blue boxes indicate items that were not
included in the BI. The grey box encompasses all the results that were used to create sweet spot
maps.
Using geophysical wireline logs and core data as inputs, a volumetric petrophysical model that has
been corrected for clay and kerogen (Vclay + Vdol + VQFM + Vker + fE = 1) is built, and placed into
petrophysical zones (isochores). These properties are used to calculate brittleness indices (BI) for
each allomember. Lastly, each property is placed into an allostratigraphic context and overlain to
create maps of potential sweet spots.
82
Divestco supplied a license to their EnerGISite™ web portal in support of this research.
EnerGISite™ is an online database that contains tens of thousands of LAS curves for oil
and gas wells drilled in western Canada. Within the project area, 227 vertical wells had
LAS data available in the Second White Specks interval – this is a small fraction of the
>20 000 wells drilled in this area. To supplement this, and increase spatial control, raster
log data for a further 217 wells was sourced using geoSCOUT™. LAS data is essential
for petrophysical model development; wells with raster data only can only be used for
stratigraphic correlation. In total, 444 individual wells were correlated.
Raw LAS data was uploaded into the Techlog™ 2015 program (Fig. 5.2), where it was
standardized into log curves with universal names and units. Despite various generations
of data, differing standards, and several types of tools that were used by the different
logging companies, there are fundamental similarities in how the logging data can be used
for petrophysical analysis. Although this is a major simplifying assumption, it allows for
bulk processing of petrophysical data. In this format, LAS data can be easily manipulated
for large groups of wells and processed quickly either with Techlog’s petrophysical
workflows or by custom Python code for functions outside of Techlog’s core capabilities.
Python code written in support of this project is available as an attachment to this thesis
(Appendix B).
On average, a density of approximately three wells per township were used in cross-
sections (refer to Fig. 1.2). This afforded reasonable confidence in allomember geometry
and thickness when correlating stratigraphic surfaces across the study area. In areas like
Willesden Green (Township 41 to 42, Range 4W5 to 6W5), well density was increased to
greater than 10 wells to resolve major discrepancies (e.g., unexpected changes in
allomember thickness) over close well spacing (<1 km) and to simulate production well
spacing. Near the edges of the study area, where not many wells were drilled to sufficient
depth to reach the Second White Specks Formation, well density is typically less than one
well per township.
Due to the highly diachronous nature of some portions of the lower Colorado Group, the
Second White Specks interval is defined lithostratigraphically within geoSCOUT’s
83
Fig. 5.2: A screenshot of the Techlog software used in this project, with individual wells in the left-hand column, the properties of any selected
parameters in the second column, and a cross-section visualized across the main section of the screen.
84
stratigraphic database. This made it difficult to select cores that fell within the interval of
interest (lower interval of the Second White Specks and Upper Belle Fourche) without
first performing low-resolution lithostratigraphic correlations. Once the preliminary
correlations had been completed, it was determined that eight cores vertically intersected
the interval of interest; of those, only one covered the entire interval (07-19-45-06W5).
Only three cores from that subtotal had core analyses available within geoSCOUT. To
supplement the project database of core, five cores were logged and sampled in detail by
the author at the Alberta Energy Regulator (AER) Core Research Centre in Calgary,
Alberta. This process included describing the visible lithology, grain size, composition,
sedimentary and structural features, as well as the trace fossil assemblages.
Core-to-log correlation
To reduce the uncertainty associated with extrapolating sparse core data, 4 cores were
sampled for LECO TOC analysis and major element X-ray fluorescence (XRF). This
helped to constrain lateral variations in organic richness and mineralogy across the study
area, to ensure that any resulting models would represent the subsurface geology.
Although X-ray diffraction (XRD) was planned to aid in the determination of volume %
mineralogy (Quirein et al. 2010), it was not performed due to logistical issues. All of the
core data gathered for this study, including detailed core logs and core analysis, is
presented in Chapter 5.
The 100/07-19-045-06W5 core was not logged by this author, as it had been previously
logged by another student and sampled for Rock-Eval 6 analysis in 2012. There is
abundant existing data for this core in the public domain: it has been extensively analyzed
by Furmann et al. (2014) and Furmann et al. (2016) for XRD and total porosity. In
addition to that work, the AGS has published Rock-Eval data for this core.
5.4.1 Total organic carbon (TOC) determination

TOC measurements are often used as a first-pass evaluation of a formation’s potential to
generate hydrocarbons (Jarvie 1991). TOC analysis was performed in-house by Dr.
Charlie Wu at the Laboratory for Geochemical Analysis at the University of Western
Ontario in London, Ontario using a LECO CS-244 carbon/sulfur analyzer on 51 mudrock
85
samples from the Second White Specks Formation. The CS-244 analyzer uses the dry
combustion-infrared absorption method to calculate total carbon. This method is a
destructive analytical technique that can be used to solve for TOC by difference using the
following relationship (Schumacher 2002):
!" = !$" + !&" [14]
where TIC is the total inorganic carbon (wt%; derived from carbonate minerals), TOC is
the total organic carbon (wt%), and TC is the total carbon in the sample.
Samples were powdered and weighed into 10 to 15 g aliquots. These aliquots are split –
one half (used for TOC) is treated with a HCl solution to digest inorganic carbon
associated with carbonate minerals, the other (used for TC) is not chemically treated.
These powdered samples were then oven-dried at 105oC overnight to remove excess
water. Once they were sufficiently dry, the samples were combusted at 1350oC in an
induction furnace where they were exposed to a stream of pure oxygen with catalysts
added to ensure complete combustion (LECO 1996; Schumacher 2002). The exposure of
the combusted samples to oxygen allowed for the carbon gases to oxidize, forming CO2.
The amount of CO2 produced during the oxidization process is directly proportional to the
carbon content of the sample, although it may be an underestimate if sulfur, water, and
carbonates have not been adequately removed prior to analysis (Law 1999). The resulting
CO2 gaseous phase flowed through scrubber tubes to remove unwanted moisture and
chlorine. The gaseous phase flowed into an nondispersive infrared (NDIR) detection cell
that was calibrated to CO2 (Bernard et al. 1995). The NDIR cell was tuned to the
absorption of infrared energy by CO2 (2.6 to 4 microns). As the gaseous phase entered the
NDIR cell, it absorbed infrared radiation and lost energy. This energy loss is converted by
the NDIR cell to a wt % TOC or TC; TIC is calculated using Equation 14 (Schumacher
2002).
Based on duplicate samples taken for quality assurance, the analytical errors for TOC
ranged from 0.3 to 2.3%. These issues may have resulted from the incomplete removal of
pyrite (which contains sulfur) or carbonate minerals from the sample, or due to
86
heterogeneities between sample splits. The raw LECO TOC analyses are available as an
attachment to this thesis.
5.4.2 X-ray fluorescence (XRF) analyses

Whole rock major oxide XRF was performed on 51 pulverized mudrock samples with a
Philips PW-1480/10 wavelength dispersive spectrometer (WDS or WDX) unit by Dr.
Charlie Wu at the Laboratory for Geochemical Analysis at the University of Western
Ontario in London, Ontario using the fusion bead preparation method. The fusion bead
method, first introduced by Claisse (1957), is an effective sample preparation technique
for accurate XRF analysis of powdered rock samples because it eliminates effects
associated with heterogeneities in grain size and mineralogy (Yamada 2010; Homma et
al. 2012). Samples were powdered and weighed into 1.0 to 2.0 g aliquots. These
powdered samples were then oven-dried at 105oC overnight to remove excess water.
After drying, the samples were placed into crucibles and weighed. These crucibles were
heated in a furnace at 1000oC for several hours, combusting the sample. Following
cooling, the crucible and now combusted LOI sample were reweighed; the difference in
weight was recorded as the percentage of sample lost on ignition, or LOI (King & Vivit
1988). LOI represents the organic carbon compounds and hydrous minerals such as
carbonate or clay that were volatized during this stage (Yamada 2010). A 0.50 g aliquot
of the combusted sample was fused with lithium borate (Li2B4O7) flux to form a fused
disc for XRF analysis.
Modern x-ray fluorescence (XRF) spectrometry is a non-destructive method often used to

obtain concentrations for major and trace elemental rock analyses due to its relative
simplicity and cost-effectiveness over other geochemical processes (La Tour 1989; Henry
& Goodge 2016). Abbott (1948) built the first commercial x-ray fluorescence
spectrometer, which has since been improved upon (Arai 2007). When electrons with
sufficient energy (15 to 20 keV) impact a target, they generate x-rays (electromagnetic
radiation in the 0.01 to 10 nanometer wavelength band) and associated electrons with
wavelengths proportional to the atomic number of the target (Roentgen 1896; Moseley
1913; Henry & Goodge 2016).
87
In WDS XRF spectroscopy, a single crystal is used to disperse fluorescent X-rays induced
in a rock sample via a collimated electron beam. The crystal is selected such that only
those induced X-rays with specific wavelengths can travel in the spaces between crystal
lattice planes (a phenomenon known as “Bragg diffraction”) and are reflected by the
crystal towards a detector (Bragg & Bragg 1913; Kawahara & Shoji 2007). An X-ray
detector is used to convert the energy released by the reflected X-ray into an electrical
signal, which corresponds to the concentration of a particular element in the sample
(Longoni & Fiorini 2007). Crystals with different lattice spacings are used to isolate the
concentrations of different elements (Henry & Goodge 2016).
The 51 fused samples were analyzed using the Philips PW-1480/10 WDS XRF unit for
the following major oxides: SiO2, TiO2, Al2O3, Fe2O3, MnO, MgO, CaO, K2O, Na2O,
P2O5, Cr2O3, BaO, and SrO. 4 synthetic standards from certified reference materials (SY-
2, R.V., JB-2, and JG-1A) were used as standard samples for calibration.
Based on duplicate samples taken for quality assurance, the analytical errors for XRF
ranged from 0.2 to 3.1%. Similar to the LECO TOC analyses, these issues may have
arisen from heterogeneities between sample splits. The raw XRF analyses are available
as an attachment to this thesis.
5.4.3 Normalized mineralogy

XRF provides major oxide element concentration, not volume % mineralogy. Due to
unforeseen circumstances, XRD could not be performed and another method for
determining volume % mineralogy was required that could utilize the data that had
already been gathered. Normative mineralogical calculation, such as the calculation of a
CIPW norm, is a routine procedure in igneous geochemistry wherein bulk geochemical
analyses are inverted to estimate the mineral assemblage for a particular rock sample
(Cross et al. 1902; Verma et al. 2003). In this process, it is assumed that the composition
of the magma that the igneous rock crystallized constrains its mineralogy rock to a series
of idealized mineral components (Sen 2014). In contrast, normative mineralogical
calculation is rarely done for clastic sedimentary rocks. This is because clastic mineralogy
is affected by diverse independent variables, including sediment provenance, weathering
88
reactions, erosion processes, transport and depositional processes, pore water and brine
chemistry; and early through late diagenetic reactions (Kackstaetter 2014). Despite this,
Rosen et al. (2004) stated:
“[there are] significant statistical regularities in the

mineralogical compositions of sedimentary rocks […] that can
be used to provide pointers to [their] mineralogical
compositions.”
Kackstaetter (2014) developed a normative mineralogy method that quantifies and applies
the assumptions made by Rosen et al. (2004) to sedimentary rocks. The Kackstaetter
(2014) method partitions major element concentrations from XRF into sedimentary
minerals with idealized chemistry. This method is not proven, but it provides a first-pass
estimation of mineral composition.
Although the Kackstaetter (2014) method is fundamentally limited because the mineral
compositions it uses are underdetermined (Spencer & Weedmark 2015), it provided a
consistent match with published mineral compositions from Furmann et al. (2016) in the
key well, 07-19-45-06W5. For this reason, the Kackstaetter (2014) method was used to
convert all XRF data points into relative mineral compositions. The spreadsheet that
Kackstaetter (2014) developed and was used in this thesis is available from
http://earthscienceeducation.net/SEDMINSedimentaryMineralCalculator.xlsx. Ideally, the
Kackstaetter (2014) method should be supplemented with sulfur content from LECO
TOC method to estimate pyrite concentration. The Second White Specks and Belle
Fourche Formations are known to contain some amount of pyrite (~5%), but pyrite could
not be reliably modelled. As a consequence, pyrite was not included in the multimineral
model.
Allostratigraphic correlation
The allomembers mapped in this study are extensions of and subdivide the framework
developed by Tyagi et al. (2007). This work extended their allomember VII (lower
Second White Specks) throughout the area and is bounded below by the Bighorn River
“red” bentonite. Tyagi et al. (2007)’s allomember VI (Upper Belle Fourche), which is
bounded above by the “red” bentonite and below by the K1 unconformity, is subdivided
89
in this work into three separate allomembers (Upper Belle Fourche 1, 2, and 3).
Petrophysical zonations were created by importing the allostratigraphic tops from
geoSCOUT into the Techlog project – these zones served as the vertical constraints for
the petrophysical models described below.
Petrophysical modelling workflow

This section outlines the petrophysical modelling process implemented in this thesis,
drawn from the petrophysical best-practices outlined in Chapter 3. A petrophysical model
was developed by adapting previous models made for unconventional mudstones;
Ambrose et al. (2010) was used as a template with some simplifications (Fig. 5.3 and
5.4); most notably, the model used here does not include pyrite or immobile
hydrocarbons. Selecting this petrophysical model restricted the number of unknowns and
identified the components that could be calculated. The petrophysical model used here
only has seven components – in comparison, the unconventional petrophysical model
illustrated in Fig. 3.2 can have greater than nine components. Petrophysical complexity
increases with the number of unknowns, increased porosity modality, the introduction of
microporosity, and by introducing matrix and fluid components with varied density and
conductivity.
All of the procedures described below were implemented using built-in Techlog modules,
or custom Python code written by the author for this project. Environmental corrections
were applied when borehole conditions were poor, and where feasible with the available
log data. The petrophysical modelling process was heuristic and iterative – methods were
improved and altered as new problems were encountered and new data was introduced.
The Techlog Quanti.Elan module was used to calculate weighted averages of each
parameter within the allostratigraphic zones established in section 6.2. The project
specific parameters and techniques are summarized in Table 5.1 where they can be easily
accessed by the reader but will be described in detail in the following section.
LAS data is recorded every 0.1 to 0.2 m. The petrophysical calculations listed here, unless
derived from core data, have the same sampling rate as the raw data.
90
Fig. 5.3: Modified from Ambrose et al. (2010) – this is the petrophysical model used for the Second White
Specks in this study.
91
Fig. 5.4: Modified from Asquith (1990) – this represents the previous petrophysical model (Fig. 5.3)
within a very simplified geological context.
92
Equations used for petrophysical modelling in the 2WS
Type Equation Remarks
Total organic carbon A = -0.0125

Issler et al. (2002) !&" = ' ∗ (∆+ + 195 ∗ log23 4) − 31.86
[19]
Kerogen volume @<=> K = 0.8
Guidry et al. (1990) ;<=> =
A<=> A<=> = 1.15 g/cm3
Crain and Holgate (2014) BCD∗EFGH
where @<=> = and K <1
[20] E∗I
]4JK=^_ selected using

Clay volume ;JK = 1.7 − M3.38 − (;NO + 0.7)Q 5th percentile cutoff
Clavier et al. (1971) RSTUV WRSXTGYZ
where ;NO = RS ]4NO^K= selected using
[3] and [4] [\YTG WRSXTGYZ
95th percentile cutoff
Total porosity Ab^ = 2.74 g/cm3

e
Sondergeld et Ab^ − Ac ∗ dAb^ ∗ A IaS − eIaS + 1f
al. (2010) `B IaS = IaS AgK = 1.05 g/cm3
Ab^ −AgK
[24] AIaS = 1.15 g/cm3
Void
volume Effective Used mixed-layer
porosity (illite) density.
Ai>jJK^j − Ac
Doveton à = `B IaS − ;JK h m
Ak=lJK^j − AgK Ai>jJK^j = 2.7 g/cm3
(2014)
[26] Ak=lJK^j = 1.7 g/cm3
Carbonate volume eJ^>c

;J^>c =
Adapted from AJ^>c
Jiang et al. (2017) Qn
where eJ^>c = n
!$" AJ^>c = 2.87 g/cm3
[28]
Linear regressions ;$$ |;ipK = 7.01 × ;<=> + 0.0168
[29] and [30] rs3 |;ipK = 24.8 × ;<=> + 0.476
Matrix volume Model assumes a 5-

[31] ;vwx = 1 − ;JK − ;J^>c − ;<=> component system
where porosity is zero
Brittleness index Assumes only QFM

Adapted from Mathia et al. ;vwx + ;J^>c and carbonate
r$ =
(2016) ;vwx + ;J^>c + ;JK + ;<=> contribute to brittle
[33] behaviour
Table 5.1: Table of equations and parameters used for petrophysical modelling in this project. All
variables in red font are measurements taken directly from the well logs.
93
5.6.1 Environmental corrections

All wireline logging companies provide unique environmental correction charts for their
tools to account for changes in borehole environment (Dresser 1979; Glover 2000;
Schlumberger 2009). These corrections, in theory, adjust raw log measurements to bring
them back to standard conditions. The accuracy of these corrections decreases with
increased borehole diameter, mud cake thickness, and magnitude of invasion (Glover
2000). The Second White Specks Formation is notorious for bad hole conditions, making
borehole corrections an essential part of petrophysical assessment. Previous studies of the
Second White Specks Formation have not adequately corrected for borehole
environmental effects. This reduces the reliability and precision of models build using
said data (Knox 1975; Patchett & Coalson 1979; Howard & Hunt 1986; Guidry et al.
1990; Pursell et al. 1998). Data scarcity, however, presents a significant barrier to
extensive application of borehole environmental corrections. Certain variables (e.g.,
drilling mud density, mud filtrate resistivity) are required to compute correction factors.
This data, which is inconsistently recorded in log headers, is difficult to access.
5.6.1.1 Gamma ray corrections

Where caliper logs and drilling mud data were available, raw total GR was corrected for
hole size ("'y, in mm) and mud weight (z@!, in kg/m3) using the following general
equation (Dresser 1979; Crain 2018):
]4D = ]4[1 + 0.000322(z@! − 1000)][1 + 0.0024("'y − 203)] [15]
where ]4D is the corrected GR in API units. The magnitude of corrections ranged from 0
to 60 API units. Where a gamma ray correction has been applied, deflections in the
character of the gamma ray curve are more likely to represent bulk lithological changes
rather than borehole breakout. This provides increased confidence to use gamma ray logs
for correlation even in poor borehole conditions.
94
5.6.1.2 Density log corrections

If the computed density correction reading exceeded ±0.15 g/cm3, the log was considered
invalid and not used for porosity calculation. No further corrections were applied to the
density logs as mudcake density and mudcake thickness were not known (Glover 2000).
5.6.1.3 Resistivity corrections

The “shoulder-bed effect” in shales can be corrected via a nonlinear joint inversion
process (Cosmo et al. 1991; Warrilow et al. 1995; Heidari et al. 2012; Heidari & Torres-
Verdin 2014), but this process can introduce significant uncertainty if calibration data is
not available (Yin 2011). For this reason, no environmental corrections were applied to
the resistivity log.
5.6.2 Volume modelling

Mineral components in a particular reservoir are preferably determined using core data,
but core data is limited. The availability of core data and lithological logs (e.g.,
lithodensity and PEF) are limiting factors on mineralogical analysis. Methods for the
compositional analysis of rocks using wireline logs have evolved concurrently with
improvements in logging technology. To extend mineralogical models from key wells
with core to the maximum number of possible wells, petrophysical models must be as
simple as possible while still respecting the geology (Moore et al. 2016). For
multimineral compositional analysis, it is important to find a technique that balances the
true lithological heterogeneity of the reservoir while simultaneously respecting the dataset
limitations (i.e., log suite).
When a sufficient variety of geophysical well logs are available, all mineral component
volumes can be confidently resolved (Doveton 1994b). More often, however, there are
insufficient logs available to solve for the volumes of all the minerals present in a
formation. The number of linear equations based on geophysical well log responses, in
relation to the number of desired unknowns (mineral volumes), designates the system as
“underdetermined” (more volumes than well logs), “overdetermined” (fewer volumes
than well logs), or “balanced” (equal number of well logs and volumes). In balanced and
95
overdetermined systems, a unique compositional solution can always be found. This is

not the case for underdetermined systems, which have infinite possible solutions
(Mitchell & Nelson 1988). In unconventional reservoirs, which have many mineralogical
components, petrophysical models are usually underdetermined. This problem also arises
in older (pre-1980s) fields where mineralogical logging was never implemented. When
the geophysical well log database contains insufficient data to define a balanced or
overdetermined system, either linear inversion, nonlinear joint inversion, or deterministic
(sequential) methods can be used to estimate composition (Mitchell & Nelson 1988; Ellis
& Singer 2008; Sondergeld et al. 2010).
Many algorithms exist for linear inverse estimation of mineral compositions (Mayer &
Sibbit 1980; Quirein et al. 1986; Doveton 1994b; Rabaute et al. 2003). Implicit in these
programs is the assumption that geophysical wireline log measurements (i.e., density,
PEF, neutron porosity, and sonic transit time) are linear functions of the mineral and fluid
volume concentrations in the wellbore. Light hydrocarbons, organic matter, and clay
minerals render this assumption invalid (Heidari et al. 2012), as the nonlinear responses
of those lithologies to certain well logs (e.g., GR and neutron porosity) are well
established (see section 3.6.2). This can be mitigated through the use of nonlinear
minimization and statistical algorithms (e.g., cluster analysis and principal component
analysis) to better constrain lithology (Mitchell & Nelson 1988; Doveton 1994b; Rabaute
et al. 2003). Linear inverse estimation methods also require a priori constraints (i.e., core
data) to achieve unique compositional solutions, as the system of equations used is almost
always underdetermined (McCarty et al. 2015).
Nonlinear joint inversion methods (Liu et al. 2007; Sanchez-Ramirez et al. 2009; Heidari
2011; Heidari et al. 2012; Gao et al. 2013; Heidari & Torres-Verdin 2014; Klenner et al.
2014; McCarty et al. 2015) modify the generalized nonlinear least squares method,
developed by Tarantola and Valette (1982), to obtain estimates of petrophysical
parameters from geophysical well logs. These methods have been used to estimate
mineralogy where linear methods had previously failed. Heidari et al. (2012), for
example, demonstrated that nonlinear joint inversion is viable way to estimate complex
lithology in a thinly-bedded invaded reservoir. Despite these recent successes, nonlinear
96
joint inversion methods are not always recommended for petrophysical modelling –
particularly when calibration data is scarce (McCarty et al. 2015). Application of these
methods may result in model instability (i.e., when small changes in inputs significantly
affect model outputs) and non-unique results (Tarantola & Valette 1982).
Deterministic petrophysical methods estimate each component in a semi-independent

fashion (Sondergeld et al. 2010). After each individual component is solved, usually in a
specific sequence, the results are used as input into the next model (Mitchell & Nelson
1988). These methods, such the processes described by Poupon et al. (1971) and Clavier
and Rust (1976), are economical but are often only valid in one type of formation (e.g.,
shaley sands) – this restricts their wider application. It is also difficult to modify the
logical steps in a sequential petrophysical modelling method to accommodate new types
of data (Ellis & Singer 2008).
Developing a linear or nonlinear inversion method for mineralogy was beyond the scope
of this project. A deterministic method was applied instead: after determining other
volume components (kerogen volume, clay volume, porosity), quartz and carbonate
volumes were calibrated to core data and then solved by difference for the total volume.
Due to the limited geophysical log data set available for this study area, it was only
feasible to solve for volumetric components in a reservoir with five or fewer assumed
components. The petrophysical model from Fig. 4.3 was further simplified to a five-
component system because water saturation and hydrocarbon saturation could not be
reliably modelled. Each volume component (Vclay, Vdol, VQFM, Vker, and fT) was
independently calculated in Techlog using methods that will be described below. The
volume components of a formation with the petrophysical model from Fig. 5.3 (clay
volume, Vclay; kerogen volume, Vker; dolomite volume, Vdol; quartz-feldspar-mica
volume, VQFM; total porosity, fT) are expressed as fractions of unit bulk volume. This
gives rise to the following equation (Guidry et al. 1990):
1 = ;vwx + ;JK + ;ipK + ;<=> + `B IaS [16]
97
5.6.2.1 Kerogen volume from logs

Kerogen is solid organic matter, rich in carbon, that is found in sedimentary rocks. The
physical properties of kerogen can confound standard log analysis methods, providing
overly optimistic estimates of porosity and fluid saturations (Crain & Holgate 2014).
Quantifying TOC is essential for constraining petrophysical models because the total
amount of kerogen (TOC, wt%; ;IaS , v/v) impacts unconventional play economics
(Haecker et al. 2016). Kerogen that has thermally degraded and produced hydrocarbons
may develop porosity. Producible hydrocarbons can be stored in this porosity(Ambrose et
al. 2010, 2012; Zee Ma et al. 2016). TOC, however, can only be measured directly using
LECO TOC and Rock-Eval methods on rock samples. Geochemical analyses like this are
relatively expensive, time consuming, and subject to sampling bias (Heslop 2010).
Alternatively, petrophysicists can use in situ methods derived from log measurements to
estimate vertical changes in TOC (Labani & Rezaee 2015). In situ methods are
economical, have superior data availability, and sample the vertical succession
continuously (Schmoker 1981). Previous workers have attempted to characterize the
relationship between organic matter and one or more petrophysical well logs (e.g. gamma
ray, density, sonic, and resistivity) with varying degrees of success. These models are
summarized and compared in Table 4.2; the D}~4 and Issler et al. (2002) methods are
discussed in detail in the following section.
5.6.2.1.1 DlogR or Exxon method

The presence of organic matter in mudstones increases the apparent sonic porosity and
elevates resistivity (see section 3.6.1). Passey et al. (1990) developed an empirical
relationship between TOC, elevated formation resistivity, and apparent porosity. In their
method, apparent porosity (usually using ∆+ as a proxy) is plotted over resistivity and then
scaled so that the two curves overlay as a baseline in zones with organic-poor shales
above and below the organic shale. Zones that do not lie along this baseline appear as a
curve separation. This curve separation, called D}~4, is related to total organic content.
The algorithm for calculation of D}~4 from the overlay of sonic and resistivity is
(Passey et al. 1990):
98
TOC equations
Type Equation Physical basis Remarks

AÄ − A
;IaS = Bulk density of Sensitive to severe
Density 1.378 kerogen is low borehole rugosity,
Schmoker where AÄ is the density of an organic- relative to matrix which is common in
(1979) free zone and A is the bulk density density (Tissot & shaley formations
measurement from the logs Welte 1978) (Haecker et al. 2016)
Assumes inorganic
154.497 Bulk density of
Density !&" = density = 2.69 g/cm3;
A − 57.261 kerogen is low
underestimates TOC in
Schmoker and relative to matrix
where A is the bulk density clay- and carbonate-
Hester (1983) density (Tissot &
measurement from the logs rich rocks (Gonzalez et
Welte 1978)
al. 2013)
ÅÄ − Å
;IaS = Total GR is Response conflates
1.378 ∗ '
GR elevated in shales radioactivity from clay
where ÅÄ is the GR intensity of an due to clay and (40K and 232Th) with
Schmoker organic-free zone and Å is the GR organic matter radioactivity from
(1981) intensity measurement from the logs (Ellis & Singer organic matter (Fertl &
2008) Chilingar 1988)
A is a locally determined calibration
SGR SGR is infrequently run

40
K and 232Th are (Ellis & Singer 2008).
Fertl and associated with
Chilingar clay, 238U occurs Correlation of 238U
Correlated [238U] with TOC and ;IaS with TOC can vary
(1988) in organic matter
(Ellis & Singer within a single
Guidry et al. formation (Cluff &
2008).
(1990) Miller 2010).
Resistivity Shales have

D}~4 = log23 d4É4 f + 0.02 Sensitive to LOM,
and sonic c^N=KÇ_= higher resistivity
∗ (∆+ − ∆+c^N=KÇ_= ) requires normalization,
and increased
Passey et al. and not calibrated to
!&" = D}~4 ∗ 10(Q.QÑÖW3.2Üáá∗àCx) ∗ " sonic porosity
(1990) local geology (Issler et
relative to non-
where C is a compaction component <1 al. 2002)
[17] and [18] shaley units
Resistivity
and sonic Shales have
higher resistivity Calibrated to local
Issler et al.
and increased geology, no
(2002); Crain !&" = ' ∗ (∆+ + 195 ∗ log23 4) − 31.86
sonic porosity normalization, and
and Holgate
relative to non- independent of LOM.
(2014)
shaley units
[19]
Table 5.2: Table of TOC equations, with remarks related to their physical basis and limitations.
99
D}~4 = log23 d4É4 f + 0.02 ∗ (∆+ − ∆+c^N=KÇ_= ) [17]

c^N=KÇ_=
where D}~4 is the curve separation measured in logarithmic resistivity cycles, 4 is the
resistivity measured in ohm×m by the logging tool, ∆+ is the measured transit time in µs/ft,
4c^N=KÇ_= is the resistivity corresponding with the ∆+c^N=KÇ_= value when the curves overlie
in organic-poor (non-source), clay-rich rocks.
D}~4 separation is linearly related to TOC and is a function of the level of organic
metamorphism (LOM), as well as kerogen type. The level of organic metamorphism
(LOM) represents the degree of thermal metamorphism a particular mudstone has
undergone, which increases with burial depth and increased geothermal gradients (Hood
et al. 1975). Good estimates of TOC from DlogR require a priori knowledge of y&z and
calibration to TOC from rock samples – this is difficult when geochemical analyses are
scarce. Sondergeld et al. (2010) suggested a modification to the Passey equation for y&z
to prevent underestimates of TOC in overmature gas shales, which have LOM > 14 :
!&" = D}~4 ∗ 10(Q.QÑÖW3.2Üáá∗àCx) ∗ " [18]
where " is a multiplier > 1 that is calibrated to measured !&" from core.
The Passey et al. (1990) method is practical and easily applied. It does have drawbacks: it
requires similar clay mineralogy and conductive mineral components (i.e., pyrite) in the
source and non-source intervals. The method also assumes that the baseline shale and
organic shale have similar matrix properties (i.e., resistivity and apparent porosity). This
may not be representative of the formation’s true vertical heterogeneity. Graphical log
overlay methods like D}~4 require log normalization, and the baseline determination
can be subjective (Issler et al. 2002).
The D}~4 method can produce false positives (Passey et al. 1990). Anomalous D}~4
separation not associated with organic shales can be caused by:
1) hydrocarbon-bearing reservoir intervals,
2) poor borehole conditions,
3) uncompacted sediments,
100
4) low porosity intervals,

5) volcanics, and
6) evaporites.
1), 2) and 4) are characteristic of Lower Colorado Group shales – it is likely that using the
D}~4 method would overestimate TOC in this interval. Due to a lack of geochemical
analyses and the uncertainty introduced by the D}~4 method, this method was not used
in this thesis for evaluation of the lower Colorado Group.
5.6.2.1.2 Issler et al. method

Using core measurements of WCSB Colorado Group rocks, Issler et al. (2002) developed
a porosity-resistivity method for calculating TOC that does not require baselines,
normalization, or a priori knowledge of y&z. Crain and Holgate (2014) provide a
multiple regression of the Issler et al. (2002) graphs, which gives the following general
equation:
!&" = ' ∗ (∆+ + 195 ∗ log23 4) − 31.86 [19]
where A is a local calibration that is < 0.1.
5.6.2.1.3 Application of TOC modelling to the lower Colorado

Group
The Issler et al. (2002) method for TOC calculation was chosen to calculate TOC. The
general equation derived by Crain and Holgate (2014) from multiple regression of the
Issler et al. (2002) graphs was coded into Techlog (Equation 19). A was set to -0.0125 for
this study area. This value for A was chosen on the basis of optimal modelled TOC curve
matching with Rock-Eval 6 TOC data (Fig. 5.5) collected by another student from the
Western University Petroleum Geology Laboratory. This core, from the 100/07-19-045-
06W5 or “7-19” well, vertically samples the entire interval of interest for this study. Fig.
5.5 shows a strong linear relationship between the Issler model predictions and Rock-Eval
6 calibration data. This relationship is also visible in the vertical profile of TOC shown in
Fig. 5.6. On the basis of this relationship, the Issler et al. (2002) method was applied to all
101
Fig. 5.5: Issler model TOC vs. Rock-Eval TOC analyses from the core study interval in the 100/07-
19-045-06W5 well. The regression shows a strong linear relationship with a moderately high
coefficient of determination (r2 = 0.811).
102
Fig. 5.6: A log display of the 7-19 well, showing the sonic and resistivity curves that were used as
input for the TOC model as well as the porosity model.
The track second from the right shows the Rock-Eval core data (TOC_G) as black circles overlaid
on the modelled TOC (blue and red fill) in the Belle Fourche and Second White Specks
Formations; the TOC curve reflects the log character of both the sonic and resistivity curves.
On the far right, black circles indicate the helium porosimetry data points taken from Furmann et
al. (2014). These are plotted against the kerogen-corrected total porosity model (PHIT_CORR)
and the effective porosity model, shown as blue-filled areas. The kerogen-corrected total porosity
is a good fit for the core porosities, generally agreeing within a percent.
103
eligible wells with LAS, deep resistivity, and sonic; this constituted 84 of the 227 wells
with LAS data in the study area. Modelled TOC values ranged from 0.5 to 5% in the area.
Petrophysical models require components reported in volume percent, not weight percent.
To do this, TOC (weight %) must be converted to kerogen volume (;<=> ). Organic
petrography is a recommended step in developing robust kerogen density measurements
because organic maceral chemistry and LOM affect kerogen density (Stankiewicz et al.
2015). Crain and Holgate (2014) use the following equation for kerogen volume %
(;<=> ), which is derived from the work of Guidry et al. (1990):
!&" ∗ A<=> [20]

A∗â ä
;<=> = A<=>
where A<=> is the density of kerogen, and â is a kerogen conversion factor <1. Although
A<=> can be derived from geochemical data, Okiongbo et al. (2005) cautioned that
kerogen in organic shales is difficult to chemically separate from pyrite, which results in a
wide range of reported densities for kerogen in specific formations. Although estimating
A<=> becomes more difficult without organic geochemical data, Ward (2010) suggested
the following relationship between vitrinite reflectance (%4C ) and A<=> in the Marcellus
Shale:
A<=> = 0.342 ∗ %4C + 0.972 [21]
A<=> was set to 1.15 g/cm3 based on %4C data from Furmann et al. (2014) and Equation
21. This value for A<=> falls within the theoretical limits for kerogen density of 1.1 to 1.4
g/cm3 set by Tissot and Welte (1978). Lacking more %4C data for local calibration, this
A<=> value was used for the entire Willesden Green field.
5.6.2.2 Clay volume from logs

Clay quantification in any reservoir is key to building a good petrophysical model,
because virtually all log measurements are affected by the volume fraction and type of
clay minerals present in a formation (see section 3.6.1). This is especially true of the
Second White Specks Formation, which can contain greater than 40% clay by volume
(Furmann et al. 2014). Clay volume would ideally be determined via x-ray diffraction
104
(XRD) – unfortunately, this type of mineralogical data is slow to process (i.e., in time to
influence production geology decisions) and is relatively expensive (Ellis & Singer 2008;
Doveton 2014). Similar to other geochemical data sources, XRD results may not be
representative due to vertical sampling bias. Total gamma ray scaling techniques,
developed for the purposes of shaley sand corrections (see section 3.5.2), are more
economical. These scaling techniques utilize the ability of the gamma ray log to detect
potassium associated with clay minerals (Asquith 1990; Krygowski 2004).
Total gamma ray clay volume techniques can be highly subjective and err by
overestimating clay volume. The total gamma ray curve includes uranium counts that are
not associated with clay minerals. This, however, is often the only alternative if core is
not available (Ellis & Singer 2008). Clay volume overestimation can be mitigated through
scaling techniques that use the uranium free (Th/K) curve from SGR logging when
available. Th/K is a better indication of clay volume because the uranium-free curve (see
section 3.6.2.1.2) isolates GR intensity to potassium and thorium associated with clays
and accessory zircons (Krygowski 2004).
5.6.2.2.1 Application of clay volume to the lower Colorado Group

After applying environmental corrections to the gamma ray logs, shale volume (;NO ) was
calculated using 5th percentile cutoffs for ]4bÇ_ and 95th percentile cutoffs for ]4b^å
over the entire lower Colorado Group interval. These cutoffs were individually set for
each well in Techlog, ensuring that ;NO would be correctly calibrated for each data point.
;NO was calculated using Equation 3 and corrected to estimate clay volume (;JK ) using
Equation 4. Published clay XRD data from Furmann et al. (2014) in the 100/07-19-045-
06W5 well was used to calibrate the ;JK model.
5.6.2.3 Porosity from logs

Porosity (`) is the total nonsolid volume in a reservoir. This includes all pores, fractures,
vugs, intercrystalline and intracrystalline volumes (Glover 2000). All porosity equations
are of the form:
105
;çp>= ;céK< − ;b^l>Çå [22]

` = =
;céK< ;céK<
where ;çp>= , ;céK< , and ;b^l>Çå are pore volume, bulk volume, and matrix volume,
respectively. An organic-rich mudstone like the Second White Specks is normally
composed of solid clay, non-clay, and organic components, with pore space between and
within them (Ambrose et al. 2012). Ideally, porosity models should be calibrated with
porosities measured from core and log-derived porosities to ensure accuracy (Glover
2000). This is relatively easy to do in conventional plays, where core analysis is
frequently performed and readily available. The fundamental physical properties of
unconventional shales (i.e., low porosity, <10%; and low permeability, <100 nD) make
them very difficult to measure using conventional core analysis methods (Ramirez et al.
2011; Handwerger et al. 2012; Busch et al. 2017). As a result, the porosity of
unconventional reservoirs is often approximated using geophysical wireline logs.
5.6.2.3.1 Total porosity

