Groundwater Hydrology: Third Edition

Groundwater Hydrology
Third Edition
David Keith Todd

University of California, Berkeley
and
Todd Engineers
Larry W. Mays
Arizona State University
~
WILEY
John Wiley & Sons, Inc.
Executive Editor Bill Zobrist
Project Editor Jennifer Welter p
Senior Production Editor Valerie A. Vargas
Marketing Manager Jennifer Powers
Senior Designer Dawn L. Stanley
New Media Editor Thomas Kulesa
Cover Design David Levy
Cover Image Photograph by Robert Morris, U.S. Geological Survey
Production Management Services Argosy Publishing
This book was set in 10/12 Times by Argosy Publishing and printed and bound by Hamilton Printing.
The cover was printed by Phoenix Color Corporation.
On the Cover: Comal Springs. The highly productive Edwards aquifer, the first aquifer to be desig-
nated as a sole source aquifer under the Safe Drinking Water Act, is the source of water for more than
1 million people in San Antonio, Texas, some military bases and small towns, and for south-central
Texas farmers and ranchers. The aquifer also supplies water to sustain threatened and endangered
species habitat associated with natural springs in the region and supplies surface water to users down-
stream from the major springs. These various uses are in direct competition with ground-water devel-
opment and have created challenging issues of ground-water management in the region. In
Photo and description from the USGS website:
http://water. usgs.gov/pubs/circ/circ 1186/html/gw_effect.html
This book is printed on acid-free paper. §
Copyright 2005© John Wiley & Sons, Inc. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Sec-
tions 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Pub-
lisher or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470. Requests to the Publisher for per-
mission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken,
NJ 07030, (201)748-6011, fax (201)748-6008, E-mail: [email protected].
To order books or for customer service please call1(800)-CALL WILEY (225-5945).
ISBN 0-471-05937-4 (cloth)

ISBN 0-471-45254-8 (WIE)
Printed in the United States of America

10987654321
PREFACE to the 3rd Editio1t
Water is essential for life on Earth, and the largest source of fresh water lies under the Earth's surface.
Increased demands for water have stimulated efforts to further develop, <and in many cases over develop,
our groundwater resources. In addition we have created pollution problems resulting from over develop-
ment of groundwater resources and the mismanagement of wastes. As a result investigations of the occur-
rence and movement of groundwater have advanced, drilling equipment has improved, new models have
been developed, techniques for managing groundwater have advanced, and research has extended our
understanding of the resource.
Intended Audience and Introduction

This book is intended for a groundwater hydrology course in civil and environmental engineering, geol-
ogy, or hydrogeology. It is appropriate for either a one or two term course at the junior, senior, or gradu-
ate level.
The first edition of this book by David K. Todd was published in 1959 and the second edition was
published in 1980. This third edition of Groundwater Hydrology has the same purpose as the previous
two editions, to present the fundamentals of groundwater hydrology in a manner understandable to those
most concerned with such knowledge. Few people specialize in the subject, yet, because groundwater is
a major natural resource, the subject becomes important for students and professionals in many fields:
water supply, the environment, agriculture, economics, mining, and the law, to name only the obvious
ones. Although it is impossible to present a subject fitted to such a diversity of students, the common need
is an understanding of the fundamental principles, methods, and problems in the groundwater field.
Thus, this book endeavors to make available a unified presentation of groundwater hydrology.
Since the earlier editions of this book, the groundwater resource field has made tremendous strides.
Awareness of the environment, concerns and competition for water supplies, contamination of ground-
water, and enhanced regulation of water resources have all focused attention on the subject. As a result
educational programs, research funding, and professionals actively involved in developing and managing
groundwater have grown dramatically. The National Ground Water Association together with journals
such as Ground Water and Water Resources Research has been instrumental in advancing groundwater
knowledge.
New to this edition

We have attempted to develop this book, not only as a stand-alone work, but in partnership with the Inter-
net to serve as a portal to the vast resources on groundwater hydrology that now exist on the Internet.
Hopefully this book will help guide the student, the professional, and the researcher to the general doc-
uments and other publications, program plans, field project details, software, and other information
found on the Internet.
v
vi Preface
Internet References
Most of the web sites in the book are maintained by government agencies and nongovernmental organi-
zations so that they will they continue in the future. Several end-of-chapter problems throughout the book
are based upon Internet exercises.
• Example Problems, Case Studies, Problem Sets The most significant change in this edition has
been the addition of example problems and case studies in the text, and the addition of problem
sets at the end of each chapter. For educational purposes these problems enable a student to apply
material in the text to realistic everyday situations.
• Modeling with MODFLOW The second most noticeable change has been an entirely new chapter
on modeling (Chapter 9) featuring the U. S. Geological Survey MODFLOW model, which has
become the standard modeling tool in the groundwater field. Chief among the new developments in
the groundwater field is the role of computers, not only for organizing data and solving problems but
also in managing groundwater resources on a basin-wide basis for known or anticipated inputs and
outputs.
• New Figures and Photos We have included over 300 new figures and photos. Many of the new
figures and tables within the book have been taken from various Web pages, in particular the U.
S. Geological Survey. These items not only provide specific information but serve as samples of
graphics and tabulations available on the Internet making the style of this book more variable than
that of most textbooks.
• References References have been updated and selected on the basis of significance and avail-
ability.
• Metric and U.S. Units Both metric and U.S. customary units have been used in the example
problems. Metric units are used in keeping with their growing acceptance in the United States as
well as in recognition of the wide use of this book throughout the World. U.S. customary units
have also been used because of the continued use by many in the United States.
Student and Instructor Resources

The website for the book, located at www. wiley.com/college/todd, includes resources for both students
and faculty: corrections to the book, updates on changes in web addresses and new web pages used in the
book, and other updates on material in the book that are of value to students and faculty. These resources
are also included on the author's website at www.public.asu.edu/-lwmays/. Larry Mays can be contacted
at [email protected]
For instructors who have adopted the book for their course, a complete Solutions Manual for all
homework problems in the text is available for download. Selected figures from the text, in PowerPoint
format, are also available for easy creation of lecture slides. Visit the Instructor Companion Site portion
of the book's website to register for a password for these assets available only to instructors.
For instructors who have adopted the book for their course, a complete Solutions Manual for all
homework problems in the text is available for download. Selected figures from the text, in PowerPoint
format, are also available for easy creation of lecture slides. Visit the Instructor Companion Site portion
of the book's website to register for a password for these assets available only to instructors.
Acknowledgements
We are indebted to personnel of the U.S. Geological Survey for their numerous excellent publica-
tions on so many aspects of groundwater from which we have borrowed freely. We are deeply indebted
to Sukru Ozger who prepared many of the problems and their solutions for the solutions manual. Students
at Arizona State University have been exposed to advanced drafts of this new edition and offered several
Preface vii
constructive comments. We want to thank the reviewers Scott Wolcott of Rochester Institute of Technol-
ogy, Rameshwar Singh of San Jose State University, Rao S. Govindaraju of Purdue University, Albert J.
;ani- Valocchi of the University of Illinois at Urbana-Champaign, Jeffrey D. Caulfield of the University of Mis-
took souri at Rolla, and Mark Widdowson of Virginia Tech. for their helpful comments and suggestions.
David W. Abbott, Maureen Reilly, and Dan Rothman, P.E. provided useful reviews of selected chapters.
has
~I em David Keith Todd Larry W. Mays
lply Piedmont, California Scottsdale, Arizona
pter
has
:sin
but
and
1ew
~ U.
s of
han
ail-
.pie
; as
1its
nts
the
;es
ted
all
int
on
all
int
on
:a-
ed
tts
·al
Contents
Introduction 1
1.1 Scope
1.2 Historical Background 2
1.2.1 Qanats 2
1.2.2 Groundwater Theories 2
1.2.3 Recent Centuries 4
1.3 Trends in Water Withdrawals and Use 5
1.4 Utilization of Groundwater 9
1.5 Groundwater in the Hydrologic Cycle 13
1.5.1 Hydrologic Cycle 13
1.5.2 The Groundwater System in the Hydrologic Cycle 15
1.6 Hydrologic Budget 20
1.7 Publication Sources 27
1. 7.1 Internet Resources 27
1.7.2 U.S. Geological Survey Publications 27
1.7.3 Publications 29
1.8 Data Sources 29
1.8.1 NWISWeb Data for the Nation 30
1.8.2 Real-Time Data 31
Problems 32
References 33
Example Publications of Organizations and Government Agencies 34
2 Occurrence of Groundwater 35
2.1 Origin and Age of Groundwater 35
2.2 Rock Properties Affecting Groundwater 36
2.2.1 Aquifers 36
2.2.2 Porosity 37
2.2.3 Soil Classification 40
2.2.4 Porosity and Representative Elementary Volume 41
2.2.5 Specific Surface 42
2.3 Vertical Distribution of Groundwater 45
2.4 Zone of Aeration 46
2.4.1 Soil-Water Zone 46
2.4.2 Intermediate Vadose Zone 46
2.4.3 Capillary Zone 47
2.4.4 Measurement of Water Content 49
2.4.5 Available Water 50
2.5 Zone of Saturation 50
2.5.1 Specific Retention 50
2.5.2 Specific Yield 50
2.6 Geologic Formations as Aquifers 52
2.6.1 Alluvial Deposits 52
2.6.2 Limestone 53
2.6.3 Volcanic Rock 55
2.6.4 Sandstone 55
ix
x Contents
2.6.5 Igneous and Metamorphic Rocks 55

2.6.6 Clay 55 3.11
2.7 Types of Aquifers 56 3.12
2.7.1 Unconfined Aquifer 56 Prot
2.7.2 Confined Aquifers 56 Refe
2.7.3 Leaky Aquifer 57
2. 7.4 Idealized Aquifer 57 4
2.8 Storage Coefficient 58
2. 9 Groundwater Basins/Regional Groundwater Flow Systems 59 4.1
2.9.1 High Plains Aquifer 59
2.9.2 Gulf Coastal Plain Aquifer System 61
2.10 Springs 67
2.1 0.1 What Are Springs? 67 4.2
2.1 0.2 Edwards Aquifer-Discharge of Springs 70
2.11 Hydrothermal Phenomena 73
2.11.1 Thermal Springs 73
2.11.2 Geothermal Energy Resources 75 4.3
2.12 Groundwater in Permafrost Regions 76 4.4
2.13 Groundwater in the United States 77
Problems 83
References 84
3 Groundwater Movement 86 4.5

3.1 Darcy's Law 86 4.6
3.1.1 Experimental Verification 86 4.7
3.1.2 Darcy Velocity 89
3.1.3 Validity of Darcy's Law 90
3.2 Permeability 91
3.2.1 Intrinsic Permeability 91
3.2.2 Hydraulic Conductivity 91 4.8
3.2.3 Transmissivity 92 4.9
3.2.4 Hydraulic Conductivity of Geologic Materials 92 4.10
3.3 Determination of Hydraulic Conductivity 94 Prob
3.3 .1 Formulas 94 Refe
3.3.2 Laboratory Methods 95
3.3.3 Tracer Tests 98 5
3.3.4 Auger Hole Tests 99
3.3.5 Pumping Tests of Wells 100 5.1
3.4 Anisotropic Aquifers 100 5.2
3.5 Groundwater Flow Rates 103
3.6 Groundwater Flow Directions 105
3.6.1 Flow Nets 105
3.6.2 Flow in Relation to Groundwater Contours 106
3.6.3 Flow Across a Water Table 112 5.3
3.6.4 Flow Across a Hydraulic Conductivity Boundary 114
3.6.5 Regional Flow Patterns 115
3.7 Dispersion 119
3. 7.1 Concept 119
3.7.2 Dispersion and Groundwater Hydrology 121
3.8 Groundwater Tracers 122 5.4
3.9 General Flow Equations 122
3.10 Unsaturated Flow 125
3.10.1 Flow Through Unsaturated Soils 126
3.10.2 Unsaturated Hydraulic Conductivity 128
Contents xi
3.10.3 Vertical and Horizontal Flows 133

3.11 Kinematic Wave 133
3.12 Infiltration: The Green-Ampt Method 134
Problems 140
References 142
4 Groundwater and Well Hydraulics 146

4.1 Steady Unidirectional Flow 146
4.1.1 Confined Aquifer 146
4.1.2 Unconfined Aquifer 147
4.1.3 Base Flow to a Stream 149
4.2 Steady Radial Flow to a Well 152
4.2.1 Confined Aquifer 152
4.2.2 Unconfined Aquifer 156
4.2.3 Unconfined Aquifer with Uniform Recharge 159
4.3 Well in a Uniform Flow 161
4.4 Unsteady Radial Flow in a Confined Aquifer 163
4.4.1 Nonequilibrium Well Pumping Equation 163
4.4.2 Theis Method of Solution 164
4.4.3 Cooper-Jacob Method of Solution 167
4.4.4 Chow Method of Solution 169
4.4.5 Recovery Test 170
4.5 Unsteady Radial Flow in an Unconfined Aquifer 172
4.6 Unsteady Radial Flow in a Leaky Aquifer 177
4. 7 Well Flow Near Aquifer Boundaries 180
4.7.1 Well Flow Near a Stream 180
4.7.2 Well Flow Near an Impermeable Boundary 186
4.7.3 Well Flow Near Other Boundaries 189
4.7.4 Location of Aquifer Boundary 191
4.8 Multiple Well Systems 192
4.9 Parti,ally Penetrating Wells 195
4.10 Well Flow for Special Conditions 198
Problems 198
References 204
5 Water Wells 206

5.1 Test Holes and Well Logs 206
5.2 Methods for Constructing Shallow Wells 206
5.2.1 Dug Wells 208
5.2.2 Bored Wells 209
5.2.3 Driven Wells 211
5.2.4 Jetted Wells 212
5.3 Methods for Drilling Deep Wells 214
5.3.1 Cable Tool Method 218
5.3.2 Rotary Method 220
5.3.3 Air Rotary Method 220
5.3.4 Rotary-Percussion Method 221
5.3.5 Reverse-Circulation Rotary Method 222
5.4 Well Completion 222
5.4.1 Well Casings 222
5.4.2 Cementing 223
5.4.3 Screens 223
5.4.4 Gravel Packs 226
xii Contents I
5.5 Well Development 228
5.5 .1 Pumping 228 I
5.5.2 Surging 228
5.5.3 Surging with Air 228
5.5.4 Backwashing with Air 229
5.5.5 Hydraulic Jetting 230
5.5.6 Chemicals 231
5.5.7 Hydrau]ic Fracturing 231
5,,;5.8 Explosives 231
5.6 Testing Wells for Yield 232
5.7 Pumping Equipment 232
5.7.1 Tota] Pumping Head 232
5.7.2 Pumps for Shallow Wells 232
5.7.3 Pumps for Deep Wells 232
5.8 Protection of Wells 234
5.8.1 Sanitary Protection 234
5.8.2 Frost Protection 236
5.8.3 Abandonment of Wells 237
5.9 Well Rehabilitation 237
5.10 Horizontal Wells 239
5.1 0.1 Infiltration Galleries 239
5.10.2 Horizontal Pipes 240
5.10.3 Collector Wells 240
5.11 Characteristic Well Losses 242
5.11.1 Well Losses 242
5.11.2 Evaluation of Well Loss 243
5.12 Specific Capacity and Well Efficiency 248
5.13 Slug Tests 250
5.13.1 Definition 250
5.13. 2 Design Guidelines 250
5.13 .3 Perfonnance of Slug Tests 251
5.13.4 Methods for Analyzing Slug Test Data 253
5.14 Slug Tests for Confined Fonnations 255
5.14.1 Cooper, Bredehoeft, and Papadopulos Method 255
5.14.2 Hvorslev Method 259
5.15 Slug Tests for Unconfined Fonnations 261
5.15.1 Bouwer and Rice Method · 262
5.15.2 Dagan Method 268
5.15.3 KGS Model 271
5.16 Slug Tests for High Conductivity Fonnations 272
5.17 Well-Skin Effect 273
Problems 273
References 276
6 Groundwater Levels and Environmental Influences 279

6.1 Time Variations of Levels 279
6.1.1 Secular Variations 279
6.1.2 Seasonal Variations 280
6.1.3 Short-Tenn Variations 280
6.2 Streamflow and Groundwater Levels 284
6.2.1 Bank Storage 285
6.2.2 Base Flow 286
6.2.3 Base Flow Recession Curve 289
6.3 Fluctuations Due to Evapotranspiration 291
Contents xiii
6.3.1 Evaporation Effects 291

6.3.2 Transpiration Effects 291
6.3.3 Evapotranspiration Effects 292
6.4 Fluctuations Due to Meteorological Phenomena 295
6.4.1 Atmospheric Pressure 295
6.4.2 Rainfall 297
6.4.3 Wind 298
6.4.4 Frost 298
6.5 Fluctuations Due to Tides 299
6.5.1 Ocean Tides 299
6.5.2 Earth Tides 302
6.6 Urbanization 303
6.7 Earthquakes 306
6.8 External Loads 308
6.9 Land Subsidence and Groundwater 308
6.9.1 Lowering of Piezometric Surface 311
6.9.2 Hydrocompaction 315
6.9.3 Dewatering of Organic Soils 317
6.9.4 Sinkhole Formation 317
6.9.5 Crustal Uplift 319
6.9.6 Monitoring of Land Subsidence 319
6.10 Effects of Global Climate Change on Groundwater 321
Problems 324
References 325
7 Quality of Groundwater 329

7.1 Natural Groundwater Quality 329
7.2 Sources of Salinity 330
7.3 Measures of Water Quality 335
7.4 Chemical Analysis 337
7.4.1 Concentrations by Weight 337
7.4.2 Chemical Equivalence 338
7.4.3 Total Dissolved Solids by Electrical Conductance 339
7.4.4 Hardness 339
7.5 Graphic Representations 340
7.6 Physical Analysis 345
7. 7 Biological Analysis 345
7.8 Groundwater Samples 345
7.9 Water Quality Criteria 346
7.9.1 Drinking Water Standards 346
7.9.2 Industrial Water Criteria 346
7.9.3 Irrigation Water Criteria 347
7.10 Changes in Chemical Composition 348
7.11 Dissolved Gases 349
7.12 Temperature 351
7.13 Saline Groundwater 354
Problems 356
References 357
8 Pollution of Groundwater 359