Total porosity (` B ) is the ratio of total void space in a reservoir to its bulk volume (see
Figs. 5.3 and 5.4). It is the combination of intergranular connected porosity, isolated
(inaccessible) porosity, and apparent clay porosity (Byrnes 1997; Moore et al. 2016). A
linear relationship between bulk density and porosity is commonly utilized to create a
density porosity curve by rearranging Equation 8 (Davis 1954; Baker 1957):
Ac − Ab^ [23]
`è =
AgK − Ab^
The matrix and fluid densities, Ab^ and AgK , respectively, need to be assigned values to
calculate `è . A quartz matrix density of Ab^ = 2.65 g/cm3 and a fluid density of AgK =
1.00 g/cm3 (water) is commonly assigned. This, however, can be inappropriate in
unconventional reservoirs like the Second White Specks Formation, which are
mineralogically complex and contain unknown fluids. Matrix density must be determined
from core, when available, and calibrated to cross-plots of other logging measurements
(i.e., neutron porosity and bulk density). Fluid density, AgK , should approximate mud
106
filtrate density (³ 1.00 g/cm3) due to filtrate invasion into the formation. Using this
approach can be effective for generating reasonable porosity estimates (Byrnes & Castle
2000), although its application is restricted in petrophysical modelling approaches that
rely on basic log suites (Moore et al. 2016). Barring borehole environmental concerns,
organic content and clay minerals are the primary sources of petrophysical uncertainty in
the Second White Specks when trying to calculate total porosity. Calculating total
porosity with multiple independent parameters, such as with the neutron porosity and
sonic logs, can help reduce this uncertainty to some extent (Ellis & Singer 2008).
Sonic logs can also be used to calculate porosity, with some caveats (Table 4.3). Organic
mudstones can influence sonic porosity, depending on the density and volume of the clay
minerals in the reservoir, as well as their level of compaction (Wyllie et al. 1956). Lower
density and lower compaction rates increase sonic transit times, and result in higher sonic
porosities (Glover 2000). This effect could be difficult to parse in an unconventional
reservoir like the Second White Specks, which is compacted, but also rich in low-density
organic material. Zee Ma et al. (2016) noted that acoustic logs are notoriously difficult to
calibrate to reservoir properties compared to neutron and density logs. This suggests that
sonic porosity should be used with caution in organic-rich reservoirs. Finally, neither
Wyllie et al. (1956) or Raymer et al. (1980)’s method for calculating sonic porosity
should be used without numerous core data points for calibration of matrix sonic transit
time – this is not an option when core data is limited (Glover 2000).
The effect of clay minerals on neutron log porosity is well understood – emitted neutrons
are slowed by contact with hydroxyls found in the clay structure, causing an increase in
neutron porosity (see section 3.6.2.1.4). Mud filtrate invasion and hydrocarbons have
similar effects (Glover 2000). Organic matter and thermal maturity impact apparent
neutron porosity, but their influence is less pronounced on neutron porosity measurements
compared to sonic porosity readings (Zee Ma et al. 2016). In organic-poor and clay-rich
units, neutron porosity is typically greater than density in the range of 3 to 10%; however,
organic-rich mudstones can have elevated neutron porosity in addition to elevated density
porosity (Krygowski 2004). Without a priori knowledge of TOC and clay content, this
107
Porosity equations
Type Equation Remarks
Total porosity ;çp>= ;céK< − ;b^l>Çå

`B = = Porosity is a volume fraction
[22] ;céK< ;céK<
Total porosity Ac − Ab^ Formula is derived from a mass-

`è =
AgK − Ab^ balance relationship
Density
where Ac is the measured density from Ab^ and AgK inputs are commonly 2.65
Davis (1954) logs, Ab^ is the assumed matrix density g/cm3 and 1.00 g/cm3 in sandstone
and AgK is the assumed fluid density density porosity logs, this is not
[23] (usually mud filtrate) necessarily representative
∆+ − ∆+b^ 1 Measures connected interparticle

`ê = ∗
∆+gK − ∆+b^ "ç porosity, works best in consolidated
Total porosity
and compacted formations; cannot
∆l
[\ ∗D handle vugs
Compressional where "ç = 233 (compaction correction
sonic factor) and 1.0 < " < 1.3 depending on
"ç added for shaley rocks, shale
regional geology,
increases ∆+ relative to clay volume
Wyllie et al.
(1956) and ∆+NO is the sonic transit time in the
adjacent shale Good for carbonates when porosity is
relatively uniform
Calculated by tool
Total porosity Neutrons are slowed by collisions with
Tool measures size of neutron cloud by hydroxyls (found in clays, pore fluid,
Neutron characterizing falloff of neutrons between drilling fluid)
two detectors
108
Corrects for observed anomalies and

Total porosity shortcomings of the Wyllie equation;
no need for compaction correction
2É
Compressional `N = −í − [í Q + (∆+b^ ⁄∆+) − 1] Q
Based on larger sample of
sonic
∆+b^ measurements and accommodates the
where í = î ä2∆+ ï − 1 entire porosity range
gK
Raymer et al.
(1980) Second-order model to be used when
improved accuracy is key
Total porosity
`B IaS
e
Density, Ab^ − A ∗ dAb^ ∗ AIaS − eIaS + 1f
IaS
kerogen- =
Ab^ −AgK Corrects for kerogen contribution to
corrected
density log, resulting in a lower
where A<=> = 0.342 ∗ 4C + 0.972 porosity estimate
Sondergeld et
al. (2010) BCD∗EFGH
and ;<=> = E∗I
where K <1
[24]
Effective
porosity Corrects total porosity for porosity
à = ` B − ;JK (`JK ) contribution from clay
[25]
Effective
porosity
Ai>jJK^j − A Corrects total porosity for porosity

Density
`è a = ` B − ;JK h m contribution from specific clay
Ak=lJK^j − AgK minerals
Doveton (2014)
[26]
Table 5.3: Table of total and effective porosity equations, with remarks related to their usefulness for
unconventional reservoirs.
109
effect can be difficult to isolate without extensive local calibration in nearby organic-poor
intervals. Neutron porosities, therefore, were not used in this study.
Some lithological effects on total porosity can be corrected for, like the presence of
kerogen. Kerogen can have a significant impact on total porosity measurements that
utilize the bulk density log – resulting in `è values that are unrealistically high (Crain &
Holgate 2014). Sondergeld et al. (2010) demonstrated that the kerogen contribution to
total porosity could be suppressed by isolating the contribution of kerogen to the bulk
density measurement. They did this by combining Equations 20 and 21 with Equation 23:
e [24]
Ab^ − A ∗ dAb^ ∗ A IaS − eIaS + 1f
IaS
` B IaS =
Ab^ −AgK
where `B IaS is the kerogen-corrected total porosity.
5.6.2.3.2 Total porosity for the lower Colorado Group

For wells in which bulk density logs were available, total porosity was calculated using a
matrix density of 2.74 g/cm3. This matrix density was selected using a neutron porosity-
density cross-plot (Fig. 5.7), because core bulk density measurements were not available.
Both the bulk density log (g/cm3; y-axis) and the neutron porosity log (v/v, x-axis) are
measurements of total porosity – points of constant porosity in a given pure lithology will
fall along a diagonal line. When a linear regression is plotted through the point cloud of
porosity measurements and extrapolated back to a neutron porosity reading of 0% (the y-
intercept), the line should intersect the y-axis at a bulk density measurement equivalent to
a rock volume of 100%. The y-intercept should approximate the matrix density (Glover
2000). A matrix density of 2.74 g/cm3 is higher than quartz (2.65 g/cm3) – it approaches
either the bulk density of dry illite clay (2.79 g/cm3), or a mixture of matrix minerals that
could include dolomite (2.87 g/cm3) and clay minerals in addition to low density organic
matter (~1.15 g/cm3).
110
r2 = 0.178
Fig. 5.7: A multi-well crossplot with 1946 bulk density and neutron porosity data points from 5 wells
(102/14-31-41-06W5, purple; 100/15-36-40-6W5, pink; 100/16-28-37-05W5, brown; 100/14-18-37-
7W5, yellow; and 100/10-16-042-06W5, light green) throughout the Second White Specks and Belle
Fourche intervals. The y-intercept of the regression is an estimate of the matrix density (100% rock
volume) of the Second White Specks and Belle Fourche Formations.
111
The fluid density variable, AgK , was set to a constant value of 1.05 g/cm3 to approximate
mud filtrate invasion into the formation. Fluid densities in the range of 1.02 to 1.125
g/cm3 were reported in the log headers of wells in the project area, and 1.05 g/cm3
approximated an average fluid density value. The total porosity was corrected for low-
density kerogen using Equation 24 (Fig. 5.8). Furmann et al. (2014) provided the only
available helium porosimetry data for the Second White Specks in the 100/07-19-45-
06W5 well – this data was compared with the kerogen-corrected porosity model to test
the model’s validity. The kerogen-corrected total porosity was a good approximation for
the core data (Fig. 5.7), providing the necessary confidence to use `B IaS going forward.
5.6.2.3.3 Effective porosity

Effective porosity (à ) is the intergranular connected porosity, which ranges from 2 to
12% in tight gas sandstones (Byrnes 1997). Effective porosity does not include the
immovable water bound to clays (Fig. 5.4), so effective porosity is always less than total
porosity (Glover 2000). Doveton (2014) uses the following equation to reduce ` B to à :
à = ` B − ;JK (`JK ) [25]
where `JK is the porosity associated with the immovable water bound to the clay volume
fraction, ;JK . To calculate `JK for a specific clay mineralogy, `JK can be calculated by
using the dry (Ai>jJK^j ) and wet (Ak=lJK^j ) clay densities (Doveton 2014):
Ai>jJK^j − Ac [26]
à = ` B − ;JK h m
Ak=lJK^j − AgK
Clay densities vary depending on clay mineral type (esp. smectite) due to differences in
surface area and clay-bound water content (Chitale 2010). Dry and wet clay densities can
be determined in the laboratory on core samples by using XRD, or by using SGR log data
(Schlumberger 2009).
5.6.2.3.4 Effective porosity for the lower Colorado Group

Equation 26 requires a valid estimate of wet and dry clay densities to calculate effective
porosity – this can be problematic due to the wide range of densities that are possible for
112
Fig. 5.8: Histograms of uncorrected (above) and kerogen-corrected (below) modelled total density
porosities from a selection of wells in the study area within the Second White Specks and Belle
Fourche intervals, presented logarithmically.
Prior to the correction, the porosity was lognormally distributed. After the kerogen correction is
applied, fewer porosities exceed 8%, so it appears that anomalously high total porosities have been
largely suppressed.
113
clay minerals (Deer et al. 1966). Without detailed clay mineralogical data, (e.g., XRD),
the only way to estimate clay mineralogy was with the Th/K curve from SGR logging
inthe target interval. This data was only available in the 100/02-14-38-06W5 well.
Plotting this data on Schlumberger Chart Lith-2 (Fig. 5.9) shows that the Th/K data from
the 100/02-14-38-06W5 well clusters largely within the mixed-layer clay field, which is
defined by 12 < Th/K < 3.5 (Schlumberger 2009). Figure 5.9 indicates that the clay
mineralogy of the lower Colorado Group in well 100/02-14-38-06W5 approaches a
mixed-layer (illite and smectite) composition, which corresponds to a Ak=lJK^j of 1.7
g/cm3 and a Ai>jJK^j of 2.7 g/cm3 (Chitale 2010). The calculated effective porosity is
consistently less than the kerogen-corrected total porosity. Unlike the total porosity,
which can be verified against routine core porosimetry analyses, verification of effective
porosity requires special core analytical methods such as mercury injection tests (Burdine
1953). This type of core analysis was not available for this study.
5.6.2.4 QFM and carbonate mineral volumes

Jiang et al. (2017) demonstrated that mineral carbon (%) values from Rock-Eval could be
used with confidence to estimate total carbonate in sedimentary rock samples:
;ipK =
kñUT [27]
EñUT
Qn
where eipK = n
!$" [28]
where TIC is the total inorganic carbon (wt %), ;ipK is the carbonate volume fraction
(v/v), eipK is the weight percent dolomite, and AipK is the density of dolomite (2.87
g/cm3). Using Equation 28, the modelled carbonate volumes from TIC published
carbonate data from Furmann et al. (2014), but only when the mineral assemblage was
simplified to only include one carbonate mineral – dolomite. Although other carbonate
minerals are contained within the lower Colorado Group, dolomite is the most
volumetrically significant (5 to 30%) in the study area (Canadian Discovery 2014).
The Jiang et al. (2017) method cannot be applied to TOC data. Only TOC had been
modelled for non-cored wells, so there was no way to calculate TIC for those wells. In
114
Fig. 5.9: This is a thorium vs. potassium % chart with data plotted in green from well 100/02-14-38-06W5. This data plots largely in the mixed-layer clay
field, providing a basis for using a mixed-layer clay density for effective porosity.
115
order to calculate carbonate volume for non-cored wells, a relationship between carbonate
volume and kerogen volume was established in wells with core (Fig. 5.10), which was
calibrated to specific allomembers:
%&&'()(*)+ !-- |!/01 = 7.01 × !89: + 0.0168 [29]

>?3 |!/01 = 24.8 × !89: + 0.476 [30]
The remaining mineral volume (!"#$ ) was solved by rearranging Equation 16 to subtract
the previously acquired volumes from unity:
!"#$ = 1 − !D1 − !DE:F − !89: [31]
5.6.3 Brittleness
Most unconventional reservoirs require flow capacity created by natural fracturing of
brittle rocks to produce oil at commercial rates (Gale et al. 2007). Rocks exhibit brittle
behaviour when they break without deformation under sufficient stress (Cho & Perez
2014). Knowing how brittleness varies with reservoir properties is key to identifying
sweet spots where natural fractures are more likely to be present, and if hydraulic
fracturing will effectively connect the borehole to the microporosity of the formation
(Jarvie et al. 2007). Brittleness, however, is difficult to quantify (Fox et al. 2013; Cui et
al. 2017) – it is a complex function of rock strength, lithology, texture, effective stress,
pressure, temperature, fluid type, diagenesis, and TOC (Wang & Gale 2009; Rybacki et
al. 2016). Determining an appropriate value for brittleness is a challenging task, because
no universally accepted, standardized definition or measurement method for brittleness
currently exists (Yang et al. 2013).
Since the first brittleness index was developed by Jarvie et al. (2007), a multitude of
brittleness indices have emerged – each with different definitions and different results,
even on identical samples (Yang et al. 2013). Three main varieties exist:
1) Static mechanical indices, based on direct measurements of rock strength on core (i.e.,
compressive, tensile, and residual strength; and fracture toughness) and strain
relations (Heidari et al. 2013; Yang et al. 2013; Hu et al. 2015);
116
Fig. 5.10: The above graph displays a linear regression of carbonate volume computed from Jiang et al.
(2017) in addition to carbonate XRD data from Furmann et al. 2014, versus modelled kerogen for wells
100/02-14-038-05W5, 100/07-19-045-06W5/00, 102/08-15-036-05W5/00, and 102/08-36-041-07W5/00. .
117
2) Elastic moduli (Young’s Modulus and Poisson’s ratio), which are calculated from
downhole measurements (P- and S-wave sonic logs) or seismic (Rickman et al. 2008;
Jin et al. 2014a; Jin et al. 2015) and;
3) Lithological indices, based on weight or volume ratios of brittle minerals (Jarvie et al.
2007; Wang & Gale 2009; Katz et al. 2016; Mathia et al. 2016; Rybacki et al. 2016) –
these indices are summarized and compared in Table 5.4.
5.6.3.1 Lithological brittleness indices

Lithological brittleness indices carry the assumption that brittleness is related to the
relative abundance of brittle rock framework components and ductile components (Jarvie
et al. 2007). Since fractures are supported by the matrix framework, matrix composition
must be a fundamental part of any brittleness calculation (Cho & Perez 2014). It is
generally accepted that brittle behaviour is enhanced in mudstone reservoirs with
increased silica or carbonate content, because siliceous and calcareous minerals impart
their brittle characteristics to their host rocks (Gale et al. 2014). In comparison, reservoirs
rich in clay minerals and organic matter fail in a ductile manner when hydraulically
stimulated, deforming the matrix and inhibiting fracture propagation (Fox et al. 2013).
Mineralogical brittleness indices are popular due to their relative simplicity and
associated with lithology, which can be determined from core, cuttings, or logs (Cui et al.
2017). Lithological brittleness indices do not incorporate texture or rock fabric, which can
impact brittle rock behaviour (Katz et al. 2016). These types of indices are, in effect,
more of a brittle rock-type indicator than a direct measure of mechanical brittleness. All
mineralogical-based definitions result in a higher brittleness index assigned to quartz-rich
rocks than clay-rich lithologies (Herwanger & Mildren 2015), thereby highlighting zones
that are more easily hydraulically fractured (Mathia et al. 2016). Carbonate minerals,
especially dolomite, can further increased brittle rock behaviour, although the relative
contribution of carbonate minerals to brittleness is less than that of quartzose lithologies
(Mathia et al. 2016). Brittleness indices that include dolomite as a ductile component
rather than a brittle component (e.g., Jarvie et al. 2007), therefore, will underestimate the
brittleness of dolomitic reservoirs.
118
Mineralogical brittleness indices

Author Equation Remarks
First lithological brittleness

index
G"
Jarvie et al. (2007) >- = Only quartz is considered
G" + GDE1 + GD1
brittle
Uses wt %
Considered dolomite brittle,

limestone ductile
Wang and Gale G" + G/01
>- = Added TOC wt %
(2009) G" + G/01 + G1H + GD1 + GIJK
Does not account for low
density of organic matter
Included all effective

G"#$ + GDE:F minerals (feldspar and mica)
Jin et al. (2014a); Jin
>- =
et al. (2014b) GI0LE1 All siliceous and carbonate
minerals considered brittle
!"#$ + !/01 + !DE1 Uses volume % instead of

Katz et al. (2016);
>- = wt % to compensate for low
Mathia et al. (2016) !"#$ + !/01 + !DE1 + !D1 + !MNO
density of kerogen
P(!RF ) Includes pyrite as a brittle

>- = mineral
P(!RF ) + *(!DE:F ) + T(!U/ ) + VW
Uses weighting factors to
where !RF = !"#$ + !XY:
Rybacki et al. (2016) compensate for deformation
conditions and relative
!U/ = !D1EY + !IJ$ and mechanical strength
P, *, T, V = weighting factors where P > * > T Includes porosity (v/v)
Table 5.4: Table of mineralogical brittleness indices in chronological order, with their associated
innovations and limitations. Compiled from Jarvie et al. (2007), Wang and Gale (2009), Jin et al. (2014a);
Jin et al. (2014b), Katz et al. (2016); Mathia et al. (2016) and Rybacki et al. (2016).
119
Most of the brittleness indices shown in Table 5.4 do not account for differences in the
relative contributions of different minerals (i.e., carbonate) to brittle rock behaviour (Gale
et al. 2014; Hu et al. 2015). This suggests that weighting factors, such as those used by
Rybacki et al. (2016), should be included to compensate for the relative differences in
mechanical strength between different minerals. Defining these weighting factors would
require extensive calibration to rock mechanical measurements from core in the study
area – this type of data was not available, so a weighted brittleness index was not used.
Although Mathia et al. (2016) did not observe a difference between brittleness values that
were calculated using volume versus weight percent, Rybacki et al. (2016) noted that in
organic-rich shales, the difference in mineral fractions in volume versus weight percent
may reach up to 8% in the case of organic matter (low density) and pyrite (high density).
Moreover, the volume fractions and relative distributions of minerals with different
brittleness likely exert control on the ability of the rock matrix to develop and maintain
open fractures (Rybacki et al. 2016). This suggests that brittleness indices that rely on
weight percentages (e.g., Jarvie et al. 2007; Wang & Gale 2009; Jin et al. 2014a,b) should
not be used in reservoirs that contain large proportions of low density or high density
minerals (i.e., organic-rich mudstones).
If the volume and distribution of minerals with different brittleness throughout a reservoir
impacts fracture propagation and preservation, the role of porosity in brittle behaviour
must also be considered. The role of porosity in brittle behaviour (poroelasticity) is
controversial: Heidari et al. (2013) and Mathia et al. (2016) did not observe any evidence
for a correlation between brittleness indices and porosity. Jin et al. (2014a,b), however,
observed a negative relationship between neutron porosity (W\ ) and brittleness (>-).
They proposed a global correlation of:
>- = −1.8748 ∗ W\ + 0.9679 [32]
where BI is the brittleness index in volume %. This relationship suggests that lower
porosity rocks have higher brittleness. Similarly, Cho and Perez (2014) showed that
highly porous rocks are geomechanically weak, as they easily fail and fracture. A
mineralogy-based brittleness index that incorporates porosity is in the early stages of
120
development (Rybacki et al. 2016), but its practicality and applicability need to be better
established due to its reliance on weighting factors. Uncertainty clouds the role of
porosity in brittle behaviour. Porosity, therefore, was not included in the brittleness
calculation used in this study. Despite their problems, mineralogical brittleness indices
can provide necessary insight into fracability when data is limited, provided that
volumetric mineralogy can be estimated and adequately calibrated to core.
5.6.3.2 Application of brittleness indices to the lower Colorado

Group
The lithological brittleness index described by Mathia et al. (2016) was the best fit for the
lower Colorado Group dataset, as it included carbonate minerals as a brittle component.
Moreover, this index includes components that can be estimated using the limited log
suite available in this study. The index was modified to include dolomite as the sole
carbonate mineral and makes the assumption that only quartz and carbonates contribute to
brittle behaviour. Using the petrophysical models calculated in section 4.6.2, the
following brittleness index was calculated, on a zonal basis, for wells in the study area:
!"#$ + !/01 [33]

>- =
!"#$ + !/01 + !D1EY + !89:
Mapping
5.7.1 Spatial interpolation

For the purposes of mapping, the petrophysical model parameters were reduced to a
single value over each allomember via a thickness weighted average in the Techlog
Summaries module. The weighted averages were imported as a comma-separated values
file (.csv) into geoSCOUT to create irregularly spaced grids of data points that were
exported to the Golden Software™ Surfer 14 program, a data visualization and mapping
software. To ensure each map was generated with identical parameters and could be used
for consistent grid calculations, identical grid geometries were used for each map (Fig.
5.11).
Raw, irregularly spaced data must be spatially interpolated into a regularly spaced grid
prior to geological mapping (Dubrule 1983). Gridding algorithms estimate values at grid
121
Fig. 5.11: A screenshot from Surfer of the grid geometry parameters used for mapping in
this study. Using identical grid geometries for each contour map ensured they could be
accurately overlain and compared to each other.
122
nodes based on weighting functions that vary according to the interpolation method. Once
grid node values (z-coordinates) are assigned to regularly spaced x and y coordinates, the
mapping software treats the grid file (x, y, z) as a continuous surface. These surfaces can
be used for mathematical operations and contour mapping (Jones et al. 1986). Common
spatial interpolation methods include kriging, nearest neighbour, natural neighbour, and
spline methods (Davis 2002). This study utilized a kriging algorithm.
First introduced by Krige (1951), kriging is a gridding method that is widely utilized for
computer mapping in the earth sciences and other fields. The characteristics of semi-
variograms, a measure of the variance between observations as a function of the distance
between them, are utilized by kriging algorithms to estimate values for the regularly
spaced grid nodes based on the weighting of irregularly spaced well data (Jones et al.
1986). Kriging methods explicitly account for the degree of spatial dependence between
data points, known as spatial autocorrelation (Clark 1979). A spatially autocorrelated data
point is assumed to be closely related to values at nearby points, and less closely related
to points that are further away (Davis 2002).
Nearest neighbour interpolation methods assign grid node values by using the closest data
point, or “nearest neighbour”, to interpolate data. This procedure extends the area of
influence of each value halfway to the closest adjacent data point, thus defining polygons
with a value at the center (Jones et al. 1986). This method is most effective for regularly-
spaced dataset, as each polygon is equally weighted. Otherwise, local variations may be
over- or under-represented (Davis 2002). Well data is not equally spaced in this study, so
the nearest neighbour method was not recommended.
By developing a weighted average from neighbouring points, natural neighbour methods

can more accurately represent local spatial variation. Natural neighbour methods are
superior to nearest neighbour algorithms for irregularly spaced data (Sibson 1980). As the
natural neighbour method cannot be used to extrapolate data (Sibson 1980), natural
neighbour methods were not used in this study.
Spline functions interpolate between data points by connecting them with a smooth,
continuous line that minimizes overall surface curvature (Davis 2002). This is a
123
nongridding interpolation method that can predict values that are less or greater than the
values of existing control points, due to the flexure of the spline function applied
(Burrough 1986). Despite honouring all control points, spline interpolation methods have
issues delineating local variation (Burrough 1986). Additionally, when used for surface-
to-surface operations, spline functions can produce artefacts that may not be based in
reality, such as negative values for thickness or porosity (Davis 2002).
Although kriging has been historically criticized due to its supposed subjectivity and
perceived inability to provide accurate predictions compared to other interpolation
methods (Philip & Watson 1986), it is now broadly accepted and is an option in most
geostatistical mapping programs. Error variances of kriging estimates are the minimum
possible of any existing linear interpolation method, and the error can be approximated at
the location of each kriging estimate (Nagy et al. 1999; Davis 2002). This provides a
concrete expression of the uncertainty associated with a contoured surface. Finally,
kriging can be used to estimate grid node values in a way that simulates manual or hand
contouring, because both methods account for spatial autocorrelation (Dahlberg 1975;
Dutton-Marion 1988). Using a kriging algorithm afforded increased confidence in the
reliability of the contoured data.
5.7.2 Surface-to-surface operations

The four allomembers mapped in this study are presented as isochore (equal vertical
thickness) and structure maps. Isochore maps were generated in Surfer by subtracting
bounding surfaces (structural surfaces) from each other, as follows:
-_'Tℎ'+) = *P_) − a'b [34]
where the isochore thickness is the difference in the subsea elevations of the top and base
of a stratigraphic unit. Isochore maps were used in this study to identify areas of increased
or decreased stratigraphic thickness, thereby establishing the geometry of specific
allomembers Isochore maps were also used to create porosity-thickness maps by
multiplying the thickness values by the weighted average of effective porosity.
124
Isochore maps are often used in lieu of isopach (equal true thickness) maps because, for
vertical wells, the isochore is equivalent to the drilled distance along the well trajectory
from the top to the bottom of the stratigraphic interval of interest. If a vertical well
intersects a dipping (sub-horizontal) unit, the measured apparent thickness of a geological
unit will increase – using isochore thickness rather than isopach thickness allows for
thickness map generation without needing to correct for apparent bed thickness. Equation
35 shows the conversion from isochore thickness to isopach thickness:
-_'bPTℎ = cos f(-_'Tℎ'+)) [35]
For units with shallow dip (<15o) – as is the case in the study area – isochore thickness
approximates isopach thickness because cos f is greater than 0.97 (Hesthammer 1998;
Lisle 2004). Once property maps were generated, anomalous values (“busts” or
“bullseyes”) were isolated and either verified or excluded on the basis of geophysical data
quality.
Storage capacity can be estimated by calculating reservoir or porosity thickness using the
following relationship (Tiab & Donaldson 2012):
W ∗ ℎ = WN ∗ -_'Tℎ'+) [36]
where WN is the thickness weighted average of effective porosity over an interval and h is
the isochore thickness.
Faults were not mapped in this study due to a lack of spatial resolution; areas that are
faulted are not recognized by the kriging algorithm as discrete bounding surfaces (Davis
2002).
Analysis
The goal of this thesis was to investigate the possible connection between well
performance from Second White Specks-targeted wells and variations in mappable
petrophysical properties via sweet spot identification. Sweet spots are arbitrarily defined
regions of a mudstone reservoir that have higher reservoir quality and production
125
potential (Glaser et al. 2013). Sweet-spotting is a subjective technique that is driven

largely by the desired result of the interpreter, and can overlook the value of the entire
area if the focus is too small (Haskett 2014). If the study area is too small, regional
geologic controls on reservoir properties may not be adequately characterized. By
evaluating the entire study area prior to sweet spot identification, this thesis intended to
avoid these issues.
Co-location pattern mining is an algorithm that can identify regions of different physical
properties that are closely located or overlapping (i.e., co-located), with the goal of
modelling the spatial correlation between the two features (Sengstock et al. 2012). This
process has clear parallels to sweet spot identification, which seeks to identify areas
where favorable reservoir properties overlap with increased hydrocarbon production.
Analyses of co-location are frequently done with categorical datasets in the fields of
natural science and geography – but in this study the modelled data is in the form of
continuous random variables, so features are not discrete (Eick et al. 2008). Isolating
“interesting” regions of petrophysical parameters requires meaningful boundary
conditions on the continuously modelled data, like cutoffs related to enhanced reservoir
quality (e.g., total porosity < 8%). Once these cutoffs are applied, isolated regions of
interest (instances of co-location) can be recognized via overlapping polygons (Eick et al.
2008). Cartographic representations like this are an effective and easily communicable
data visualization method for co-location pattern identification (Desmier et al. 2011).
In this study, co-location pattern mining was simulated by using cutoffs to identify
overlapping sweet spots – regions with enhanced reservoir quality (porosity-thickness and
brittleness) that were spatially associated with improved well performance. The resulting
polygons are effectively Boolean AND operators for well data within them. For a well to
reside within a sweet spot, it must meet at least three conditions: it must have enhanced
porosity-thickness, brittleness, and hydrocarbon production.
Highlights
• Sparse core data and a limited geophysical wireline log dataset present the main
procedural challenges to petrophysical modelling in this thesis.
126
• Mineral volumes were modelled using a deterministic method wherein each volume
was independently calculated and input into a five-component petrophysical model.
A lithology-based brittleness index was used to estimate variations in brittle rock
behaviour.
• Sweet spot delineation was achieved by overlaying areas with enhanced porosity-
thickness and brittleness with oil production results.
127
Solving the Second White Specks 6. Core-to-log correlation and allostratigraphy
Chapter 6
6 Allostratigraphic correlation of the Lower Second White
Specks and Upper Belle Fourche alloformations in the
Willesden Green – Gilby region, Alberta, Canada
Overview
This chapter presents the results of core logging and high-resolution allostratigraphic
correlation of the Lower Second White Specks and Belle Fourche alloformations in the
Willesden Green area of west-central Alberta. A shallow muddy ramp is introduced as a
depositional model for the Second White Specks and Belle Fourche alloformations.
Allostratigraphic framework
The study area overlaps with Tyagi et al. (2007)’s high-resolution allostratigraphic study
of the Blackstone Formation. Tyagi et al. (2007) mapped nine regionally correlative
flooding surfaces that bound eight allomembers in the Sunkay and Vimy Members of the
Blackstone Formation over an area of 200 000 km2. Of those allomembers, three fell
within the stratigraphic interval of interest for this study (Fig. 6.1). The surfaces used to
constrain those allomembers are the X-flooding surface, the K-1 unconformity, the “red”
bentonite, and an unnamed flooding surface above the red bentonite. These surfaces
exhibited negligible topographic relief over Tyagi et al. (2007)’s study area, suggesting
that unit was deposited on a sea floor with extremely low gradient. This is inferred from
the lack of shelf-slope physiography and little to no clinoform development, which is
typical of a shallow, epicontinental ramp setting (Asquith 1970; Varban & Plint 2005;
Midtkandal et al. 2007; Varban & Plint 2008a).
The X-flooding surface, K1 disconformity, and the “red” bentonite were traced by Tyagi
et al. (2007) from Townships 23 to 58, originating northeast of Grande Cache in British
Columbia and extending through the Alberta Foothills, and terminating in western
Saskatchewan. The X and K1 surfaces, defined by Plint (2000), were traced into the
Foothills of British Columbia and the Peace River area by Kreitner (2002). The X-
flooding surface is an important regional transgressive surface that is linked to major
128
Fig. 6.1: Correlation of the allostratigraphic framework used in this thesis (left) to the framework used by Tyagi et al. (2007), using the K1 disconformity as a
datum. Dashed lines indicate the extension of the framework used in this thesis into Tyagi et al. (2007), and vice versa.
129
lowstand units of the Kaskapau Formation that were mapped by Plint and Kreitner
(2005). The K1 surface is a major beveling unconformity in northwest Alberta, but
becomes a transgressive disconformity within central Alberta (Plint 2000; Kreitner &
Plint 2006) – it coincides with the top of the Pouce Coupe sandstone in northeastern
Alberta (Varban & Plint 2005).
The “red” bentonite, which approximates the Cenomanian-Turonian boundary, has been
traced by Tyagi et al. (2007) and others throughout 235 000 km2 in Alberta and British
Columbia (Stott 1963; Varban 2004; Varban & Plint 2005; Varban & Plint 2008a; Tyagi
2009; Plint et al. 2012a). Allomembers of Unit II (lower Kaskapau alloformation) from
Varban and Plint (2005) progressively downlap onto the top of the “red” bentonite
towards west-central Alberta. Extending the correlations of Varban and Plint (2005),
Tyagi (2009) interpreted the “red” bentonite surface as a basin-wide hiatal surface that
occurred during the onset of the Early Turonian – he estimated that the hiatal surface
represents approximately 500 Kyr of missing time. These observations were corroborated
by Plint et al. (2012a): they observed the expression of this hiatal surface as a minor
unconformity that was correlated from Mount Robert in northeastern British Columbia
into northwestern Alberta (well 15-34-77-1W6).
The 100/07-19-045-06W5 core and the type well extracted from by Tyagi et al. (2007)
were used to extend the allostratigraphic framework developed by Tyagi et al. (2007) into
the lower Colorado allogroup in the Willesden Green and Gilby areas by correlating the
major bounding surfaces (the X-flooding surface, the K1 disconformity, and the “red”
bentonite) through a coarse grid (1 well per township) of wells. These correlations were
used to determine the vertical stratigraphic placement of Second White Specks and Belle
Fourche Formation cores in the study area.
Tyagi (2009) subdivided the lower Colorado Group was subdivided into three
alloformations above the Fish Scales Formation: the Lower and Upper Belle Fourche
alloformations, and the Second White Specks alloformation. The Upper Belle Fourche
alloformation was further subdivided by this author into three allomembers (BF1, BF2,
and BF3). Similarly, the Second White Specks alloformation was subdivided by this
130
author into two allomembers (allomembers VII and VIII). Of these units only
allomembers VII, BF2, and BF3 were completely present in cores that were logged within
the study area. Emphasis was placed on these allomembers due to their spatial association
with the typical “Second White Specks” reservoir interval.
Lower Colorado Group facies

Dalrymple (2010a) defines facies as:
“…[bodies] of rock characterized by a particular combination

of lithology, physical, and biological structures that bestow an
aspect different from the bodies above, below, and laterally
adjacent.”
Facies are controlled by the sedimentary processes operating in the depositional

environment where they are deposited. Describing facies, therefore, is essential for the
interpretation of depositional processes (Catuneanu 2006).
Many authors have described facies of the Belle Fourche and Second White Specks
Formation and their equivalents (Stott 1963, 1984; Leckie et al. 1994; Plint 2000;
Kreitner 2002; Fraser 2005; Varban & Plint 2005; Tyagi et al. 2007; Varban & Plint
2008a; Tyagi 2009; Yang & Miall 2009, 2010; Plint et al. 2012b; Jiang 2013; Jiang &
Cheadle 2013). Although high-resolution allostratigraphy of the entire lower Colorado
allogroup was out of scope for this project, defining facies and determining their
distribution across the study area was an essential step in establishing reservoir geometry.
Five facies were identified in core (Table 5.2), and were differentiated on the basis of
composition, clay content, grain size, sedimentary structures, and bioturbation (Figs. 5.2
and 5.4).
The facies of the Second White Specks and Belle Fourche alloformations in this thesis
(Table 5.1) were derived from observations made from six cored wells: 102/08-15-036-
05W5, 100/02-14-038-05W5, 100/04-27-39-06W5, 102/08-36-041-07W5, 100/16-28-
041-04W5, and 100/07-19-045-06W5. The first five of these cores were logged and
sampled in detail at the Alberta Energy Regulator (AER) Core Research Centre in
Calgary, Alberta (Fig 5.2). This process included describing the visible lithology, grain
131
Core cross-section through the Second White Specks and Belle Fourche alloformations in Willesden Green, Alberta
Study area index map
B GR DT LLD
Fig. 6.2: An approximately south-to-north oriented cross section displaying the five Second White 100/07-19-045-06W5
DEN
Specks and Belle Fourche cores logged by this author. Core photos are included for reference to facies. LLD
102/08-36-041-07W5 GR
2124
GR
DT LLD
102/08-36-041-07W5 2125
100/16-28-041-04W5
2126 100/07-19-045-06W5
100/16-28-041-04W5
2127
100/04-27-039-06W5 1806
100/04-27-39-06W5 1803
2129
Lithofacies code 100/02-14-038-05W5

2128
1807
1804
2130 Lithic clast lag on top
2129
LLD of erosional surface 1808
1 2131
GR H 1805
102/08-15-036-05W5
DT 2130
1809
1806
Allomember VII
A 2132
2131
1810
1807
2 2133
100/02-14-038-05W5 2132
1811
1808
Allomember VII
2052 2134
2133 1812
1809
3 2053
GR LLD 2135
2134 40% silt & vfU sand, wavy to lenticular bedded; 1813 Fining upwards cm-scale units with mm-scale laminae,
2054
DT bases show ss. def and have small ripups (<1cm)
~20% silt and 80% mudstone
1810
2136 Fish scale lag
2135
1814
1811
4
Allomember VII
2055 2137 Shell fragment lag
2136 Pyritized sand lens
1815
GR 2056 2138
Wavy to undulating laminae,
Shell fragment lag 1812
20% silt and 80% mudstone Lithic clast lag, 1 cm tall and 5-6 cm wide clasts
DT 2137
1816 Shell fragment lag
5
1813
>30% silt, wavy to undulating laminated, interbedded w/ silt
2057 25% vfU sand and silt, 10% flat lithic clasts, 65% mudstone
2139 moderate ss def.
LLD 2138 coal hz.
Allomember VII
1817 x x x x x
E Bentonite (Bighorn River “red’ bentonite) 1814

x x x x
2058 2140 Wavy to undulating laminae, Thin, disc. coal hz
G
some very large siderite noducles (6 cm), well rounded 2139
1818
1815
ss. def
Allomember VII
2059 2141
Dominated by shell fragments and cc. sand, 2140
C 1x1 cm to 2x5 cm lithic clasts in cc. sand,

D 20% lithic rip-up clasts,
coarsening upwards
F Hz cc-filled fracture, 3 mm wide, minor vertical fracturing
“sand dike” 1 cm wide and 2 cm long
1819
1816
2060 cm-scale sets are topped by black mud 2142
102/08-15-036-05W5 and increasingly bioturbated (subvertical burrows) towards the top
with soft sediment deformation at the bases
Shell fragment lag 2141 Slickensides, cc-filled vertical & hz fractures 1820
B
1817
2201
2061 2143 20% silt, 80% mudstone
planar laminated 2142
H 1821
Belle Fourche 3
x
1818
cm-scale fining upwards lamina sets with scoured bases,
2062
Belle Fourche 3
Allomember VII
rare sand beds are pyritized and oil stained 2144 Bentonite (Bighorn River “red’ bentonite)
x x x x x
2202 2143
1822
1819
2063 x x x x x
2145
2203 2144
<5% silt, 95% mudstone 1823
fL
vfL
fU
mud
vfU
silt
planar laminate with rare shell fragments 1820
2064
Pyrite and shell fragment lag
Belle Fourche 3
2146
2204 2145
1824
20-40% silt, brownish grey mudstone f 1821
po I
2065 2147 To lo VI
2205 x 2146 Al
x x
Bentonite (Bighorn River “red’ bentonite) 1825
Belle Fourche 3
x
Belle Fourche 3
x 1822
x x
ag
2066 2148 ticl
Structureless 100% mudstone c las
2206
bio
fL
vfL
fU
mud
vfU
silt
2147 1826
Fish bone lag 3 cm scoured sand lens Shell fragment lag 1823
Belle Fourche 3
2067 2149
2207 2148
1827
Very dark grey, <5% silt, 95% mudstone 1824
2068 planar laminate with rare shell fragments 2150
2208
2149
1828
1825
2069 2151
2209 2150
1829 <5% silt, 95% mudstone
planar laminate with rare shell fragments 1826
Fish bone lag 2070
2152
2210
fL
vfL
fU
mud
vfU
silt
2151
1830
1827
20% silt, 80% mudstone 2071

2153
2129.4 m - top of Allo VII, erosional
2211
2152 scour with lag 1831
1828
fL
vfL
fU
mud
vfU
silt
2072
2154
A D G
2212
2153 1832
1829
B
Alternating fining-upwards and coarsening upwards sequences,
5 cm thickness with scoured bases 2073 2155
2213
tops are moderately bioturbated 2154 1833
1830
Trough-crossbedded sand horizon with subvertical
A
2074
C 2156
E F
2214 mud-filled burrows at the top 2155 1834
Belle Fourche 2
1831
2075 2157
ss-def
2215
Belle Fourche 2
2156 1835
1832
Belle Fourche 2
Rip up clasts (<4 mm)
2076
BF2
2158
2216
2157 `836
Belle Fourche 2
Belle Fourche 2
1833
Belle Fourche 2
Fish bone lag 2077 2159 disturbed bedding
2217
2158 1837
1834
Fish bone lag 2078 ss-def
2160
2218
2159 1838
1835
Fining upwards cm-scale units with mm-scale laminae
scoured base 2079 2161
2219
disturbed bedding 2160 1839
starved rippes
vFU sand 1836
Rare coccoliths 2080 2162
2220 2161 1840
ase
concretion scoured b 1837
2081 2163 ss-def
2221 Mottled to indistinctly laminated 2162 1841
1838
scoure 2082 2164

2222 d base
fL
vfL
mud
vfU
silt
2163 1842
coal hz, fragmented and discontinuous
2223 2083 2165 sub-horizontal burrows

Thinly laminated (mm-scale beds), 75% mud and 25% silt, very dark grey scoure
d base 2164
1843
Rare shell fragments <2 cm long, 5 mm wide 2059.8 m - soft-sediment deformation and starved
sub-horizontal burrow
2224 2084
(?) sand lenses
2166
fL
vfL
fU
mud
vfU
silt
2165
2140.14 m - erosional scour with vfU infill 2139.6 m - sub-vertical burrows
2212.13 m - alternating FU and CU beds, 2085
2160.8 m - oil-stained carbonate sand with 2141.8 m - laminae appear to have been partially
2167
majority are fining upward, bases are scoured soft-sediment deformation lithified prior to deformation, some bioturbation 2139.75 m - large, rounded concretion (siderite?)
Fig. 6.3: An approximately south-to-north oriented cross section displaying the available LAS data for cores in the study area.
A A Author: Kienan MARION

(ID: kmarion2)
Date: 13/12/2017
Well: 102081503605W500 Well: 100021403805W500 Well: 100042703906W500 Well: 102083604107W500 Well: 100060704202W500 Well: 100071904506W500
UWI: 102081503605W500 UWI: 100021403805W500 UWI: 100042703906W500 UWI: 102083604107W500 UWI: 100060704202W500 UWI: 100071904506W500
Short name: Short name: Short name: GREEN Short name: Short name: Short name:
Long name: Long name: Long name: Long name: Long name: Long name:
ENERGY INC.
Mineral components = 1 Cumulated variables Mineral components = 1 Mineral components = 1 Mineral components = 1
Clay volume from core

0 1 0 1 0 1 0 1 0 1
Gamma ray core Gamma ray core PHIT_E_KER Gamma ray core Gamma ray core PHIT_E_KER Gamma ray core PHIT_E_KER
75 (GR) 150 75 (GR_K) 150 0.15 v/v 0 75 (GR) 155 75 (GR) 155 0.15 v/v 0 75 (GR_K) 155 0.15 v/v 0
Q+F Norm
Q+F Norm
Q+F Norm
Clay Norm
Clay Norm
Clay Norm
TOC Norm
TOC Norm
TOC Norm
Q+F Norm
TOC Norm
Clay Norm
CALI Vshale TOC PHIT_E_KER CALI Vshale TOC PHIT_CORR Gamma ray core CALI Vshale toc_i PHIT_E_KER CALI Vshale PHIT_CORR CALI Vshale toc_i PHIT_CORR
Q+F Norm
C Norm
C Norm
C Norm
TOC Norm
BI = 0.855 BI = 0.855 BI = 0.855 BI = 0.855 BI = 0.855
C Norm
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
150 MM 550 0 (VSH_GR) 1 0 v/v 0.05 0.15 v/v 0 150 MM 550 0 (VSH_GR) 1 0 v/v 0.05 0.15 v/v 0 75 (GR) 155 150 MM 550 0 (VSH_GR) 1 0 v/v 0.05 0.15 v/v 0 150 MM 550 0 (VSH_GR) 1 0.15 v/v 0 150 MM 550 0 (VSH_GR) 1 0 v/v 0.05 0.15 v/v 0
C Norm
Reference Reference Reference Reference Reference Reference
(M) GR VSH_GR DT RES_DEP toc_i PHIT_CORR bi (M) GR_K VSH_GR DT RES_DEP toc_i CPOR bi (M) GR DT TOC CPOR (M) GR VSH_GR DT RES_DEP TOC PHIT_CORR bi (M) GR VSH_GR DT RES_DEP toc_i CPOR bi (M) GR_K VSH_GR DT RES_DEP TOC_G CPOR bi
1:200 0 GAPI 200 0 v/v 1 550 US/M 400 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 0 v/v 1 1:200 0 gAPI 250 0 v/v 1 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 V/V 0 0 v/v 1 1:200 0 GAPI 250 350 US/M 200 0 v/v 0.05 0.15 V/V 0 1:200 0 GAPI 250 0 v/v 1 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 0 v/v 1 1:200 0 GAPI 250 0 v/v 1 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 0 0 v/v 1 1:200 0 gAPI 250 0 v/v 1 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 0 v/v 1
2125 1650
1800
2125
Allo VII
2050
Allo VII
Allo VII
Allo VII
Allo VII
2200
Allo VII
BF_3
BF_3
BF_3
BF_3
BF_3
2150 1675
BF_3
1825
2150
BF_2
2075
BF_2
BF_2
BF_2
BF_2
BF_2
BF_1
BF_1
2225
BF_1
BF_1
BF_1
BF_1
2175 1700
1850
2175
2100
Depositional
Relative advection
Facies Description environment
energy
(Facies association)
Bentonite ashfall
1 Bentonite beds Dispersed by wind and -
water currents
Distal offshore
Very dark grey, planar- shallow ramp
laminated mudstone Influenced by
2 interbedded with combined flows, ~70 m
Low
siltstone water depth
Anoxic bottom water
Offshore shallow
Weakly bioturbated, ramp
medium-grey laminated
Influenced by
3 mudstone, interbedded
combined flows, 40 to
Intermediate to low
siltstone, and very fine- 70 m water depth
grained sandstone
Dysoxic bottom water
Offshore shallow
ramp
Wavy to planar-
laminated calcareous influenced by
4 mudstone, siltstone, and
combined flows, 40 to Intermediate
<70 m water depth
fine-grained sandstone
Variably dysoxic
bottom water
Marginal offshore
to mid-shelf shallow
Wavy to undulating ramp
laminated calcareous
influenced by
5 siltstone and fine-
combined flows, <40 m
High
grained sandstone with
water depth
frequent lag deposits
Variably oxic to
dysoxic bottom water
Table 6.1: The facies codes used in this thesis, along with their descriptions and associated depositional
environments. Facies colours reflect the colours used in later cross-sections to indicate those facies. Vertical
shifts in facies indicate lateral shift in depositional environment, which are interpreted to be driven by
changes in relative advection energy.
134
size, composition, sedimentary and structural features, as well as the trace fossil
assemblages. The graphic logs from the logged cores are illustrated in Fig. 6.2. The
geophysical well logs associated with these cores (Fig. 6.3) were used to help identify
major compositional changes in the core and tie those changes to their respective log
signatures. This evidence will be discussed in section 6.5, after the facies have been
described.
6.3.1 Facies observed in core

The following observations of facies and their accompanying descriptions are derived
from the five cores logged for this study and was supplemented with data from the
100/07-19-045-06W5 core. From these observations, five facies were identified and
related to a shallow marine muddy ramp environment.
6.3.1.1 Facies 1: bentonite beds

Facies 1 contains bentonite beds that are white to grey in colour, and range in thickness
from 5 to 40 cm. The bentonites were typically soft and very powdery, suggesting that the
measured thicknesses may not be representative of their true vertical thickness in the
subsurface (the core may not have achieved 100% recovery).
Elder (1988) determined that his B bentonite, which correlates with the Bighorn River
“red” bentonite of Tyagi et al. (2007), was derived from the western Cordilleran orogenic
belt near the border between the United States and Canada. The time interval around the
Cenomanian-Turonian boundary coincided with a major period of pluton emplacement in
the western Cordillera (Chen & Moore 1982). Elder (1988) interpreted his B bentonite as
the reflection of an ashfall event that was primarily dispersed by wind in an overall north-
to northwest wind direction, based on its geometry (uniform thickness) and a lack of
evidence for physical reworking of the bentonite.
Facies 1 is interpreted as the record of volcanic ashfall from an eruption in the western
Cordillera that dispersed smectite-rich clays across a shallow marine ramp in west-central
Alberta. Although the bentonite did not appear to be physically reworked or bioturbated,
it does vary in thickness between the cores from the study area – suggesting that currents
135
may have contributed to bentonite redistribution and therefore were within storm wave
base for silt (<70 m water depth). Ash dispersion was likely facilitated by a combination
of wind- and current-related processes. Because the “red” bentonite is concomitant with
OAE-II, the bentonite facies was likely deposited during a marine transgression when
biotic activity was relatively low.
6.3.1.2 Facies 2: Very dark grey, planar laminated mudstone

Facies 2 is a mudstone facies that contains a relatively low proportion of silt (<5 to 25%),
as well as rare disseminated pyrite grains (<1%) – sand is not present in this facies. Rocks
of facies 2 are typically very dark grey in colour, consist of very thin (1 to 2 cm) normally
graded beds, and lack evidence of bioturbation. Normally graded siltstone laminae are
either planar laminated or discontinuous. Discontinuous fish scale and shell fragment lags
were occasionally observed and were always less than 5 mm thick. Reverse graded beds
were encountered in 102/08-15-036-05W5 but were not observed in other wells.
Individual beds are sharp-based, occasionally with sharp scours. Bivalve shells and shell
fragments (< 2 cm long, 5 mm wide) were frequently encountered. Isolated siderite
concretions, ranging from 5 mm to 1 cm in size and ellipsoidal in shape, were also
observed. Measured TOC in this facies ranged from 1.75 to 2.5%.
Planar lamination in sandstones is typically developed directly above the erosional,

scoured bases of storm beds (Cheel 1991). Petrographical studies such as Macquaker et
al. (2010) and Plint et al. (2012a) have demonstrated that planar laminae in mudstones are
often internally cross-laminated. These beds, therefore, represent cycles of advective
sedimentation – possibly during storms – that are delimited by periods of erosion or non-
deposition, rather than passive settling in a low-energy environment (Trabucho-Alexandre
2015).
In shallow marine environments where oxygen is readily available, laminated sediments

will only be preserved when sedimentation rates are especially high (1.0 to 10 cm/y) –
such as in a delta front or inner shelf setting (Nittrouer et al. 1986). Tyagi (2009),
however, demonstrated that sedimentation rates for his laminated mudstone facies in the
Blackstone Formation were exceedingly low (< 5 cm/Kyr) and therefore could not have
136
been deposited in an oxic setting. In anoxic marine environments, the mudstone

laminations are preferentially observed when bioturbation is absent (Byers 1974).
Laminated, nonbioturbated, and organic-rich mudstone successions are typically
interpreted as the record of fine-grained sediment deposition in variably anoxic, low-
energy settings (Demaison & Moore 1980), although Plint et al. (2012a) showed that
similar successions can be deposited by advective means.
Authigenic minerals are derived from pore water solutes, so their compositions (e.g,
siderite and pyrite) should reflect the pore water chemistry where they precipitate (Curtis
& Coleman 1986). Pyrite, a common sulfide mineral in lower Colorado Group mudstones
(Bloch et al. 1999), is an early diagenetic mineral that forms depending on the amount of
available organic matter, dissolved sulfate, and reactive iron minerals (Berner 1984).
Diagenetic pyrite is most commonly formed and preserved in anoxic environments
(Berner 1984), although pore waters were likely only anoxic at a depth of a few
millimeters below the sediment/water interface (Hudson & Martill 1991). Siderite is an
iron carbonate mineral that occurs in lower Colorado Group mudstones as concretions
(Bloch et al. 1999). Siderite concretions document the activity of anaerobic, methane-
producing bacteria at the sediment/water interface (Curtis et al. 1972). Similar to pyrite,
the precipitation of siderite concretions requires the development of strongly reducing, or
dysaerobic, bottom water conditions (Fritz et al. 1971). Other early diagenetic features,
such as early cementation, have been interpreted as the record of significant hiatuses in
sediment accumulation related to sediment bypass and winnowing of the seafloor
(Macquaker et al. 2007).
Facies 2 is interpreted as the distal expression of mud and silt deposited by combined
flows along a distal shallow ramp setting. Deposition occurred in an anoxic marine
environment that was below storm wave base for sand, and variably within storm wave
base for silt and mud (70 to 100 m water depth). This is evidenced by the lack of sandy
beds and bioturbation. The presence of pyrite and siderite, which are early diagenetic
minerals, suggests that the bottom water conditions were at least dysoxic if not anoxic.
137
6.3.1.3 Facies 3: Weakly bioturbated, medium-grey laminated

mudstone, interbedded siltstone, and very fine-grained
sandstone
Facies 3 is coarser-grained than facies 2. It is a mudstone facies that is interbedded with
25 to 40% silt, and 0 to 5% very fine sand. Rocks of facies 3 are lighter grey than facies 2
and comprise thin (1.5 to 3 cm) normally graded beds with scoured bases. Beds are
typically planar laminated but may exhibit subtle cross-stratification. The tops of some
individual beds are partially or weakly churned by sub-horizontal burrowing. Lenses of
very fine sand are rare, but when they appear they contain subtle cross-laminations, mud
flasers, and are typically oil-stained (Fig. 6.2F). Like facies 2, fish scale and shell
fragment lags were occasionally observed and were always less than 1 cm thick – these
lags are more continuous than those observed in facies 2. Sideritic concretions were
common throughout this facies. Small, ellipsoid rip-up clasts (<4 mm) were observed in
this facies as a mm-thick lag horizon in the 102/08-15-036-05W5 well. Bivalve shells and
shell fragments (< 2 cm long, 5 mm wide) were frequently encountered. Measured TOC
in facies 3 ranged from 2 to 3.75%.
Across mud-dominated shelves, sand is rare and confined to storm-associated laminae on

the inner shelf (Plint 2010). This is because higher oscillatory wave velocities are
required to entrain sand grains, relative to finer-grained silt or mud-sized particles. Low
unidirectional flow velocities (<10 cm/s) and intermediate oscillatory current velocities
(20 to 50 cm/s) are required to form symmetrical and asymmetrical wave ripples in very
fine sand (Dumas et al. 2005).
Fluid mud deposition inhibits benthic colonization, so bioturbation of muddy shelf

sediments is normally rare. Combined flows associated with geostrophic flows and storms
can erode partially consolidated mud, which can then be burrowed – this produces a
succession of mud sequences that are capped by scoured and burrowed surfaces (Rine &
Ginsburg 1985; Kuehl et al. 1996).
The presence of very fine sandy lenses and interbeds with subtle cross-stratification
suggest that deposition of facies 3 occurred between the storm wave base for sand and silt
(40 to 70 m water depth). Sandy interbeds indicate sand emplacement during storms,
138
whereas silty and muddy laminae record the waning phase of storms and periods of time
when slightly lower-energy advective flows dominated. Sub-horizontal burrows at the
tops of some beds suggest sedimentation was punctuated by periods of erosion, during
which the sediment/water interface became dysaerobic and benthic colonization could
take place. Overall, facies 3 is interpreted as a slightly shallower water facies than facies
2 and was deposited within an offshore shallow ramp setting.
6.3.1.4 Facies 4: Wavy laminated calcareous mudstone, siltstone,

and fine-grained sandstone
Facies 4 is coarser-grained than both facies 2 and 3: it is a calcareous mudstone that is
interbedded with greater than 40% silt, and 5 to 10% very fine and upper fine-grained
sand. Sequences within facies 4 comprise packages (~3 cm thick) of alternating fining-
upwards and coarsening upwards beds, although the majority of beds are fining upwards.
Like facies 2 and 3, these beds have scoured bases (Fig. 6.2A) that may develop into
small-scale gutter casts (5 by 10 cm). Fig. 6.2H depicts a steeply-dipping erosional scour
in this facies that contains a relatively coarse, bioclastic lag horizon. Some individual
beds are well churned by sub-vertical and vertical burrows (Fig. 6.2G) – in those cases,
primary sedimentary structures are partially obliterated. In other cases, however,
bioturbation is weak or non-existent (Figs. 6.2A, 6.2B). Lenses and interbeds of very fine
sand and exceptionally upper fine-grained sand are more common than in facies 3,
comprising more than 40% of this facies. Weakly convoluted or disturbed bedding, which
sometimes appeared to have been partially lithified prior to deformation was occasionally
observed, as well as small load casts (< 5 mm) at the bases of some beds. Like facies 2
and 3, fish scale, shell fragment, and sideritic nodule lags were occasionally observed and
were always less than 1 cm thick. Bivalve shells and shell fragments (< 2 cm long, 5 mm
wide) were frequently encountered. Measured TOC in facies 4 ranged from 1.5 to 3.5%.
Soft-sediment deformation structures in clastic sediments are defined by Shanmugam

(2017) as “features that form before lithification of sediments and that form under a state
of liquidization.” For soft-sediment deformation to occur, a threshold value must be
exceeded: for instance, the gradual accumulation of dense sediment on top of less dense,
water-saturated sediment can initiate water-escape and the formation of load casts (van
139
Loon 2009). Soft-sediment deformation is associated with a wide range of sedimentary

environments worldwide, and over 120 types of soft-sediment deformation have been
characterized (Shanmugam 2017). Convolute bedding, in particular, is a type of soft-
sediment deformation associated with the response of cohesive plastic sediment to bottom
currents (Dzulynski & Smith 1963; Shipboard Scientific Party 1998).
Facies 4 comprises a higher percentage of coarser-grained sediment than facies 2 and 3

and is more frequently bioturbated. This suggests that deposition of facies 4 occurred in
shallower water than the preceding facies while remaining within the transition zone
between storm wave base for sand and silt (40 to 70 m water depth). Convolute bedding
and soft-sediment deformation structures could indicate partial liquidization of
unconsolidated muddy substrate. For the same reasons discussed for facies 3, facies 4 is
interpreted as a slightly shallower water facies than facies 2 and 3 and was deposited in an
offshore shallow ramp setting.
6.3.1.5 Facies 5: Wavy to undulating laminated calcareous

siltstone and fine-grained sandstone
Facies 5 consists of the coarsest-grained sediments of all the described lower Colorado
Group facies. It is a calcareous siltstone interbedded with upper fine-grained quartz sand
as well as occasional carbonate sand grains. The soft-sediment deformation structures
described for facies 4 are also present in facies 5, but are more common (Figs. 6.2C,
6.2D). Sequences within facies 5 comprise cm-scale packages of fining-upwards beds.
The tops of these sequences are frequently burrowed and are typically capped with a thin
(< 5 mm) layer of black mudstone. Like facies 2 through 4, 1 to 2 cm thick fish scale,
shell fragment, and sideritic nodule lags were frequently observed, and sometimes
occured repeatedly. A large siderite nodule was encountered in the 100/04-27-39-06W5
well (Fig. 6.2E). Measured TOC in facies 4 ranged from 2.5 to 4%.
Relative to the preceding facies, facies 5 comprises the highest percentage of coarse-
grained sediment and is the most frequently bioturbated. This suggests that deposition of
facies 5 occurred in shallower water than the preceding facies and approached storm
wave base for sand (~40 m water depth), where the bottom water was variably
140
oxygenated. Numerous lag horizons suggest repeated periods of sediment winnowing and
erosion by storms and combined flows. Convolute bedding and soft-sediment
deformation structures could indicate partial liquidization of unconsolidated muddy
substrate during advective flow. Facies 5 is interpreted as a shallower water facies than
facies 2 through 4 and was deposited in a marginal offshore to mid-shelf shallow ramp
setting.
6.3.2 Facies association

Facies can be grouped into successions, which are recurring, genetically-related
associations of lithological characteristics (Collinson 1969). When defined in three
dimensions, a facies association constitutes a depositional sequence (Brown & Fisher
1977). Facies associations can have significant predictive power when the consequences
of Walther’s Law are considered (Middleton 1973). Walther’s Law states that a vertical
succession of facies that conformably grade into one another were once laterally
juxtaposed; in contrast, chronologically disconformable facies (e.g., separated by
erosional or hiatal surfaces) are not genetically related. Ultimately, all surfaces across
which there is evidence of abrupt facies dislocation are potentially important (Galloway
& Hobday 1996).
The relationship between grain size and variations in sediment transport energy is the
cornerstone of fluvial, deltaic, and turbidite facies models. Gravity drives sediment
dispersal in these systems, primarily via selective deposition and preservation. This
results in an overall “down-system fining” trend, where particles tend to become finer
with distance along a particular sediment transport path (Aigner & Reineck 1982). The
energy needed to maintain grain motion decreases with particle size and specific gravity –
as a result, fine-grained particles can be transported farther by low-energy advective
conveyance mechanisms like combined flows or WESGF’s (Dumas et al. 2005;
Dalrymple 2010b; Plint et al. 2012a; Allen & Allen 2013; Plint 2014).
Using Walther’s Law, the facies association observed in core is interpreted in the context
of a depositional environment (shallow muddy ramp; Table 6.1 and Fig. 6.4). This
141
established a relationship between laterally and vertically adjacent facies, which helped to
constrain the allostratigraphic framework.
6.3.2.1 Offshore to mid-shelf shallow muddy ramp setting

Figure 6.4 is a cartoon of a paleoenvironmental reconstruction of the mud-dominated
shallow ramp environment interpreted for the Second White Specks and Belle Fourche
alloformations. During times of relative sea-level rise (transgressive and highstand time),
muddy heterolithic deposits were dispersed seaward via combined flows. Times of sea-
level fall, comparatively, promoted progradation of the Kaskapau delta across the inner
ramp. The mid-shelf and offshore environments experienced progressive erosion by wave
scouring, which may have carved out gutter casts as the geographic position of sand and
silt storm wave base shifted seaward. This exposed early diagenetic products, such as
siderite nodules, to winnowing. The gutter casts became filled with coarser-grained storm
deposits like sand and rip-up clasts. In this model, facies 2 through 5 developed as
laterally adjacent facies.
Fine-grained sediment transport in ancient epicontinental seas

Epicontinental seas are partially enclosed, shallow seas (water depth in 10s of metres)
with normal marine salinity within continental areas, and form large expanses when they
are substantially flooded by oceans (Shaw 1964; Johnson & Baldwin 1996). Ancient
epicontinental seas, for which there are no modern analogues (Hay et al. 1993), differed
from modern shelf environments due to the large lateral extent and low shelf gradients of
the former (Shaw 1964). Modern continental shelves are typically 10 to less than 100
kilometres wide with bottom slopes ranging from 0.02 to 0.1o, whereas ancient
epicontinental seas were commonly 1000’s of kilometres wide and had bottom slopes in
the range of 0.005 to 0.001o (Shaw 1964; Johnson & Baldwin 1996). Ancient
epicontinental seas, therefore, had gently sloping ramp-like profiles without defined shelf-
slope physiography (Keulegan & Krumbein 1949). As turbidity currents require sufficient
slope to form, turbidity currents were likely not an important mechanism for sediment
transport across shallow epicontinental shelves (Curray 1960). The rock record of ancient
epicontinental seas, like the WIS, contain fine-grained clastic sediments (e.g., the lower
Colorado Group) that can be traced over 1000 km offshore from their respective basin
142
Fig. 6.4: Cartoons summarizing the paleogeography and paleobathymetry of the Kaskapau delta and its
basinal equivalent Second White Specks and Belle Fourche alloformations – vertical scale is extremely
exaggerated. This approximates a shallow marine muddy ramp setting with a very low gradient. The
approximate location of facies 2 through 5 are shown as circled numbers.
Adapted from Varban and Plint (2008a), Plint et al. (2012a), and Plint (2014).
143
margins (Schieber 2016). This presents an intriguing problem: how could mud have been
mobilized over long distances offshore across a shallow, low-gradient ramp?
Classic models of mud deposition in low-gradient offshore settings dictate that organic-
rich mud is deposited in relatively deep water (> 100 m water depth), where continuous
pelagic rainout of sediment dominates in a relatively low-energy setting (McCave 1984;
Alldredge & Silver 1988; Potter et al. 2005). Newer studies have dispelled this notion:
low-energy depositional conditions for fine-grained sediments are the exception, rather
than the rule (Schieber et al. 2007; Bhattacharya & MacEachern 2009; Schieber 2010;
Aplin & Macquaker 2011; Laycock 2014; Plint 2014). It is now generally accepted that
mudstones can be deposited in a variety of depositional settings via relatively high-energy
processes, as well as at different water depths.
Formation of a marine pycnocline – a boundary separating two layers of different

densities due to changes in water salinity or temperature – can influence the redistribution
of fine-grained sediment. Fine-grained particles entering shelf seas via riverine deltas
readily fall out of suspension from hypopycnal plumes as aggregates formed due to
flocculation and coagulation (Plint 2010, 2014). Flocculation denotes the adhesive
clustering and attachment of clay minerals, organic polymers, and clastic debris
(Grabowski et al. 2011). Large cations like sodium and calcium can reduce the thickness
of the electric double layer in clay minerals, facilitating coagulation – the agglomeration
of clay minerals with similar compositions – at higher salinities (Bennett et al. 1991;
Grabowski et al. 2011). Aggregation may be further enhanced by van der Waals forces,
as well as extracellular polysaccharides from biotic excretion (McCave 1984; Plint 2014).
Aggregates settle into a turbid mud “fluff” layer (nepheloid layer) that exists within 10
kilometres of river mouths (Grabowski et al. 2011). This nepheloid layer may flow
downslope and offshore as a hyperpycnal density current (Schieber et al. 2007). If
resuspension via storms occurs when the bottom nepheloid layer has begun to
consolidate, intraclastic aggregates (IAs) may be advected further offshore via along-
shore bottom currents like geostrophic flows (Plint 2010). IAs were observed by Plint
(2014) in rocks of the Cenomanian Dunvegan delta and by Jiang and Cheadle (2013) in
the Second White Specks Formation – suggesting that erosional working of partially-
144
lithified muddy substrate by storms was a widespread phenomenon during the Late
Cretaceous.
Evidence for advective mud transport in the Late Cretaceous WIS in relatively shallow
water (<100 m water depth) is well documented (Varban & Plint 2005; Varban & Plint
2008a, b; Bhattacharya & MacEachern 2009; Plint et al. 2012b; Shank & Plint 2013; Plint
2014). Plint (2014) determined that mudstone deposition on the distal prodelta, which was
greater than 120 km offshore from the Dunvegan delta of the Upper Cretaceous Colorado
Group, occurred at water depths of 20 to 70 m (SWB for mud). This process was aided by
the resuspension of fine-grained sediment aggregates by storms – it cannot be explained
by classic “marine snow” models of mud deposition. Storm waves can initiate offshore
mud deposition by triggering downslope flows of dense, fluid mud (McCave 1984;
Grabowski et al. 2011). Fluid muds are high particle-density and low wet bulk density
fluids with suspended sediment concentrations exceeding 10 g/L (Nishida et al. 2013).
Density flows of fluid mud are an important transport mechanism for fine-grained
sediment in the marine domain. Termed “wave-enhanced sediment gravity flows”
(WEGSFs) by Macquaker et al. (2010), WESGFs are recognized by a “triplet” pattern of
silt, intercalated silt and clay, followed finally by a clay drape. This pattern reflects an
initial wave-induced turbulent flow that gradually wanes in strength, eventually allowing
suspension settling to dominate.
In marine depositional settings, storms commonly generate combined flows. Combined

flows have both unidirectional and oscillatory components – these are produced by the
interaction of geostrophic flows and increased wave activity (Murray 1970; Nittrouer &
Wright 1994). Geostrophic flows are downwelling bottom currents that are generated by
gravity-driven relaxation flows from coastal set-up during storm surges. In the northern
hemisphere, the return flow is deflected to the right (Heezen et al. 1966; Swift et al.
1986). Varban and Plint (2008a) invoked storm-driven geostrophic flows as the primary
transport mechanism for mudstones of the Kaskapau Formation over 300 km offshore.
These bottom currents can efficiently entrain sediment and advect it obliquely or parallel
to the shoreline, up to hundreds of kilometres offshore. In this manner, muddy clinoforms
can experience substantial increases in environmental energy in their bottom set regions
145
while maintaining basinward progradation (Cattaneo et al. 2007; Laycock 2014). Formed
during highstand system tracts, mud can develop into prismatic, shore-parallel detached
“mud wedges” or “subaqueous deltas”. These are characterized by wave-ravined muddy
clinoforms that downlap onto a sediment-starved shelf (Abbott 2000; Cattaneo et al.
2003; Cattaneo et al. 2007; Plint 2010).
Core-to-log correlation
As previously discussed in Chapter 3, cores need to be correlated to geophysical well logs
in order to ground-truth petrophysical responses. This provides a theoretical basis for
extrapolating rock properties from sparsely located control points across a large area,
particularly in the absence of outcrop information – sedimentary successions can be
associated with distinct electrofacies, or “…[a] set of log parameters that characterizes a
sediment and permits the sediment to be distinguished from others” (Serra & Abbott
1982). As this study only has limited core data, it can be difficult to establish an
association between core-derived parameters (e.g., porosity and TOC) and log-derived
attributes (Amaefule et al. 1993). Amaefule et al. (1993) recognized that by dividing
reservoirs into statistically-defined hydraulic flow units, porosity and permeability could
be estimated reasonably well with limited core data. Instead, this study uses the
qualitative definition of hydraulic flow units described by Ebanks (1987), where it is
assumed that coeval rock units (in this case, allomembers) have defined relationships
between their petrophysical properties because they are genetically related. This provides
the interpreter with a way to extrapolate the available core data to distant wells.
Figure 6.2 illustrates a S-N oriented cross-section with all the cores logged by this author
(with the exception of the 7-19 core, which was included for reference) with their
corresponding log signatures beside them. Figure 6.3 illustrates the same cross-section,
with the full suite of petrophysical parameters that were calculated for this thesis (clay
volume, TOC, total and effective porosity, volumetric mineralogy, and brittleness). In
certain places (esp. BF1), the clay volume (VSH_GR) curve was a particularly useful
addition to the normal logs used for allostratigraphy (GR, DT, and RES) as it provided a
standardized measure of clay content in a particular interval.
146
6.5.1 Electrofacies
6.5.1.1 Facies 1
Facies 1 is most easily identified by its spiky, high GR reading, which is associated with
40
K in bentonite clays. This response may be subdued if the bentonite is particularly thin
(< 10 cm), as is the case in well 100/07-19-45-06W5, due to the limited vertical
resolution of standard GR tools ( ~0.4 to 0.9 m).
6.5.1.2 Facies 2
The geophysical wireline log response to facies 2 is characteristic of a clay-rich
composition: it has a relatively high GR response (lower than facies 1), reduced bulk
density, and faster sonic travel times. Despite the high proportion of organic content (a
lithology with high resistivity) in this facies, the elevated conductivity of clay minerals
present in the rock matrix likely contributes to a resistivity signature that is relatively low.
6.5.1.3 Facies 3
The GR tool response to facies 3, which is intermediately high, reflects a slightly reduced
proportion of clay minerals relative to facies 2. The upward transition of facies 2 into the
coarser-grained sediments of facies 3 is reflected as a gradually suppressed GR response.
Siderite (iron carbonate) nodules have a significant impact on the log character of facies
3, as the GR response sharply decreases – although this effect is dependent on the
concentration of nodules within the particular horizon. Sonic travel times in this facies are
slightly increased relative to facies 2, which may represent a marginal increase in porosity
that is associated with reduced clay content. The resistivity reading in facies 3 is slightly
elevated relative to facies 2, potentially responding to an increase in organic content and a
simultaneously decrease in clay content.
6.5.1.4 Facies 4 and 5

Several compositional factors contribute to facies 4 and 5 having a significantly
suppressed GR signature relative to facies 2 and 3. Of the facies described in this study,
facies 4 and 5 are the most heterolithic: they contain a higher proportion of calcareous
matrix, elevated organic content, reduced clay, and also exhibit a marked increase in
147
siderite nodule concentration. These mineralogical components, which are variably

interbedded and distributed, result in a GR signal that is reduced overall but also
inconsistent.
6.5.1.5 Electrofacies summary

The upward transition from the Upper Belle Fourche alloformation to the Second White
Specks alloformation is marked by an abrupt shift to a more calcareous matrix, reduced
clay content, a suppressed gamma ray profile, as well as elevated resistivity and TOC
(Bloch et al. 1993, Furmanm et al. 2014). To ensure allostratigraphic tops were picked
consistently across the entire area and could be easily utilized by other geologists, a
petrophysical tops guidance was developed (Table 6.2). Due to the limited petrophysical
contrast between some of the allomembers, multiple logs were used to confirm picks
(GR, DT, DEN, LLD, and neutron if it was available).
Allostratigraphic correlation
The facies from section 6.3, the electrofacies from section 6.5.1, and the allostratigraphic
framework described in section 6.2 were integrated in order to guide regional
allostratigraphic correlations and to estimate the spatial distribution of facies throughout
the area. Fig. 6.5 summarizes the relationship between these three datasets.
Figure 6.6 illustrates the four west-east oriented and three north-south oriented
allostratigraphic cross-sections that have been included for reference (Figs. 6.7 through
6.13). In cross-section, the Upper Belle Fourche and Second White Specks allomembers
mapped in this study exhibit varied geometries. BF1 appears to fill in a limited amount of
erosional topography onto the K1 disconformity and thickens overall from southwest to
northeast. BF2, which appears to be preserved in its entirety, is relatively sheet-like
overall, but thins from west to east. In contrast, BF 3 exhibits wedge-like geometry,
experiencing pronounced thickening northwards to townships 40-43 before becoming
thinner north of township 42-43. Finally, allomember VII also exhibits a northwards-
thickening trend, without the pronounced bulge exhibited by BF3.
148
Allomember name Marker Log character Common pitfalls

Often impacted by
borehole breakout
Elevated GR relative to
Allo VIII -
Allo VII
which makes the true
GR character difficult
Second to discern
White
Specks High porosity units are
Suppressed GR and
very thin (<1 m) and
Flooding surface at slow sonic related to
Allo VII the top increased carbonate
can be difficult to
correlate due to limited
content
lateral extent
“Coarse” facies may

not be present at the top
of the well due to
“Red” bentonite “Coarsening upwards”
lateral variation or
BF3 from Tyagi et al. pattern in GR with high
subtle erosion, use
(2007) GR unit at base
offsetting wells to
confirm if erosion has
occurred
Towards the north of

the study area, coarser
Contains two lower-
facies appears
Upper order “coarsening
“transitional” into BF3
Belle BF2 Flooding surface upward” cycles –
due to limited vertical
visible in sonic and
Fourche petrophysical contrast;
resistivity
use other logs to
confirm pick
Highest GR of all
Upper BF units, Drapes over K1
borehole breakout is unconformity, which
common, elevated has some topographic
BF1 Flooding surface resistivity; serrated GR irregularity – this unit
pattern with no can be very thin in
apparent coarsening places but is always
upward or fining present
upward pattern
Double “spike” in
GR, DT, DEN, and Elevated GR, lower “Spike” may be subtle
Lower Belle Fourche LLD; correlates to density relative to in certain wells, can
(Lower BF) K1 unconformity Upper BF from higher confirm pick with
from Varban and organic content resistivity and sonic
Plint (2008a)
Table 6.2: Petrophysical tops guidance for the Second White Specks and Belle Fourche alloformations in
the Willesden Green – Gilby area of west-central Alberta. This includes distinguishing log characteristics in
addition to problems the interpreter may encounter while picking tops.
149
Fig. 6.5: A composite cross-section, using the “red” bentonite as a datum, of the Lower Second White Specks and Belle Fourche alloformations across the study
area. Net reservoir intervals are shaded in purple (Belle Fourche) and green (Second White Specks)
150
6.6.1 Upper Belle Fourche (BF2 and BF3)