8.1 Pollution in Relation to Water Use 359
8.2 Municipal Sources and Causes 361
8.2.1 Sewer Leakage 361
xiv Contents
8.2.2 Liquid Wastes 361

8.2.3 Solid Wastes 362
8.3 Industrial Sources and Causes 364
8.3.1 Liquid Wastes 364 9.:
8.3.2 Tank and Pipeline Leakage 364
8.3.3 Mining Activities 364
8.3.4 Oil-Field Brines 366
8.4 Agricultural Sources and Causes 367
8.4.1 Irrigation Return Flows 367 9.(
8.4.2 Animal Wastes 367
8.4.3 Fertilizers and Soil Amendments 367
8.4.4 Pesticides, Insecticides, and Herbicides 368
8.5 Miscellaneous Sources and Causes 368
8.5.1 Urbanization 368
8.5.2 Spills and Surface Discharges 370
8.5.3 Stockpiles 370 9.i
8.5.4 Septic Tanks and Cesspools 370
8.5.5 Roadway Deicing 370
8.5.6 Saline Water Intrusion 371
8.5.7 Interchange through Wells 372 9.8
8.5.8 Surface Water 372
8.6 Attenuation of Pollution 373
8.6.1 Filtration 373 9.9
8.6.2 Sorption 373 9.1
8.6.3 Chemical Processes 374
8.6.4 Microbiological Decomposition 374
8.6.5 Dilution 374 9.1
8.7 Distribution of Pollution Underground 375
8.7.1 Hanford Site in Richland, Washington 377
8.8 Mass Transport of Pollutants 378
8.8.1 Transport Processes 378
8.8.2 Advection-Dispersion Equation for Solute Transport in Saturated Porous Media 382 Pro
8.8.3 Analytical Procedure 384 Ref
8.8.4 Transport of Reactive Pollutants 385
8.9 Monitoring Groundwater Quality 386
8.10 Remediation of Contaminated Groundwater 392 10
8.10.1 Remediation Goals 392
8.10.2 System Design 394 10.:
8.10.3 Hydraulic Containment of Groundwater 394
8.10.4 Groundwater Extraction Systems 396 10.:
8.10.5 Treatment of Contaminated Groundwater 396
8.11 Conventional Pump-and-Treatment Systems 401
8.11.1 Pump-and-Treat Remediation Strategies 401
10.~
8.11.2 Characterizing Sites for Pump and Treatment Design 403
8.11.3 Capture Zone Analysis 406
8.11.4 Extraction/Injection Scheme Design 407
Problems 407
10.4
References 409
10.5
9 Groundwater Flow Modeling Techniques 413

9.1 Why Develop Groundwater Models? 413 10.6
9.2 Types of Groundwater Models 414
9.3 Steps in the Development of a Groundwater Model 415
9.4 Simulation of Two-Dimensional Groundwater Systems 415 10.7
9.4.1 Governing Equations 415
Contents xv
9.4.2 Finite Difference Equations 417

9.4.3 Solution 419
9.4.4 Case Study 419
9.5 Three-Dimensional Groundwater Flow Model 420
9.5.1 Derivation of Finite Difference Equations 420
9.5.2 Simulation of Boundaries 427
9.5.3 Vertical Discretization 427
9.5.4 Hydraulic Conductance Equations 429
9.6 MODFLOW-2000 Description 431
9.6.1 Model Introduction 431
9.6.2 Space and Time Discretization 432
9.6.3 External Sources and Stresses 433
9.6.4 Hydraulic Conductance-Layer-Property Flow Package (LPF) 436
9.6.5 Solver Packages 440
9.6.6 Telescopic Mesh Refinement 440
9.7 Case Study: Using MODFLOW: Lake Five-0, Florida 444
9.7.1 Finite Difference Grid and Boundary Conditions 444
9.7.2 Model Calibration and Sensitivity Analysis 444
9.7.3 Model Results 448
9.8 Particle Tracking-MODPATH 449
9.8.1 What Is Particle Tracking? 449
9.8.2 Particle Tracking Analysis-An Application 449
9.9 Example Applications and Input of MOD FLOW 450
9.10 Solute Transport Modeling-MOC3D 455
9.10.1 Solute Transport Equation 455
9.10.2 MOC3D Model 456
9.11 Groundwater Modeling Software Support 459
9.11.1 U.S. Geological Survey 459
·9 .11.2 U.S. EPA Center for Exposure Assessment Modeling (CEAM) 459
9.11.3 International Groundwater Modeling Center (IGWMC) 459
9.11.4 Processors for MODFLOW 459
Problems 460
References 461
10 Management of Groundwater 464

10.1 Concepts of Basin Management 464
10.1.1 Managing Groundwater Resources 465
10.2 Groundwater Basin Investigations and Data Collection 466
10.2.1 Topographic Data 467
10.2.2 Geologic Data 468
10.2.3 Hydrologic Data 468
10.3 Yield 469
10.3.1 Alternative Basin Yields 469
10.3.2 Evaluation of Perennial Yield 471
10.4 Conjunctive Use and Watershed Management 473
10.5 Groundwater Management: Water Laws and Policies 477
10.5.1 Water Law and Policy 477
10.5.2 Arizona's Groundwater Management Code 477
10.5.3 Texas Groundwater Law 478
10.6 Case Study: Groundwater Management: Examples 479
10.6.1 Edwards Aquifer Management 479
10.6.2 High Plains Aquifer: Conjunctive Water Use on the High Plains 480
10.7 Groundwater Management Using Models 481
10.7.1 What Are Groundwater Management Models? 481
xvi Contents
10.7 .2 Optimization Methods 482

10.7.3 Types of Groundwater Management Models 482
10.8 Groundwater Management Modeling: Hydraulic Management Models 484
10.8.1 Steady-State One-Dimensional Problems for Confined Aquifers 484
10.8.2 Steady-State One-Dimensional Problems for Unconfined Aquifers 486
10.8.3 Steady-State Two-Dimensional Model for Confined Aquifers 487
10.8.4 Transient One-Dimensional Problem for Confined Aquifers 489
10.8.5 Steady-State Two-Dimensional Problem for Unconfined Aquifers 490
10.9 Policy Evaluation and Allocation Models: Response Matrix Approach 492.
10.10 Optimal Control Groundwater Management Modeling 495
10.11 Case Studies: Groundwater Management Modeling Applications 496
10.11.1 Optimal-Control Model for Barton Springs-Edwards Aquifer, Texas 496
10.11.2 Heuristic Optimization-Simulation Model for Groundwater Remediation Design:
N-Springs Site, Hanford, Washington 497
10.11.3 Groundwater Management Model Using Response Matrix Approach:
Santa Clara-Calleguas Basin, California 499
Problems 504
References 506
11 Surface Investigations of Groundwater 509

11.1 Geologic Methods 509
11.2 Remote Sensing 510
11.3 Geophysical Exploration 512
11.4 Electrical Resistivity Method 513
11.5 Seismic Refraction Method 517
11.6 Gravity and Magnetic Methods 520
Problems 520
References 521
12 Subsurface Investigations of Groundwater 523

12.1 Test Drilling 523
12.1.1 Geologic Log 524
12.1.2 Drilling-Time Log 525
12.2 Water Level Measurement 526
12.3 Geophysical Logging 527
12.4 Resistivity Logging 529
12.5 Spontaneous Potential Logging 533
12.6 Radiation Logging 533
12.6.1 Natural-Gamma Logging 533
12.6.2 Gamma-Gamma Logging 534
12.6.3 Neutron Logging 535
12.7 Temperature Logging 535
12.8 Caliper Logging 536
12.9 Fluid-Conductivity Logging 537
12.10 Fluid-Velocity Logging 537
12.11 Miscellaneous Logging Techniques 538
12.11.1 Television Logging 538
12.11.2 Acoustic Logging 538
12. i 1.3 Casing Logging 539
12.12 Other Subsurface Methods 539
12.13 Case Study: Oasis Valley, Nevada 539
Problems 541
References 545
Contents xvii
13 Artificial Recharge of Groundwater 547

13.1 Concept of Artificial Recharge 547
13.2 Recharge Methods 548
13.2.1 Methods 548
13.2.2 Recharge Rates 556
13.3 Wastewater Recharge for Reuse 559
13.4 Soil Aquifer Treatment (SAT) Systems 560
13.4.1 What Are SAT Systems? 560
13.4.2 Operation of SAT Systems 563
13.4.3 Modeling SAT Systems 565
13.5 Recharge Mounds 567
13.5.1 Perched Groundwater Mounds 567
13.5.2 Steady-State Equations for Groundwater Mounds 568
13.5.3 Hantush Equation 570
13.6 Storm water Infiltration Basin Mound Development 576
13.6.1 Potential Flow Model for a Trench 576
13.6.2 Potential Flow Model for Circular Basin 578
13.6.3 Mound Growth 580
13.6.4 Mound Recession 582
13.7 Innovative Approaches 583
13.7.1 Wildwood, New Jersey 583
13.7.2 Orange County, California 584
13.7.3 Long Island, New York 584
13.7.4 Orlando, Florida 584
13.7.5 Dayton, Ohio 584
13.8 Induced Recharge 584
Problems 585
References 586
14 Saline Water Intrusion in Aquifers 589

14.1 Occurrence of Saline Water Intrusion 589
14.2 Ghyben..:.Herzberg Relation Between Fresh and Saline Waters 590
14.3 Shape of the Fresh-Salt Water Interface 592
14.4 Structure of the Fresh-Salt Water Interface 596
14.5 Effect of Wells on Seawater Intrusion 597
14.6 Upconing of Saline Water 600
14.7 Fresh-Salt Water Relations on Oceanic Islands 603
14.8 Seawater Intrusion in Karst Terrains 605
14.9 Control of Saline Water Intrusion 606
14.10 Examples of Seawater Intrusion 608
14.10.1 Long Island, New York 608
14.10.2 Southern California 608
Problems 611
References 612
Appendix A 615
Appendix B 619
Appendix C 620
Index 625
Chapter 1
Introduction
Groundwater hydrology may be defined as the science of the occurrence, distribution, and
movement of water below the surface of the earth. Geohydrology has an identical connotation,
and hydrogeology differs only by its greater emphasis on geology. Utilization of groundwater
dates from ancient times, although an understanding of the occurrence and movement of sub-
surface water as part of the hydrologic cycle is recent.
The U.S. National Research Council (1991) presented the following definition of hydrology:
Hydrology is the science that treats the waters of the Earth, their occurrence, circulation, and
distribution, their chemical and physical properties, and their reaction with the environment,
including the relation to living things. The domain of hydrology embraces the full life history
of water on Earth.
Section 1.5 describes in further detail the concepts of the hydrologic cycle.
The importance of groundwater (hydrology) in the hydrologic cycle has been the subject of
extensive technical research and publishing by many investigators over the past decades. Many
of these publications are introduced in this book. Also, many books written on the subject of the
fate of water have caught the attention of the general public, especially those interested in saving
our resources. These include books by Carson 18 and de Villiers, 28 among others.
1.1 SCOPE
Groundwater (referred to without further specification) is commonly understood to mean water
occupying all the voids within a geologic stratum. This saturated zone is to be distinguished
from an unsaturated, or aeration, zone where voids are filled with water and air. Water con-
tained in saturated zones is important for engineering works, geologic studies, and water sup-
ply developments; consequently, the occurrence of water in these zones will be emphasized
here. Unsaturated zones are usually found above saturated zones and extend upward to the
ground surface; because water here includes soil moisture within the root zone, it is a major
concern of agriculture, botany, and soil science. No rigid demarcation of waters between the two
zones is possible, for they possess an interdependent boundary, and water can move from zone
to zone in either direction. The interrelationships are described more fully in Chapter 2.
Groundwater plays an important part in petroleum engineering. Two-fluid systems,
involving oil and water, and three-fluid systems, involving gas, oil, and water, occur frequently
in development of petroleum. Although the same hydrodynamic laws govern flows of these
systems and groundwater, the distinctive nature of water in petroleum reservoirs sets it apart
from other groundwater. Major differences exist in water quality, depth of occurrence, and
methods of development and utilization, all of which contribute to a separation of interests and
1
2 Chapter 1 Introduction
applications. Therefore, groundwater in petroleum reservoirs will not be treated specifically in

this book. It should be noted, however, that groundwater hydrology has gained immeasurably
from research conducted by the petroleum industry.
1.2 HISTORICAL BACKGROUND

1.2.1 Qanats
Groundwater development dates from ancient times. 15•47• The Old Testament contains numerous
references to groundwater, springs, and wells. Other than dug wells, groundwater in ancient
times was supplied from horizontal wells known as qaiUlts. t These persist to the present day and
can be found in a band across the arid regions of Southwestern Asia and North Africa extending
from Afghanistan to Morocco. A cross section along a qanat is shown in Figure 1.2.1.Typically,
a gently sloping tunnel dug through alluvial material leads water by gravity flow from beneath
the water table at its upper end to a ground surface outlet and irrigation canal at its lower end. 13
I
Vertical shafts dug at closely spaced intervals provide access to the tunnel. 81 Qanats are labori-
ously hand constructed by skilled workers employing techniques that date back 3,000 years.t
Iran possesses the greatest concentration of qanats; here some 22,000 qanats supply 75
percent of all water used in the country. Lengths of qanats extend up to 30 km, but most are
less than 5 km. 13 The depth of the qanat mother well (see Figure 1.2.1) is normally less than
50 m, but instances of depths exceeding 250 m have been reported. Discharges of qanats vary
seasonally with water table fluctuations and seldom exceed 100m3/hr. Indicative of the den-
sity of qanats is the map in Figure 1.2.2. Based on aerial photographs of the Varamin Plain,
located 40 km southeast of Tehran, this identifies 266 qanats within an area of 1,300 km2.
1.2.2 Groundwater Theories

Utilization of groundwater greatly preceded understanding of its origin, occurrence, and move-
ment. The writings of Greek and Roman philosophers to explain origins of springs and ground-
Mountain
foothills
Mother, well
Figure 1.2.1. Vertical cross section along a qanat (after Beaumont 12).
' Superscript numbers refer to references at the end of the chapter.

t Qanar comes from a Semitic word meaning "to dig." There are several variants of the name, including karez, fog-
gara, andfalaj, depending on location; in addition, there are numerous differences in spelling.26
*Illustrative of the tremendous human effort expended to construct a qanat is a calculation by Beaumont. 13 The
longest qanat near Zarand, Iran, is 29 km long with a mother well depth of 96 m and with 966 shafts along its
length; the total volume of material excavated is estimated at 75,400 m3.
1.2 Historical Background 3
ly in water contain theories ranging from fantasy to nearly correct accounts. 1• 6 As late as the sev-
ably enteenth century it was generally assumed that water emerging from springs could not be
derived from rainfall, for it was believed that the quantity was inadequate and the earth too
impervious to permit penetration of rainwater far below the surface. Thus, early Greek philoso-
phers such as Homer, Thales, and Plato hypothesized that springs were formed by seawater
conducted through subterranean channels below the mountains, then purified and raised to the
surface. Aristotle suggested that air enters cold dark caverns under the mountains where it con-
denses into water and contributes to springs.
rous
The Roman philosophers, including Seneca and Pliny, followed the Greek ideas and con-
:ient
tributed little to the subject. An important step forward, however, was made by the Roman archi-
and
tect Vitruvius. He explained the now-accepted infiltration theory that the mountains receive large
Hng
amounts of rain that percolate through the rock strata and emerge at their base to fonn streams.
:illy,
!ath
d.l3 """"""Mountain front 2 3
t I
ori- -Oanat kM
t ----···Surface canal
Oanat outlet
'75 D VIllage
are 0 Town
han Contours
It20 in Interval
•ary
len-
ain,
ve-
nd-
r-
Figure 1.2.2. Map of qanats on the Varamin Plain, Iran (after Beaumont9).
The Greek theories persisted through the Middle Ages with no advances until the end of
the Renaissance. The French potter and philosopher Bernard Palissy (c. 151 0-1589) reiterated
the infiltration theory in 1580, but his teachings were generally ignored. The German
astronomer Johannes Kepler (1571-1630) was a man of strong imagination who likened the
earth to a huge animal that takes in water of the ocean, digests and assimilates it, and dis-
charges the end products of these physiological processes as groundwater and springs. The
seawater theory of the Greeks, supplemented by the ideas of vaporization and condensation
processes within the earth, was restated by the French philosopher Rene Descartes
(1596-1650).
A clear understanding of the hydrologic cycle was achieved by the latter part of the sev-
enteenth century. For the first time, theories were based on observations and quantitative data.
Three Europeans made notable contributions, although others contributed to and supported
these advances. Pierre Perrault* (1611-1680) measured rainfall during three years and esti-
mated runoff of the upper Seine River drainage basin. He reported in 1674 that precipitation
on the basin was about six times the river discharge, thereby demonstrating as false the early
assumption of inadequate rainfall. 57 The French physicist Edme Mariotte (c. 1620-1684) made
measurements of the Seine at Paris and confirmed Perrault's work. His publications appeared
in 1686, after his death, and contained factual data strongly supporting the infiltration theory.
Meinze~ 4 once stated, "Mariotte ... probably deserves more than any other man the distinc-
tion of being regarded as the founder of groundwater hydrology, perhaps I should say of the
entire science of hydrology." The third contribution came from the English astronomer
Edmund Halley (1656-17 42), who reported in 1693 on measurements of evaporation, demon-
strating that sea evaporation was sufficient to account for all springs and stream flow.
1.2.3 Recent Centuries