In core, the Upper Belle Fourche contains two several coarsening up sequences
(allomembers BF2 and BF3) that are bounded above by sharp-based erosional contacts or
shelly lags (Fig. 6.2A, 6.2G, 6.2H). These units are clay-rich, and display increases in silt-
and carbonate-content towards the tops of individual parasequences (Fig. 6.3). BF3
thickens north of township 36, reaching maximum thickness in 102/14-31-041-06W5.
North of township 42, BF3 is clearly truncated by the overlying Second White Specks. In
BF2, facies 3 and 4 are thicker towards the south of the study area and appear to gradually
transition into facies 2 laterally, disappearing entirely north of township 41. These
allomembers are interpreted to record an overall rise in base level, which was punctuated
by brief base level falls that allowed for coarser sediment supply from the Kaskapau delta
to be transported farther into the basin via combined flows.
BF 3 is anomalously thick in townships 41 and 42; this could be due to tectonic

thickening and/or differential erosion of the top of the unit. The appearance of facies 3
suggests that higher energy processes dominated in this sequence over BF2, possibly due
to shallower water. The disappearance of facies 3 and 4 in 100/04-27-39-06W5 appears to
suggest a subtle erosional unconformity at the top of BF 3, but it could also be caused by
a lateral facies change. This facies change could be facilitated by a channelized or
lenticular combined flow preferentially moving coarse sediment to other parts of the
study area, such as townships 41-42.
Figure 6.5 illustrates the irregular nature of the boundary between the Second White
Specks and Upper Belle Fourche alloformations, as well as its potential relationship with
production. The boundary between Second White Specks allomember VII (green) and the
Belle Fourche is unconformable – the lower gamma ray unit at the top of the Belle
Fourche 3 allomember appears to truncate against the “red” bentonite both to the north
and south of township 41, reappearing in a condensed form south of township 38. This
electrofacies – outlined in purple – does not reappear in the 4-27 or 7-19 cores.
The 14-31 well, which is one of the best producing wells in the area, is included for
comparison to the 102/08-36 offset well – in 14-31, allomember BF3 has developed a
151
Study area with core-cross section locations
C D
B
E
E’
F
F’
G
G’
Ed
ge
H’
of
M
es
oz
oi
H
c
de
fo
B’
rm
at
C’ D’
io
n
fro
nt
metres
Wells in cross-section
Wells with core
Fig. 6.6: Reference map for summary allostratigraphic cross-sections, with core locations for reference.
152
Fig. 6.7: Allostratigraphic cross-section B to B’, which runs north-south.
B B’ Aut hor: Kienan MARI ON

(I D: kmar ion2)
Scale: 1: 250 Dat e: 13/ 12/ 2017
Well: 100060304305W500 Well: 100081504205W500 Well: 100162604105W500 Well: 102060704005W500 Well: 100062303905W502 Well: 100082803805W500 Well: 102061003805W500 Well: 100072603705W500 Well: 100140103705W500 Well: 100062203605W500
UWI: 100060304305W500 UWI: 100081504205W500 UWI: 100162604105W500 UWI: 102060704005W500 UWI: 100062303905W500 UWI: 100082803805W500 UWI: 102061003805W500 UWI: 100072603705W500 UWI: 100140103705W500 UWI: 100062203605W500
Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name:
Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name:
Empty variable Empty variable

CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500
( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT ( M) GR DT ( M) GR RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP
1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100
1825 1850
2050
2075
2075
2200
1975
1875
2025
Allo VII
2075
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
1875
Allo VII
Allo VII
1850
2075
BF_3 Allo VII

2100
2100
2225
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
2000
BF_3
BF_3
BF_3
1900
2050
BF_2
BF_2
BF_2
BF_2
2100
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
1875 1900
2100
BF_1
BF_1
2125
BF_1
2125
BF_1
BF_1
2250
BF_1
BF_1
BF_1
BF_1
BF_1
2025
1925
2075
2125
1900 1925
2125
2150
Fig. 6.8: Allostratigraphic cross-section C to C’, which runs north-south.
C
Scale: 1: 250
C’ Aut hor: Kienan MARI ON
(I D: kmar ion2)
Dat e: 13/ 12/ 2017
Well: 100141604404W500 Well: 100161704304W500 Well: 100160904204W500 Well: 100162804203W500 Well: 100072004104W500 Well: 100101104104W500 Well: 100082804004W500 Well: 100012103904W500 Well: 100102303804W500 Well: 100062303704W500 Well: 100031703704W500 Well: 100112503605W500
UWI: 100141604404W500 UWI: 100161704304W500 UWI: 100160904204W500 UWI: 100162804203W500 UWI: 100072004104W500 UWI: 100101104104W500 UWI: 100082804004W500 UWI: 100012103904W500 UWI: 100102303804W500 UWI: 100062303704W500 UWI: 100031703704W500 UWI: 100112503605W500
Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name:
Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name:
CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500
( M) GR DT RES_DEP ( M) GR RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP
1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100
1900 2025
1725 1675
Lorem ipsum
1775 1900
1800
1875
1750
Allo VII
Allo VII
Allo VII
1850
Allo VII
Allo VII
Allo VII
Allo VII
1975 2125
Allo VII
Allo VII
BF_3 Allo VII

1925 2050
BF_3 Allo VII
BF_3 Allo VII

BF_3
BF_3
BF_3
1750 1700
BF_3
BF_3
BF_3
BF_3
BF_3
1925
BF_3
1800
1825
1900
1775
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
1875
BF_2
BF_2
BF_2
BF_2
BF_2
2000 2150
BF_1
BF_1
1950 2075
Zone_1BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
1775 1725
1825 1950
1850
1925
1800
1900
2025 2175
1975 2100
Fig.
Fig.6.10:
6.9: Allostratigraphic cross-section D to D’, which runs north-south.
D D’
Scale: 1:240 Date: 13/12/2017
Well: 100080904403W500 Well: 100020904303W500 Well: 100122004303W500 Well: 100162804203W500 Well: 100050404203W500 Well: 100141604203W500 Well: 100033504103W500 Well: 100163604103W500 Well: 100050304003W500 Well: 100072804003W500 Well: 100160504003W500 Well: 100032703903W500 Well: 100080403903W500 Well: 100060403903W500 Well: 100061703803W500 Well: 100141803803W500 Well: 100132703703W500 Well: 100010603703W500 Well: 100060603703W500 Well: 100042603603W500
UWI: 100080904403W500 UWI: 100020904303W500 UWI: 100122004303W500 UWI: 100162804203W500 UWI: 100050404203W500 UWI: 100141604203W500 UWI: 100033504103W500 UWI: 100163604103W500 UWI: 100050304003W500 UWI: 100072804003W500 UWI: 100160504003W500 UWI: 100032703903W500 UWI: 100080403903W500 UWI: 100060403903W500 UWI: 100061703803W500 UWI: 100141803803W500 UWI: 100132703703W500 UWI: 100010603703W500 UWI: 100060603703W500 UWI: 100042603603W500
Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name:
Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name:
Empty variable Empty variable Empty variable
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX CAX
Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4 Reference 152.4 MM 406.4
(M) GR RES_DEP (M) GR RES_DEP (M) GR RES_DEP (M) GR DT RES_DEP (M) GR RES_DEP (M) GR DT RES_DEP (M) GR DT RES_DEP (M) GR DT RES_DEP (M) GR DT (M) GR RES_DEP (M) GR DT RES_DEP (M) GR RES_DEP (M) GR RES_DEP (M) GR RES_DEP (M) GR DT RES_DEP (M) GR RES_DEP (M) GR DT RES_DEP (M) GR DT RES_DEP (M) GR DT RES_DEP (M) GR RES_DEP
1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 350 US/M 200 2 OHMM 100 1:240 0 GAPI 250 2 OHMM 100
1800 1925
1850
1775 1750 1800
1675 1775
1675 1675
1750 1800
1775 1875 1875
1600
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
1675
BF_3 Allo VII
BF_3 Allo VII

1775
BF_3 Allo VII

BF_1 BF_2 BF_3 Allo VII
1950
BF_3Allo VII
1725
BF_3Allo VII
BF_3Allo VII
BF_2 BF_3Allo VII
Allo VII
1950
Allo VII
1825
Allo VII
BF_3
1875
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
1800 1775 1825
BF_3
1700 1800
1700 1700
1775
BF_2 BF_3
1825
BF_2 BF_3
BF_3
BF_2 BF_3
1800 1900 1900
1625
BF_1 BF_2
BF_1 BF_2
BF_1 BF_2
BF_1 BF_2
BF_1 BF_2
BF_1 BF_2
BF_1 BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
1700
1800
1750 1975
BF_1
1975
BF_1
1850
BF_1
BF_1
BF_1
BF_1
BF_1
1900
BF_1
BF_1
BF_1
1825 1800 1850
1725 1825
BF_1
BF_1
1725 1725
1800 1850
1825 1925 1925
1650
1725
1825
1775 2000
1875 2000
1925
1850 1825 1875
Fig. 6.10: Allostratigraphic cross-section E to E’, which runs east-west.
E’
Scale: 1: 250
E (I D: kmar ion2)
Dat e: 13/ 12/ 2017
Well: 100130704110W500 Well: 100113504109W500 Well: 100040304208W500 Well: 100093404108W500 Well: 100123304107W500 Well: 100073404107W500 Well: 102083604107W500 Well: 100040604206W500 Well: 102143104106W500 Well: 100163304102W500 Well: 100162604105W500 Well: 100062504105W500 Well: 100113004104W500 Well: 100163604103W500
UWI: 100130704110W500 UWI: 100113504109W500 Country: UWI: 100040304208W500 UWI: 100093404108W500 UWI: 100123304107W500 UWI: 100073404107W500 UWI: 102083604107W500 UWI: 100040604206W500 UWI: 102143104106W500 UWI: 100163304102W500 UWI: 100162604105W500 UWI: 100062504105W500 UWI: 100113004104W500 UWI: 100163604103W500
Short name: Short name: Field: FERRIER Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name:
Long name: Long name: State: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name:
Company: CANADIAN NATURAL
RESOURCES LIMIT
CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Reference 150 MM 550 Reference 100 MM 1500 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550
( M) GR DT ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP
1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100
2325 2150
1875 1875
2125 2125 1850
1600
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
1675
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
2125
Allo VII
2675 2175
2175
2175
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
2175
BF_3
BF_3
BF_3
BF_3
2350
BF_3
BF_3
BF_3
1900 1900
2150 2150 1875
1625
BF_3
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
1700
BF_2
BF_2
2150
BF_2
BF_2
2700
BF_2
2200
2200
BF_2
BF_1
2200
BF_1
BF_1
BF_2
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
2375 2200
BF_1
BF_1
BF_1
1925 1925
2175 2175 1900
BF_1
1650
1725
2175
2725 2225
2225
2225
2225
Fig. 6.11: Allostratigraphic cross-section F to F’, which runs east-west.
F’
Scale: 1: 250
F Aut hor: Kienan MARI ON
(I D: kmar ion2)
Dat e: 13/ 12/ 2017
Well: 100130704110W500 Well: 100113504109W500 Well: 100040304208W500 Well: 100093404108W500 Well: 100123304107W500 Well: 100073404107W500 Well: 102083604107W500 Well: 100040604206W500 Well: 102143104106W500 Well: 100163304102W500 Well: 100162604105W500 Well: 100062504105W500 Well: 100113004104W500 Well: 100163604103W500
UWI: 100130704110W500 UWI: 100113504109W500 UWI: 100040304208W500 UWI: 100093404108W500 UWI: 100123304107W500 UWI: 100073404107W500 Country: UWI: 102083604107W500 UWI: 100040604206W500 UWI: 102143104106W500 UWI: 100163304102W500 UWI: 100162604105W500 UWI: 100062504105W500 UWI: 100113004104W500 UWI: 100163604103W500 Country:
Short name: Short name: Short name: Short name: Short name: Short name: Field: WILLESDEN GREEN Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Field: GILBY
Long name: Long name: Long name: Long name: Long name: Long name: State: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: State:
Company: YANGARRA RESOURCES Company: MOSAIC ENERGY L
CORP.
CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI CALI
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Reference 150 MM 550 Reference 100 MM 1500 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550 Reference 150 MM 550
( M) GR DT ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP
1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100
2325 2150
1875 1875
2125 2125 1850
1600
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
1675
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
2125
Allo VII
2675 2175
2175
2175
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
2175
BF_3
BF_3
BF_3
BF_3
2350
BF_3
BF_3
BF_3
1900 1900
2150 2150 1875
1625
BF_3
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
1700
BF_2
BF_2
2150
BF_2
BF_2
2700
BF_2
2200
2200
BF_2
BF_1
2200
BF_1
BF_1
BF_2
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
BF_1
2375 2200
BF_1
BF_1
BF_1
1925 1925
2175 2175 1900
BF_1
1650
1725
2175
2725 2225
2225
2225
2225
Fig. 6.12: Allostratigraphic cross-section G to G’, which runs east-west.
G’ G
Scale: 1: 250 Dat e: 13/ 12/ 2017
Well: 100101404009W500 Well: 100022004008W500 Well: 100113104006W500 Well: 100070604105W500 Well: 100103104005W500 Well: 100030204105W500 Well: 100082804004W500 Well: 100060504103W500 Well: 100040204103W500 Well: 100063504002W500
UWI: 100101404009W500 UWI: 100022004008W500 UWI: 100113104006W500 UWI: 100070604105W500 UWI: 100103104005W500 UWI: 100030204105W500 UWI: 100082804004W500 UWI: 100060504103W500 UWI: 100040204103W500 UWI: 100063504002W500
Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name: Short name:
Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name: Long name:
GRLD
0 GAPI 250
CALI CALI CALI CALI CALI CALI CALI CALI
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference 100 MM 1500 Reference Reference
( M) GR DT ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) GR DT ( M) GR DT RES_DEP ( M) GR DT RES_DEP ( M) RES_DEP ( M) DT
1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 1:250 2 OHMM 100 1:250 350 US/M 200
1850 1750
1650
2425
2125
1800
2350 2025
2075
2075
1875 1775
1675
2450
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
Allo VII
2150
Allo VII
1825
2375 2050
2100
2100
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
BF_3
1900 1800
1700
2475
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
2175
1850
2400
BF_1
2075
BF_1
BF_1
BF_1
2125
BF_1
BF_1
BF_1
BF_1
2125
BF_1
BF_1
1925 1825
1725
2500
2200
1875
2425 2100
2150
2150
1950 1850
1750
2525
Fig. 6.13: Allostratigraphic cross-section H to H’, which runs east-west.
H’
Scale: 1:250
H Author: Kienan MARION
(ID: kmarion2 )
Date: 13/12/2017
Well: 100050303809W500 Well: 100082903708W500 Well: 100141803707W500 Well: 100082703606W500 Well: 100062203605W500 Well: 102081503605W500 Well: 100112503605W500 Well: 100010603703W500 Well: 100061203702W500
UWI: 100050303809W500 UWI: 100082903708W500 UWI: 100141803707W500 UWI: 100082703606W500 UWI: 100062203605W500 UWI: 102081503605W500 untry: UWI: 100112503605W500 UWI: 100010603703W500 UWI: 100061203702W500
Short name: Short name: Short name: Short name: Short name: Short name: Short name:
eld: GARRINGTON Short name: Short name:
Long name: Long name: Long name: Long name: Long name: Long name: ate: Long name: Long name: Long name:
mpany: NAL RESOURCES LIMITED
PHIT_E_KER PHIT_E_KER PHIT_E_KER PHIT_E_KER PHIT_E_KER PHIT_E_KER PHIT_E_KER

Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
Z_FINAL
CALI PHIT_E_KER CALI CALI PHIT_E_KER CALI PHIT_E_KER CALI CALI TOC PHIT_E_KER CALI PHIT_E_KER CALI PHIT_E_KER CALI PHIT_E_KER
Reference 150 MM 550 0.15 v/v 0 Reference 150 MM 550 Reference 150 MM 550 0.15 v/v 0 Reference 150 MM 550 0.15 v/v 0 Reference 150 MM 550 Reference 150 MM 550 0 v/v 0.05 0.15 v/v 0 Reference 150 MM 550 0.15 v/v 0 Reference 150 MM 550 0.15 v/v 0 Reference 150 MM 550 0.15 v/v 0
(M) GR DT RES_DEP toc_i PHIT_CORR (M) GR DT RES_DEP toc_i (M) GR DT RES_DEP toc_i PHIT_CORR (M) GR DT RES_DEP toc_i PHIT_CORR (M) GR DT RES_DEP toc_i (M) GR RES_DEP toc_i PHIT_CORR (M) GR DT RES_DEP toc_i PHIT_CORR (M) GR DT RES_DEP toc_i PHIT_CORR (M) GR DT RES_DEP toc_i PHIT_CORR
1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 1:250 0 GAPI 250 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0 1:250 0 GAPI 250 350 US/M 200 2 OHMM 100 0 v/v 0.05 0.15 v/v 0
1700
2625 2125
2350
BF_3 Allo VII

2200 1950
BF_3Allo VII
BF_3Allo VII
VII
BF_3Allo VII
II
BF_3Allo V
II
2825
BF_2 BF_3Allo VII

2225
2525
BF_3
BF_3
1725
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
BF_2
2650 2150
2375
2225 1975
BF_1
BF_1
2850
BF_1
BF_1
BF_1
BF_1
BF_1
2250
2550
BF_1
BF_1
1750
2675 2175
2400
2250 2000
2875
thicker section of reservoir (purple). Allomember BF3 is thinner in 102/08-36 and was
not perforated, potentially indicating a bypassed pay scenario. Notably, this unit
disappears south of township 41 and north of township 42, only reappearing south of
township 38. Because the underlying Belle Fourche units are not significantly altered in
the 4-27 well, where the reservoir interval is missing, it is suspected that there has been
some type of lateral facies change or subtle regional erosion, rather than a major uplift
and/or erosional event. In contrast, in the 7-19 well, all Belle Fourche units “jump”
upwards and the top of the unit is truncated, potentially indicating erosion that occurred
prior to Second White Specks deposition.
BF3 is mantled by a bentonite (the Bighorn River “red” bentonite), facies 1. It is

interpreted to record a volcanic eruption or a series of volcanic eruptions from the western
Cordillera, which spread volcanic ash across the WIS. This ash fell out of suspension and
was eventually deposited on the seafloor. The presence of the bentonite in all wells
suggests that the bentonite ashfall was likely deposited after the sedimentary hiatus that
occurred during the Cenomanian-Turonian boundary.
6.6.2 Second White Specks (Allomember VII)

Allomember VII of the Second White Specks thins from north to south. Compositionally,
allomember VII contains less clay and more carbonate content than the Belle Fourche;
this is consistent with core analysis from Tuzo Energy (Canadian Discovery 2014). This
allomember is internally complex. Three groups of “reservoir facies” units subdivide this
unit in Fig. 6.4 (pale green) and were correlated on the basis of similar grain size patterns.
The presence of multiple siderite pebble lags and the vertical stacking of the “reservoir
facies” units suggests some degree of amalgamation, possibly by storms. Storm-deposited
sediments on a shallow marine shelf may develop into laterally continuous, amalgamated
sheets that have enhanced permeability and porosity at the top of each upwards
coarsening sequence (Atkinson et al. 1986; Gaynor & Schiehing 1988).
Allomember VII represents a shift to higher energy transport mechanisms in a slightly

shallower environment, represented by the predominance of coarse-grained facies 4 and
5, as well as the absence of facies 2. The frequent presence of soft-sediment deformation,
160
siderite lags, and erosional surfaces are interpreted to represent many cycles of combined
flow winnowing, erosion, and general predominance over settling processes. At least five
different packages can be identified in allomember VII on the basis of grain size patterns;
these may reflect cyclicity in ocean currents and climate patterns, as well as pulses of sea-
level fall.
Allomember VII is bounded by an erosive transgressive lag and a major flooding surface,
which was likely generated by a marine transgression.
6.6.3 Overall interpretation

The facies sequence of Cenomanian to Turonian-aged mudstones of the Lower Colorado
Group in west-central Alberta are consistent with a shallow marine environment affected
by combined flows during a global highstand, which was punctuated by brief periods of
relative base level fall and largely remained within storm wave base. The Second White
Specks and Upper Belle Fourche-equivalent Kaskapau alloformation represents the
majority of the Greenhorn transgressive-regressive cycle, which peaked during the
Turonian (Kauffman & Caldwell 1993). During that time, several small-scale base level
fluctuations occurred (Haq 2014), which initiated lower-order regressions and basinward
progradation.
Highlights
• The allostratigraphic interval of interest in this study is constrained below by the K1
disconformity, and above by the allomember VII flooding surface. By extending the
allostratigraphic correlations of Tyagi et al. (2007) and subdividing his allomembers
on the basis of previously uncorrelated flooding surfaces.
• Facies observed in this study are consistent with a distal to mid-shelf, marine, low-
gradient ramp setting where water depths ranged between 40 and 70 m. Fine-grained
sediments were transported offshore by storm-generated combined flows.
• The contact between the Belle Fourche and Second White Specks alloformation,
indicated by the presence of the “red” bentonite, is unconformable and occasionally
erosional across the study area. This is consistent with the work of Tyagi et al. (2007)
and Plint et al. (2012a).
161
Solving the Second White Specks 7. Depositional trends
Chapter 7
7 Depositional trends of the Second White Specks and
Belle Fourche alloformations in the Willesden Green –
Gilby region, Alberta, Canada
Overview
Structure, thickness, and isolith maps of the Lower Second White Specks and Belle
Fourche alloformations in the Willesden Green area of west-central Alberta are presented.
Structure mapping
A structure map establishes the subsurface geometry of a geologic surface, including
regional dip and areas that may be deformed or faulted (Nelson et al. 1999). In this study,
structure maps were primarily used to delineate regional dip and structural features.
Contour lines on a structure map represent lines of equal subsurface elevation or depth of
a particular allomember. The shape and location of contour lines on a structure map can
be used to infer the shape of three-dimensional structures (Lisle 2004):
• Parallel, equally spaced contours represent a uniformly dipping surface;
• Closely spaced contours represent a steeply dipping surface;
• Closed concentric arrangements of contours reveal isolated hills or bowl structures;
• Valleys and ridges, which may indicate fault locations, give rise to V-shaped patterns.
Structure maps for each allomember were generated from subsurface stratigraphic picks
using the kriging algorithm in Surfer. Changes in the geometry of surfaces bounding
individual allomembers can delineate structural features that may have influenced
deposition, and potentially identify structural traps for future petroleum exploration.
In the lower Colorado Group of west-central Alberta, the changes between structural
maps for each allomember are extremely subtle and thus only one has been included for
reference. Figures 7.1 and 7.2 illustrate the subsea elevation of allomember VII in two-
and three-dimensions, respectively. As evidenced by subparallel, approximately
equidistant contour lines of constant subsea elevation, this surface dips relatively
uniformly to the southwest up until the edge of the Mesozoic deformation front.
162
Fig. 7.1: 2D structure map of the top of allomember VII, with the Mesozoic deformation front for reference.
163
5
Allomember VII surface in 3D
T4
4
T4
3
T4
2
T4
1
T4
0
T4
9
T3
8
T3
7
T3
6
T3
5
T3
W5
R1
W5
R2
W5
R3
W5
50 R4
-5 00
- 6 50
-6 00 W5
-7 50 R5
- 7 00
-8 50
-8 00 W5
- 9 50 R6
-9 000
-1 050
-1 100 W5
-1 150 R7
De
- 1 200
-1 250
pth
-1 300 W5
- 1 350 R8
(su
-1 400
-1 450
bs
- 1 500 W5
R9
ea
-1 550
-1 600
,m
-1 650 5
-1 700 0W
)
- 1 750
-1 800 R1
-1 850
-1
Fig. 7.2: 3D structure map of the top of allomember VII.
164
In three dimensions, the effect of the Laramide orogeny on lower Colorado Group strata
is clearly seen (Fig 7.2) – the Mesozoic deformation front is oriented approximately
NNW and dips towards the southwest. This is the prevailing dip direction and structural
style of the entire region. Localized perturbations associated with small-scale structural
relief are more easily visible in three-dimensions, particularly in townships 41 to 42 from
R6 to 7W5.
Isochore mapping
A “depocentre” refers to an area of maximum stratigraphic thickness where deposition
was concentrated (Gary et al. 1974). Basins can have multiple depocentres and their
locations may shift over time if they experience lateral changes in sediment supply, have
multiple drainage areas, or have complicated subsidence histories (Armentrout 1999). The
spatial distribution of facies within a particular depocentre in a shelf setting is controlled
by the ratio of accommodation to sediment supply (Van Wagoner et al. 1988), as well as
by the mechanism of sedimentary transport across the shelf (Armentrout 1999). By
mapping coeval stratigraphic thicknesses (i.e., allostratigraphically-defined isochores),
shifts in the extent and locations of depocentres can temporally and spatially constrained
within a particular basin (Armentrout 1999).
Contour lines on an isochore map represent lines of equal true vertical thickness (Lisle
2004). Isochore maps can be interpreted on the basis of contour spacing and geometry, as
follows:
• Parallel, widely spaced contours represent a sheet-like stratigraphic unit whose
thickness is relatively constant
• Parallel, equally spaced contours represent a wedge-like stratigraphic unit whose
thickness is changing at a constant rate;
• Parallel, closely spaced contours represent a wedge-shaped stratigraphic unit whose
thickness is changing at a variable rate;
• Closed concentric arrangements of increasing contours reveal isolated regions of
increased stratigraphic thickness, which could represent sedimentary depocentres;
165
• Closed concentric arrangements of decreasing contours reveal isolated regions of

decreased stratigraphic thickness, which could reveal eroded regions or areas with
decreased available accommodation;
• Valleys and ridges indicate irregular stratigraphic geometry (i.e., not sheet-like).
Figures 7.3, 7.4, 7.5 and 7.6 illustrate the isochore thickness of allomembers BF1, BF2,
BF3 and VII, respectively.
7.3.1 Allomember BF1

Allomember BF1 (Fig 7.3), which is bounded below by the K1 disconformity and above
by a major flooding surface, illustrates an overall southward-thickening trend. It ranges
from a maximum thickness of 24 m in T37-R6W5 to a minimum thickness of 0.3 m in the
northwest corner of the study area. This allomember has two NNW-oriented zones of
increased thickness outlined by the 12 m contour. These features, interpreted as
depocentres, extend from township 36 to township 40 and are approximately 20
kilometres wide. These depocentres do not follow the westward-thickening trend that is
expected for the lower Colorado Group due to flexural loading from the Laramide
orogeny during the Late Cenomanian (cf. Figs. 2.1 and 2.2). The location of these
depocentres is likely related to pockets of accommodation that developed during the
deposition of allomember BF1, which could have been generated by localized subsidence
or by erosional topography on the K1 surface.

Allomember BF2 (Fig. 7.4), bounded below by BF1 and above by a major flooding
surface, depicts an overall westward-thickening trend – emulating the expected
depositional pattern of the lower Colorado Group (cf. Fig 2.1). This thickness pattern
suggests clockwise rotation and a northwest shift in the location of maximum
accommodation from allomember BF1, which had reached maximum thickness in the
southern part of the study area. Allomember BF2 ranges from a maximum thickness of
over 30 m in T41 RW5 to approximately 5 m in the northeast corner of the study area, but
averages 10 to 12 m thick across most of the study area. East of R6W5, the geometry of
this allomember is relatively sheet-like – contours are widely spaced (particularly
166
Fig. 7.3: Isochore map of allomember BF1.
167
168
169
Fig. 7.6: Isochore map of allomember VII.
170
between 12 and 8 m). Allomember BF2 develops a north-south oriented depocentre west
of R8W5 that stretches from T37 to 43 – although well control is poor in this region.

Allomember BF3 (Fig. 7.5), which is bounded below by the BF2 flooding surface and
bounded above by the “red” bentonite, reveals an overall northwest thickening trend
across the study area – it ranges from a minimum thickness of approximately 4 m in T35
R5W5 to a maximum thickness of 19 m in T43 R3W5. The 12 m thickness contour
reveals a large, oblong depocentre, approximately 25 km wide and 90 km long, that is
oriented 60 degrees east of north. This depositional pattern represents an overall
clockwise and northeast sense of rotation to the location of maximum accommodation in
the study area, relative to allomembers BF2 and BF1. The westward edge of this
depocentre, defined by T44 R8W5 and T43 R9W5, exhibits a NNE-trending thin that
extends ~40 km south into T41 R9W5 – it is possible that this “thinning” trend may
represent erosion of BF3 prior to “red” bentonite deposition.
7.3.4 Allomember VII

Allomember VII (Fig. 7.6), which is bounded below by the “red” bentonite and above by
a major flooding surface, depicts an overall northeast thickening trend. This allomember
ranges from a maximum thickness of approximately 16 m in T41 R7W5 to a minimum
thickness of 3 m in the southwest corner of the study area. The 9.5 m contour delineates a
depocentre in the northeast that trends at approximately 30 degrees west of north; this
depositional pattern represents an overall clockwise rotation in maximum accommodation
in the study area relative to allomembers from the Belle Fourche alloformation. An
irregular, “finger” shaped departure from the main depocentre extends from T41 R7W5
just over 40 km towards the south. Similar to allomember BF3, this depocentre is defined
to the west by a NNE-trending thin that extends 40 km south from T43 R8W5 into T39
R10W5.
171
Isolith mapping
An isolith map made within a coeval interval can be used to reconstruct paleogeography
and establish the distribution of reservoir facies (Armentrout 1999; Tearpock & Bischke
2002). Isolith mapping are typically generated by creating a map of the “net to gross” or
“sand to shale” ratio where the thickness of “net sand” is divided by the gross isochore
thickness of the entire interval (Tearpock & Bischke 2002). Net sand is determined by
adding up the total thickness of sand, which is determined using a GR cutoff (e.g., 50
API) in a stratigraphic unit across a series of wells (Dikkers 1985). By constructing an
isolith map in this way, only geologic units that exceed a threshold GR value are
contoured (see Varban and Plint, 2008b). GR logs have varied responses to common
lithological components of unconventional reservoirs (e.g., clay minerals and kerogen
content) like the Second White Specks Formation – applying a GR cutoff, therefore, does
not adequately capture the amount of non-clay minerals in a stratigraphic zone. The net
sand method is used to establish the distribution of a particular lithology (i.e., sand)
across an area. This implies that net sand contours are associated with variations in grain
size (le Roux 1993) – grain size can only be accurately determined through observations
of measured sections, core, or drill cuttings. GR logs record variations in radioactivity,
which does not always correlate with grain size (see section 3.6.2.1.1).
Composite isolith maps can be generated by superimposing different base maps (Sloss et
al. 1960) – this is most effective when no more than two or three maps are overlain
(Merriam & Jewett 1989). In this study, composite isolith maps were generated by
overlapping the isochore maps with weighted interval average maps of clay volume. Clay
volume is an important part of reservoir assessment as the presence of clay minerals in
the rock matrix can reduce porosity (see Fig. 5.4). Areas with low clay volume may have
higher porosity, and therefore have increased storage capacity. The composite isolith
maps are interpreted as follows:
• Areas where low clay volume values (orange) overlap with regions that have elevated
isochore thickness represent thick accumulations of clay-poor facies, where
accommodation increased and coarse sediment supply was available (i.e., facies 3
through 5);
172
• Regions with low clay volume overlapping with regions of reduced isochore thickness
represent thin accumulations of clay-poor facies, where accommodation was reduced
but coarse-grained sediment supply was still available;
• Areas where high clay volume values (grey) overlap with greater stratigraphic
thickness represent accumulations of clay-rich facies (i.e., facies 2), where
accommodation was increased but coarse sediment supply was not available;
• Regions with high clay volume that overlap with regions of reduced isochore
thickness represent thin accumulations of clay-rich facies, where accommodation was
limited, and coarse-grained sediment supply was not available;
• Shifts in the location of clay-poor facies between allomembers should indicate shifts
in sediment supply locations.
Figures 7.7, 7.8, 7.9 and 7.10 illustrate the composite isolith maps of allomembers BF1,
BF2, BF3 and VII, respectively, with clay volume in grey and orange and isochores
overlaid as transparent contours. As fewer wells were used to calculate clay volume than
were used to calculate isochore thickness, the data points that were used to generate these
maps do not perfectly overlap – although their grid geometries are identical. Only wells
that were completed in the Second White Specks and Belle Fourche Formations have
been included on these maps.

The weighted average clay volume of allomember BF1 (Fig. 7.7) is lowest in the
southeast corner of the study area, which largely overlaps with the major depocentre of
this unit (Fig. 7.3). Clay volume in this interval ranges from 0.32 in the south east corner
of the study area to a maximum of 0.84 at the northern edge of the study area. Because
low clay volume overlaps with the region of increased isochore thickness, it is interpreted
that a source of coarse sediment supply for allomember BF1 was located in the south and
sufficient accommodation was available in this region for sediment to accumulate.

The clay volume of allomember BF2 (Fig. 7.8) is highest (0.7) in the northeast corner of
the study area and reaches a minimum value of 0.06 in T41 R10W5. The clay volume of
173
thin
thick
Fig. 7.7: Isolith (isochore thickness overlaid on clay volume) of allomember BF1.
174
thick
thin
175
thick
thick
thin
176
thick
thin
Fig. 7.10: Isolith (isochore thickness overlaid on clay volume) of allomember VII.
177
allomember BF2 increases from west to east across the study area. The region of lowest
clay volume is best constrained by clay volume contours less than 0.22, which are
concentrated in an area west of R7W5 that is oriented approximately north-south. This
region of lowest clay volume largely overlaps with the depocentre indicated in Fig. 7.4,
which is approximated by an isochore thickness of 13 m. There is a region in the southern
part of the study area with relatively low clay volume (<0.22) and reduced isochore
thickness of BF2, which overlies with the depocentre and source of coarse sediment
indicated for BF1. This indicates that coarse sediment was still being supplied to that area
but was no longer the main sedimentary depocentre. The main depocentre and source of
coarse-grained sediment for BF2 is located in the west, representing an overall clockwise
sense of rotation of maximum accommodation and sediment supply.