During the eighteenth century, fundamentals in geology were established that provided a basis
for understanding the occurrence and movement of groundwater. During the first half of the
nineteenth century many artesian wells were drilled in France, stimulating interest in ground-
water. The French hydraulic engineer Henry Darcy (1803-1858) studied the movement of
water through sand. His treatise of 1856 defined the relation, now known as Darcy's law, gov-
erning groundwater flow in most alluvial and sedimentary formations. Later European contri-
butions of the nineteenth century emphasized the hydraulics of groundwater development.
Significant contributions were made by J. Boussinesq, G. A. Daubree, J. Dupuit, P. Forch-
heimer, and A. Thiem. In the twentieth century, increased activity in all phases of groundwa-
ter hydrology occurred. Many Europeans have participated with publications of either
specialized or comprehensive works. There are too many people to mention them all, but R.
Dachler, E. Imbeaux, K. Keilhack, W. Koehne, J. Kozeny, E. Prinz, H. Schoeller, and G. Thiem
are best known in the United States.
* Pierre Perrault was a lawyer by profession and held administrative and financial positions in the French govern-
ment; hence he is not well known in scientific circles. His interest in groundwater, leading to publication of
De l'Origine des Fontaines in 1674, can be traced to the stimulus of the Dutch mathematician, astronomer, and
physicist, Christiaan Huygens, who was then living in Paris and to whom the book is dedicated. Also, Pierre
Perrault is often overshadowed by his four distinguished brothers: Jean (c. 1610-1669), a lawyer; Nicolas
(1624-1662), a noted theologian; Claude (1613-1688), a physician, architect, and scientist, who is regarded as one
of the most eminent French scholars of his time; and Charles (1628-1703), author and critic, who is best known for
his Mother Goose fairy tales.
1.3 Trends in Water Withdrawals and Use 5
l of American contributions to groundwater hydrology date from near the end of the nine-
ted teenth century. In the past 110 years, tremendous advances have been made. Important early
1an theoretical contributions were made by A. Hazen, F. H. King, and C. S. Slichter, while detailed
the field investigations were begun by men such as T. C. Chamberlin, N.H. Darton, W. T. Lee, and
lis- W. C. Mendenhall. 0. E. Meinzer, through his consuming interest in groundwater and his
~he dynamic leadership of groundwater activities of the U.S. Geological Survey, stimulated many
lOll individuals in the quest for groundwater knowledge. In recent decades the publications of
tes M.S. Hantush, C. E. Jacob, G. B. Maxey, C. L. McGuinness, and R. W. Stallman are note-
worthy. Within the last 40 years the surge in university research on groundwater problems, the
ev- establishment of professional consulting firms specializing in groundwater, the advent of the
Lta. digital computer, and the extensive development of computer software have jointly produced
ted a competence for development and management of groundwater resources.
;ti-
iOll
rly 1.3 TRENDS IN WATER WITHDRAWALS AND USE
Lde
The U.S. Geological Survey65 estimated the total fresh and saline withdrawals in the United
~ed
States during 1995 to have been 402,000 million gallons per day (MgaVd) for all off-stream
ry.
water-use categories (public supply, domestic, commercial, irrigation, livestock, industrial,
lC-
mining, thermoelectric power). This estimate is nearly two percent less than the withdrawal
:he
estimate for 1990. Table 1.3.1 and Figure 1.3.1 provide a comparison of total water with-
1er
drawals by water-resources region. This comparison indicates that the California, South
m-
Atlantic-Gulf, and Mid-Atlantic regions account for one-third of the total water withdrawn in
the United States.
National water-use compilations began in 1950 and are conducted at five-year intervals. 65
Estimates in Table 1.3.2 and Figure 1.3.2 summarize the water use-withdrawals, source of
water, reclaimed wastewater, consumptive use, and in-stream use (hydroelectric power)-at
sis five-year intervals. Figure 1.3.3 illustrates the trends in water withdrawals by water-use cate-
he gory and total withdrawals for 1960-1995. Table 1.3.2 also illustrates the percentage change
td- in the 1990 and 1995 summary estimates. Estimates indicate that the general increase in water
of use from 1950 to 1980 and the decrease from 1980 to 1995 can be attributed in part to the fol-
IV- lowing major factors: 65
ri-
nt. • Most of the increases in water use from 1950 to 1980 were the result of expansion of
:h- irrigation systems and increases in energy development.
'a- • The development of center-pivot irrigation systems and the availability of plentiful and
inexpensive groundwater resources supported the expansion of irrigation systems.
~m
• Higher energy prices in the 1970s, and large drawdown in groundwater levels in some
areas increased the cost of irrigation water. In the 1980s, improved application tech-
niques, increased competition for water, and a downturn in farm economy reduced
demands for irrigation water.
• The transition from water-supply management to water-demand management encour-
aged more efficient use of water.
• New technologies in the industrial sector that require less water, improved plant effi-
ciencies, increased water recycling, higher energy prices, and changes in laws and reg-
ulations to reduce the discharge of pollutants resulted in decreased water use and less
water being returned to natural systems after use.
e
)f
• The enhanced awareness by the general public of water resources and the active con-
servation programs in many states have contributed to reduced water demands.
Table 1.3.1 Total Offstream Water Use by Water-Resources Region, 1995 0\
Per Withdrawals (Mgal/d) (includes irrigation conveyance losses) Consump- n

:::r
~
capita Convey- tive use, "0
By source and type ~
"""1
use, Reclaimed ance fresh-
Groundwater Surface water Total
Population freshwater wastewater losses water
Region (thousands) (gal/d) Fresh Saline Total Fresh Saline Total Fresh Saline Total (Mgal/d) (Mgal/d) (Mgal/d) 5"
l=i
0
0..
New England 12,849 289 725 0 726 2,980 8,800 11,800 3,710 8,800 12,600 0 0 388 c::
~
Mid-Atlantic 42,412 509 2,690 2,690 18,900 20,300 39,200 21,600 20,300 41,900 72 1.9 1,170 c;·
::::1
South Atlantic 37,845 848 7,110 16 7,120 25,000 12,700 37,700 32,100 12,700 44,800 237 33 5,570
-Gulf
Great L~es 21,836 1,500 1,510 4.6 1,520 31,100 6.5 31,100 32,700 11 32,700 0 .1 1,580
Ohio 22,631 1,330 1,980 22 2,000 28,100 .6 28,100 30,100 23 30,100 1.1 .7 1,870
Tennessee 4,198 2,140 258 0 258 8,730 0 8,730 8,980 0 8,980 .3 0 289
Upper Mississippi 22,268 1,050 2,570 4.2 2,570 20,700 0 20,700 23,300 4.2 23,300 11 0 1,660
Lower Mississippi 7,324 2,720 9,180 0 9,180 10,800 0 10,800 20,000 0 20,000 .7 553 7 ,740
Souris-Red 693 364 115 0 115 138 0 138 253 0 253 0 1.8 122
-Rainy
Missouri Basin 10,664 3,380 9,320 38 9,360 26,700 0 26,700 36,000 38 36,100 22 7,840 14,200
Arkansas-White 8,931 1,800 7,490 284 7,780 8,590 0 8,590 16,100 284 16,400 37 944 8,190
-Red
Texas-Gulf 16,755 1,050 5,960 324 6,280 11,700 4,860 16,600 17,700 5,190 22,900 71 390 7,340
Rio Grande 2,566 2,600 1,930 61 1,990 4,740 0 4,740 6,670 61 6,730 7.2 1,360 2,960
Upper Colorado 714 10,400 116 14 130 7,310 0 7,310 7,420 14 7,440 1.7 1,940 2,520
Lower Colorado 5,318 1,500 3,000 12 3,010 4,970 2.3 4,970 7,960 14 7,980 187 1,090 4,520
Great Basin 2,405 2,510 1,610 56 1,660 4,420 143 4,560 6,030 199 6,230 33 1,140 3,260
Pacific Northwest 9,948 3,220 5,500 0 5,500 26,500 38 26,500 32,000 38 32,000 .1 8,050 10,600
California 32,060 1,140 14,600 185 14,800 21,900 9,450 31,300 36,500 9,640 46,100 330 1,860 25,300
Alaska 604 350 58 75 132 154 43 196 211 117 329 0 .I 25
Hawaii 1,187 853 515 16 531 497 906 1,400 1,010 922 1,930 6.2 98 542
Caribbean 3,858 152 156 .2 156 433 2,450 2,880 588 2,450 3,040 0 15 189
Total 267,068 1,280 76,400 1,110 77,500 264,000 59,700 324,000 341,000 60,800 402,000 I ,020 25,300 100,000
Source: Solley65
Figures may not add to totals because of independent rounding. Mgal/d = million gallons per day; gal/d = gallons per day
Water withdrawa ls
(million gallons per day)
Do - 11,999
• 12,000 - 23,999
- 2 4 ,000 - 35,999
- 36,000 - 47,000
-- <..>
CARIBBEAN .... ~
-
~ ::s
..
'(; .._ HAWAII
....:,_~~..""
c..
V>
.~ ~~- ···- -" ·.:;:"...

~ ~
s·
··- ·.... .
'(-
"<\.. f. ·· J·ALASKA "' ~

1.~
:..-- (b
.....
""'"~....... ~
.f;~ ;:;.
c..
~
":. -. t ' '.
'. ..... _.;ft10f;.;··' .
.......... - _,..; -
I»
v;-
...::s
65 a.
Figure 1.3.1. Total water withdrawals by water-resources region, 1995. c:
V>
0
-l
8 Chapter I Introduction
Table 1.3.2 Trends of Estimated Water Use in the United States, 1950-95
Percentage
Year change
19501 1955 1 19602 19652 19703 19754 19804 19854 19904 19954 1990-95
Population (millions) 150.7 164.0 179.3 193.8 205.9 216.4 229.6 242.4 252.3 267.1 +6
Offstream use
Total withdrawals 180 240 270 310 370 420 4405 399 408 402 -2
Public supply 14 17 21 24 27 29 34 36.5 38.5 40.2 +4
Rural domestic 3.6 3.6 3.6 4.0 4.5 4.9 5.6 7.79 7.89 8.89 +13
and livestock
Irrigation 89 110 110 120 130 140 150 137 137 134 -2
Industrial
Thermoelectric 40 72 100 130 170 200 210 187 195 190 -3
power use
Other industrial use 37 39 38 46 47 45 45 30.5 29.9 29.1 -3
Source of water
Ground
Fresb 34 47 50 60 68 82 835 73.2 79.4 76.4 -4
Saline (6) .6 .4 .5 .9 .652 1.22 1.11 -9
Surface
Fresh 140 180 190 210 250 260 290 265 259 264 +2
Saline 10 18 31 43 53 69 71 59.6 68.2 59.7 -12
Reclaimed wastewater (6) .2 .6 .7 .5 .5 .5 .579 .750 1.02 +36
Consumptive use (6) (6) 61 77 877 967 1007 92.37 947 1007 +6
Instream use
Hydroelectric power 1,100 1,500 2,000 2,300 2,800 3,300 3,300 3,050 3,290 3,160 -4
1 5
48 States and District of Columbia Revised
2
50 States and District of Columbia 6 Data not available.
3 7
50 States and District of Columbia, and Puerto Rico Freshwater only
4
50 States and District of Columbia, Puerto Rico, and VIrgin Islands
Source: Solley65
400 300
[ ] Groundwater
350
lL
• Surface water 250
o-
"'>-
o(§ 300 • Total
~a: 200 0
"'z
"'w
::JCI. 250 3
O<ll
l:Z ~
1-0
-..J 200 150 z
~
~z 150 :5
~Q 100
::::>
Cl.
oJ 0
a.
l:::::!: 100
t:
3: 60
50
0 0
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995
Figure 1.3.2. Trends in fresh groundwater and surface-water withdrawals, and population, 1950-95.65
1.4 Utilization of Groundwater 9
•
300 500
Public supply
1tage • Rural 450
,ge 250
Total withdrawals
0 Irrigation
~>
u. 400
o~
0-95 (/)~ Z4;
gjo
oo 350
::>0:
Zo: 200
6 <w ::: ow
~0.. ::: :rO..
300 1-(IJ
O"' ' ~z
:rZ
1-0 :;: f~: ~g
~-' 150 !.• i 250 <-'
2 -'
~~
Ill<
-'0
I
4 ~z 200 oo
3 ~g 100 .; I
:r-
1-:::l
o:::! 150 ~~
~::;;
<l!lJ..
~ 1-0
2 100
~
50
50
3
0 0
196() 1965 1970 1975 1980 1985 1990 1995
3
Figure 1.3.3. Trends in water withdrawals (fresh and saline) by water-use category and total (fresh
and saline) withdrawals, 1960-95.65
4
9 1.4 UTILIZATION OF GROUNDWATER
2 Groundwater is an important source of water supply throughout the world. Its use in irrigation,
2 industries, municipalities, and rural homes continues to increase. Figure 1.4.1 strikingly illus-
6 trates the dependence of an Indian village on its only water source-groundwater from a sin-
6 gle dug well. Cooling and air conditioning have made heavy demands on groundwater because
of its characteristic uniformity in temperature. Shortages of groundwater in areas where exces-
sive withdrawals have occurred emphasize the need for accurate estimates of the available sub-
surface resources and the importance of proper planning to ensure the continued availability
of water supplies.
There is a tendency to think of groundwater as being the primary water source in arid
regions and of surface water in humid regions. But a study of groundwater use in the United
States, for example, reveals that groundwater serves as an important resource in all climatic
Figure 1.4.1. Villagers laboriously lifting and carrying water from a deep dug well in northern India
(photo by David K. Todd).
zones. 55 Reasons for this include its convenient availability near the point of use, its excellent
quality (which typically requires little treatment), and its relatively low cost of development.
Furthennore, in humid locales such as Barbados, Jamaica, and Hawaii, groundwater predom-
inates as the water source because the high infiltration capacity of the soils sharply reduces sur-
face runoff.
Figure 1.4.2 illustrates the relative proportion of water source and disposition and the gen-
eral distribution of water from source to disposition for 1995. Table 1.4.1 lists the total off-
stream water use by state for 1995-breaking down the withdrawals into groundwater and
surface water and subdividing these into fresh and saline water. The total groundwater with-
drawal was 77,500 Mgalld and the total surface-water withdrawal was 324,000 Mgalld. Table
1.4.2 lists the groundwater withdrawals by water-use category and water-resources region for
1995. The significant proportion ofgroundwater used for irrigation purposes is clearly indi-
cated by the fact that 49,000 Mgal/d of the total 77,500 Mgal!d is used for irrigation.
SOURCE USE DISPOSITION

SURFACE WATER DOt.IESTIC.COMMERCIA!. CONSUIIIPTIVE USE
Figure 1.4.2. Source, use, and
disposition of freshwater in the
United States, 1995. For each
water-use category, this diagram
shows the relative proportion of
water source and disposition and
the general distribution of water
from source to disposition. The
lines and arrows indicate the distri-
bution of water from source to dis-
position for each category; for
example, surface water was 77.6
percent of total freshwater with-
drawn, and going from "Source" to
"Use" columns, the line from the
surface-water block to the domes-
tic and commercial block indicates
that 0.8 percent of all surface water
withdrawn was the source for 4.8
percent of total water (self-sup-
plied withdrawals, public-supply
deliveries) for domestic and com-
mercial purposes. In addition,
going from the "Use" to
"Disposition" columns, the line
from the domestic and commercial
block to the consumptive use block
indicates that 19.2 percent of the
water for domestic and commercial
purposes was consumptive use;
this represents 8.0 percent of total
consumptive use by all water-use
categories. 65
I:.. ;;>';"" ,_.
0 j~ I
9 ;a g
t:.-1 C/)
7-sr~¥"5.7l¥ ~
Table 1.4.1 Total Offstream Water Use by State, 1995

Per Withdrawals (Mgal/d) (includes irrigation conveyance losses) Consump-
capita Convey- tive use,
By source and type
use Reclaimed ance fresh-
Groundwater Surface water Total
Population freshwater wastewater losses water
Region (thousands) (gal/d) Fresh Saline Total Fresh Saline Total Fresh Saline Total (Mgalld) (Mgalld) (Mgalld)
Alabama 4,253 1,670 436 9.1 445 6,650 0 6,650 7,090 9.1 7,100 0.1 0 532
Alaska 604 350 58 75 132 154 43 196 211 117 329 0 .1 25
Arizona 4,218 1,620 2,830 12 2,840 3,980 2.3 3,990 6,820 14 6,830 180 1,030 3,830
Arkansas 2,484 3,530 5,460 0 6,460 3,310 0 3,310 8,770 0 8,770 0 416 4,760
California 32,063 1,130 14,500 185 14,700 21,800 9,450 31,300 36,300 9,640 45,900 334 1,670 25,500
Colorado 3,747 3,690 2,260 17 2,270 11,600 0 11,600 13,800 17 13,800 11 3,770 5,230
Connecticut 3,275 389 166 0 166 1,110 3,180 4,290 1,280 3,180 4,450 0 0 97
Delaware 717 1,050 110 0 110 642 743 1,390 752 743 1,500 0 0 71
D.C. 554 18 .5 0 .6 9.7 0 9.7 10 0 10 0 0 15
Florida 14,166 509 4,340 4.6 4,340 2,880 11,000 13,800 7,210 11,000 18,200 236 32 2,780
Georgia 7,201 799 1,190 0 1,190 4,560 64 4,630 5,750 64 5,820 .6 0 1,170
Hawaii 1,187 853 515 16 531 497 906 1,400 1,010 922 1,930 6.2 98 542
Idaho 1,163 13,000 2,830 0 2,830 12,300 0 12,300 15,100 0 15,100 0 5,480 4,340
Illinois 11,830 1,680 928 25 953 19,000 0 19,000 19,900 25 19,900 2.0 0 857
Indiana 5,803 1,570 709 0 709 8,430 0 8,430 9,140 0 9,140 0 0 505
Iowa 2,842 1,070 528 0 528 2,510 0 2,510 3,030 0 3,030 0 0 290
Kansas 2,565 2,040 3,510 0 3,510 1,720 0 1,720 5,240 0 5,240 6.8 143 3,620
Kentucky 3,860 1,150 226 0 226 4,190 0 4,190 4,420 0 4,420 0 .5 318
Louisiana 4,342 2,270 1,350 0 1,350 8,500 0 8,500 9,850 0 9,850 0 166 1,930
Maine 1,241 178 80 0 80 141 105 248 221 105 326 0 0 48
Maryland 5,042 289 246 0 246 1,210 6,270 7,480 1,460 6,270 7,730 70 0 150
Massachusetts 6,074 189 351 0 351 795 4,370 5,160 1,150 4,370 5,510 0 0 180
Michigan
Minnesota
Mississippi
9,549
4,610
2,697
1,260
736
1,140
858
714
2,590
4.4
0
0
862
714
2,590
11,200
2,680
502
0
0
112
11,200
2,680
614
12,100
3,390
3,090
4.4
0
112
12,100
3,390
3,200
0
0
0
0
0
17
667
417
1,570
-
:p..
Missouri 5,324 1,320

c
891 0 891 6,140 0 6,140 7,030 0 7,030 11 0 692 §
N
Montana 870 10,200 204 13 217 8,640 0 8,640 8,850 13 8,860 0 4,410 1,960 ~
Nebraska 1,637 6,440 6,200 4.7 6,200 4,350 0 4,350 10,500 4.7 10,500 2.0 906 7,020
g·
::s
Nevada 1,530 1,480 855 42 896 1,400 0 1,400 2,260 42 2,300 24 473 1,340 _.,
0
New Hampshire 1,148 388 81 0 81 364 877 1,240 446 877 1,320 0 0 35 a
>-t
0
t:
(continues) ::s
0..
~
~
~
>-t
......
......
Table 1.4.1 (continued) Total Offstream Water Use by State, 1995 .......
N
Per Withdrawals (Mgal/d) (includes irrigation conveyance losses) Consump- n