The clay volume of allomember BF3 (Fig. 7.9) ranges from 0.1 in T40 R9W5 to 0.55 in
T37 R7W5. There are two trends of low clay volume that are relatively linear and trend at
60 degrees west of north. The first begins in T44 R4W5 and is 30 km wide, extending 60
km southeast to T41 R1W5. The other begins in T41 R10W5, ranges from 20 to 40 km
wide, and stretches southeast to T38 R1W5. These two low clay volume trends are
separated by a region of high clay volume that passes through the middle of the main
depocentre in Fig. 7.5. This is notable as the high clay volume trend passes through the
middle of the main depocentre – this region had available accommodation for sediment,
but coarse-grained sediment was not evenly distributed across the area of increased
thickness. Two separate bands of coarse-grained facies developed from west to east,
possibly driven by paleobathymetry or subtle changes in sediment sources. A third, lobate
region of low clay volume encompasses a region that stretches from T38 R4W5 to T35
R7W5 – this area is significant because the isochore thickness of allomember BF3 is
relatively low in this area. This suggests that accommodation was reduced, but coarse-
grained sediment was still available.
178
7.4.4 Allomember VII

The clay volume of allomember VII (Fig. 7.10) trends at approximately 80 degrees east of
north and decreases towards the northern half of the study area – it ranges from a
maximum clay volume of 0.56 in the south to 0.04 in T41 R10W5. Lowest clay volumes
are found north of T40, and largely coincide with regions with increased isochore
thickness. The only exception to this is found in the northwest corner of the study area,
west of the 9.5 m thickness contour – in this area, low clay volumes coincide with
reduced isochore thickness. Similar to allomember BF3, this region may represent an area
where accommodation was reduced, but coarse-grained sediment was still available.
Highlights
• The regional structure of the study area dips gently towards the Mesozoic deformation
front in the southwest.
• Isochore maps reveal a sense of clockwise rotation and an overall northward shift in
the location of Belle Fourche and Second White Specks depocentres across the basin.
• Isolith maps, constructed by overlaying clay volume with isochore thickness,
demonstrate lateral facies heterogeneity across the study area and between
allomembers.
179
Solving the Second White Specks 8. Petrophysical property mapping
Chapter 8
8 Petrophysical property mapping of the Lower Second
White Specks and Upper Belle Fourche alloformations
in the Willesden Green – Gilby region, Alberta, Canada
Overview
This chapter presents total organic carbon, total porosity, effective porosity, porosity-
thickness, and brittleness models of the Lower Second White Specks and Belle Fourche
alloformations in the Willesden Green area of west-central Alberta.
TOC
The hydrocarbon generation potential of a rock interval can be assessed by measuring
TOC (Law 1999). Hydrocarbon generation is caused by the thermal decomposition of
organic matter over time (Jarvie 1991). In tight oil petroleum systems like the Second
White Specks Formation, oil is generated in place and stored within low permeability
shale – so high TOC indicates elevated source and reservoir quality (Jarvie 2012b).
Organic matter enrichment is a function of biotic productivity in the photic zone, minus
any destruction or dilution of organic matter that may occur due to biotic activity or
clastic input (Pedersen & Calvert 1990; Bohacs et al. 2005). Within an individual
mudstone formation, organic richness may vary vertically on a sub-metre scale (Bohacs
1998; Bohacs et al. 2005; Passey et al. 2010; Aplin & Macquaker 2011; Jarvie 2012a).
Variations in TOC can be due to depositional environmental conditions, organofacies
differences, thermal maturity, and stratigraphic architecture (Passey et al. 2010; Jarvie
2012a). Within coeval strata, lateral TOC variations are associated with changes in the
predominance of organic matter production, destruction, and dilution (Bohacs & Lazar
2008; Guthrie & Bohacs 2009; Passey et al. 2010).
Creaney and Passey (1993) suggested that high TOC values in proximally-located source
rocks were associated with increased sediment starvation and organic matter deposition
during maximum flooding events. Decreases in TOC are associated with clastic dilution
and degradation of organic matter (Passey et al. 2010). In the Creaney and Passey (1993)
model, TOC profiles for most source rocks should exhibit a “higher at the base” pattern.
180
Jarvie (2012a) cautions that laboratory measurements of TOC strictly represent present-
day TOC – in thermally mature source rocks, TOC is decreased due to the generation and
migration of hydrocarbons out of the reservoir. Decreases in measured TOC, therefore,
could be affected by subtle differences in thermal maturity.
TOC was modelled from resistivity and sonic using the Crain and Holgate (2014) and
Issler et al. (2002) method described in section 5.6.2.1. Figure 8.1 illustrates the lateral
and vertical variations of TOC between allomembers in the study area: individual
allomembers have different vertical TOC profiles. The TOC of allomembers BF2 and
BF1 exhibit decreasing upwards profiles, whereas allomembers BF3 and VII demonstrate
increasing upwards profiles. Highest TOC values (2.75 to 3.5%) were recorded in
allomember VII and near the top of allomember BF3. Between wells 102/14-31-041-
06W5 and 102/08-36-041-07W5, 1.5 m of high TOC (red) at the top of allomember BF3
in 102/14-31 is no longer present in 102/08-36. This supports previous interpretations
from chapters 6 and 7 that the “red” bentonite caps an unconformity between the Belle
Fourche and Second White Specks alloformations. Most significantly, there is a stark
difference in productivity between the two wells, which can be attributed to increased
porosity-thickness and a greater thickness of high TOC in the 102/14-31 well relative to
the 102/08-36 well. The difference in the thickness of this petrophysical zone between the
two wells is less than 1.5 m, underscoring the importance of observing subtle changes
within and between allomembers.
Lower Colorado Group strata have vertical TOC profiles that are constrained to specific
allomembers, suggesting that vertical variations in TOC are primarily controlled by
genetic depositional conditions rather than thermal maturity. Allomembers BF3 and VII,
however, do not exhibit the expected “higher at the base” TOC profile described by
Creaney and Passey (1993): allomembers BF3 and VII contain more of the coarser-
grained and more proximal facies 4 and 5, and should therefore have a TOC profile that is
higher at the base and decreases upward; they instead record upwards increases in TOC.
This upwards increase in TOC may have been caused by the initiation of OAE-II, which
increased nutrient upwelling in the WIS and enhanced organic carbon burial (see section
2.2.2).
181
Fig. 8.1: Comparison of corrected GR (green and yellow colour fill, first track on the left), TOC (blue and red colour fill, second from the right), total porosity
(black line with white fill, far right) and effective porosity (far right, blue fill) for 102/14-31-041-06W5, 102/08-36-041-07W5, and 100/07-19-045-06W5.
The 14-31 and 8-36 wells were both completed and fracked by the same company within 10 days of one another. The 14-31 well was productive, whereas the 8-
36 well was not. Thinning of the high effective porosity interval in allomember BF3, indicated as an orange overlay, between these wells suggest that subtle
differences in effective porosity and porosity-thickness (m-scale) may have impacted the productivity of these wells.
182
Maps of modelled TOC variation across the study area are presented for the two reservoir
allomember BF3 and VII (Figs. 8.2 and 8.3). Allomembers BF3 (Fig. 8.2) and VII (Fig.
8.3) both exhibit similar pattern of modelled TOC. TOC is highest (> 2.23 wt%) in a
north-south oriented fairway in the middle of the study area (R4W5 to R7W5) and
decreases both to the west and east of that fairway. Lateral increases in TOC may be
associated with increases in biotic production as well as decreases in clastic dilution and
organic matter degradation – these maps, therefore, demonstrate that TOC preservation
was highest in the centre fairway.
Porosity
8.3.1 Total porosity

Low density kerogen impacts bulk density readings, and therefore influences total density
porosities. After converting modelled TOC to kerogen, the kerogen contribution to total
porosity calculations was suppressed by using the density porosity correction method
(Equation 24) described by Sondergeld et al. (2010). Using this method removed
anomalously high total porosities associated with high kerogen volume (see Fig. 5.7).
Figure 8.1 illustrates the lateral and vertical variations of total porosity (curve labelled
PHIT_CORR) between allomembers in the study area. The 100/07-19-045-06W5 well
data was supplemented with helium porosimetry data from Furmann et al. (2014) – this
data is visible as black points in the far-right hand column that have been overlaid over
the total porosity model from this project. The log-derived porosities matched this core
data, providing confidence to use this model for other wells across the study area.
Figures 8.4 and 8.5 illustrate the distributions of total porosity for allomembers BF3 and
VII, respectively. Both allomembers have highest total porosities (>9%) in a north-south
oriented central fairway that stretches from T43 R7W5 southwards to T36 R6W5. This
trend is more laterally continuous for allomember BF3 than allomember VII. This total
porosity fairway exhibits some similarities to the geometry of modelled TOC shown in
Figs. 8.2 and 8.3. In addition to the central fairway, total porosities of both allomembers
are elevated west of R9W5 and east of R3W5, where well control is relatively poor
183
Fig. 8.2: TOC of allomember BF3.
184
Fig. 8.3: TOC of allomember VII.
185
Fig. 8.4: Total porosity of allomember BF3.
186
Fig. 8.5: Total porosity of allomember VII.
187
compared to the rest of the study area.
8.3.2 Effective porosity

High clay volumes reduce the porosity that can effectively transmit fluids, known as
effective porosity (see section 5.6.2.3.3). After total porosity had been calculated, it was
reduced to effective porosity using equation 27. Using this method removed the
“ineffective” porosity associated with high clay volumes, as can be seen in the right-hand
columns of well plots in Fig. 8.1.
Figure 8.1 illustrates the lateral and vertical variations in effective porosity
(PHIT_E_KER in Fig. 8.1) across a small section of the study area and between
allomembers. Their associations with clay volume can be seen in Fig. 6.3; highest
effective porosities are associated with low clay volumes (approximately <0.2). Across
the study area, effective porosity is generally highest at the top of allomember BF3 and
throughout allomember VII, although in places (e.g., 100/07-19-045-06W5 and 102/08-
36-041-07W5) the effective porosity may be only present in streaks that are less than 0.20
m thick.
Figures 8.6 and 8.7 illustrate the distributions of effective porosity for allomembers BF3
and VII, respectively. Although the distribution of effective porosity is relatively similar
to the total porosity, their geometry is influenced by high clay volumes. For example, the
effective porosity of allomember BF3 is significantly reduced in allomember BF3 in a
west-east oriented region that stretches from T41 R7W5 to T41 R4W5. This area
corresponds with the region of high clay volume identified in Fig. 7.9. A similar effect is
observed in the effective porosities of allomember VII (Fig. 8.7), wherein effective
porosity is removed in T36 R6W5 to 5W5: this area has relatively high clay volumes (Fig
7.10).
8.3.3 Porosity-thickness
Porosity-thickness (Equation 36) represents the net porous reservoir present in an
allomember, thereby quantifying its storage capacity. Reduced porosity-thickness
indicates decreased porosity and/or a decline in isochore thickness. Increased porosity-
188
Fig. 8.6: Effective porosity of allomember BF3.
189
Fig. 8.7: Effective porosity of allomember VII.
190
thickness, alternatively, signifies either an increase in porosity or an increase in isochore

thickness.
Figures 8.8 and 8.9 depict the effective porosity-thickness of allomembers BF3 and VII,
respectively. The porosity-thickness of allomember BF3 closely resembles the effective
porosity map shown in Fig. 8.6; this is because the isochore thickness of this allomember
(Fig. 7.5) only experiences gradual changes in thickness across the study area. In contrast,
the porosity-thickness of allomember VII is significantly reduced south of T39 – this is
due to the dramatic thinning of allomember VII past this point (Fig. 7.6).
Brittleness
The brittleness index used in this study was chosen to reflect the proportion of brittle
minerals (quartz and carbonate) versus non-brittle or ductile minerals (clay and kerogen)
present in the rock matrix (Equation 33). Figures 8.10 and 8.11 illustrate the model of
brittleness that was generated for allomembers BF3 and VII, respectively, across the
study area in west-central Alberta. These maps closely resemble the maps of clay volume
that were generated in Chapter 7 (see Fig. 7.8 and 7.9); high clay volumes correspond
with lower brittleness values, whereas low clay volumes are associated with higher
brittleness values (>0.81 for allomember BF3 and >0.855 for allomember VII). There is
also a strong association between higher brittleness and low TOC (see Figs. 8.2 and 8.3).
Highlights
• Vertical changes in TOC within coeval strata record depositional conditions that
favoured organic matter preservation. Allomembers BF3 and VII exhibit upwards
increases in TOC that may reflect the initiation of OAE-II. Modelled TOC is highest
in a north-south oriented fairway in the centre of the study area.
• Modelled total porosity, effective porosity, and porosity-thickness for allomembers
BF3 and VII indicate that the highest proportion of porous and clay-free reservoir is
found in a NNW-oriented fairway in the centre of the study area. Low clay content
and increased stratigraphic thickness have the most positive effect on net porous
reservoir development.
• Lateral variations in brittleness are controlled by clay volume and organic richness.
191
Fig. 8.8: Porosity thickness for allomember BF3.
192
Fig. 8.9: Porosity thickness for allomember VII.
193
Fig. 8.10: Brittleness of allomember BF3.
194
Fig. 8.11: Brittleness of allomember VII.
195
Solving the Second White Specks 9. Sweet spot mapping
Chapter 9
9 Sweet spot mapping of the Second White Specks and
Belle Fourche alloformations in the Willesden Green –
Gilby region, Alberta, Canada
Overview
This chapter presents sweet spot maps of the Lower Second White Specks and Belle
Fourche alloformations in the Willesden Green area of west-central Alberta.
Sweet spot maps
9.2.1 Location of producing wells

This thesis sought to determine the degree of co-location between reservoir quality
indicators and oil production from Second White Specks wells. Figures 2.9 and 2.10
illustrate the spatial distribution and productivity of successful wells that targeted strata
from the Second White Specks Formation. The number of wells in these maps were
reduced to only include wells that exclusively produced from the desired target interval –
commingled wells were excluded. Second White Specks wells with the highest
cumulative oil production are found in a region broadly constrained by T43 to 35, R7 to
3W5 that trends approximately NNW (Fig. 2.10). As cumulative production maps like
Fig. 2.10 are biased in favour of older wells – they have had more time to produce greater
hydrocarbon volumes – the IP4 map shown in Fig. 2.9 likely better illustrates the true
distribution of higher performing Second White Specks wells. Figure 2.9 shows three
loosely defined groups of producing wells:
1) T40 to 43, R4 to 7W5
2) T39 to 40, R8 to 9W5
3) T36 to 39, R3 to 7W5.
9.2.2 Overlay maps

The contour maps of petrophysical properties that were generated in Chapter 8 (TOC,
clay volume, porosity-thickness, and brittleness) are overlain for allomembers BF3 and
VII in Figs. 9.1 and 9.2, respectively. Cutoffs for these petrophysical properties, which
196
represent different measures of reservoir quality (i.e., storage and flow capacity) were
selected on the basis of defining major geometric features like fairways. Areas where
multiple reservoir quality indicators overlap are designated “sweet spots” if they coincide
with increased lower Colorado Group oil production. If no producing wells are present in
the overlapping regions, those areas delineate regions where the Second White Specks
and Belle Fourche alloformations have the potential to produce oil at economic rates and
should be investigated further. Sweet spots demarcate areas where the probability of a
vertical well intersecting a natural fracture network located in porous and permeable
reservoir is increased.
9.2.2.1 Allomember BF3

1) T40 to 43, R4 to 7W5: elevated porosity-thickness (>0.1) and brittleness generally
overlap with each other and co-locate with better producing wells. This is not the case
in T41 to 42 R4W5 and T42 to 43, R7W5; in these areas, brittleness is reduced but
porosity-thickness is elevated. The opposite is observed (brittleness is elevated but
porosity-thickness is reduced) in T41 R4W5 to 7W5. Brittleness and porosity-
thickness appear to define the geometry of this group of successful wells. A weaker
association with lower clay volumes (<0.26) is also observed (see T41 and 42, R6W5
and T41 and 42, R4W5).
2) T39 to 40, R8 to 9W5: elevated porosity-thickness and brittleness coincide and co-
locate with producing wells. The non-producing well located in T39 R8W5 has
elevated brittleness but no developed porosity-thickness.
3) T36 to 39, R3 to 7W5: elevated porosity-thickness and brittleness only overlap in a
NNW-oriented region east of R5W5, and do not appear to define the boundaries of
this group of producing wells. TOC and porosity-thickness, instead, appear to best
define the boundaries of this group of wells.
9.2.2.2 Allomember VII

Sweet spots for allomember VII can be broadly divided into two categories: those north
of township 38, and those south of township 39. North of township 39, elevated porosity-
197
Fig. 9.1: Sweet spot map for allomember BF3.
198
Fig. 9.2: Sweet spot map for allomember VII.
199
thickness and brittleness overlap with low clay content. These polygons and define an
irregular V-shape that contains many of the wells located in group 1 (T40 to 43, R4 to
7W5). Similar to allomember BF3, wells in group 2 (T39 to 40, R8 to 9W5) are
delineated by elevated brittleness and low clay content, although the spatial association
with increased porosity-thickness is weaker for allomember VII than for allomember
BF3. South of township 39, however, only elevated TOC appears to be associated with
better producing wells.
Highlights
• North of township 39, sweet spots are defined by regions where elevated porosity-
thickness and brittleness coincide with reduced clay content. These petrophysical
properties define regions where well productivity from the Second White Specks
interval is historically higher. Vertical wells that experienced increased production in
these areas are attributed to thicker units of porous and permeable reservoir that was
likely naturally fractured.
• For allomember BF3, sweet spots south of township 39 are defined either by the
intersection of elevated brittleness and increased porosity-thickness, or by elevated
TOC coinciding with increased porosity-thickness.
• For allomember VII, the sweet spot south of township 39 is best defined by modelled
TOC, although this spatial association is somewhat weaker than in the northern half of
the study area.
200
Solving the Second White Specks 10. Discussion
Chapter 10
10 Discussion
Defining hydraulic flow units
10.1.1 Significance of allostratigraphic bounding surfaces

Hydraulic flow units within the Second White Specks and Belle Fourche alloformations
were defined on the basis of allostratigraphic bounding surfaces, upon which contrasting
facies were vertically juxtaposed. These surfaces include the K1 disconformity, major
flooding surfaces, and the erosional disconformity at the top of allomember BF3 that is
mantled by the “red” bentonite (Fig. 6.1). The “red” bentonite surface has been previously
recognized as a hiatal surface of nondeposition between the Belle Fourche and Second
White Specks alloformations (Tyagi 2009) and later as a subtle unconformity by Plint et
al. (2012a). The presence and nature of this unconformity is not acknowledged by
Furmann et al. (2014) or Zajac (2016) – they place the Cenomanian-Turonian boundary
in a stratigraphically lower position (within allomember BF3) on the basis of bulk
geochemistry.
As Fig. 6.5 demonstrates, the amount of erosion experienced by allomember BF3 is

relatively subtle; generally, less than a few metres of allomember BF3 are removed.
Towards the north of the study area, however (i.e., between wells 06-07-042-07W5 and
07-19-45-06W5), greater than six metres of allomember BF3 are removed. This can be
seen in Fig. 7.5; north of T42 R10W5, the isochore thickness of allomember BF3
significantly decreases. In places where the coarser-grained facies of allomember BF3 has
been eroded, the coarser-grained facies of allomember VII unconformably overlie finer-
grained strata from allomember BF3. Because this erosional surface represents a
significant amount of missing time, this study takes a position similar to Tyagi (2009) and
Plint et al. (2012a) that the Cenomanian-Turonian boundary is best approximated by this
combined hiatal and erosional unconformity.
The origin of the Belle Fourche/Second White Specks unconformity is somewhat

enigmatic. If this unconformity only recorded nondeposition (i.e., no erosion), the most
201
likely scenario for its generation is sediment starvation during maximum sea level, such
as the global highstand associated with OAE-II. Because erosion of allomember BF3 is
sometimes observed, sediment winnowing and advective transport of fine-grained
sediment must have occurred. Figure 6.4 illustrates a scenario during a relative fall in sea-
level (generated by eustatic fall or tectonic uplift) – in this depositional model, the sea
floor can be progressively eroded by storm-driven bottom currents. The development of
the Belle Fourche/Second White Specks unconformity across a shallow epicontinental
marine shelf setting could, therefore, be generated by apparently conflicting mechanisms.
As OAE-II was a global highstand event, the mechanism for lowering base level was
likely eurybatic (local) in nature. Orogenic loading of the western margin near the
Cenomanian-Turonian boundary could have initiated subtle uplift of the sea floor into
local bathymetric highs, which could then be eroded by geostrophic flows during storms.
Small basins between these bathymetric highs would experience no sediment deposition
during lowstand or falling-stage system tracts – in these areas, a disconformity correlative
with the erosive unconformity would be recorded. Because the epicontinental shelf
gradient was so shallow, even subtle uplift or sea level fall could have significantly
impacted sedimentation in this manner. Within the study area, this unconformity
represents non-deposition and erosion that is associated with some type of relative base
level fall.
Defining the mechanism and extent of the Belle Fourche/Second White Specks
unconformity was out of scope for this study. The recognition of this allostratigraphic
surface revealed regions where erosion of allomember BF3 had removed potentially
economic, porous reservoir rock (see Figs. 6.5 and 8.1), which has not been observed in
other studies of the same rocks. The degree of erosion along the Belle Fourche/Second
White Specks unconformity, therefore, exerts control on the vertical stacking of reservoir
facies within the Belle Fourche and Second White Specks alloformations. This
observation has major implications for future exploitation of the Second White Specks
stratigraphic interval – if potential reservoir has been eroded, less net pay will be present
within that well.
202
10.1.2 Quantifying lateral facies variation

In this study, the “red” bentonite surface marks the vertical transition between the finer-
grained facies of allomember BF3 and the highly amalgamated coarser-grained facies
(facies 3 through 5) of allomember VII (see Fig. 6.5). Across the Belle Fourche/Second
White Specks unconformity, the degree of facies amalgamation increases. Within the
Belle Fourche alloformation, facies stacking patterns reflect a gradual vertical transition
into coarser-grained, shallower water facies. Allomember VII, in contrast, exhibits a
significant increase in coarse-grained facies amalgamation. This suggests that deposition
of allomember VII took place in higher energy, shallower water conditions than
allomembers from the Belle Fourche alloformation. This unconformity, therefore,
represents a period of environmental change that occurred near the Cenomanian-Turonian
boundary. Laterally continuous, amalgamated sheets of sediment may be deposited by
storms on shallow marine shelves (Atkinson et al. 1986; Gaynor & Schiehing 1988) – this
appears to be analogous to the facies stacking patterns and geometry of allomember VII
that is observed in this study.
Plint et al. (2012b) demonstrated that changes in the orogenic thrust load could cause
regions of maximum accommodation (depocentres) to shift laterally between
allomembers, thereby influencing paleogeography and sedimentation patterns. Previous
studies utilized thickness (isopach or isochore) maps to estimate the location of lower
Colorado Group mudstone depocentres across Alberta and British Columbia (Kreitner
2002; Plint & Kreitner 2005; Varban & Plint 2005; Kreitner & Plint 2006; Tyagi et al.
2007; Varban & Plint 2008a,b; Tyagi 2009; Plint et al. 2012b; Zajac 2016). On their own,
thickness maps cannot quantify the lateral distribution of facies within allomembers, as
there is no way to establish lateral changes in mineralogy.
Zajac (2016) observed significant lateral facies changes within individual allomembers of
the Belle Fourche and Second White Specks alloformations in west-central Alberta. He
interpreted these variations as the result of subtle changes in water depth and sedimentary
transport processes across a shallow epicontinental shelf, which may have included
combined flows generated by storms. In his study, resistivity and gamma ray values were
considered good proxies for lithology variation within lower Colorado Group
203
allomembers. Similarly, Varban and Plint (2008b) used a gamma ray cutoff of 50 API to
estimate the distribution of net sand (rocks with >50% sand) within the Kaskapau
alloformation. These approaches do not adequately capture the mineralogical variation of
organic-rich shaley rocks: resistivity measurements and gamma ray logs are strongly
affected by organic matter and clay content as well as borehole environmental effects
(e.g., invasion and borehole breakout). To parse the individual contributions of
lithological components (e.g, clay content, TOC), these components must be modelled in
three dimensions.
Lateral facies variations were quantified in three dimensions in this study by creating
isolith maps. In these maps, isochore thicknesses (Figs 7.3 through 7.6) were overlain
with clay volume (Figs. 7.7 through 7.10). By constructing isolith maps in this way, it
was possible to discern the relative contributions of fine-grained sediment supply (clay)
and accommodation (thickness) to the facies within individual allomembers. The isochore
maps revealed a previously unrecognized clockwise and northwards shift in maximum
accommodation across the basin, which may have been caused by tectonic flexure from
subtle changes in orogenic loading (Plint et al. 2012b). Shifts in the location of clay-poor
facies (low clay volume) between allomembers were interpreted to record changes in
sediment supply locations.
Comparison of the clay volume and isochore maps revealed regions where clay volumes
were low but isochore thicknesses were reduced (e.g., the northwest corner of the study
area in Fig. 7.10) – these regions delineate places where clay was not deposited, and
accommodation space was limited. This resulted in deposition of thin and clay-poor
rocks. Alternatively, regions where isochore thicknesses were increased and clay volumes
are higher (e.g., T41 to 42 and R5 to 6W5 in Fig. 7.9) demarcate areas where
accommodation was increased but coarse sediment supply was not available. These
observations suggested there is significant lateral facies heterogeneity across the study
area and between allomembers. These maps may also be used to show the distribution of
net reservoir across west-central Alberta; increased isochore thickness and decreased clay
volume correspond with the preferred reservoir facies (facies 3 through 5) shown in Fig.
6.5. Using the repeatable and petrophysically sound methods developed in this thesis,
204
these maps have helped to establish – for the first time – the lateral facies heterogeneity of
the lower Colorado Group in west-central Alberta.
Composite isolith maps are commonly made in industry but are not frequently utilized in
the literature. This is likely because petrophysical and stratigraphic studies are most often
performed independently. These maps are relatively simple to construct and have
significant interpretive power – they should be more widely utilized in future stratigraphic
studies where lateral facies changes are under investigation.
Less than half of the wells used for isochore thickness mapping were usable for clay
volume calculation, and only one well (100/07-19-045-06W5) had clay XRD data
available for petrophysical model adjustment. This means that a denser average well
spacing was used for thickness mapping relative to clay volume calculation – the two
maps do not perfectly correspond with one another. The mismatch in resolutions could
have been rectified by removing data points associated with wells that did not have LAS
data.
Petrophysical property modelling

Prior to this work, there were no publicly available static petrophysical models of the
Second White Specks and Belle Fourche alloformations. Regional variations in reservoir
quality and lithology had been characterized on the basis of maps of average wireline log
values (Zajac 2016) and core analyses (Bloch et al. 1999; Furmann et al. 2014; Zajac
2016).
This study sought to reduce uncertainty associated with petrophysical modelling of

Second White Specks reservoirs. The magnitude of error, however, remains unknown.
Wireline logging tools are proprietary – there is no way to access the code that was used
to transform the signal from the reservoir into a recorded measurement. Log
measurements have some amount of error associated with this data transformation: they
represent imperfect proxies of rock properties such as TOC or clay volume. Moreover,
the wireline logging tools are individually calibrated by the logging company to each well
and may be further calibrated during separate logging runs. There are also differences in
geophysical wireline log tools depending on the vintage of the tool and the operating
205
company it was developed by – the tool physics and design may vary on the basis of
sensitivity, accuracy, precision, or depth of penetration. The tool specifications discussed
in section 3.6.2 represent average tool configurations, and only approximate the most
likely setup of a particular sonde. To adequately characterize measurement errors from
the primary acquisition of petrophysical data, the well log dataset would have to be
restricted to wells where all the logging parameters were known – this is not feasible due
to the extremely limited nature of the well log dataset in this study.
Although some of the uncertainties mentioned above could have been resolved through a
wireline log normalization process, normalization may have potentially suppressed lateral
changes in lithology across the study area. Quantifying the error associated with
geophysical wireline logging is beyond the scope of this project – this uncertainty,
however, can be mitigated by setting reasonable constraints for well log responses in
expected lithologies for individual allomembers (Table 6.2 and Fig. 6.3). This establishes
commonalities between wells that may have had different logging configurations (e.g.,
different operators, well vintages, tool physics, calibration), thereby allowing for
correlation of hydraulic flow units between wells (Fig. 6.5) that can be used for
petrophysical modelling.
10.2.1 TOC
Although the thermal maturity of the Second White Specks and Belle Fourche
alloformations is well constrained (e.g., Furmann et al. 2014), variations in source rock
quality had not been previously investigated prior to this study. Figure 8.1 illustrates the
estimated TOC profiles through the Belle Fourche and Second White Specks
alloformations for three wells in the study area. The Issler et al. (2002) method was used
on the basis of a strong linear relationship between modelled values and Rock-Eval TOC
analyses from the 7-19 core (Fig. 5.5). Highest modelled TOC values (2.75 to 3.5%) were
recorded at the top of allomember BF3 and throughout allomember VII. This upwards
increase in TOC relative to the preceding allomembers (BF1 and BF2) may have been
caused by the initiation of OAE-II, which increased nutrient upwelling in the WIS and
enhanced organic carbon burial. Figures 8.2 and 8.3 illustrate the modelled distribution of
TOC throughout the study area in allomembers BF3 and VII, respectively – highest TOC
206
values occur in a north-south oriented fairway in the middle of the study area (R4W5 to
R7W5) and decreases both to the west and east of that fairway.
Total organic carbon (TOC) was modelled in this study using the Issler et al. (2002) and
Crain and Holgate (2014) method. This method relies on elevated resistivity and slow
sonic responses to detect intervals with elevated organic content. Implicit in this method
is the assumption that these well log responses are uniquely related to kerogen volume –
this is not necessarily true. Increased resistivity measurements can be produced by an
increase in carbonate mineral matrix or by non-conductive liquid hydrocarbons stored in
matrix or fracture porosity, in addition to their association with elevated organic content.
Sonic travel times, similarly, can be slowed by hydrocarbon-filled porosity, clay-bound
water, natural fractures, or low-density organic matter. Consequently, the combined Issler
method may not provide a unique solution for TOC – the TOC model presented in this
study may actually delineate the locations of open, hydrocarbon-filled natural fractures in
addition to areas with increased organic richness.
10.2.2 Porosity
Furmann et al. (2014) observed that the highest reservoir quality (elevated total porosity)
interval in the 7-19 core occurred at the top of their Belle Fourche Formation, which
approximates the top of allomember BF2 in Fig. 8.1. Although this observation is true for
the 7-19 core, it is not representative of most of west-central Alberta. Highest modelled
effective porosities (>2.5 volume %), which best approximate the potential storage
capacity of the Second White Specks and Belle Fourche alloformations due to their high
clay volumes, were observed in this study in allomember VII and in the coarser-grained
facies of allomember BF3. The coarse-grained facies of allomember BF3 is not present in
the 7-19 core due to removal by the Second White Specks/Belle Fourche erosional
unconformity (Figs. 6.5 and 8.1). This unconformity was not detected by Furmann et al.
(2014) because they placed the Second White Specks – Belle Fourche boundary
stratigraphically lower compared to the “red” bentonite used by Tyagi et al. (2007) and
others, including this study.
207
Lateral variations in total porosity within the Second White Specks and Belle Fourche
alloformations had previously been constrained by helium porosimetry data from several
cores spread across Alberta, British Columbia, and Saskatchewan (Bloch et al. 1993,
1999; Furmann et al. 2014; Zajac 2016). Using the allostratigraphic framework to define
petrophysical zones permitted the application of a kerogen-corrected total density
porosity model across the study area, which was calibrated on the basis of helium
porosimetry data taken from one well (7-19) in the study area by Furmann et al. (2014).
Using a kerogen-corrected porosity model suppressed anomalous density porosities
associated with low-density organic matter (Fig. 5.8), resulting in a porosity model that
agreed closely with the total porosity measurements from Furmann et al. (2014). When
modelled total porosity, effective porosity, and porosity-thickness were extended across
the study area, the highest proportion of porous and clay-free reservoir within
allomembers BF3 and VII were found in a NNW-oriented fairway in the centre of the
study area.
Relying on the 7-19 core to constrain lateral variations in total porosity introduced some
uncertainty with greater distance away from the 7-19 well. In this study, lateral
constraints of porosity in this study rely on log-based calculations of density porosity,
rather than porosity analyses from other cores in the area as none were available. The
total porosity model used a matrix density (2.74 g/cm3) derived from a neutron porosity
and bulk density cross plot that used log data from four wells (Fig. 5.7) – a 1% shift in the
y-intercept (matrix density) introduces a ±0.027 g/cm3 shift in uncorrected matrix density
(no kerogen correction), which translates to calculated total density porosity variations of
±1.4%. A 1% shift in matrix density is well within the limit of neutron and density
porosity variation observed within the Second White Specks and Belle Fourche intervals
in Fig. 5.7. Without core data for calibration, however, 2.74 g/cm3 is the best available
estimation of matrix density for the study area. Establishing the confidence interval in this
chart would have helped to improve confidence in the selected matrix density value.
The kerogen-corrected density porosity equation used in this study (Equation 24) is
derived from an assumption that low bulk density readings are associated with increased
matrix porosity and reduced fluid density (i.e., the presence of hydrocarbons). This
208
assumption is not correct, because porosity modes associated with open natural fractures
and organic matter cannot be discriminated from matrix porosity using this method.
Clarkson and Pedersen (2011) demonstrated that Second White Specks wells have
production histories that reflect dual porosity modes that are most likely associated with
increased depositional porosity and fracture porosity. Porosity within organic matter in
magnetic resonance (NMR) logs are required to assess the contribution of fracture
porosity to the modelled porosities – this type of log data was not available. As fracture
porosity cannot be deconvolved from total or effective porosities without NMR log data,
the kerogen-corrected density porosity could reflect a combination of fracture and matrix
porosity.
Calculating kerogen-corrected density porosity required an estimate of kerogen density,

which was not available in the study area. Kerogen density was assumed to be a constant
value of 1.15 g/cm3 throughout the study area, on the basis of vitrinite reflectance data
from Furmann et al. (2014). If this estimate was too low, the kerogen-corrected density
porosity will have underestimated the total density porosity of the Second White Specks
and Belle Fourche alloformations. An overestimate of kerogen density, conversely, would
have caused an overestimate of total density porosity. In the 7-19 well, the kerogen-
corrected density porosity agrees with the helium porosimetry data (Fig. 8.1) – providing
a basis for using this kerogen density in other wells. This, however, is an
oversimplification. Kerogen density varies with thermal maturity and organic matter type
and, therefore, may vary both laterally between wells and vertically within allomembers.
Due to the model agreement with the porosity data from core, the kerogen density of 1.15
g/cm3 presented a best possible fit for the available data.
Effective porosity (Equation 26) removes the contribution of clay porosity to total
porosity, thereby establishing the storage capacity available for hydrocarbons in the
reservoir. No core calibration data was available to constrain the effective porosity model
(e.g., mercury injection tests). Higher effective porosities calculated in Chapter 8 and the
coarser-grained reservoir facies mapped in Chapter 6 overlap (Fig. 8.1), providing some
assurance that the effective porosity maps (Figs. 8.6 and 8.7) are actually capturing
variations in depositional effective porosity.
209
Clay volume, Vclay, impacts effective porosity (Equations 3 and 4). Higher Vclay values
reflect a larger differential between the measured GR values and GRclean (5th percentile
clay-free gamma ray value). In organic-rich shales, the measured GR value can be
significantly increased by uranium in organic matter, thereby causing an overestimate of
Vclay and an underestimation of the effective porosity in that interval. Although this effect
can be resolved through the use of the uranium-free (CGR) curve, SGR log data was only
available for one well (2-14). As the lateral distribution of uranium in the Second White
Specks and Belle Fourche alloformations is not currently known, the modelled effective
porosity may be inaccurate in organic-rich zones. The magnitude of this effect, however,
cannot be determined with currently available core and log data.
Effective porosity is also affected by clay mineralogy (Equation 26). By Equation 26,
clay minerals with increased differences between their dry and wet densities (i.e., mixed
layer clays with more bound water) have higher clay porosities, and therefore reduce
effective porosity by a larger amount. Detailed clay mineralogy from core was not
available, so the Th/K ratio from SGR logging of the 2-14 was utilized for clay mineral
determination instead (Fig. 5.9). Figure 5.9 indicates that the clay mineralogy approaches
a mixed-layer (illite and smectite) composition, which corresponds to a !"#$%&'( of 1.7
g/cm3 and a !)*(%&'( of 2.7 g/cm3 (Chitale 2010). If no smectite is present and illite clays
dominate the rock matrix, !"#$%&'( increases to 2.5 g/cm3 and the calculated clay porosity
decreases. As Furmann et al. (2014) reported between 12 and 36 wt% illite in the 7-19
core, the !"#$%&'( used in this study may have been too low. Using an illite clay density
would significantly improve the effective porosity-thickness present in the Belle Fourche
and Second White Specks alloformations in this study, but there is no basis for changing
this until calibration to core is established.
10.2.3 Brittleness
No DSSI logs or geomechanical tests run or core were available to establish the
brittleness of the Second White Specks or Belle Fourche alloformations in the study area.
To compensate for this, Furmann et al. (2014) calculated brittleness using the lithological
index defined by Wang and Gale (2009), which uses wt %. They reported that the BI for
210
their Second White Specks Formation averaged 47% and ranged from 39 to 61%,
whereas the BI for their Belle Fourche Formation was slightly lower on average (42%)
and ranged from 25 to 46%. The slight difference in brittleness between formations was
attributed to a higher proportion of carbonate minerals in the Second White Specks
Formation relative to the Belle Fourche (Gale et al. 2014). As noted in section 5.6.3,
brittleness indices that use volume % (e.g. Katz et al 2016; Mathia et al. 2016; Rybacki et
al. 2016) are preferred for use in organic-rich reservoirs over those that use wt % (e.g.,
Jarvie et al. 2007; Wang & Gale 2009; Jin et al. 2014a,b) because the contribution of
low-density organic matter to ductile behaviour may be otherwise overlooked (Rybacki et
al. 2016). In this study, the volumetric brittleness index described by Mathia et al. (2016)
was used for the Belle Fourche and Second White Specks alloformations, thereby
resulting in average BI of 0.79 and 0.82 for those formations, respectively. The results
from Furmann et al. (2014) and this study cannot be directly compared, as the indices
used were fundamentally different and their stratigraphic zonation also differs. They do,
however, support the incremental increase in the brittleness of allomember VII relative to
BF3.
The Second White Specks and Belle Fourche alloformations contain pyrite, multiple
carbonate mineral phases (calcite, siderite, and dolomite) as well as different types of clay
minerals. These mineral phases may all have different contributions to brittleness
(Rybacki et al. 2016). This study assumed that all carbonate minerals were associated
with dolomite, and all clay minerals were mixed-layer clays (illite or smectite). This is a
simplification of the real mineral assemblage. XRD analysis of mineral volumes would
have enabled more accurate estimation of clay volume (ductile), pyrite (brittle) and
multiple carbonate (brittle) mineral phases, as well as their associated volumes.
The carbonate volume model used in this study (Equation 28) only works for cored wells
that have total inorganic carbon data available (Jiang et al. 2017). Because inorganic
carbon was not available for non-cored wells, kerogen volume was used to calculate
carbonate volume % using a linear regression (Fig. 5.10). In general practice, PEF logs
would be used to establish carbonate mineral volumes, so kerogen was used in their
absence. Based on the LECO TOC data that is attached to this thesis, inorganic carbon
211
was often twice as high as the TOC. As a result, the regressions shown in Fig. 5.10 likely
underestimate the percentage of dolomite present in the reservoir. The maps of brittleness
for allomembers BF3 and VII (Figs .8.10 and 8.11), however, mostly reflect the geometry
of the clay volume maps from Chapter 7 and the TOC model from Chapter 8 – thereby
suggesting that lateral variations in brittleness mostly correspond to the volume of ductile
components.
Sweet spot mapping

Zajac (2016) identified reservoir fairways in the Second White Specks Formation on the
basis of gamma ray and resistivity heat maps. As Chapter 3 demonstrated, the responses
of these well logs to organic, clay-rich lithologies are complex. High resistivities may
correspond to the presence of hydrocarbons, solid organic matter, and organic maturity
levels, whereas high gamma ray values are associated with formation radioactivity that
originates from organic matter or clay minerals. Consequently, these measurements do
not correspond with reservoir quality indicators like clay volume, porosity, organic
richness, hydrocarbon saturation, or brittleness. By applying a rigorous approach to
petrophysical modelling, this thesis sought to deconvolve the signals of reservoir quality
indicators embedded within a basic suite of well logs in order to establish changes in rock
properties that could impact well production.
The sweet spot maps in Chapter 9 reveal regions where modelled petrophysical properties
co-locate with successful producing wells from allomembers BF3 and VII. Producing
wells for allomembers BF3 and VII can be broadly separated by township 39. North of
township 39, elevated porosity-thickness and brittleness overlap with reduced clay
content and improved well performance – this is likely due to the presence of thick,
relatively porous and brittle reservoir that became naturally fractured. South of township
39, sweet spots are constrained best by modelled TOC. The sweet spots in this region do
not exhibit the same degree of overlap between modelled petrophysical properties as is
seen north of township 39. These fairways have not been previously delineated in any
studies that are currently publicly available.
212
The sweet spot maps included in this work suffer from a certain degree of confirmation
bias, as there are not many wells located outside the delineated fairways. This is partially
due to the exclusion of commingled wells: these are wells where the Second White
Specks Formation was being co-produced with other formations in order to improve the
overall well production. In general practice, formations are commingled when any single
formation does not have sufficient reservoir quality to produce hydrocarbons at economic
rates. From an oil production perspective, these wells are problematic to deal with as the
individual contributions of each formation cannot be easily separated.
The producing wells in this study could not be separated on the basis of perforation
locations and hydraulic fracturing styles within specific allomembers due to the limited
number of producing wells in the area. Moreover, the reported position of perforations
and hydraulic fracture treatments in geoSCOUT may not actually capture the location and
extent of induced fractures: they may extend into other allomembers, particularly if
relative brittleness values between allomembers favour vertical fracture propagation.
The relative contributions of individual reservoir quality indicators to well performance

from the Second White Specks interval could not be established, likely due to the
somewhat non-unique nature of the model responses. It is possible that the maps of TOC,
brittleness, and effective porosity and, by extension, porosity-thickness presented in this
thesis may have detected enhanced reservoir quality associated with open, hydrocarbon-
filled natural fractures.
Recommendations for future Second White Specks exploration