::r
capita
By source and type
Convey- tive use, .g
use
Total
Reclaimed ance fresh- ....~
Population freshwater Groundwater Surface water wastewater losses water
Region (thousands) (gal/d) Fresh Saline Total Fresh Saline Total Fresh Saline Total (Mgalld) (Mgalld) (Mgal/d) .........
:::s
r::;
0
New Jersey 7,945 269 580 0 580 1,560 3,980 5,530 2,140 3,980 6,110 1.1 0 210 0...
c
New Mexico 1,686 2,080 1,700 0 1,700 1,800 0 1,800 3,510 0 3,510 0 628 1,980 ~
c;·
New York 18,136 567 1,010 1.5 1,010 9,270 6,500 15,800 10,300 6,500 16,800 0 0 469 :::s
North Carolina 7,195 1,070 535 2.1 535 7,200 1,550 8,750 7,730 1,560 9,290 1.0 0 713
North Dakota 641 1,750 122 0 122 1,000 0 1,000 1,120 0 1,120 0 5.1 181
Ohio 11,151 944 905 0 905 9,620 0 9,620 10,500 0 10,500 0 .2 791
Oklahoma 3,278 543 959 259 1,220 822 0 822 1,780 259 2,040 0 4.9 716
Oregon 3,140 2,520 1,050 0 1,050 6,860 0 6,860 7,910 0 7,910 0 1,300 3,210
Pennsy I vania 12,072 802 860 0 860 8,820 0 8,820 9,680 0 9,680 1.1 0 565
Rhode Island 990 138 27 0 27 109 275 383 136 275 411 0 0 19
South Carolina 3,673 1,690 322 0 322 5,880 0 5,880 6,200 0 6,200 0 0 321
South Dakota 729 631 187 0 187 273 0 273 460 0 460 0 54 249
Tennessee 5,256 1,920 435 0 435 9,640 0 9,640 10, 100 0 10,100 .5 0 233
Texas 18,724 1,300 8,370 411 8,780 16,000 4,860 20,800 24,300 5,280 29,600 109 540 10,500
Utah 1,951 2,200 776 14 790 3,530 143 3 ,670 4,300 157 4,460 14 612 2,200
Vermont 585 967 50 0 50 515 0 515 565 0 565 0 0 24
Virginia 6,618 826 358 0 358 5,110 2,800 7,900 5,470 2,800 8,260 0 2.9 218
Washington 5,431 1,620 1,760 0 1,760 7,060 38 7,100 8,820 38 8,860 0 1,090 3,080
West Virginia 1,828 2,530 146 .5 146 4,470 0 4,470 4,620 .5 4 ,620 0 0 352
Wisconsin 5,102 1,420 759 0 759 6,490 0 6,490 7,250 0 7 ,250 0 0 443
Wyoming 480 14,700 317 18 335 6,720 0 6,720 7,040 18 7,060 9.1 2,470 2,800
Puerto Rico 3,755 154 155 0 155 422 2,260 2,680 576 2,260 2,840 0 15 187
Virgin Islands 103 113 .5 .2 .7 11 190 201 12 190 202 0 0 1.9
Total 267,068 1,280 76,400 1,110 77,500 264,000 59,700 324,000 341,000 60,800 402,000 1,020 25,300 100,000
Figures may not add to totals because of independent rounding. Mgal/d = million gallons per day; galld = gallons per day
Source: Solley 65
_ _..._ _lliiillll_ _ _ _ _ _ _ _ _ _ _......_..._ _ _ _ _ _ _ _ _ _............._ ....................__"'""-..........-............

--..~--- .......
~- --"- -·_ .. ·-· -~~-~-- ------ ~ -- ·- ----
~ ~ I~O~~D~9~~~~?~~~~~Q9~~? ~
Table 1.4.2 Groundwater Withdrawals by Water-Use Category and Water-Resources Region, 1995
Public Commer- Live- Thermo-
Region supply Domestic cia! Irrigation stock Industrial Mining electric Total
----
Fresh Fresh Fresh Fresh Fresh Fresh Saline Fresh Saline Fresh Fresh Saline
New England 335 168 64 47 6.4 53 0 2.9 0 48 725 0

Mid-Atlantic 1,270 485 217 128 79 344 0 159 1.0 11 2,690 1.0
South Atlantic-Gulf 2,760 719 114 2,280 188 787 0 177 9.1 79 7,110 16
Great Lakes 585 354 44 170 50 270 3.6 34 1.0 7.6 1,510 4.6
Ohio 880 323 91 61 60 379 0 115 22 70 1,980 22
Tennessee 125 64 3.6 8.7 19 35 0 3.7 0 0 258 0
Upper Mississippi 1,150 311 94 430 216 328 0 22 4.2 24 2,570 4.2
Lower Mississippi 741 73 15 6,930 740 611 0 3.1 0 69 9,180 0
Souris-Red-Rainy 34 17 .2 45 17 1.7 0 .4 0 0 115 0
Missouri Basin 643 137 19 8,030 253 102 0 104 38 30 9,320 38
Arkansas-White-Red 378 105 16 6,660 190 78 0 30 284 37 7,490 284
Texas-Gulf 978 115 34 4,370 82 214 .5 118 324 50 5,960 324
Rio Grande 356 25 17 1,420 27 10 0 53 60 16 1,930 61
Upper Colorado 35 11 5.6 38 4.2 2.4 0 20 14 0 116 14
Lower Colorado 476 44 22 2,210 33 42 0 126 12 45 3,000 12
Great Basin 350 13 10 1,090 9.2 60 .1 71 19 2.6 1,610 56
Pacific Northwest 917 253 37 4,030 44 215 0 6.5 0 .5 5,500 0
California 2,730 112 77 10,900 231 522 10 16 151 3.6 14,600 185
Alaska 30 8.3 11 .1 .1 3.8 0 0 75 4.2 58 75
Hawaii 200 2.4 45 173 7.5 19 .9 .5 0 67 515 16
Caribbean 95 6.4 1.3 33 4.5 10 .2 3.4 0 2.2 156 .2
Total 15,100 3,350 939 . 49,000 2,260 4,090 15 1,070 1,010 565 76,400 1,110
Figures may not add to totals because of independent rounding. All values in million gallons per day
Source: Solley 65
The largest single demand for groundwater is irrigation, amounting to 67.3% percent of
all groundwater used in 1995. More than 90 percent of this water is pumped in the western
states, where arid and semiarid conditions have fostered extensive irrigation development.
1.5 GROUNDWATER IN THE HYDROLOGIC CYCLE

1.5.1 Hydrologic Cycle
The central focus of hydrology is the hydrologic cycle consisting of the continuous processes
shown in Figure 1.5.1. Water evaporates from the oceans and land surfaces to become water
vapor that is canied over the earth by atmospheric circulation. The water vapor condenses and
precipitates on the land and oceans. The precipitated water may be intercepted by vegetation,
become overland flow over the ground surface, infiltrate into the ground, flow through the soil
as subsurface flow, or discharge as surface runoff. Evaporation from the land surface comprises
evaporation directly from soil and vegetation surfaces, and transpiration through plant leaves.
Collectively these processes are called evapotranspiration. Infiltrated water may percolate
deeper to recharge groundwater and later become spring/low or seepage into streams to also
become streamflow.
14 Chapter l Introduction
~\_
~< ¢= ~< 0;1'