• Careful delineation of the Second White Specks/Belle Fourche unconformity may
help identify areas with vertically-stacked pay in both allomembers BF3 and VII.
• The sweet spots identified in this study, which likely delineate areas with increased
reservoir quality, should be high-graded as targets for three-dimensional seismic
acquisition and geophysical inversion. This should confirm if the sweet spots
identified in this thesis actually capture regions with increased reservoir quality.
213
• NMR and DSSI logs should be run in any new wells drilled in the Second White
Specks Formation in order to assess fracture porosity and elastic moduli, respectively.
This would better establish brittleness variation and potentially help avoid areas
where wellbore stability in long horizontal wells had previously caused issues.
Additionally, running borehole XRF logs would help constrain mineralogical
variations.
• Over 2200 wells in the study area penetrate the Second White Specks interval, of
which less than 5% are completed in the lower Colorado Group. Existing, non-
producing wells should be re-logged with cased hole tools to try and identify missed
pay in allomembers BF3 and VII.
214
Solving the Second White Specks 11. Conclusions
Chapter 11
11 Conclusions
The primary goal of this research was to find a way to reliably predict what geographic
areas and stratigraphic zones are more likely to produce consistently well from the
Second White Specks Formation. It was hypothesized that areas with higher relative
brittleness (increased silica and carbonate content) and increased porosity would co-locate
with improved inflow performance from the Second White Specks Formation. To
adequately test this hypothesis, this study sought to answer three key questions. The
answers to these conclusions are drawn from the discussion in Chapter 10.
1) Can the Second White Specks and Belle Fourche Formations in the study area be
subdivided into allomembers that provide a framework for mapping coeval
hydraulic flow units?
• Coeval depositional units (allomembers) were defined by extending and

subdividing the regional allostratigraphic framework established by Tyagi et al.
(2007). This was accomplished by correlating previously unrecognized flooding
surfaces throughout the study area, which were tied to core data and geophysical
wireline log signatures. The interval of interest, which was constrained below by
the K1 disconformity and above by the allomember VII flooding surface, was
subdivided into four allomembers.
• Facies observed in this study are consistent with a distal to mid-shelf, marine,
low-gradient ramp setting in a shallow (40 to 70 m water depth) epicontinental
sea. Sediments were transported offshore by storm-generated combined flows.
Isolith maps, constructed by overlaying clay volume with isochore thickness,
demonstrate lateral facies heterogeneity across the study area and between
allomembers.
• The contact between the Belle Fourche and Second White Specks alloformation,
capped by the “red” bentonite and spanning the Cenomanian-Turonian boundary,
215
is unconformable across the study area. The top of allomember BF3 was, at times,
truncated by this unconformity. Across this unconformity, the depositional style of
lower Colorado Group allomembers subtly change: in the Belle Fourche
alloformation, coarser-grained “reservoir” facies do not exhibit a high degree of
amalgamation; in the Second White Specks Formation, by comparison, a higher
degree of amalgamation and vertical stacking of coarser-grained facies was
observed. This results in increased reservoir facies thickness in allomember VII
relative to allomembers from the Belle Fourche (BF1, BF2, and BF3).
• Second White Specks and Belle Fourche allomembers exhibit significant lateral
facies heterogeneity across the study area. This was established on the basis of
isolith maps. These maps illustrate changes in clay content associated with
sediment supply, and stratal thickness variations that were attributed to shifts in
the location of sedimentary depocentres.
• Allomembers of the Second White Specks and Belle Fourche represent coeval
depositional units that function as hydraulic flow units for petrophysical analysis
and sweet spotting.
2) Can the petrophysical properties of the Second White Specks and Belle Fourche
Formations be reliably modelled, using limited geophysical well data and sparse
core control for calibration?
• Using the Issler method, TOC was modelled for hydraulic flow units within
allomembers from the Second White Specks and Belle Fourche Formations. This
model was in agreement with LECO TOC and Rock-Eval data collected from
cores in the study area. Elevated TOC in allomembers BF3 and VII reflect the
initiation of OAE-II near the Cenomanian-Turonian boundary. Allomembers
exhibit lateral variations in organic richness (TOC) that are likely associated with
increased organic carbon preservation in the centre of the study area.
• Using a kerogen-corrected porosity method, total density porosity was modelled

for lower Colorado Group allomembers. This model provided a close match to
216
existing core porosity measurements. The kerogen-corrected porosity method

effectively suppressed anomalous porosities associated with increased organic
richness. In addition to illustrating lateral and vertical variations in matrix
porosity, this porosity model may have captured regions with enhanced natural
fracture porosity.
• By assuming a mixed-layer clay mineral assemblage, effective porosity was

modelled for the Belle Fourche and Second White Specks Formations. Units with
effective porosity coincide with coarser-grained “reservoir” facies at the top of
individual allomembers. Effective porosity was particularly well developed in
allomembers BF3 and VII, suggesting that these units have improved storage
capacity.
• Applying a volumetric brittleness index to allomembers BF3 and VII revealed

significant lateral variability in brittle rock behaviour across the study area. These
variations appear to be controlled by elevated ductile mineral components (clay
content and TOC).
• Reasonable estimates of reservoir quality can be developed even when data is

sparse. Depositionally controlled variations in reservoir quality were revealed
when petrophysical parameters (lithology, porosity, and brittleness) were
confined to allomembers.
3) Is there a spatial association (co-location) between reservoir quality indicators

(e.g., brittleness and porosity thickness) and oil production?
• Combining brittle mineral models with porosity-thickness, TOC, and clay volume
trends into sweet spot maps revealed potentially economic fairways in the Second
White Specks and Belle Fourche alloformations that co-located with increased
historical oil production. These trends have not been previously recognized in the
literature.
This thesis used petrophysical principles and theory to apply petrophysical methods to
real geological data. The integrated petrophysical and allostratigraphic method of sweet
217
spot mapping established in this thesis provides a framework for delineating fairways in
underexplored unconventional basins where well log data is scarce. Although the
petrophysical models developed in this work are imperfect and potentially non-unique
responses to reservoir quality indicators, they provide critical information about
variations in reservoir quality and brittle behaviour. Sequence stratigraphic methods and
petrophysical analyses are often combined when seismic data is available for nonlinear
joint inversions. The practice of integrating petrophysics and allostratigraphy for
exploration purposes, however, is not commonly attempted. The method presented in this
thesis provides a way to estimate lateral and vertical variations in reservoir quality
without needing seismic data.
Future work
The core analysis database developed within this study should be supplemented with
several key analyses to improve the accuracy of the petrophysical models herein:
• Helium porosimetry data taken from other cores and at higher vertical resolution (<
0.5 m) to ground-truth modelled lateral and vertical variations in total porosity within
coeval strata;
• Core bulk density and core matrix density data to calibrate the matrix density used for
total porosity estimation, and to better establish pore fluid density;
• Local measurements of kerogen density to improve the total porosity correction for
organic content;
• Total sulphur measurements to estimate the distribution of pyrite;
• XRD analysis of matrix and clay mineralogy to constrain the lithological variations of
individual allomembers, especially with regards to carbonate mineralogy as it is not
easily modelled without PEF logs.
Further improvements to the petrophysical model could be achieved through more

sophisticated well logging:
• The addition of uranium-free SGR log data would assist in improved calibration of
clay volumes across the area, reducing uncertainty associated with the calculation of
effective porosity.
218
• Mineralogical logs are rarely available. Regional, field-scale studies of brittleness

within the Second White Specks and Belle Fourche alloformations would benefit
significantly from robust mineralogical modelling because vertical and lateral
variations in apparent brittleness could be investigated.
• The addition of DSSI logs would assist in the development of static geomechanical
models of brittleness, which could be compared and contrasted with lithological
brittleness indices. This would help establish an association, if any, between changes
in brittle mineralogy and elastic moduli in organic-rich mudstones. Trend surface
mapping (e.g., residual mapping) could help constrain the localized perturbations
associated with small-scale structural relief along major bounding surfaces – these
small variations may be associated with small faults or natural fractures.
Successful vertical Second White Specks Formation wells have historically accessed
natural fracture networks with enhanced flow capacity. Although this work has better
constrained lateral variations in the brittleness of the Second White Specks Formation
across west-central Alberta, the locations of those natural fracture networks remain
somewhat uncertain. High brittleness values only indicate possible natural fracture
development within an area. If 3D seismic data is available for the Willesden Green
region in the lower Colorado Group interval, an integrated petrophysical, stratigraphic,
and structural study should be undertaken within the high-graded “sweet spots” shown in
Chapter 9 to ascertain the location of fault networks and natural fractures. Trend surface
mapping (e.g., residual mapping) could help constrain the localized perturbations
associated with small-scale structural relief along major bounding surfaces – these small
variations may be associated with small faults or natural fractures. A 3D seismic dataset
would enable the use of nonlinear joint inversion methods for estimation of reservoir
quality parameters (e.g., clay volume, porosity, total organic content, and brittleness),
which could help constrain the otherwise sparsely located well data and resulting
petrophysical models of the lower Colorado Group. This could serve as a case study for
the comparison of deterministic and inverse petrophysical modelling methods. This could
also help separate the relative contributions of different petrophysical properties to
improved well performance.
219
Solving the Second White Specks CV
Curriculum Vitae
Name: Kienan P. Marion
Post-secondary The University of Western Ontario
Education and London, Ontario, Canada
Degrees: 2010 – 2014 Bachelor of Science (B.Sc.)
Honours Geology for Professional Registration
London, Ontario, Canada
2014 – 2018 Master of Science (M.Sc.)
Geology
Honours and Winston Karel Memorial Student Award for Best Abstract
Awards: Canadian Well Logging Society (CWLS)
2016 and 2017
Best Student Geological Oral Presentation
CSPG GeoConvention 2017: May 15 – 17, Calgary, AB, Canada
Canadian Society of Petroleum Geologists (CSPG)
2017
SPWLA Foundation Grant
Society of Petrophysicists and Well Log Analysts (SPWLA)
2016
Martin D. Hewitt Named Grant
AAPG Foundation Grants-In-Aid
2016
Honourable Mention, Best Student Geological Poster Presentation
CSPG GeoConvention 2015: May 4 – 8, Calgary, AB, Canada
2015
Bronze Medal
AAPG Imperial Barrel Award Canada Region
2015
L. Austin Weeks Undergraduate Grant
American Association of Petroleum Geologists (AAPG)
2014
CSPG Undergraduate Award: Central/Ontario
2014
Related Work Graduate Teaching & Research Assistant
Experience The University of Western Ontario
London, Ontario, Canada
2014 – 2018
220
Solving the Second White Specks CV
Geology Intern – Oil Sands, Lease Geology

Shell Canada Ltd.
Calgary, Alberta, Canada
Summer 2016
Petrophysics Intern – Alberta Deep Basin
Shell Canada Ltd.
Summer 2015
Geology Intern – Unconventional Exploration
Imperial Oil Ltd.
Summer 2014
Geology Intern – Exploration
NAL Resources Ltd.
Summer 2013
Geology Intern – Exploration
NAL Resources Ltd.
Summer 2012
Presentations:
Marion, K.P. and Cheadle, B.A. 2017. Solving the Second White Specks: Integrating
petrophysics and allostratigraphy to find sweet spots (oral presentation). Presented
at CSPG GeoConvention 2017: 15-17 May 2017, Calgary, Alberta.
Marion, K.P. and Cheadle, B.A. 2016. Petrophysical characterization of the Upper
Cretaceous Second White Specks alloformation, Willesden Green – Gilby area,
Alberta: An updated workflow for anisotropic fine-grained reservoirs (poster
presentation). Presented at 2016 AAPG Annual Convention and Exhibition: 21
June 2016, Calgary, Alberta.
Marion, K.P. and Cheadle, B.A. 2015. Petrophysical characterization of the Upper
Cretaceous Second White Specks alloformation, Willesden Green – Gilby area,
Alberta (poster presentation). Presented at CSPG GeoConvention 2015: 4-8 May
2015, Calgary, Alberta.
Other work:
Marion, K.P. 2014. Microstructure of Cretaceous mudrocks from a variety of sedimentary
environments. B.Sc. thesis, Department of Earth Sciences, University of Western
Ontario, London, Ontario.
221
Solving the Second White Specks References
References
Abbott, J.L. 1948. X-ray fluorescence analysis. The Iron Age: pp. 58-62 and 121-124.
Abbott, S.T. 2000. Detached mud prism origin of highstand system tracts from mid-
Pleistocene sequences, Wanganui Basin, New Zealand. Sedimentary Geology
47: pp. 15-29.
Adams, J.A.S., and Gasparini, P. 1970. Gamma-Ray Spectrometry of Rocks. Elsevier

Science.
Aigner, T., and Reineck, A. 1982. Proximality trends in modern storm sands from the
Helgoland Bight (North Sea) and their implications for basin analysis.
Senckenbergiana Maritima 14: pp. 183-215.
Alldredge, A.L., and Silver, M.W. 1988. Characteristics, dynamics and significance of
marine snow. Progress in Oceanography 20: pp. 41-82.
Allen, P.A., and Allen, J.R. 2013. Basin analysis: principles and application to petroleum
play assessment. Third ed. Wiley-Blackwell.
Amaefule, J.O., Altunbay, M., Tiab, D., Kersey, D.G., and Keelan, D.K. 1993. Enhanced
reservoir description: Using core and log data to identify hydraulic (flow)
units and predict permeability in uncored intervals/wells. Presented at 68th
Annual Technical Conference and Exhibition of the Society of Petroleum
Engineers, Houston, Texas, 3-6 October 1993. SPE.
Ambrose, R.J., Hartman, R.C., Diaz-Campos, M., Akkutlu, I.Y., and Sondergeld, C.H.
2010. New pore-scale considerations for gas shale in place considerations,
SPE131772. Presented at SPE Unconventional Gas Conference, Pittsburgh,
Pennsylvania, 23-25 February 2010. SPE.
Ambrose, R.J., Hartman, R.C., Diaz-Campos, M., Akkutlu, I.Y., and Sondergeld, C.H.
2012. Shale gas-in-place calculations Part I: New pore-scale considerations.
SPE Journal 17(1).
Andersen, M.A. 2011. Introduction to Wireline Logging. Oilfield Review 23(1): pp. 58-
59.
Aplin, A.C., and Macquaker, J.H.S. 2011. Mudstone diversity: Origin and implications
for source, seal, and reservoir properties in petroleum systems. AAPG Bulletin
95(12): pp. 2031-2059. doi: 10.1306/03281110162.
Arai, T. 2007. The discovery of x-rays and origin of x-ray fluorescence analysis. In
Handbook of practical x-ray fluorescence analysis. Edited by B. Beckhoff and
B. Kanngießer, et al. Springer, New York, New York.
222
Archie, G.E. 1942. The electrical resistivity log as an aid in determining some reservoir
characteristics. Transactions of the AIME 146(1): pp. 54-62. doi:
10.2118/942054-G.
Archie, G.E. 1950. Introduction to Petrophysics of Reservoir Rocks. AAPG Bulletin

34(5): pp. 943-961.
Armentrout, J.M. 1999. Sedimentary basin analysis. In Handbook of Petroleum Geology:

Exploring for Oil and Gas Traps. Edited by E.A. Beaumont and N.H. Foster.
American Association of Petroleum Geologists. pp. 4.1-4.123.
Arthur, M.A., and Philip, G.M. 1986. Cyclicity in the Milankovitch band through
geologic time: an introduction. Paleoceanography and Paleoclimatology 1(4):
pp. 369-372. doi: 10.1029/PA001i004p00369.
Arthur, M.A., and Sageman, B.B. 2004. Sea-level control on source-rock development:
Perspectives from the Holocene Black Sea, the mid-Cretaceous Western
Interior Basin of North America, and the Late Devonian Appalachian Basin.
SEPM Special Publication 82: pp. 35-59.
Arthur, M.A., Schlanger, S.O., and Jenkyns, H.C. 1987. Palaeoceanographic controls on
organic-matter production and preservation. In Marine Petroleum Source
Rocks, Geological Society Special Publication no. 26. Edited by J. Brooks and
A.J. Fleet. pp. 401-420.
Asquith, D.O. 1970. Depositional topography and major marine environments, Late
Cretaceous, Wyoming. AAPG Bulletin 54(7): pp. 1184-1124.
Asquith, G.B. 1990. Log evaluation of shaly sandstones: a practical guide. Continuing
Education Course Note Series 31. AAPG. Tulsa, Oklahoma. 59 p.
Atkinson, C.D., Goesten, M.J.B.G., Speksnuder, A., and van der Vlugt, W. 1986. Storm-
generated sandstone in the Miocene Miri Formation, Seria Field, Brunei (N.W.
Borneo). In Shelf sands and sandstones. Edited by R.J. Knight and J.R.
McLean. CSPG Memoir 11, Calgary, Alberta. pp. 213-240.
Baker, P.E. 1957. Density logging with gamma rays. SPE-301-G 210: pp. 289-294.
Barker, I.R., Davis, W.J., Moser, D.E., Kamo, S.L., and Plint, A.G. 2011. High-precision
U–Pb zircon ID–TIMS dating of two regionally extensive bentonites:
Cenomanian Stage, Western Canada Foreland BasinThis article is one of a
series of papers published in this Special Issue on the theme of Geochronology
in honour of Tom Krogh. Canadian Journal of Earth Sciences 48(2): pp. 543-
556. doi: 10.1139/e10-042.
Bateman, R.M. 2009. Petrophysical data acquisition, transmission, recording and

processing: a brief history of change from dots to digits. Presented at SPWLA
50th Annual Logging Symposium, The Woodlands, Texas, June 21-24 2009.
223
Beaumont, C. 1981. Foreland Basins. Geophysical Journal of the Royal Astronomical

Society 65(2): pp. 291-329. doi: 10.1111/j.1365-246X.1981.tb02715.x.
Becquerel, H. 1896. Sur les radiations émises par phosphorescence. Comptes rendus
hebdomadaires des séances de l’Académie des sciences 122(8): pp. 420-421.
Bennett, R.H., O’Brien, N.R., and Hulbert, M.H. 1991. Determinants of clay and shale
microfabric signatures: processes and mechanisms. In Microstructure of fine-
grained sediments, from mud to shale. Edited by R.H. Bennett and W.R.
Bryant, et al. Springer-Verlag, New York, New York. pp. 5-32.
Bernard, B.B., Bernard, H., and Brooks, J.M. 1995. Determination of total carbon, total
organic carbon and inorganic carbon in sediments.
Berner, R.A. 1984. Sedimentary pyrite formation: an update. Geochimica et

Cosmochimica Acta 48: pp. 605-615.
Bhattacharya, J.P. 1989. Allostratigraphy and river- and wave-dominated depositional

systems of the Upper Cretaceous (Cenomanian) Dunvegan Formation,
Alberta. PhD thesis, School of Geography & Earth Sciences, McMaster
University, Hamilton, Ontario.
Bhattacharya, J.P. 2001. Allostratigraphy versus sequence stratigraphy. Presented at

AAPG Hedberg Research Conference, 26-29 August, 2001, Dallas, Texas.
Bhattacharya, J.P. 2011. Practical problems in the application of the sequence

stratigraphic method and key surfaces: integrating observations from ancient
fluvial–deltaic wedges with Quaternary and modelling studies. Sedimentology
58: pp. 120-169.
Bhattacharya, J.P., and MacEachern, J.A. 2009. Hyperpycnal rivers and prodeltaic
shelves in the Cretaceous Seaway of North America. Journal of Sedimentary
Research 79(4): pp. 184-209.
Bhattacharya, J.P., and Posamentier, H.W. 1994. Sequence stratigraphy and

allostratigraphic applications in the Alberta Foreland Basin. In Geological
Atlas of the Western Canada Sedimentary Basin. Edited by G.D. Mossop and
I. Shetsen. CSPG. pp. 407-413.
Bhattacharya, J.P., and Walker, R.G. 1991. Allostratigraphic subdivision of the Upper
Cretaceous Dunvegan, Shaftesbury, and Kaskapau Formations in the
northwestern Alberta subsurface. Bulletin of Canadian Petroleum Geology 39:
pp. 145-164.
Blackett, P.M.S., and Occhialini, G.P.S. 1933. Some photographs of the tracks of
penetrating radiation. Proceedings of the Royal Society of London. Series A,
Containing Papers of a Mathematical and Physical Character 139(839): pp.
699-726. doi: 10.1098/rspa.1933.0048.
224
Bloch, J., Schröder-Adams, C.J., Leckie, D.A., Mcintyre, D.J., Craig, J., and Staniland,
M. 1993. Revised Stratigraphy of the Lower Colorado Group (Albian to
Turonian), Western Canada. Bulletin of Canadian Petroleum Geology 41(3):
pp. 325-348.
Bloch, J.D., Schröder-Adams, C.J., Leckie, D.A., Craig, J., and McIntyre, D.J. 1999.
Sedimentology, micropaleontology, geochemistry, and hydrocarbon potential
of shale from the Cretaceous Lower Colorado Group in Western Canada.
Natural Resources Canada, Geological Survey of Canada.
Bohacs, K.M. 1998. Contrasting expressions of depositional sequences in mudstones

from marine to non-marine environs. In Mudstones and Shales, vol. 1,
Characteristics at the Basin Scale. Edited by J. Schieber and W. Zimmerle, et
al. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart, Germany. pp. 32-77.
Bohacs, K.M., Grabowski, G.J., Carroll, A.R., Mankeiwitz, P.J., Miskell-Gerhardt, K.J.,
Schwalbach, J.R., Wegner, M.B., and Simo, J.A. 2005. Production,
destruction, and dilution - the many paths to source-rock development. SEPM
Special Publication 82: pp. 61-101.
Bohacs, K.M., and Lazar, O.R. 2008. The role of sequence stratigraphy in unraveling and
applying the complex controls from mudstone reservoir properties. Presented
at AAPG Annual Convention and Exhibition, 20-23 April 2008, San Antonio,
Texas.
Bornemann, A., Norris, R.D., Friedrich, O., Beckmann, B., Schouten, S., Damsté, J.S.S.,
Vogel, J., Hofmann, P., and Wagner, T. 2008. Isotopic evidence for glaciation
during the Cretaceous supergreenhouse. Science 319(5860): pp. 189-192.
Bragg, W.H., and Bragg, W.L. 1913. The reflexion of x-rays by crystals. Proceedings of
the Royal Society of London, Series A. 88(605): pp. 428-438. doi:
10.1098/rspa.1913.0040.
Brown, L.F., Jr., and Fisher, W.L. 1977. Seismic stratigraphic interpretation of
depositional systems: examples from Brazilian rist and pull apart basins. In
Seismic stratigraphy - Applications to hydrocarbon exploration. Edited by C.E.
Payton. American Association of Petroleum Geologists Memoir 26. pp. 213-
248.
Burdine, N.T. 1953. Relative permeability calculations from pore size distribution data.
Journal of Petroleum Technology 5(3): pp. 71-78. doi: 10.2118/225-G.
Burrough, P.A. 1986. Principles of geographical information systems for land resources
assessment. Oxford University Press. New York, New York. 193 p.
Busch, A., Schweinar, K., Kampan, N., Coorn, A., Pipich, V., Feoktystov, A., Leu, L.,
Amann-Hildenbrand, A., and Bertier, P. 2017. Determining the porosity of
mudrocks using methodological pluralism. In Geomechanical and
225
petrophysical properties of mudrocks. Edited by E.H. Rutter and J.

Mecklenburgh, et al. Geological Society of London, London. pp. 15-38.
Bust, V.K., Majid, A.A., Oletu, J.U., and Worthington, P.F. 2013. The petrophysics of
shale gas reservoirs: Technical challenges and pragmatic solutions. Petroleum
Geoscience 19(2): pp. 91-103. doi: 10.1144/petgeo2012-031.
Byers, C.W. 1974. Shale fissility: relation to bioturbation. Sedimentology 21(479-484).
Byrnes, A.P. 1997. Reservoir characteristics of low-permeability sandstones in the Rocky

Mountains. Mountain Geologist 34: pp. 39-51.
Byrnes, A.P., and Castle, J.W. 2000. Comparison of core petrophysical properties
between low-permeability sandstone reservoirs: Eastern U.S. Medina Group
and Western U.S. Mesaverde Group and Frontier Formation. 60304-MS SPE
Conference Paper.
Cadrin, A.A.J. 1992. Geochemistry and paleoenvironmental reconstruction of the

Cretaceous Greenhorn Cyclothem in the Western Interior Basin of Canada.
Ph.D. thesis, Department of Geological Sciences, University of Saskatchewan,
Saskatoon, Saskatchewan. 191 p.
Caldwell, W.G.E., Diner, R., Eicher, D.L., Fowler, S.P., North, B.R., Stelck, C.R., and
von Holt Wilhelm, L. 1993. Foraminiferal biostratigraphy of the Cretaceous
marine cyclothems. In Evolution of the Western Interior Basin. Edited by
W.G.E. Caldwell and E.G. Kauffman. Geological Association of Canada,
Special Paper.
Caldwell, W.G.E., North, B.R., Stelck, C.R., and J.H., W. 1978. A foraminiferal zonal
scheme for the Cretaceous System in the Interior Plains of Canada. In Western
and Arctic Canadian Biostratigraphy. Edited by C.R. Stelck and B.D.E.
Chatterton. Geological Association of Canada, Special Paper Number. pp.
495-575.
Canadian Discovery. 2011. Second White Specks oil - another Colorado Group resource
play? Canadian Discovery Digest 1: 82-101. Available from
http://digest.canadiandiscovery.com/exploration-review/second-white-specks-
oil [accessed March 13, 2018].
Canadian Discovery. 2014. 6-20-38-3W5: Tuzo Energy, Significant Well Report:

Canadian Discovery Digest. Available from
http://digest.canadiandiscovery.com/significant-well/6-20-38-3w5-second-
white-specks-oil [accessed March 20, 2018].
Cao, W., Lee, C.-T.A., and Lackey, J.S. 2017. Episodic nature of continental arc activity
since 750 Ma: A global compilation. Earth and Planetary Science Letters 461:
pp. 85-95. doi: 10.1016/j.epsl.2016.12.044.
226
Cattaneo, A., Correggiari, A., Langone, L., and Trincardi, F. 2003. The late-Holocene
Gargano subaqueous delta, Adriatic shelf: Sediment pathways and supply
fluctuations. Marine Geology 193: pp. 61-91.
Cattaneo, A., Trincardi, F., Asioli, A., and Correggiari, A. 2007. The Western Adriatic
shelf clinoform: energy-limited bottomset. Continental Shelf Research 27(3-
4): pp. 506-525. doi: 10.1016/j.csr.2006.11.013.
Catuneanu, O. 2002. Sequence stratigraphy of clastic systems: concepts, merits, and

pitfalls. Journal of African Earth Sciences 35(1): pp. 1-43.
Catuneanu, O. 2006. Principles of sequence stratigraphy. Elsevier. Amsterdam, The

Netherlands.
Catuneanu, O., Abreu, V., Bhattacharya, J.P., Blum, M.D., Dalrymple, R.W., Eriksson,
P.G., Fielding, C.R., Fisher, W.L., Galloway, W.E., Gibling, M.R., Giles,
K.A., Holbrook, J.M., Jordan, R., Kendall, C.G.S.C., Marcuda, B., Martinsen,
O.J., Miall, A.D., Neal, J.E., Nummedal, D., Pomar, L., Posamentier, H.W.,
Pratt, B.R., Sarg, J.F., Shanley, K.W., Steel, R.J., Strasser, A., Tucker, M.E.,
and Winker, C. 2009. Towards the standardization of sequence stratigraphy.
Earth Science Reviews 92(1-2): pp. 1-33.
Catuneanu, O., Galloway, W.E., Kendall, C.G.S.C., Miall, A.D., Posamentier, H.W.,
Strasser, A., and Tucker, M.E. 2011. Sequence Stratigraphy: Methodology and
Nomenclature. Newsletters on Stratigraphy 44(3): pp. 173-245. doi:
10.1127/0078-0421/2011/0011.
Cheel, R.J. 1991. Grain fabric in hummocky cross-stratified storm beds: Genetic
implications. Journal of Sedimentary Petrology 61(1): pp. 102-110.
Chen, J.H., and Moore, J.G. 1982. Uranium-lead isotopic ages from the Sierra Nevada
batholith, California. Journal of Geophysical Research 87: pp. 4761-4784.
Chitale, V.D. 2010. Simplified and more accurate clay typing enhances the value added
by petrophysical evaluation of shale and tight gas sand plays. Presented at
AAPG Convention. AAPG, Denver, Colorado, 7-10 June 2009.
Cho, D., and Perez, M. 2014. Rock quality assessment for hydraulic fracturing: a rock
physics perspective. Presented at SEG Denver 2014 Annual Meeting, Denver,
Colorado, 26-31 October 2014. SEG. doi: 10.1190/segam2014-1624.1
Claisse, F. 1957. Accurate x-ray fluorescence analysis without internal standards. Norelco
Reporter 4: pp. 3-7.
Clark, I. 1979. Practical geostatistics. Applied Science Publishers Ltd. 129 p.
227
Clarkson, C., and Pedersen, P.K. 2011. Production analysis of Western Canadian
unconventional light oil plays. Presented at Canadian Unconventional
Resources Conference, Calgary, Alberta, 15-17 November 2011. SPE.
Clavier, C., Coates, G.R., and Dumanoir, J. 1984. Theoretical and experimental bases for
the “dual water” model for the interpretation of shaly sands. Society of
Petroleum Engineers Journal 24: pp. 153-167.
Clavier, C., Hoyle, W., and Meunier, D. 1971. Quantitative interpretation of thermal
neutron decay time logs: Part I. Fundamentals and Techniques. Journal of
Petroleum Technology 23(6): pp. 743-755. doi: 10.2118/2658-A-PA.
Clavier, C., and Rust, D.H. 1976. MID-plot: a new lithology technique. The Log Analyst
17(6): pp. 16-24.
Cluff, R., and Miller, M. 2010. Log evaluation of gas shales: a 35-year perspective.
DWLS Luncheon April 2010.
Cohen, K.M., Finney, S.C., Gibbard, P.L., and Fan, J.-X. 2013. The ICS International
Chronostratigraphic Chart. Episodes 36(3): pp. 199-204.
Collinson, J.D. 1969. The sedimentology of the Grindslow Shales and the Kinderscout
Grit: a deltaic complex in the Namurian of northern England. Journal of
Sedimentary Research 39: pp. 194-221.
Compton, A.H. 1923. A quantum theory of the scattering of x-rays by light elements.
Physical Review 21(5): pp. 483-502. doi: 10.1103/PhysRev.21.483.
Cosmo, C., Spalburg, M.R., and Looyestijn, W.J. 1991. Fast Deconvolution Of
Laterologs By Direct Inverse Filtering. Presented at SPWLA 32nd Annual
Logging Symposium, 16-19 June 1991. SPWLA.
Crain, E.R. 2018. Crain’s Petrophysical Handbook: Shale volume from gamma ray log
models. Available from https://www.spec2000.net/11-vshgr.htm [accessed
April 4, 2018].
Crain, E.R., and Holgate, D. 2014. A 12-step program to reduce uncertainty in kerogen-
rich reservoirs. Presented at CSPG GeoConvention, Calgary, Alberta, 12-16
May 2014. CSPG.
Creaney, S., and Allan, J. 1990. Hydrocarbon generation and migration in the Western
Canada Sedimentary Basin. In Classic Petroleum Provinces: Geological
Society of London Special Publication 50. Edited by J. Brooks. pp. 189-202.
Creaney, S., Allan, J., Cole, K.S., Fowler, M.G., Brooks, P.W., Osadetz, K.G.,
MacQueen, R.W., Snowdon, L.R., and Riediger, C. 1994. Petroleum
generation and migration in the Western Canada Sedimentary Basin. In
228
Geological Atlas of the Western Canada Sedimentary Basin. Canadian Society

of Petroleum Geologists and Alberta Research Council. pp. 455-469.
Creaney, S., and Passey, Q.R. 1993. Recurring Patterns of Total Organic-Carbon and
Source Rock Quality within a Sequence Stratigraphic Framework. AAPG
Bulletin 77(3): pp. 386-401.
Cross, W., Iddings, J.P., Pirsson, L.V., Washington, H.S. 1902. A quantitative chemico-
mineralogical classification and nomenclature of igneous rocks. Journal of
Geology, 10(6): pp. 555-690.
Cross, T.A., and Lessenger, M.A. 1988. Seismic Stratigraphy. Annual Review of Earth
and Planetary Sciences 16: pp. 319-354.
Cui, A., Wust, R., and Nassichuk, B.R. 2017. A comparison study of existing and new
mineralogical, (geo)mechanical, and petrophysical brittleness indices of the
Alberta Montney and their applicability for optimizing well stimulations.
Presented at CSPG Geoconvention, Calgary, Alberta, 15-19 May 2017.
Curie, P., Curie, M.S., and Bémont, G. 1898. Sur une nouvelle substance fortement radio-
active, contenue dans la pechblende. Comptes rendus de l’Academie des
Sciences 127: pp. 1215-1217.
Curray, J.R. 1960. Sediments and history of Holocene transgression, continental shelf,
northwest Gulf of Mexico. In Recent sediments, northwest Gulf of Mexico.
Edited by F.P. Shepard and F.B. Phleger, et al. AAPG. pp. 221-266.
Curtis, C.D., and Coleman, M.L. 1986. Controls on the precipitation of early diagenetic
calcite, dolomite and sideritic concretions in complex depositional sequences.
Roles of Organic Matter in Sediment Diagenesis, SEPM Special Publication
38.
Curtis, C.D., Petrowski, C., and Oertel, G.F. 1972. Stable carbon isotope ratios within
carbonate concretions: a clue to place and time of formation. Nature 235: pp.
98-100.
Dahlberg, E.C. 1975. Relative effectiveness of geologists and computers in mapping

potential hydrocarbon exploration targets. Journal of the International
Association for Mathematical Geology 7(5-6): pp. 373-394.
Dalrymple, R.W. 2010a. Interpreting sedimentary successions: facies, facies analysis and
facies models. In Facies Models. 4th ed. Edited by N.P. James and R.W.
Dalrymple. Geological Association of Canada, St. John’s, Newfoundland.
Dalrymple, R.W. 2010b. Introduction to siliciclastic facies models. In Facies Models. 4th
ed. Edited by N.P. James and R.W. Dalrymple. Geological Association of
Canada, St. John’s, Newfoundland.
229
Davies, G.R., Fockler, M., and Fox, A. 2013. Structural fabrics in unconventional
reservoirs - geomechanical discontinuities in mudrocks and limestone.
Canadian Discovery Digest 4: 17-54. Available from
http://digest.canadiandiscovery.com/exploration-review/structural-fabrics-
unconventional-reservoirs [accessed March 20, 2018].
Davis, D.H. 1954. Estimating porosity of sedimentary rocks from bulk density. The
Journal of Geology 62(1): pp. 102-107. doi: 10.1086/626136.
Davis, J.C. 2002. Statistics and data analysis in geology. Third ed. John Wiley & Sons.
New York, New York. 638 p.
De Witte, L. 1954. Resistivity Logging in Thin Beds. Presented at Petroleum Branch Fall
Meeting, Dallas, Texas, 19-21 October 1953. Society of Petroleum Engineers.
DeCelles, P.G. 2012. Foreland basin systems revisited: variations in response to tectonic
settings. In Tectonics of Sedimentary Basins: Recent Advances. Edited by C.
Busby and A.A. Pérez. Blackwell Publishing Ltd. pp. 405-426.
DeCelles, P.G., and Giles, K.A. 1996. Foreland basin systems. Basin Research 8: pp. 105-
123.
Deer, D.A., Howie, R.A., and Zussman, J. 1966. An introduction to the rock forming
minerals. Longman Scientific & Technical. Essex, England.
Demaison, G.J., and Moore, G.T. 1980. Anoxic environments and oil source bed genesis.
Organic Geochemistry 2(9): pp. 9-31.
Desmier, E., Flouvat, F., Gay, D., and Selmaoui-Folcher, N. 2011. A clustering-based
visualization of colocation patterns. Proceedings of the 15th Symposium on
International Database Engineering & Applications, Lisbon, Portugal, 21-23
September 2011. pp. 70-78. doi: 10.1145/2076623.2076633.
Dikkers, A.J. 1985. Geology in petroleum production: A primer in production geology.

Developments in Petroleum Science 20. Edited by G.V. Chilingar. Elsevier.
The Hague, The Netherlands. 239 p.
Donovan, D.T., and Jones, E.J.W. 1979. Causes of world-wide changes in sea level.
Journal of the Geological Society, London 136: pp. 187-192. doi:
10.1144/gsjgs.136.2.0187.
Doveton, J.H. 1994a. 1: Wireline logging measurements of subsurface geology. In

Geological Log Interpretation. University of Kansas. pp. 1-7.
Doveton, J.H. 1994b. 13: Computer Methods for Rock Composition Analysis from Logs.
In Geological Log Interpretation. University of Kansas. pp. 121-133.
230
Doveton, J.H. 2001. All Models Are Wrong, but Some Models Are Useful: "Solving" the
Simandoux Equation. Presented at 2001 Annual Conference of the
International Association for Mathematical Geology, Cancún, Mexico, 6-12
September 2001. pp. 1-15.
Doveton, J.H. 2014. Principles of Mathematical Petrophysics. Oxford University Press.

New York, New York. 288 p.
Dresser. 1979. Dresser Atlas log interpretation charts. Houston, Texas. 108 p.
Dubrule, O. 1983. Journal of the International Assocation for Mathematical Geology.

Two methods with different objectives: splines and kriging 15(2): pp. 245-
257.
Dumas, S., Arnott, R.W.C., and Southard, J.B. 2005. Experiments on oscillatory-flow and
combined-flow bed forms: Implications for interpreting parts of the shallow-
marine sedimentary record. Journal of Sedimentary Research 75(3): pp. 501-
513.
Dutton-Marion, K.E. 1988. Principles of interpolation procedures in the display and

analysis of spatial data: A comparative analysis of conceptual and computer
contouring. Ph.D. thesis, Department of Geography, University of Calgary,
Calgary, Alberta.
Dzulynski, S., and Smith, A.J. 1963. Convolute lamination, its origin, preservation, and
directional significance. Journal of Sedimentary Petrology 33(3): pp. 616-627.
Ebanks, W.J. 1987. Flow unit concept - integrated approach to reservoir description for
engineering projects. Presented at AAPG Annual Convention and Exhibition,
Los Angeles, California, 7-10 June 1987.
Edwards, J.M. 1959. Quantitative evaluation of the density log in the Rocky Mountain
area. Journal of Petroleum Technology 12(12): pp. 29-34. doi: 10.2118/1229-
G.
Eick, C.F., Parmar, R., Ding, W., Stepinski, T.F., and Nicot, J.-F. 2008. Finding regional
co-location patterns for sets of continuous variables in spatial datasets.
Proceedings of the 16th ACM SIGSPATIAL International Conference on
Advances in Geographic Information Systems, Irvine, California, 5-7
November 2008. ACM. doi: 10.1145/1463434.1463472.
Einstein, A. 1905. Über einen dir Ezeugung und Versandlung des Lichtes betreffenden
heuristischen Gesichtspunkt. Annalen der Physik 17: pp. 132-148.
Elder, W.P. 1988. Geometry of Upper Cretaceous bentonite beds: Implications about
volcanic source areas and paleowind patterns, western interior, United States.
Geology 16: pp. 835-838.
231
Ellis, D.V., and Singer, J.M. 2008. Well Logging for Earth Scientists. Springer.
Dordrecht, The Netherlands.
Eriksen, M.C., and Slingerland, R. 1990. Numerical simulations of tidal and wind driven
circulation in the Cretaceous Interior Seaway of North America. Geological
Society American Bulletin 102: pp. 1499-1516.
Evenchick, C.A., McMechan, M.E., McNichol, V.J., and Carr, S.D. 2007. A synthesis of
the Jurassic-Cretaceous tectonic evolution of the central and southeastern
Canadian Cordillera: exploring links across the orogen. In Whence the
Mountains? Inquiries into the Evolution of Orogenic Systems: a volume in
Honor of Raymond A. Price. Edited by J.W. Sears and T.A. Harms, et al.
Geological Society of America, Special Paper 433. pp. 117-145.
Fertl, W.H., and Chilingar, G.V. 1988. Total Organic Carbon Content Determined From
Well Logs. SPE Formation Evaluation 3(2): pp. 407-419. doi: 10.2118/15612-
pa.
Fox, A., and Garcia, L. 2015. Common geomechanical drilling problems in Alberta.
Canadian Discovery Digest 1: 1-26. Available from
http://digest.canadiandiscovery.com/feature-article/common-geomechanical-
drilling-problems-alberta [accessed March 20, 2018].
Fox, A., Snelling, P., McKenna, J., Neale, C., Neuhaus, C., and Miskimmins, J. 2013.
Geomechanical principles for unconventional reservoirs. Available from
http://www.microseismic.com/wp-
content/uploads/2017/07/2013_Geomechanical_Principles_For_Unconvention
al_Resources.pdf [accessed March 30, 2018].
Fraser, F.J., McLearn, F.H., Russell, L.S., Warren, P.S., and Wickenden, R.T.D. 1935.
Geology of southern Saskatchewan. Geological Survey of Canada Memoir
176.
Fraser, K.H.A. 2005. The sedimentology and geochemistry of the Second White Specks
Formation, north/central Alberta, Canada. M.Sc. thesis, Department of Earth
and Atmospheric Sciences, University of Alberta, Edmonton, Alberta.
Fritz, P., Binda, P.L., Folinsbee, F.E., and Krouse, H.R. 1971. Isotopic composition of
diagenetic siderites from Cretaceous sediments in Western Canada. Journal of
Sedimentary Petrology 41(1): pp. 282-288.
Furmann, A., Mastalerz, M., Bish, D., Schimmelmann, A., and Pedersen, P.K. 2016.
Porosity and pore size distribution in mudrocks from the Belle Fourche and
Second White Specks Formations in Alberta, Canada. AAPG Bulletin 100(8):
pp. 1265-1288. doi: 10.1306/02191615118.
Furmann, A., Mastalerz, M., Schimmelmann, A., Pedersen, P.K., and Bish, D. 2014.
Relationships between porosity, organic matter, and mineral matter in mature
232
organic-rich marine mudstones of the Belle Fourche and Second White Specks
formations in Alberta, Canada. Marine and Petroleum Geology 54: pp. 65-81.
doi: 10.1016/j.marpetgeo.2014.02.020.
Gadeken, L.L., Gartner, M.L., Sharbak, D.E., and Wyatt, D.F. 1991. The interpretation of
radioactive-tracer logs using gamma-ray spectroscopy measurements. The Log
Analyst 32(1): pp. 24-34.
Gale, J.F.W., Laubach, S.E., Olson, J.E., Eichhubl, P., and Fall, A. 2014. Natural
fractures in shale: A review and new observations. AAPG Bulletin 98(11): pp.
2165-2216. doi: 10.1306/08121413151.
Gale, J.F.W., Reed, R.M., and Holder, J. 2007. Natural fractures in the Barnett Shale and
their importance for hydraulic fracture treatments. AAPG Bulletin 91(4): pp.
603-622. doi: 10.1306/11010606061.
Galloway, W.E. 1989. Genetic stratigraphic sequences in Basin Analysis I: Architecture

and Genesis of Flooding-Surface Bounded Depositional Units. AAPG Bulletin
73(2): pp. 125-142.
Galloway, W.E., and Hobday, D. 1996. Terrigenous clastic depositional systems:

applications to fossil fuel and groundwater resources. Springer-Verlag.
Heidelberg. 489 p.
Gao, G., Abubakar, A., and Habashy, T. 2013. Borehole petrophysical imaging using
induction and acoustic measurements. Petrophysics 54(4): pp. 383-394.
Gary, M., McAfee Jr., R., and Wolf, C.L. 1974. Glossary of Geology. American
Geosciences Institute. Washington, D.C. 805 p.
Gaymard, R., and Poupon, A. 1968. Response of neutron and formation density logs in
hydrocarbon bearing formations. The Log Analyst 9(5): pp. 3-20.
Gaynor, G.C., and Schiehing, M.H. 1988. Shelf depositional environments and reservoir
characteristics of the Kuparuk River Formation (Lower Cretaceous), Kuparuk
Field, north slope, Alaska. In Giant oil and gas fields - A core workshop.
Edited by A.J. Lomando and P.M. Harris. SEPM Core Workshop 12. pp. 333-
389.
Gilboy, C.F. 1988. Geology and natural gas production of the Upper Cretaceous Second
White-Speckled Shale, Southwestern Saskatchewan. Saskatchewan Geological
Survey, Saskatchewan Energy and Mines.
Glaser, K.S., Miller, C.K., Johnson, G.M., Toelle, B., Kleinberg, R.L., Miller, P., and
Pennington, W.D. 2013. Seeking the sweet spot: reservoir and completion
quality in organic shales. Oilfield Review 25(4).
233
Glass, D.J. 1990. CSPG Lexicon of Canadian Stratigraphy, vol. 4, Western Canada,
including eastern British Columbia, Alberta, Saskatchewan, and southern
Manitoba. Edited by D.J. Glass. Canadian Society of Petroleum Geologists.
Glover, P.W.J. 2000. Formation Evaluation: MSc. Petroleum Geology course notes.
Department of Geology and Petroleum Geology, University of Aberdeen, UK.
Available from
http://homepages.see.leeds.ac.uk/~earpwjg/PG_EN/CD%20Contents/GGL-
66565%20Petrophysics%20English/ [accessed April 4, 2018].
Gonzalez, J., Lewis, R., Hemingway, J., Grau, J., Rylander, E., and Schmitt, R. 2013.
Determination of formation organic carbon content using a new neutron-
induced gamma ray spectroscopy service that directly measures carbon.
Presented at SPWLA 54th Annual Logging Symposium, New Orleans,
Louisiana, 22-26 June 2013. SPWLA.
Grabowski, R.C., Droppo, I.G., and Wharton, G. 2011. Erodibility of cohesive sediment:
the importance of sediment properties. Earth Science Reviews 105(101-120).
Guidry, F.K., Luffel, D.L., and Olszewski, A.J. 1990. Devonian Shale Formation
Evaluation Model Base on Logs, New Core Analysis Methods, and Production
Tests. Presented at SPWLA 31st Annual Logging Symposium, 24-27 June
1990, Lafayette, Louisiana.
Guthrie, J.M., and Bohacs, K.M. 2009. Spatial variability of source rocks: a critical
element for defining the petroleum system of Pennsylvanian carbonate
reservoirs of the Paradox Basin, SE Utah. In The Paradox Basin Revisited -
New Developments in Petroleum Systems and Basin Analysis. Edited by W.S.
Houston and L.L. Wray, et al. Rocky Mountain Association of Geologists. pp.
95-130.
Haecker, A., Carvajal, H., and White, J. 2016. Comparison of organic matter correlations
in North American shale plays. Presented at SPWLA 57th Annual Logging
Symposium, Reykjavik, Iceland, 25-29 June 2016.
Hall, M. 2017. 90 years of well logs [Blog]. Available from

https://agilescientific.com/blog/2017/9/5/90-years-of-well-logs [accessed
September 27 2017].
Hallam, A. 1985. A review of Mesozoic climates. Journal of the Geological Society,

London 142: pp. 433-445.
Handwerger, D.A., Suarez-Rivera, R., Vaughn, K.I., and Keller, J.F. 2012. Methods
improve shale core analysis. The American Oil & Gas Reporter.
Haq, B.U. 2014. Cretaceous eustasy revisited. Global and Planetary Change 113: pp. 44-
58. doi: 10.1016/j.gloplacha.2013.12.007.
234
Haq, B.U., Hardenbol, J., and Vail, P.R. 1987. Chronology of fluctuating sea levels since
the Triassic. Science 235: pp. 1156-1167.
Harrison, C.G.A. 1990. Long-term eustasy and epeirogeny in continents. In Sea-Level

Change. Edited by R. Revelle. National Research Council, Studies in
Geophysics, National Academy Press, Washington, D.C. pp. 141-158.
Haskett, W.J. 2014. The myth of sweet spot exploration. Presented at SPE Annual
Technical Conference and Exhibition, Amsterdam, The Netherlands, 27-29
October 2014. SPE.
Hattin, D.E. 1971. Widespread, synchronously deposited, burrow-mottled limestone beds

in Greenhorn limestone (upper Cretaceous) of Kansas and southeastern
Colorado. AAPG Bulletin 55(3): pp. 412-431.
Hay, W.H., Eicher, D.L., and Diner, R. 1993. Physical oceanography and water masses in
the Cretaceous Western Interior Seaway. In Evolution of the Western Interior
Basin. Edited by W.G.E. Caldwell and E.G. Kauffman. Geological
Association of Canada, Special Paper.
Hay, W.H., and Floegel, S. 2012. New thoughts about the Cretaceous climate and oceans.
Earth Science Reviews 115: pp. 262-272. doi:
10.1016/j.earscirev.2012.09.008.
Heezen, B.C., Hollister, C.D., and Ruddiman, W.F. 1966. Shaping of the continental rise
by deep geostrophic contour currents. Science 152: pp. 133-144. doi:
10.1126/science.152.3721.502.
Heidari, M., Khanlari, G.R., Torabi-Kaveh, M., Kargarian, S., and Saneie, S. 2013. Effect
of Porosity on Rock Brittleness. Rock Mechanics and Rock Engineering
47(2): pp. 785-790. doi: 10.1007/s00603-013-0400-0.
Heidari, Z. 2011. Estimation of static and dynamic petrophysical properties from well
logs in multi-layer formations. Ph.D. thesis, Department of Petroleum and
Geosystems Engineering, The University of Texas at Austin, Austin, Texas.
Heidari, Z., and Torres-Verdin, C. 2014. Inversion-based detection of bed boundaries for
petrophysical evaluation with well logs: Applications to carbonate and
organic-shale formations. Interpretation-a Journal of Subsurface
Characterization 2(3): pp. T129-T142. doi: 10.1190/Int-2013-0172.1.
Heidari, Z., Torres-Verdin, C., and Preeg, W.E. 2012. Improved estimation of mineral
and fluid volumetric concentrations from well logs in thinly bedded and
invaded formations. Geophysics 77(3): pp. Wa79-Wa98. doi:
10.1190/Geo2011-0454.1.
235
Helander, D.P., and Campbell, J.M. 1966. The effect of pore configuration, pressure and
temperature on rock resistivity. Presented at SPWLA 7th Annual Logging
Symposium, 9-11 May 1966, Tulsa, OK.
Helland-Hansen, W., and Gjelberg, J.G. 1994. Conceptual basis and variability in
sequence stratigraphy: A different perspectivw. Sedimentary Geology 92: pp.
31-52.
Henry, D., and Goodge, J. 2016. Wavelength-dispersive x-ray spectroscopy (WDS).

Available from
https://serc.carleton.edu/research_education/geochemsheets/wds.html
[accessed April 8, 2018].
Herwanger, J.V., and Mildren, S.D. 2015. Uses and Abuses of the Brittleness Index With
Application to Hydraulic Stimulation. Proceedings of the 3rd Unconventional
Resources Technology Conference, San Antonio, Texas, 20-22 July 2015. doi:
10.15530/urtec-2015-2172545.
Heslop, A. 1974. Gamma-ray log response of shaly sandstones. Presented at SPWLA

15th Annual Logging Symposium, McAllen, Texas, 2-5 June 1974.
Heslop, K.A. 2010. Generalized method for the estimation of TOC from GR and Rt*.
Presented at AAPG Annual Convention and Exhibition, New Orleans,
Louisiana, 11-14 April 2010.
Hesthammer, J. 1998. Integrated use of well data for structural control of seismic
interpretation. Petroleum Geoscience 4: pp. 97-109.
Hicks, B. 2010. 2WS: Where are the sweet spots and how can we convert resources to
reserves? Presented at AAPG Hedberg Conference, 5-10 December 2010,
Austin, Texas.
Higley, D.K., Lewan, M.D., Roberts, L.N.R., and Henry, M. 2009. Timing and petroleum
sources for the Lower Cretaceous Mannville Group oil sands of northern
Alberta based on 4-D modeling. AAPG Bulletin 93(2): pp. 203-230. doi:
10.1306/09150808060.
Hill, D.G. 2012. Historical review - Milestone developments in petrophysics. In

Fundamentals of the Petrophysics of Oil and Gas Reservoirs. Edited by L.
Buryakovsky and G.V. Chillingar, et al. Scrivener, Beverly, MA.
Hohn, M.E. 1988. Geostatistics and petroleum geology. Van Nostrand Reinhold. New
York, New York.
Homma, H., Inoue, H., Feeney, M., Oelofse, L., and Kataoka, Y. 2012. Strategy of fusion
bead correction in XRF analysis of powders. Advances in X-ray Analysis 55:
pp. 242-251.
236
Hood, A., Gutjahr, C.C.M., and Heacock, R.L. 1975. Organic metamorphism and the
generation of petroleum. AAPG Bulletin 59(6): pp. 986-996.
Howard, W.E., and Hunt, E.R. 1986. Travis Peak: an integrated approach to formation
evaluation. Presented at SPE Unconventional Gas Technology Symposium,
18-21 May 1986, Louisville, Kentucky. SPE. doi: 10.2118/15208-MS.
Hu, Y., Wu, K., Chen, Z., Zhang, K., Ji, D., and Zhong, H. 2015. A novel model of
brittleness index for shale gas reseroirs: confining pressure effect. Presented at
SPE Asia Pacific Unconventional Resources Conference and Exhibition,
Brisbane, Australia, 9-11 November 2015.
Hubbell, J.H. 2006. Electron-positron pair production by photons: A historical overview.

Radiation Physics and Chemistry 75: pp. 614-623. doi:
10.1016/j.radphyschem.2005.10.008.
Huber, B.T., Norris, R.D., and MacLeod, K.G. 2002. Deep-sea paleotemperature record
of extreme warmth during the Cretaceous. Geology 30(2): pp. 123-126. doi:
10.1130/0091-7613(2002)030<0123:DSPROE>2.0.CO;2.
Hudson, J.D., and Martill, D.M. 1991. The Lower Oxford Clay: production and
preservation of organic matter in the Callovian (Jurassic) of central England.
In Modern and ancient continental shelf anoxia. Edited by R.V. Tyson and
T.H. Pearson. Geological Society of London, Special Publication 58. pp. 363-
379.
Hurst, A. 1990. Natural gamma-ray spectroscopy in hydrocarbon bearing sandstones from

the Norwegian continental shelf. In Geological Applications of Wireline Logs.
Edited by A. Hurst and M.A. Lovell, et al. pp. 211-222.
Issler, D.R., Hu, K., and Bloch, J.D. 2002. Organic carbon content determined from well
logs: examples from Cretaceous sediments of Western Canada. Presented at
CSPG GeoConvention, Calgary, Alberta, 3-7 June 2002.
Jarvie, D.M. 1991. Total organic carbon (TOC) analysis: Chapter 11: Geochemical
methods and exploration. In Source and migration processes and evaluation
techniques: AAPG Special Volume Edited by M.K. Merrill. pp. 113-118.
Jarvie, D.M. 2012a. Shale resource systems for oil and gas: Part 1 - shale-gas resource
systems. . In Shale reservoirs - giant resources for the 21st century: AAPG
Memoir 97. Edited by J.A. Breyer. pp. 69-87.
Jarvie, D.M. 2012b. Shale resource systems for oil and gas: Part 2 - shale-oil resource
systems. In Shale reservoirs - giant resources for the 21st century: AAPG
Memoir 97. Edited by J.A. Breyer. pp. 89-119.
Jarvie, D.M., Hill, R.J., Ruble, T.E., and Pollastro, R.M. 2007. Unconventional shale-gas
systems: The Mississippian Barnett Shale of north-central Texas as one model
237
for thermogenic shale-gas assessment. AAPG Bulletin 91(4): pp. 475-499. doi:
10.1306/12190606068.
Jervey, M.T. 1988. Quantitative geological modeling of siliciclastic rock sequences and
their seismic expression. In Sea-level changes: an integrated approach. Edited
by C.K. Wilgus and B.S. Hasting, et al. Society of Economic Paleontologists
and Mineralogists, Special Publication 42, Tulsa, Oklahoma. pp. 47-69.
Jiang, C., Chen, Z., Lavoie, D., Percival, J.B., and Kabanov, P. 2017. Mineral carbon
MinC(%) from Rock-Eval analysis as a reliable and cost-effective
measurement of carbonate contents in shale source and reservoir rocks.
Marine and Petroleum Geology 83: pp. 184-194.
Jiang, P. 2013. Pore morphometrics and thermal evolution of organic-matter

microporosity, Colorado Group, Western Canada Foreland Basin. M.Sc.
thesis, Department of Earth Sciences, The University of Western Ontario,
London, Ontario.
Jiang, P., and Cheadle, B.A. 2013. Depositional and burial domain influences on
microporosity modalities in carbonaceous mudstones of the Upper Cretaceous
Colorado Group, Western Canada Foreland Basin. Presented at AAPG Annual
Convention and Exhibition, Pittsburgh, Pennsylvania, 19-22 May 2013.
Jin, X., Shah, S., Roegiers, J.-C., and Zhang, B. 2015. An integrated petrophysics and
geomechanics approach for fracability evaluation in shale reservoirs.
Presented at SPE Hydraulic Fracturing Conference, The Woodlands, Texas, 4-
6 February 2014.
Jin, X., Shah, S., Truax, J.A., and Roegiers, J.-C. 2014a. A practical petrophysical
approach for brittleness prediction from porosity and sonic logging in shale
reservoirs. Presented at SPE Annual Technical Conference and Exhibition,
Amsterdam, The Netherlands, 27-29 October 2014.
Jin, X., Shah, S.N., Roegiers, J.-C., and Zhang, B. 2014b. Fracability evaluation in shale
reservoirs-an integrated petrophysics and geomechanics approach. Presented
at SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas,
4-6 February 2014.
Johnson, H.D., and Baldwin, C.T. 1996. Shallow clastic seas. In Sedimentary
environments: Processes, facies, and stratigraphy. Edited by H.G. Reading.
Blackwell Science, London, UK. pp. 232-280.
Jones, T.A., Hamilton, D.E., and Johnson, C.R. 1986. 4: Simple grids and contour maps.
In Contouring geological surfaces with the computer. Van Nostrand Reinhold
Company, New York, New York. pp. 43-67.
238
Jordan, T.E., and Flemings, P.B. 1991. Large-scale stratigraphic architecture, eustatic
variation, and unsteady tectonism: A theoretical evaluation. Journal of
Geophysical Research 96(B4): pp. 6681-6699.
Juhasz, I. 1981. Normalised Qv - the key to shaly sand evaluation using the Waxman-
Smits equation in the absence of core data. Presented at SPWLA 22nd Annual
Logging Symposium, Mexico City, Mexico, 23-26 June 1981.
Kackstaetter, U.R. 2014. SEDMIN - Microsoft Excel spreadsheet for calculating fine-
grained sedimentary rock mineralogy from bulk geochemical analysis. Central
European Journal of Geosciences 6(2): pp. 170-181.
Katz, D., Jung, M., Canter, L., Sonnenfeld, M., Odegard, M., Daniels, J., Byrnes, A.,
Guisinger, M., and Forster, J. 2016. Mineralogy derived brittleness from the
Qemscan: Niobrara case study. Presented at SPE Low Perm Symposium,
Denver, Colorado, 5-6 May 2016.
Kauffman, E.G. 1969. Cretaceous marine cycles of the Western Interior. Mountain
Geologist 6(4): pp. 227-245.
Kauffman, E.G. 1977. Geological and biological overview: Western Interior Cretaceous
Basin: Cretaceous facies, faunas, and paleoenvironments across the Western
Interior Basin. Mountain Geologist 14(3): pp. 75-99.
Kauffman, E.G. 1984. Paleogeography and evolutionary response dynamic in the

Cretaceous Western Interior Seaway of North America. In Jurassic-Cretaceous
Bioclimatology and Paleogeography of North America. Edited by G.E.G.
Westerman. Geological Association of Canada, Special Paper. pp. 273-306.
Kauffman, E.G., and Caldwell, W.G.E. 1993. The Western Interior Basin in space and
time. In Evolution of the Western Interior Basin. Edited by W.G.E. Caldwell
and E.G. Kauffman. Geological Association of Canada, Special Paper. pp.
273-306.
Kauffman, E.G., Sageman, B.B., Kirkland, J.I., Elder, W.P., Harries, P.J., and Villamil, T.
1993. Molluscan biostratigraphy of the Cretaceous Western Interior Basin,
North America. In Evolution of the Western Interior Basin. Edited by W.G.E.
Caldwell and E.G. Kauffman. Geological Association of Canada, Special
Paper. pp. 397-434.
Kawahara, N., and Shoji, T. 2007. Wavelength dispersive XRF and a comparison with
EDS. In Handbook of practical x-ray fluorescence. Edited by B. Beckhoff and
B. Kanngießer, et al. Springer, New York, New York.
Keulegan, G.H., and Krumbein, W.C. 1949. Stable configuration of bottom slope in a
shallow sea and its bearing on geological processes. Transations of the
American Geophysical Union 30(6): pp. 855-861.
239
King, B.-S., and Vivit, D. 1988. Loss-on-ignition corrections in the XRF analysis of
silicate rocks. X-Ray Spectromoetry 17(4): pp. 145-147. doi:
10.1002/xrs.1300170406.
Klenner, R., Braunberger, J., Sorensen, J., Eylands, K., Azenkeng, A., and Smith, S.
2014. A formation evaluation of the middle Bakken Member using a
multimineral petrophysical analysis approach. Presented at Unconventional
Resources Technology Conference, 25-27 August 2014, Denver, Colorado.
doi: 10.15530/urtec-2014-1922735.
Knechtel, M.M., and Patterson, S.H. 1956. Bentonite deposits in marine Cretaceous
formations, Hardin district, Montana and Wyoming. USGS Bulletin 1023: pp.
1-116.
Knox, C.C. 1975. Quality control of well logs. The Log Analyst 16(3): pp. 14-20.
Kominz, M.A. 1984. Oceanic ridge volumes and sea-level change - An error analysis. In
Interregional unconformities and hydrocarbon accumulation, AAPG Memoir
36. Edited by J.S. Schlee. pp. 109-127.
Kreitner, M. 2002. Sedimentology, stratigraphy, and paleogeography of the Lower

Kaskapau Formation, Upper Cretaceous (Cenomanian), northwest Alberta and
northeast British Columbia. M.Sc. thesis, Department of Earth Sciences, The
University of Western Ontario, London, Ontario.
Kreitner, M., and Plint, A.G. 2006. Allostratigraphy and paleogeography of the Upper
Cenomanian, Lower Kaskapau Formation in subsurface and outcrop, Alberta
and British Columbia. Bulletin of Canadian Petroleum Geology 54: pp. 147-
174.
Krige, D.G. 1951. A statistical approach to some basic mine valuation problems on the
Witwatersrand. Journal of the Chemical, Metallurgical and Mining Society of
South Africa 52(6): pp. 119-139.
Krygowski, D.A. 2004. Guide to petrophysical interpretation. Austin, Texas. p. 147.
Kuehl, S.A., Nittrouer, C.A., Allison, M.A., Faria, L.E.C., Dukat, D.A., Jaeger, J.M.,
Pacioni, T.D., Figueiredo, A.G., and Underkoffler, E.C. 1996. Sediment
deposition, accumulation, and seabed dynamics in an energetic fine-grained
coastal environment. Continental Shelf Research 16(5/6): pp. 787-815.
Kuroda, J., Ogawa, N.O., Tanimizu, M., Coffin, M.F., Tokuyama, H., Kitazato, H., and
Ohkouchi, N. 2007. Contemporaneous massive subaerial volcanism and Late
Cretaceous ocean anoxic event 2. Earth and Planetary Science Letters 256: pp.
211-223.
La Tour, T.E. 1989. Analysis of rocks using x-ray fluorescence spectrometry. The Rigaku
Journal 6(1): pp. 3-9.
240
Labani, M., and Rezaee, R. 2015. Petrophysical evaluation of gas shale reservoirs. In
Fundamentals of gas shale reservoirs. Edited by R. Rezaee. John Wiley &
Sons, Hoboken, New Jersey. pp. 117-138.
Lambeck, K., Cloetingh, S., and McQueen, H. 1987. Intraplate stress and apparent
changes in sea level: the basins of northwestern Europe. In Sedimentary
Basins and Basin-forming Mechanisms. Edited by C. Beaumont and A.J.
Tankard. Canadian Society of Petroleum Geologists, Memoir 12. pp. 259-268.
Larsen, R.L. 1991. Geological consequences of superplumes. Geology 19: pp. 963-966.
Law, C.A. 1999. Evaluating source rocks. In Handbook of Petroleum Geology: Exploring
for Oil and Gas Traps. Edited by E.A. Beaumont and N.H. Foster. American
Association of Petroleum Geologists. pp. 308-348.
Laycock, D.P. 2014. Stratigraphy, sedimentology, and geochemistry of mudstone-

dominated clinoforms and ther depositional environments, Carlile Formation,
Eastern Alberta, Canada. PhD thesis, Department of Geoscience, University of
Calgary, Calgary, Alberta.
le Roux, J.P. 1993. Palaeogeographic reconstruction of sandstones using weighted mean

grain-size maps, with exampls from the Karoo Basin (South Africa) and the
Sydney Basin (Australia). Sedimentary Geology 81: pp. 173-180.
Leckie, D.A., Bhattacharya, J.P., Bloch, J.D., Gilboy, C.F., Norris, B., Plint, A.G.,
Gilders, M., Holmstrom, G., Krause, F.F., Reinson, F.E., Safton, D., Sawicki,
J., and Sawicki, O. 1994. Cretaceous Colorado/Alberta Group of the Western
Canada Sedimentary Basin. In Geological Atlas of the Western Canada
Sedimentary Basin. Edited by G.D. Mossop and I. Shetsen. Canadian Society
of Petroleum Geologists and the Alberta Research Council.
Leckie, D.A., Schröder-Adams, C.J., and Bloch, J.D. 2000. The effect of paleotopography
on the late Albian and Cenomanian sea-level record of the Canadian
Cretaceous interior seaway. Geological Society of America Bulletin 112(8):
pp. 1179-1198. doi: 10.1130/0016-7606(2000)112<1179:TEOPOT>2.0.CO;2.
LECO. 1996. CNS-2000 Elemental Analyzer - Instruction manual. LECO Corporation.
Lee, C.-T.A., Jiang, H., Ronay, E., Minisini, D., Stiles, J., and Neal, M. 2018. Volcanic
ash as a driver of enhanced organic carbon burial in the Cretaceous. Scientific
Reports 8(1): p. 4197. doi: 10.1038/s41598-018-22576-3.
Letourneau, J., Magee, A., and Jones, L. 1996. The Willesden Green Second White
Specks pools, central Alberta: an evaluation of fluid properties and
compartments. In Oil and Gas Pools of the Western Canada Sedimentary
Basin: Program and Abstracts. Edited by J.R. Hogg. CSPG, Calgary, Alberta.
Lisle, R.J. 2004. Geological structures and maps. Third ed. Elsevier. Oxford, UK. 106 p.
241
Liu, Z., Torres-Verdín, C., Wang, G.L., Mendoza, A., Zhu, P., and Terry, R. 2007. Joint
Inversion of Density and Resistivity Logs for the Improved Petrophysical
Assessment of Thinly-Bedded Clastic Rock Formation. Presented at SPWLA
48th Annual Logging Symposium, 3-6 June 2007, Austin, Texas.
Longoni, A., and Fiorini, C. 2007. X-ray detectors and signal processing. In Handbook of
practical x-ray fluorescence analysis. Edited by B. Beckhoff and B.
Kanngießer, et al. Springer, New York, New York.
Lorenz, J.C. 1982. Lithospheric flexure and the history of the Sweetgrass Arch,
northwestern Montana. In Geologic studies of the Cordilleran Thrust Belt.
Edited by R.B. Powers. Rocky Mountain Association of Geologists, Denver,
Colorado. pp. 77-89.
Loutit, T.S., Hardenbol, J., Vail, P.R., and Baum, G.R. 1988. Condensed sections: the key
to age-dating and correlation of continental margin sequences. In Sea Level
Changes - An Integrated Approach. Edited by C.K. Wilgus and B.S. Hastings,
et al. SEPM Special Publication. pp. 183-213.
Luffel, D.L., and Guidry, F.K. 1992. New Core Analysis-Methods for Measuring
Reservoir Rock Properties of Devonian Shale. Journal of Petroleum
Technology 44(11): pp. 1184-1190.
MacKay, P. 2014. Second White Specks Formation: New concepts for understanding
fractured reservoirs. Presented at CSPG AAPG Playmaker Forum, Calgary,
Alberta, 27 May 2014.
Macquaker, J.H.S., Bentley, S.J., and Bohacs, K.M. 2010. Wave-enhanced sediment-
gravity flows and mud dispersal across continental shelves: Reappraising
sediment transport processes operating in ancient mudstone successions.
Geology 38(10): pp. 947-950.
Macquaker, J.H.S., Taylor, K.G., and Gawthorpe, R.L.G. 2007. High-resolution facies
analyses of mudstones: implications for paleoenvironmental and sequence
stratigraphic interpretations of offshore ancient mud-dominated successions.
Journal of Sedimentary Research 77: pp. 324-339.
Magoon, L.B., and Dow, W.G. 1994. The petroleum system. In The petroleum system -
from source to trap. AAPG Memoir 60. Edited by L.B. Magoon and W.G.
Dow.
Martinsen, O.J., Martinsen, R.S., and Steidmann, J.R. 1993. Mesaverde Group (Upper
Cretaceous), southern Wyoming: allostratigraphy versus sequence stratigraphy
in a tectonically active area. AAPG Bulletin 77: pp. 1351-1373.
Martinsen, O.J., Ryseth, A., Helland-Hansen, W., Flesche, H., Torkildsen, G., and Idil, S.
1999. Stratigraphic base level and fluvial architecture: Ericson Sandstone
242
(Campanian), Rock Springs Uplift, SW Wyoming, USA. Sedimentology 46:

pp. 235-239.
Mathia, E., Ratcliffe, K., and Wright, M. 2016. Brittleness index - a parameter to embrace
or avoid? Presented at Unconventional Resources Technology Conference,
San Antonio, Texas, 1-3 August 2016.
Mayer, C., and Sibbit, A. 1980. Global: a new approach to computer-processed log
interpretation. Presented at SPE Annual Technical Conference and Exhibition,
21-24 September 1980, Dallas, Texas. doi: 10.2118/9341-MS.
McCarty, D.K., Theologou, P.N., Fischer, T.B., Derkowski, A., Stokes, M.R., and Ollila,
A. 2015. Mineral-chemistry quantification and petrophysical calibration for
multimineral evaluations: A nonlinear approach. AAPG Bulletin 99(7): pp.
1371-1397. doi: 10.1306/01221514195.
McCausland, P., Symons, D.T.A., Hart, C.J.R., and Blackburn, W.H. 2006. Assembly of
the northern Cordillera: New paleomagnetic evidence for coherent, moderate
Jurassic to Eocene motion of the Intermontane belt and Yukon-Tanana
terranes. In Paleogeography of the North American Cordillera: Evidence for
and against large-scale displacements. Edited by J.W. Haggart and R.J. Enkin,
et al. Geological Association of Canada, Special Paper 46. pp. 147-170.
McCave, I.N. 1984. Size spectra and aggregation of suspended particles in the deep
ocean. Deep-Sea Research 31: pp. 329-352.
Merriam, D.F., and Jewett, D.G. 1989. Methods of thematic map comparison. In Current
trends in geomathematics. Edited by D.F. Merriam. pp. 9-18.
Miall, A.D. 2010. The geology of stratigraphic sequences. Second ed. Springer-Verlag.
Berlin, Germany. 514 p.
Middleton, G.V. 1973. Johannes Walther’s Law of the Correlation of Facies. Geological
Society American Bulletin 84: pp. 979-988.
Midtkandal, I., Nystuen, J.P., and Nagy, J. 2007. Paralic sedimentation on an

epicontinental ramp shelf during a full cycle of relative sea-level fluctuation;
the Helvetiafjellet Formation in Nordenskiöld Land, Spitsbergen. Norwegian
Journal of Geology 87: pp. 343-359.
Milankovitch, M. 1941. Canon of Insolation and the Ice-Age Problem (in German). In
Special Publications of the Royal Serbian Academy. Israel Program for
Scientific Translations.
Miller, K.G., Kominz, M.A., Browning, J.V., Wright, J.D., Mountain, G.S., Katz, M.E.,
Sugarman, P.J., Cramer, B.S., Christie-Blick, N., and Pekar, S.F. 2005. The
Phanerozoic record of global sea-level change. Science 310(5752): pp. 1293-
1298. doi: 10.1126/science.1116412.
243
Millikan, R.A. 1913. On the elementary electrical charge and the Avogadro constant.
Physical Review 2(2): pp. 109-143. doi: 10.1103/PhysRev.2.109.
Minette, D.C. 1990. Thin Bed Resolution Enhancement: Potential And Pitfalls. Presented
at SPWLA 31st Annual Logging Symposium, Lafayette, Louisiana, 24-27
June, 1990.
Mitchell, W.K., and Nelson, R.J. 1988. A practical approach to statistical log analysis.
Presented at SPWLA 29th Annual Logging Symposium, San Antonio, Texas,
5-8 June 1988.
Mitchum, R.M. 1977. Seismic stratigraphy and global changes of sea level; part 11,
glossary of terms used in seismic stratigraphy. American Association of
Petroleum Geologists Memoir 26: pp. 205-212.
Mitchum, R.M., Vail, P.R., and Thompson, S., III. 1977. Seismic stratigraphy and global
changes of sea level; Part 2, The depositional sequence as a basic unit for
stratigraphic analysis. In Seismic stratigraphy: applications to hydrocarbon
exploration. Edited by C.E. Payton. AAPG Memoir 26. pp. 53-62.
Moore, W.R., Ma, Y.Z., Urdea, J., and Bratton, T. 2011. Uncertainty analysis in well-log
and petrophysical interpretations. In Uncertainty analysis and reservoir
modeling: AAPG Memoir 96. Edited by Y.Z. Ma and P.R. La Pointe. AAPG.
pp. 17-28.
Moore, W.R., Zee Ma, Y., Pirie, I., and Zhang, Y. 2016. Tight gas sandstone reservoirs,
part 2: Petrophysical analysis and reservoir modeling. In Unconventional Oil
and Gas Resources Handbook: Evaluation and Development. Edited by Y. Zee
Ma and S.A. Holditch. Gulf Professional Publishing, Oxford, UK. pp. 429-
448.
Moseley, H.G.J. 1913. The high-frequency spectra of the elements. Philosophical

Magazine 26: pp. 1024-1034.
Murray, S.P. 1970. Bottom currents near the Coast during Hurricane Camille. Journal of
Geophysical Research 75(24): pp. 4579-4582.
Muto, T., and Steel, R.J. 1997. Principles of regression and transgression: the nature of
the interplay between accommodation and supply. Journal of Sedimentary
Research 67(6): pp. 994-1000. doi: 1073-130X/97/067-994.
Muto, T., and Steel, R.J. 2000. The accommodation concept in sequence stratigraphy:
some dimensional problems and possible redefinition. Sedimentary Geology
130: pp. 1-10.
NACSN. 1983. North American stratigraphic code. AAPG Bulletin 67: pp. 841-875.
244
NACSN. 2005. North American Stratigraphic Code. AAPG Bulletin 89(11): pp. 1547-
1591.
Nagy, D., Franke, R., Battha, L., Kalmár, J., Papp, G., and Závoti, J. 1999. Comparison of
various gridding methods. Acta Geodaetica et Geophysica Hungarica 34(1-2):
pp. 41-51.
Neidell, N.S. 1979. Stratigraphic modeling and interpretation: geophysical principles and
techniques. American Association of Petroleum Geologists Continuing
Education Course Note Series 13. 145 p.
Nelson, R.A., Patton, T.L., and Serra, S. 1999. Exploring for structural traps. In
Handbook of Petroleum Geology: Exploring for Oil and Gas Traps. Edited by
E.A. Beaumont and N.H. Foster. American Association of Petroleum
Geologists. pp. 20.21-20.70.
Nichols, G. 2009. Sedimentology and stratigraphy. Second ed. Wiley-Blackwell. 419 p.
Nielsen, K.S., Schroder-Adams, C.J., Leckie, D.A., Haggart, J.W., and Elberdak, K.
2008. Turonian to Santonian paleoenvironmental changes in the Cretaceous
Western Interior Sea: The Carlile and Niobrara formations in southern Alberta
and southwestern Saskatchewan, Canada. Palaeogeography,
Palaeoclimatology, Palaeoecology 270(1-2): pp. 64-91. doi:
10.1016/j.palaeo.2008.08.018.
Nishida, N., Ito, M., Inoue, A., and Takizawa, S. 2013. Clay fabric of fluid mud deposits
from laboratory and field observations: potential application to the
stratigraphic record. Marine Geology 337: pp. 1-8.
Nittrouer, C.A., DeMaster, D.J., Kuehl, S.A., and McKee, B.A. 1986. Association of sand
with mud deposits accumulating on continental shelves. In Shelf sands and
sandstones. Edited by R.J. Knight and C.A. Ross. CSPG Memoir 11. pp. 17-
25.
Nittrouer, C.A., and Wright, L.D. 1994. Transport of particles across continental shelves.
Reviews of Geophysics 32(1): pp. 85-113.
Nummedal, D., Riley, G.W., and Templet, P.L. 1993. High-resolution sequence
architecture: A chronostratigraphic model based on equilibrium profile
studies. In Sequence stratigraphy and facies associations. Edited by H.W.
Posamentier and C.P. Summerhayes, et al. The International Association of
Sedimentologists Special Publication 18. pp. 55-68.
O’Connell, S.C. 2003. The unknown giants - low-permeability shallow gas reservoirs of
southern Alberta and Saskatchewan, Canada. Presented at CSPG/CSEG Joint
Convention, Calgary, Alberta.
245
Obradovich, J.D. 1991. A revised Cenomanian-Turonian time scale based on studies from
the Western Interior United States. Geological Society of America Abstracts
with Program 23(5): pp. 149-153.
Obradovich, J.D. 1993. A Cretaceous time scale. In The Cretaceous System in the
Western Interior of North America. Edited by W.G.E. Caldwell and E.G.
Kauffman. Geological Association of Canada. pp. 303-331.
Obradovich, J.D., and Cobban, W.A. 1975. A time scale for the Late Cretaceous of the
western interior of North America. In The Cretaceous System in the western
interior of North America. Edited by W.G.E. Caldwell. Geological Society of
Canada Special Paper 13. pp. 31-54.
Okiongbo, K.S., Aplin, A.C., and Larter, S.R. 2005. Changes in Type II kerogen density
as a function of maturity: Evidence from the Kimmeridge Clay Formation.
Energy & Fuels 19(6): pp. 2495-2499. doi: 10.1021/ef050194.
Osadetz, K.G., Chen, Z., and Obermajer, M. 2010. Remaining conventional and
unconventional (shale oil) resource potential in the Colorado Group. Presented
at CSPG Luncheon, Calgary, Alberta, 10 September 2010.
Owen, J.E. 1952. The resistivity of a fluid-filled porous body. Journal of Petroleum
Technology 4(7). doi: 10.2118/952169-G.
Passey, Q.R., Bohacs, K.M., Esch, W.L., Klimentidis, R., and Sinha, S. 2010. From oil-
prone source rock to gas-producing shale reservoir - geologic and
petrophysical characterization of unconventional shale-gas reservoirs.
Presented at CPS/SPE International Oil & Gas Conference and Exhibition,
Beijing, China, 8-10 June 2010.
Passey, Q.R., Creaney, S., Kulla, J.B., Moretti, F.J., and Stroud, J.D. 1990. A Practical
Model for Organic Richness from Porosity and Resistivity Logs. AAPG
Bulletin 74(12): pp. 1777-1794.
Pedersen, T.F., and Calvert, S.E. 1990. Anoxia Vs Productivity - What Controls the
Formation of Organic-Carbon-Rich Sediments and Sedimentary-Rocks.
AAPG Bulletin 74(4): pp. 454-466.
Percy, E.L., and Pedersen, P.K. 2017. Depositional processes and environments of the
Belle Fourche Formation, southern Alberta, Canada. Presented at CSPG
GeoConvention, Calgary, Alberta, 15-19 May 2017.
Pérez-Rosales, C. 1982. On the relationship between formation resistivity factor and

porosity. Society of Petroleum Engineers Journal 22(4): pp. 531-536. doi:
10.2118/10546-PA.
Philip, G.M., and Watson, D.R. 1986. Matheronian geostatistics - Quo vadis?
Mathematical Geology 18(1): pp. 93-117. doi: 10.1007/BF00897657.
246
Philip, J.M., and Airaud-Crumiere, C. 1991. The demise of the rudist-bearing carbonate
platforms at the Cenomanian/Turonian boundary: a global control. Coral Reefs
10: pp. 115-125.
Pickell, J.J., and Heacock, J.G. 1960. Density Logging. Geophysics 25(4): pp. 891-094.
doi: 10.1190/1.1438769.
Plint, A.G. 1988. Sharp-based shoreface sequences and “offshore bars” in the Cardium
Formation of Alberta: their relationship to relative changes in sea level. In
Sea-Level Changes: An integrated approach. Edited by C.K. Wilgus and B.S.
Hasting, et al. Society of Economic Paleontologists and Mineralogists, Special
Publication 42. pp. 357-370.
Plint, A.G. 1991. High-frequency relative sea level oscillations in Upper Cretaceous shelf
clastics of the Alberta Foreland Basin: Evidence for a Milankovitch- scale
glacio-eustatic control? In Sedimentation, Tectonics, and Eustasy. Edited by
D.I.M. Macdonald. International Assocation of Sedimentologists, Special
Plint, A.G. 1996. Marine and non-marine systems tracts and fourth-order sequences in the
Early-Middle Cenomanian Dunvegan Alloformation, northeastern British
Columbia, Canada. In High resolution sequence stratigraphy: innovations and
applications. Edited by J.A. Howell and J.D. Aitken. Geological Society
Special Publication 104. pp. 159-191.
Plint, A.G. 2000. Sequence stratigraphy and paleogeography of a Cenomanian deltaic

complex: the Dunvegan and lower Kaskapau formations in subsurface and
outcrop, Alberta and British Columbia, Canada. Bulletin of Canadian
Petroleum Geology 48(1): pp. 43-79. doi: 10.2113/48.1.43.
Plint, A.G. 2010. Wave- and storm-dominated shoreline and shallow-marine systems. In
Facies Models. 4th ed. Edited by N.P. James and R.W. Dalrymple. Geological
Association of Canada, St. John’s, Newfoundland. pp. 167-200.
Plint, A.G. 2014. Mud dispersal across a Cretaceous prodelta: Storm-generated, wave-
enhanced sediment gravity flows inferred from mudstone microtexture and
microfacies. Sedimentology 61(3): pp. 609-647. doi: 10.1111/sed.12068.
Plint, A.G., Eyles, N., Eyles, C.H., and Walker, R.G. 1992. Controls of sea level change.
In Facies models: response to sea level change. Edited by R.G. Walker and
N.P. James. Geological Association of Canada, St. John’s, Newfoundland. pp.
15-25.
Plint, A.G., and Kreitner, M. 2005. High-frequency sequences reveal complex patterns of
plate flexure: Late Cenomanian of Western Canada Foreland Basin. Presented
at AAPG Annual Convention, 19-22 June 2005, Calgary, Alberta.
247
Plint, A.G., Macquaker, J.H.S., and Varban, B.L. 2012a. Bedload transport of mud across
a wide, storm-influenced ramp: Cenomanian-Turonian Kaskapau Formation,
western Canada foreland basin. Journal of Sedimentary Research 82: pp. 801-
822.
Plint, A.G., and Nummedal, D. 2000. The falling stage systems tract: Recognition and
importance in sequence stratigraphic analysis. In Sedimentary responses to
forced regressions. Edited by D. Hunt and R.L.G. Gawthorpe. Geological
Society of London Special Publication No. 172. pp. 1-17.
Plint, A.G., Tyagi, A., McCausland, P., Krawetz, J., Zhang, H., Roca, X., Varban, B.L.,
Hu, Y.G., Kreitner, M., and Hay, M.J. 2012b. Dynamic relationship between
subsidence, sedimentation, and unconformities in mid-Cretaceous, shallow-
marine strata of the Western Canada Foreland Basin: links to Cordilleran
tectonics. In Tectonics of Sedimentary Basins: Recent advances. Edited by C.
Busby and A. Azor. Blackwell Publishing Ltd.
Podruski, J.A., Barclay, J.E., Hamblin, A.P., Lee, P.J., Osadetz, K.G., Procter, R.M., and
Taylor, G.C. 1988. Conventional oil resources of Western Canada (light and
medium) Part I: Resource Endowment. Geological Survey of Canada Paper
87-26.
Posamentier, H.W. 2000. Seismic stratigraphy into the next millenium; a focus on 3D
seismic data. American Association of Petroleum Geologists Annual
Convention, New Orleans, Abstracts Volume 9: p. A118.
Posamentier, H.W., Allen, G.P., James, D.P., and Tesson, M. 1992. Forced regressions in
a sequence stratigraphic framework: concepts, examples, and exploration
significance. AAPG Bulletin 76: pp. 1687-1709.
Posamentier, H.W., and Allen, J.R.L. 1993. Siliciclastic sequence stratigraphic patterns in
foreland ramp-type basins. Geology 21: pp. 455-458.
Posamentier, H.W., Jervey, M.T., and Vail, P.R. 1988. Eustatic controls on clastic
deposition. I. Conceptual framework. . In Sea Level Changes - An Integrated
Approach. Edited by C.K. Wilgus, B.S. Hastings, C.G.St.C. Kendall, H.W.
Posamentier, C.A. Ross, J.C. Van Wagoner. SEPM Special Publication. pp.
125-154.
Posamentier, H.W., and Vail, P.R. 1988. Eustatic control on clastic deposition. II.
Sequence and system tracts models. In Sea Level Changes - An Integrated
Approach. Edited by C.K. Wilgus and B.S. Hastings, et al. SEPM Special
Publication. pp. 125-154.
Potter, P.E., Maynard, J.B., and Depetris, P.J. 2005. Mud and mudstones: introduction
and overview. Springer-Verlag. Berlin, Germany.
248
Poupon, A., Hoyle, W., and Schmidt, A.W. 1971. Log analysis in formations with
complex lithologies. Journal of Petroleum Technology 23(8): pp. 995-1005.
Prokoph, A., Babalola, L.O., El Bilali, H., Olagoke, S., and Rachold, V. 2013.
Cenomanian-Turonian carbon isotope stratigraphy of the Western Canada
Sedimentary Basin. Cretaceous Research 44: pp. 39-53. doi:
10.1016/j.cretres.2013.03.005.
Prokoph, A., Villeneuve, M.E., Agterberg, F.P., and Rachold, V. 2001. Geochronology
and calibration of global Milankovitch cyclicity at the Cenomanian-Turonian
boundary. Geology 29(6): pp. 523-526.
Pursell, D.A., Hunt, E.R., and Treviño, A. 1998. Integrating modern open hole logs, core
data and old electric logs: a study of the San Andres Field, Veracruz, Mexico.
Presented at International Petroleum Conference and Exhibition of Mexico, 3-
5 March 1998, Villahermosa, Mexico.
Quirein, J., Kimminau, S., LaVigne, J., Singer, J.M., and Wendel, F. 1986. A coherent
framework for developing and applying multiple formation evaluation models.
Presented at SPWLA 27th Annual Logging Symposium, 9-13 June, Houston,
Texas. SPWLA.
Quirein, J., Witkowsky, J., Truax, J.A., Galford, J., Spain, D., and Odumosu, T. 2010.
Integrating core data and wireline geochemical data for formation evaluation
and characterization of shale gas reservoirs. Presented at SPE Annual
Technical Conference and Exhibition, Florence, Italy, 19-22 September 2010.
Rabaute, A., Revil, A., and Brosse, E. 2003. In situ mineralogy and permeability logs
from downhole measurements: Application to a case study in chlorite-coated
sandstones. Journal of Geophysical Research-Solid Earth 108(B9): pp. n/a-n/a.
doi: 10.1029/2002jb002178.
Ramirez, T.R., Klein, J.D., Bonnie, R., and Howard, J.J. 2011. Comparative study of
formation evaluation methods for unconventional gas shale reservoirs:
Application to the Haynesville Shale (Texas), SPE144062. Presented at North
American Unconventional Gas Conference and Exhibition, The Woodlands,
Texas, 14-16 June 2011.
Raymer, L.L., Hunt, E.R., and Gardner, J.S. 1980. An improved sonic transit time-to-
porosity transform. Presented at SPWLA 21st Annual Logging Symposium,
Lafayette, Louisiana, 8-11 July 1980. pp. 1-12.
Read, J.F., and Goldhammer, R.K. 1988. Use of Fischer plots to define third-order sea-
level curves in Ordovician peritidal carbonates, Appalachians. Geology 16: pp.
895-899.
Rickman, R., Mullen, M., Petre, E., Grieser, B., and Kundert, D. 2008. A practical use of
shale petrophysics for stimulation design optimization: All shale plays are not
249
clones of the Barnett Shale. Presented at SPE Annual Technical Conference

and Exhibition, Denver, Colorado, 21-24 September 2008. SPE.
Rider, M.H. 1990. Gamma-ray log shape used as a facies indicator: critical analysis of an
oversimplified methodology. In Geological Applications of Wireline Logs.
Edited by A. Hurst and M.A. Lovell, et al. pp. 27-37.
Ridgley, J.L., and Gilboy, C.F. 2001. Lithofacies architecture of the Upper Cretaceous
Belle Fourche Formation, Saskatchewan, Alberta, and Montana - its
relationship to sites of shallow biogenic gas production. Saskatchewan
Geological Survey, Sask. Energy Mines.
Ridgley, J.L., McNeil, D.H., Gilboy, C.F., Condon, S.M., and Obradovich, J.D. 2001.
Structural and stratigraphic controls on sites of shallow biogenic gas
accumulations in the Upper Cretaceous Belle Fourche and Second White
Specks-Greenhorn Formations of Southern Alberta, Saskatchewan and
Northern Montana. In Gas in the Rockies. Edited by D.S. Anderson, Robinson,
J.W., Estes-Jackson, J.E., Coalson, E.B. The Rocky Mountain Association of
Geologists. pp. 241-270.
Rine, J.M., and Ginsburg, R.N. 1985. Depositional facies of a mud shoreface in
Suriname, South America - a mud analogue to sandy, shallow-marine
deposits. Journal of Sedimentary Petrology 55(5): pp. 633-652.
Roberts, L.N.R., Higley, D.K., and Henry, M.E. 2005. Input data used to generate one-
dimensional burial history models, central Alberta, Canada, Open File Report
2005-1412. USGS.
Roca, X. 2007. Tectonic and eustatic controls on the allostratigraphy and depositional
environments of the Lower Colorado Group (Upper Albian) central Foothills
and adjacent plains of Alberta, Western Canada Foreland Basin. PhD thesis,
Department of Earth Sciences, The University of Western Ontario, London,
Ontario, Canada. 323 p.
Roca, X., Rylaarsdam, J.R., Zhang, H., Varban, B.L., Sisulak, C.F., Bastedo, K., and
Plint, A.G. 2008. An allostratigraphic correlation of Lower Colorado Group
(Albian) and equivalent strata in Alberta and British Columbia, and
Cenomanian rocks of the Upper Colorado Group in southern Alberta. Bulletin
of Canadian Petroleum Geology 56(4): pp. 259-299.
Roentgen, W. 1896. On a new kind of rays. Nature 53: pp. 274-277.
Rokosh, C.D., Pawlowicz, J.G., Berhane, H., Anderson, S.D.A., and Beaton, A.P. 2009.
ERCB/AGS Open File Report 2008-09: Geochemical and sedimentological
investigation of the Colorado Group for shale gas potential: initial results.
Energy Resources Conservation Board and Alberta Geological Survey.
250
Rosen, O.M., Abbyasov, A.A., and Tipper, J.C. 2004. MINLITH - an experience-based
algorithm for estimating the likely mineralogical compositions of sedimentary
rocks from bulk chemical analyses. Computers and Geosciences 30(6): pp.
647-661.
Rybacki, E., Meier, T., and Dresen, G. 2016. What controls the mechanical properties of
shale rocks? – Part II: Brittleness. Journal of Petroleum Science and
Engineering 144: pp. 39-58. doi: 10.1016/j.petrol.2016.02.022.
Sageman, B.B., Rich, J., Arthur, M.A., Birchfield, G.E., and Dean, W.E. 1997. Evidence
for Milankovitch periodicities in Cenomanian-Turonian lithologic and
geochemical cycles, Western Interior U.S.A. Journal of Sedimentary Research
67(2): pp. 286-302.
Sanchez-Ramirez, J.A., Torres-Verdín, C., Wang, G.L., Mendoza, A., Wolf, D., Liu, Z.,
and Schell, G. 2009. Field examples of the combined petrophysical inversion
of gamma-ray, density, and resistivity logs acquired in thinly-bedded clastic
rock formaitons. Presented at SPWLA 50th Annual Logging Symposium 21-
24 June 2009, The Woodlands, Texas.
Schieber, J. 2010. Common themes in the formation and preservation of intrinsic porosity
in shales and mudstones = illustrated with examples across the Phanerozoic.
SPE Journal. doi: 10.2118/132370-MS.
Schieber, J. 2016. Mud re-distribution in epicontinental basins - Exploring likely

processes. Marine and Petroleum Geology 71: pp. 119-133. doi:
10.1016/j.marpetgeo.2015.12.014.
Schieber, J., Southard, J.B., and Thaisen, K. 2007. Accretion of mudstone beds from
migrating floccule ripples. Science 318(5857): pp. 1760-1763. doi:
10.1126/science.1147001.
Schlager, W. 1993. Accommodation and supply - a dual control on stratigraphic

sequences. Sedimentary Geology 86: pp. 111-136. doi:
10,1130/MEM390P.91.
Schlanger, S.O., and Jenkyns, H.C. 1976. Cretaceous ocean anoxic events: Causes and
consequences. Geologie en Mijnbouw 55(3-4): pp. 179-184.
Schlumberger. 2009. Log Interpretation Charts.
Schmoker, J.W. 1979. Determination of organic content of Appalachian Devonian shales

from formation-density logs. AAPG Bulletin 63(9): pp. 1504-1509.
Schmoker, J.W. 1981. Determination of organic-matter content of Appalachian Devonian

shales from gamma-ray logs. AAPG Bulletin 65(7): pp. 1285-1298.
251
Schmoker, J.W., and Hester, T.C. 1983. Organic carbon in Bakken Formation, United
States portion of Williston Basin.
Schröder-Adams, C.J. 2014. The Cretaceous Polar and Western Interior Seas:
paleoenvironmental history and paleoceanographic linkages. Sedimentary
Geology 301: pp. 26-40.
Schröder-Adams, C.J., Cumbaa, S.L., Bloch, J.D., Leckie, D.A., Craig, J., El-Dein,
S.A.S., Simons, D.-J.H.A.E., and Kenig, F. 2001. Late Cretaceous
(Cenomanian to Campanian) paleoenvironmental history of the Easter
Canadian margin of the Western Interior Seaway: bonebeds and anoxic events.
Palaeogeography, Palaeoclimatology, Palaeoecology 170: pp. 261-289.
Schröder-Adams, C.J., Herrle, J.O., and Tu, Q. 2012. Albian to Santonian carbon isotope
excursions and faunal extinctions in the Canadian Western Interior Sea:
Recognition of eustatic sea-level controls on a forebulge setting. Sedimentary
Geology 281: pp. 50-58. doi: 10.1016/j.sedgeo.2012.08.004.
Schröder-Adams, C.J., Leckie, D.A., Bloch, J.D., Craig, J., McIntyre, D.J., and Adams,
P.J. 1996. Paleoenvironmental changes in the Cretaceous (Albian to Turonian)
Colorado Group of western Canada: microfossil, sedimentological and
geochemical evidence. Cretaceous Research 17: pp. 311-365.
Schröder-Adams, C.J., Leckie, D.A., Craig, J., and Bloch, J.D. 1999. Upper Cretaceous
Colorado group in the Pasquia Hills, Northeastern Saskatchewan: a
multidisciplinary study in progress. Saskatchewan Geological Survey,
Summary of Investigations 1: pp. 52-56.
Schumacher, B.A. 2002. Methods for the determination of total organic carbon (TOC) in
soils and sediments. Available from
https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryId=63371
[accessed April 8, 2018].
Sen, G. 2014. Petrology: Principles and practice. Springer-Verlag, Berlin. 368 p.
Sengstock, C., Gertz, M., and Van Canh, T. 2012. Spatial interestingness measures for co-
location pattern mining. Presented at 2012 IEEE 12th International Conference
on Data Mining Workshops, Brussels, Belgium, 10 December 2012. IEEE.
Shank, J.A., and Plint, A.G. 2013. Allostratigraphy of the Upper Cretaceous Cardium
Formation in subsurface and outcrop in southern Alberta, and correlation to
equivalent strata in northwestern Montana. Bulletin of Canadian Petroleum
Geology 61(1): pp. 1-40. doi: 10.2113/gscpgbull.61.1.1.
Shanley, K.W., and McCabe, P.J. 1994. Perspectives on the sequence stratigraphy of
continental strata. AAPG Bulletin 78(544-568).
252
Shanmugam, G. 2017. Global case studies of soft-sediment deformation structures

(SSDS): Definitions, classifications, advances, origins, and problems. Journal
of Palaeogeography 6(4): pp. 251-320.
Shaw, A.B. 1964. Time in Stratigraphy. McGraw-Hill. 365 p.
Shipboard Scientific Party. 1998. Site 1085. In Proceedings of the Ocean Drilling
Program, Initial Reports, 134. Edited by G. Wefer and W.H. Berger, et al.,
College Station, Texas. pp. 385-428.
Sibson, R. 1980. A vector identity for the Dirichlet tessellation. Mathematical

Proceedings of the Cambridge Philosophical Society 87: pp. 151-155.
Simandoux, P. 1963. Dielectric measurements on porous media application to the

measurement of water saturations: study of the behaviour of argillaceous
formations. Revue de l’Institut Francais du Petrole 18: pp. 193-215.
Simons, D.J.H., Kenig, F., and Schroder-Adams, C.J. 2003. An organic geochemical
study of Cenomanian-Turonian sediments from the Western Interior Seaway,
Canada. Organic Geochemistry 34(8): pp. 1177-1198. doi: 10.1016/S0146-
6380(03)00064-0.
Singh, C. 1983. Cenomanian microfloras of the Peace River area, northwestern Alberta.
Alberta Research Council, Bulletin 44: p. 322.
Skuce, A.G., Goody, N.P., and Maloney, J. 1992. Passive-roof duplexes under the Rocky
Mountain Foreland Basin, Alberta. AAPG Bulletin 76(1): pp. 67-80.
Slaughter, M., and Earley, J.W. 1965. Mineralogy and geological significance of the
Mowry bentonites, Wyoming. Geological Society of America Special Paper
83: pp. 1-116.
Sloss, L.L., Dapples, E.C., and Krumbein, W.C. 1960. Lithofacies maps: an atlas of the
United States and Canada. Wiley. New York, New York. 108 p.
Sloss, L.L., Krumbein, W.C., and Dapples, E.C. 1949. Integrated facies analysis. In
Sedimentary facies in geologic history. Edited by C.R. Longwell. Geological
Society of America Memoir 39. pp. 91-124.
Snow, L.R., Duncan, R.A., and Bralower, T.J. 2005. Trace element abundances in the
Rock Canyon Anticline, Pueblo, Colorado, marine sedimentary section and
their relationship to Caribbean plateau construction and oxygen anoxic event
2. Paleoceanography and Paleoclimatology 20(3): pp. 1-14. doi:
10.1029/2004PA001093.
Sondergeld, C.H., Newsham, K.E., Comisky, J.T., Rice, M.C., and Rai, C.S. 2010.
Petrophysical considerations in evaluating and producing shale gas resources.
253
Presented at SPE Unconventional Gas Conference, Pittsburgh, Pennsylvania,

23-25 February 2010.
Spencer, R., and Weedmark, T. 2015. Application of x-ray fluorescence (XRF) analyses
to the characterization of tight reservoirs. Presented at Gas Mexico Congress,
Villahermosa, Mexico, 29-30 September 2015.
Stankiewicz, A., Ionkina, N., Motherwell, B., Bennett, B., Wint, O., and Mastalerz, M.
2015. Kerogen density revisited - lessons from the Duvernay Shale. Presented
at Unconventional Resources Technology Conference, 20-22 July 2015, San
Antonio, Texas. doi: 10.15530/URTEC-2015-2157904.
Stelck, C.R. 1962. Upper Cretaceous, Peace River area, British Columbia; Edmonton
Geological Society Fourth Annual Field Trip Guidebook.
Stott, D.F. 1963. The Cretaceous Alberta Group and equivalent rocks, Rocky Mountain
Foothills, Alberta. Geological Survey of Canada Memoir 317. Department of
Mines and Technical Surveys, Geological Survey of Canada.
Stott, D.F. 1967. The Cretaceous Smoky Group, Rocky Mountain Foothills, Alberta.
Geological Survey of Canada Bulletin 132.
Stott, D.F. 1982. Lower Cretaceous Fort St. John Group and Upper Cretaceous Dunvegan
Formation of the Foothills and Plains of Alberta, British Columbia, District
Mackenzie and Yukon Territory. Geological Survey of Canada Bulletin 328.
Stott, D.F. 1984. Cretaceous sequences of the Foothills of the Canadian Rocky
Mountains. In The Mesozoic of Middle North America. Edited by D.F. Stott
and D.J. Glass. Canadian Society of Petroleum Geology. pp. 85-107.
Struyk, C., Bishop, R., Fortune, D., Foster, E., Gordon, D., d’Haene, T., Joyce, D.,
Kenny, S., Kowalchuk, H., and Stadnyk, M. 1989. LAS: A floppy disk
standard for log data. The Log Analyst 30(5): pp. 395-396.
Sweeney, J.J., and Burnham, A.K. 1990. Evaluation of a simple model of vitrinite
reflectance based on chemical kinetics. AAPG Bulletin 74: pp. 1559-1570.
Swift, D.J.P., Han, G., and Vincent, C.E. 1986. Fluid processes and sea-floor responses
on a modern storm-dominated shelf: Middle Atlantic shelf of North America.
Part I: the storm-current regime. In Shelf sands and sandstones. Edited by R.J.
Knight and J.R. McLean. Canadian Society of Petroleum Geologists, Memoir
11, Calgary, Alberta. pp. 99-119.
Swift, D.J.P., and Thorne, J.A. 1991. Sedimentation on Continental Margins, I: A general
model for shelf sedimentation. In Shelf Sand and Sandstone Bodies:
Geometry, Facies, and Sequence Stratigraphy. Edited by D.J.P. Swift and G.F.
Oertel, et al. The International Association of Sedimentologists Special
254
Tarantola, A., and Valette, B. 1982. Generalized nonlinear inverse problems solved using
the least squares criterion. Reviews of Geophysics and Space Physics 20(2):
pp. 219-232.
Tearpock, D.J., and Bischke, R.E. 2002. Applied subsurface geological mapping with
structural methods. Prentice-Hall. 648 p.
Ter Heege, J., Zijp, M., Nelskamp, S., Douma, L., Verreussel, R., Ten Veen, J., de Bruin,
G., and Peters, R. 2015. Sweet spot identification in underexplored shales
using multidisciplinary reservoir characterization and key performance
indicators: Example of the Posidonia Shale Formation in the Netherlands.
Journal of Natural Gas Science and Engineering. doi: 10.1016/j.jngse.08.032.
Threadgold, P. 1972. Some problems and uncertainties in log interpretation. The Log
Analyst 13(2): pp. 3-11.
Tiab, D., and Donaldson, E.C. 2012. Petrophysics: Theory and practice of measuring
reservoir rock and fluid transport properties. Third ed. Elsevier. Waltham,
Massachusetts.
Tissot, B.P., and Welte, D.H. 1978. Petroleum Formation and Occurence. Springer-
Verlag. Berlin, Germany. 538 p.
Towle, G. 1962. An analysis of the formation resistivity factor-porosity relationship of

some assumed pore geometries. Presented at SPWLA 3rd Annual Logging
Symposium, 17-18 May 1962, Houston, TX.
Trabucho-Alexandre, J. 2015. More gaps than shale: erosion of mud and its effect on
preserved geochemical and palaeobiological signals. In Strata and Time:
Probing the Gaps in Our Understanding. Edited by D.G. Smith and R.J.
Bailey, et al. Geological Society, London, Special Publications 404, London,
UK. pp. 251-270.
Tyagi, A. 2009. Sedimentology and high-resolution stratigraphy of the Upper Cretaceous

(Late Albian to Middle Turonian) Blackstone Formation, Western Interior
Basin, Alberta, Canada. Ph.D. thesis, Department of Earth Sciences, The
University of Western Ontario, London, Ontario.
Tyagi, A., Plint, A.G., and McNeil, D.H. 2007. Correlation of physical surfaces,
bentonites, and biozones in the Cretaceous Colorado Group from the Alberta
Foothills to southwest Saskatchewan, and a revision of the Belle Fourche -
Second White Specks formational boundary. Canadian Journal of Earth
Sciences 44: pp. 871-888.
Vail, P.R. 1987. Seismic stratigraphy interpretation using sequence stratigraphy, part 1:
seismic stratigraphy interpretation procedure. In Atlas of Seismic Stratigraphy:
AAPG Studies in Geology 27. Edited by A.W. Bally. pp. 1-10.
255
Vail, P.R., Todd, R.G., and Sangree, J.B. 1977. Seismic stratigraphy and global changes
of sea level, Part 5: Chronostratigraphic significance of seismic reflections. In
Seismic stratigraphy - applications to hydrocarbon exploration. Edited by C.E.
Payton. American Association of Petroleum Geologists. pp. 99-116.
van Loon, A.J. 2009. Soft-sediment deformation structures in siliciclastic sediments: an

overview. Geologos 15(1): pp. 3-55.
Van Wagoner, J.C. 1985. Reservoir facies distribution as controlled by sea-level change.
Presented at Society of Economic Paleontologists and Mineralogists Mid-Year
Meeting,. pp. 91-92.
Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M.J., Sarg, J.F., Loutit, T.S., and
Hardenbol, J. 1988. An overview of sequence stratigraphy and key definitions.
In Sea level change - an integrated approach: Society of Economic
Paleontologists and Mineralogists, Special Publication 42. Edited by C.K.
Wilgus and B.S. Hastings, et al. pp. 39-45.
Varban, B.L. 2004. Sedimentology and stratigraphy of the Cenomanian-Turonian

Kaskapau Formation, northeast British Columbia and northwest Alberta. PhD
thesis, Department of Earth Sciences, The University of Western Ontario, 452
p.
Varban, B.L., and Plint, A.G. 2005. Allostratigraphy of the Kaskapau Formation
(Cenomanian-Turonian) in the subsurface and outcrop: NE British Columbia
and NW Alberta, Western Canada Foreland Basin. Bulletin of Canadian
Petroleum Geology 53(4): pp. 357-389. doi: 10.2113/53.4.357.
Varban, B.L., and Plint, A.G. 2008a. Palaeoenvironments, palaeogeography, and

physiography of a large, shallow, muddy ramp: Late Cenomanian-Turonian
Kaskapau Formation, Western Canada foreland basin. Sedimentology 55(1):
pp. 201-233. doi: 10.1111/j.1365-3091.2007.00902.x.
Varban, B.L., and Plint, A.G. 2008b. Sequence stacking patterns in the Western Canada
foredeep: influence of tectonics, sediment loading and eustasy on deposition
of the Upper Cretaceous Kaskapau and Cardium Formations. Sedimentology
55(2): pp. 395-421. doi: 10.1111/j.1365-3091.2007.00906.x.
Varhaug, M. 2016. Basic well log interpretation. Oilfield Review: E&P Defining Series.
Verma, S.P., Torres-Alvarado, I.S., and Velasco-Tapia, F. 2003. A revised CIPW norm.
Schweizerische Mineralogische und Petrographische Mitteilungen 83: pp.
197-216.
Ver Straeten, C.A. 2004. K-bentonites, volcanic ash preservation, and implications for
Early to Late Devonian volcanism in the Acadian orogen, eastern North
America. Geological Society of America Bulletin 116(3-4): pp. 474-489.
256
Wall, J.H., and Rosene, K.K. 1977. Upper Cretaceous stratigraphy and micropaleontology
of the Crowsnest Pass-Waterton area, southern Alberta Foothills. Bulletin of
Canadian Petroleum Geology 25: pp. 842-867.
Wang, F.P., and Gale, J.F.W. 2009. Screening criteria for shale-gas systems. Gulf Coast
Association of Geological Societies Transactions 59: pp. 779-793.
Ward, J. 2010. Kerogen density in the Marcellus Shale. SPE 131767. Presented at SPE
Unconventional Gas Conference, Pittsburgh, Pennsylvania, 23-25 February
2010.
Warrilow, I.M., Ibrahim, R., Chin, H.T., and Forsyth, D. 1995. Application of Resistivity
Inversion Software To Thin-Bed Evaluation. Presented at SPE Asia Pacific
Oil and Gas Conference, Kuala Lumpur, Malaysia, 20-22 March 1995. doi:
10.2118/29263-ms.
Waxman, M.H., and Smits, L.J.M. 1968. Electrical Conductivities in Oil-Bearing Shaly
Sands. Society of Petroleum Engineers Journal 8(2): pp. 107-122. doi:
10.2118/1863-A.
Wheeler, H.E. 1964. Base level, lithosphere surface, and time-stratigraphy. Geological
Society American Bulletin 75(7): pp. 599-610.
Wilson, P.A., and Norris, R.D. 2001. Warm tropical ocean surface and global anoxia
during the mid-Cretaceous period. Nature 412(6845): pp. 425-429. doi:
10.1038/35086553.
Wilson, R.D., Stromswold, D.C., Evans, M.L., Jain, M., and Close, D.A. 1979. Spectral
Gamma-Ray Logging III : Formation And Thin Bed Effects. Presented at
SPWLA 20th Annual Logging Symposium, 3-6 June, 1979. Society of
Petrophysicists and Well-Log Analysts, Tulsa, Oklahoma.
Winsauer, W.O., Shearin, H.M., Masson, P.H., and Williams, M. 1952. Resistivity of
Brine-Saturated Sands in Relation to Pore Geometry. AAPG Bulletin 36(2):
pp. 253-277.
Woodhouse, R. 1978. The laterolog Groningen Phantom can cost you money. Presented
at SPWLA 19th Annual Logging Symposium, El Paso, Texas, 13-16 June
1978, El Paso, Texas.
Worthington, P.F. 1985. The evolution of shaly-sand concepts in reservoir evaluation.

The Log Analyst 26(1).
Wyllie, M.R.J., Gregory, A.R., and Gardner, L.W. 1956. Elastic wave velocities in
heterogeneous and porous media. Geophysics 21: pp. 41-70.
Yamada, Y. 2010. X-ray fluorescence analysis by fusion bead method for ores and rocks.
The Rigaku Journal 26(2): pp. 15-23.
257
Yang, Y., Hiroki, S., Hows, A., and Zoback, M.D. 2013. Comparison of brittleness
indices in organic-rich shale formations. Presented at 47th US Rock
Mechanics / Geomechanics Symposium, San Francisco, California, 23-26
June 2013. American Rock Mechanics Association.
Yang, Y.T., and Miall, A.D. 2008. Marine transgressions in the mid-Cretaceous of the
Cordilleran foreland basin re-interpreted as orogenic unloading deposits.
Bulletin of Canadian Petroleum Geology 56(3): pp. 179-198. doi:
10.2113/gscpgbull.56.3.179.
Yang, Y.T., and Miall, A.D. 2009. Evolution of the northern Cordilleran foreland basin
during the middle Cretaceous. Geological Society of America Bulletin 121(3-
4): pp. 483-501. doi: 10.1130/B26374.1.
Yang, Y.T., and Miall, A.D. 2010. Migration and stratigraphic fill of an underfilled
foreland basin: Middle-Late Cenomanian Belle Fourche Formation in southern
Alberta, Canada. Sedimentary Geology 227(1-4): pp. 51-64. doi:
10.1016/j.sedgeo.2010.03.005.
Yin, H. 2011. Application of resistivity tool response modeling for formation evaluation;
AAPG Archie Series, No. 2. AAPG and ExxonMobil Upstream Research
Company. Tulsa, OK.
Zajac, N.A. 2016. Depositional processes and facies variability in organic-rich mudstones
of the Colorado Group, west-central Alberta, Canada. M.Sc. thesis,
Department of Geosciences, University of Calgary, Calgary, Alberta.
Zee Ma, Y., Moore, W.R., Gomez, E., Luneau, B., Kaufman, P., Gurpinar, O., and
Handwerger, D.A. 2016. Wireline log signatures of organic matter and
lithofacies classifications for shale and tight carbonate reservoirs. In
Unconventional Oil and Gas Resources Handbook: Evaluation and
Development. Edited by Y. Zee Ma and S.A. Holditch. Gulf Professional
Publishing, Oxford, UK. pp. 151-172.
258

Integrating Petrophysics and Allostratigraphy To Find Sweet Spots

Uploaded by

Copyright:

Available Formats

Integrating Petrophysics and Allostratigraphy To Find Sweet Spots

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

Integrating Petrophysics and Allostratigraphy To Find Sweet Spots

Uploaded by

Copyright:

Available Formats

Western University

Electronic Thesis and Dissertation Repository

Integrating Petrophysics and Allostratigraphy to Find Sweet Spots

Graduate Program in Geology

Follow this and additional works at: https://ir.lib.uwo.ca/etd

Part of the Geology Commons, and the Stratigraphy Commons

The Upper Cretaceous Second White Specks Formation – an organic-rich, calcareous

Keywords: allostratigraphy, brittleness, Cenomanian-Turonian boundary, formation

“It's a dangerous business, Frodo, going out of your door,” he

– J.R.R. Tolkien, The Fellowship of the Ring (1954)

I am further indebted to my thesis examination committee – Dr. Mahi Singh, Dr.

To my brother, Riley: always stay lit.

To my friends: I owe you a drink or five. May we meet again.

3.4.2 Drilling mud ............................................................................................. 41

5.6.1.2 Density log corrections ...................................................................... 95

6.5.1.4 Facies 4 and 5 .................................................................................. 147

10 Discussion .............................................................................................................. 201

Fig. 2.2: Schematic of a strike-oriented cross-section through a foreland basin. Modified

Fig. 3.1: Schematic of a centered caliper-type tool attached to an instrument assembly,

Fig. 3.4: Schematic of a bulk density logging tool in a conventional reservoir. 61

Fig. 5.8: Histograms of uncorrected (above) and kerogen-corrected (below) modelled

versus modelled kerogen for wells 100/02-14-038-05W5, 100/07-19-045-06W5/00,

Fig. 6.4: Cartoons summarizing the paleogeography and paleobathymetry of the

Fig. 6.7: Allostratigraphic cross-section B to B’, which runs north-south. 153

Fig. 6.8: Allostratigraphic cross-section C to C’, which runs north-south. 154

Fig. 6.9: Allostratigraphic cross-section D to D’, which runs north-south. 155

Fig. 6.10: Allostratigraphic cross-section E to E’, which runs east-west. 156

Fig. 6.11: Allostratigraphic cross-section F to F’, which runs east-west. 157

Fig. 6.12: Allostratigraphic cross-section G to G’, which runs east-west. 158

Fig. 6.13: Allostratigraphic cross-section H to H’, which runs east-west. 159

Fig. 7.2: 3D structure map of the top of allomember VII. 164

Fig. 7.3: Isochore map of allomember BF1. 167

Fig. 7.4: Isochore map of allomember BF2. 168

Fig. 7.5: Isochore map of allomember BF3. 169

Fig. 7.6: Isochore map of allomember VII. 170

Fig. 8.2: TOC of allomember BF3. 184

Fig. 8.3: TOC of allomember VII. 185

Fig. 8.4: Total porosity of allomember BF3. 186

Fig. 8.5: Total porosity of allomember VII. 187

Fig. 8.6: Effective porosity of allomember BF3. 189

Fig. 8.7: Effective porosity of allomember VII. 190

Fig. 8.8: Porosity thickness for allomember BF3. 192

Fig. 8.9: Porosity thickness for allomember VII. 193

Fig. 8.10: Brittleness of allomember BF3. 194

Fig. 8.11: Brittleness of allomember VII. 195

Fig. 9.1: Sweet spot map for allomember BF3. 198

Fig. 9.2: Sweet spot map for allomember VII. 199

Variables and Abbreviations

Bbl (barrels of oil equivalent)

Vqtz = quartz volume (v/v)

1.2.1 Influence of brittle behaviour

(Letourneau et al. 1996; Clarkson & Pedersen 2011).

Previous studies have not attempted to continuously model variations in petrophysical

1.2.2 Influence of porosity

To date, petrophysically-derived reservoir quality indicators have not been correlated to

1.2.3 Stratal architecture of the Second White Specks

Aims & Objectives

1.3.1 Research aim

By answering these three questions, recommendations will be made for future