~
39
~
JJ
100
Moisture
over land
Precipitation on land AI
385
Precipitation
61 on ocean
Evaporation
from land
Evaporation and
evapotranspiration
424
Evaporation
from ocean
Figure 1.5.1. Hydrologic cycle with global annual average water balance given in units relative to a value of 100 for the rate of
precipitation on land.22
A hydrologic system is defined as a structure or volume in space, surrounded by a bound- Fi

ary, that accepts water and other inputs, operates on them internally, and pwduces them as out-
puts.2l. 22 The structure (for surface or subsurface flow) or volume in space (for atmospheric
moisture flow) is the totality of the flow paths through which the water may pass as through-
put from the point it enters the system to the point it leaves. The boundary is a continuous sur-
face defined in three dimensions enclosing the volume or structure. A working medium enters
the system as input, interacts with the structure and other media, and leaves as output. Physi-
cal, chemical, and biological processes operate on the working media within the system; the
most common working media involved in hydrologic analysis are water, air, and heat energy.
The global hydrologic cycle can be represented as a system contairung three subsystems:
the atmospheric water system, the surface water system, and the subsurface water system as
shown in Figure 1.5.2. Another example is the storm-rainfall-runoff process on a watershed,
which can be represented as a hydrologic system. The input is rainfall distributed in time and
space over the watershed and the output is streamflow at the watershed outlet. The boundary
is defined by the watershed divide and extends vertically upward and downward to horizontal
planes.
Drainage basins, catchments, and watersheds are three synonymous terms that refer to the
topographic area that collects and discharges surface streamflow through one outlet or mouth.
Catchments are typically referred to as small drainage basins but no specific area lirruts have
been established. The drainage basin divide, watershed divide, or catchment divide is the line
dividing land whose drainage flows toward the given stream from land whose drainage flows
away from that stream. Think of drainage basin sizes ranging from the Mississippi River
drainage basin to a small urban drainage basin in your local community or some small valley
in the countryside near you.
Precipitation Evaporation
Atmospheric
water Interception
Transpiration
Surface Runoff to streams

water and ocean
~
-r-------~-----~~-~r-tfi~ltr_m_io_n~~--~~~----~~~----~~~
Subsurface
water
und- Figure 1.5.2. Block-diagram representation of the global hydrologic system. 22

out-
leric
lgh-
sur- 1.5.2 The Groundwater System in the Hydrologic Cycle
lters Groundwater constitutes one portion of the earth's water circulatory system known as the
tysi- hydrologic cycle. Figure 1.5.1 illustrates some of the many facets involved in this cycle. Water-
, the bearing formations of the earth's crust act as conduits for transmission and as reservoirs for
rgy. storage of water. Water enters these formations from the ground surface or from bodies of sur-
~ms:
face water, after which it travels slowly for varying distances until it returns to the surface by
n as action of natural flow, plants, or humans. The storage capacity of groundwater reservoirs com-
1ed, bined with small flow rates provide large, extensively distributed sources of water supply.
and Groundwater emerging into surface stream channels aids in sustaining streamflow when sur-
lary face runoff is low or nonexistent. Similarly, water pumped from wells represents the sole
ntal water source in many regions during much of every year.
Practically all groundwater originates as surface water. Principal sources of natural
the recharge include precipitation, streamflow, lakes, and reservoirs. Other contributions, known
uth. as artificial recharge, occur from excess irrigation, seepage from canals, and water purposely
ave applied to augment groundwater supplies. Even seawater can enter underground along coasts
line where hydraulic gradients slope downward in an inland direction. Water within the ground
JWS
moves downward through the unsaturated zone under the action of gravity, whereas in the sat-
.ver urated zone it moves in a direction determined by the surrounding hydraulic situation.
lley Discharge of groundwater occurs when water emerges from underground. Most natural
discharge occurs as flow into surface water bodies, such as streams, lakes, and oceans; flow to
the surface appears as a spring. Groundwater near the surface may return directly to the
atmosphere by evaporation from within the soil and by transpiration from vegetation. Pumpage
from wells constitutes the major artificial discharge of groundwater.
In this section we will discuss in general some of the aspects of the movement of ground-
water in the hydrologic cycle. The groundwater flow system comprises the subsurface water,
the geologic (porous) media containing the water, the flow boundaries, the sources (outcrop
areas, streams for recharge to the aquifer), and the sinks (springs, interaquifer flow, and wells (
for flow from the aquifer). Water flows through and is stored within the groundwater system. Ce1
Under natural conditions, the travel time of groundwater can range from less than a day to
more than a million years. The age of the water can range from recent precipitation to water
trapped in sediments that were deposited in geologic time. Chapter 3, Groundwater Movement,
discusses the mechanics of groundwater movement.
The water that is withdrawn from a groundwater system initially comes from storage. The
effects of the withdrawal from storage are propagated through the system, over time, as water
heads (the water table) decrease at greater distances from the point of withdrawal. Ultimately,
the effect of the withdrawal reaches a boundary such as a stream. At the stream, either
increased recharge (water added) to the groundwater system occurs or increased discharge
from the system occurs. Figure 1.5.3 shows the sources of water supplying pumpage from ten
major regional aquifer systems in the United States. The figure illustrates the variability of
aquifer response to long-term pumping and the extent to which changes in recharge and dis-
charge can exceed changes in storage. It is important to quantify recharge, despite the difficulty
of that undertaking.
Typically, most water from precipitation that infiltrates does not become recharge, but
instead is stored in the soil zone and is eventually returned to the atmosphere by evaporation
and plant transpiration. 2 The percentage of precipitation that becomes diffuse recharge is
highly variable and depends upon many factors, such as depths to the water table, properties
of surface soils, aquifer properties, and many other factors.
Interactions of surface-water systems with groundwater systems depend upon many fac-
tors, including positions of the surface-water systems relative to the groundwater systems;
characteristics of the surface-water systems and their underlying materials; and the climate set-
°
ting.2· 8 Figure 1.5.4 illustrates the effect of transient recharge from precipitation on the con-
figuration of a water table and the associated groundwater flow. The exchange of water across
the interface between surface water and groundwater can result from downstream movement
of water in and out of streambeds and banks, as illustrated in Figure 1.5.5. Other exchanges
result from tides, wave action, filling or draining of reservoirs, and transpiration from vegeta-
tion at the edges of wetlands and other surface water sources. Most studies of exchanges have
focused on streams.
Flows within groundwater systems can be on a local, intermediate, and regional basis, as
illustrated in Figure 1.5.6. The recharge and discharge areas in a local system of groundwater
flow are adjacent to each other. The recharge and discharge in an intermediate groundwater
flow system are separated by one or more topographic high and low. In regional groundwater Fig
flow systems, recharge areas are along groundwater divides and discharge areas are located at sho
the bottom of major drainage divides. Not every aquifer has each of these types of flow sys- terr.
tems.33· 71 In an aquifer system, the largest amount of groundwater flow is commonly in the Sat
sys
local flow systems which are mostly affected by seasonal variations in recharge. Recharge
is a
areas of these local systems make up the largest part of the surface of a drainage basin, are rel-
gro
atively shallow, and have transient conditions. Regional groundwater flow systems are less (rec
transient than local and intermediate flow systems. lea]
A conceptual model of an aquifer system, as illustrated in Figure 1.5.6 for the Midwest- retr
em Basins and Arches aquifer system (see Figure 1.5. 7 for location), is a simplified qualitative sys·
• the
Jage
md-
ater,
:rop
tells
tern.
y to
·ater
1ent,
The
ater
:ely,
ther
rrge
ten
r of
dis-
ulty
? 2ji0
but Miles
~on
e is
ties
fac-
ms;
set-
on-
Southern Aorldan
oss High Plains
.ent
ges Edwards-Trinity Southeastern
Gulf Coastal Plain Coastal Plaln
~ta-
ave
Increased recharge Decreased discharge
, as • Induced recharge D
1ter
!ter
!ter
• Infiltration of irrigation
water •
Decreased storage
Figure 1.5.3. Sources of water that supply withdrawals from major aquifer systems in the United States are highly variable, as
l at shown by these results from model simulations for various periods (Johnston, 1997). The Aoridan and Edwards-Trinity aquifer sys-
ys- tems, which equilibrate rapidly after pumping, were simulated as steady-state with no long-term change in storage. In contrast, the
the Southern High Plains (with most natural discharge occurring far from pumping wells) and the deeply buried Great Plains aquifer
rge system have had substantial changes in groundwater storage. The distinction between changes in recharge and changes in discharge
is a function of how the system was defined (i.e., a gain to one system may result in a Joss from an adjoining system). For example,
-el-
groundwater withdrawals from confined aquifers (Northern Atlantic Coastal Plain, Gulf Coastal Plain) can cause flow to be diverted
ess
(recharged) into the deeper regional flow regime that would otherwise discharge to streams in the outcrop areas or cause vertical
leakage across confining units. Groundwater recharge in a region can be increased as a result of human modifications, such as
:st- return flow of excess irrigation water (California Central Valley). Note that the areal extent of the Southeastern Coastal Plain aquifer
ive system overlaps the areal extents of the Aoridan and Gulf Coastal Plain aquifer systems. 2
description of the physical system. 33 Conceptual models may include a description of the
aquifers and confining units that make up the aquifer system, the boundary conditions, flow
regimes, sources and sinks of water, and general directions of the groundwater flow.
Conditions before recharge event- Some groundwater

discharges to surface water
\
I
~------------------------------~ A.
Initial effects of recharge event- Water table mounds near

surface-water bodies where infiltration distance is smallest;
groundwater discharge to surface water Increases
later effects of recharge event- Eevated water table mimics

- ·· land surface; groundwater discharge to surface water remains
elevated as compared to pre-recharge conditions
~------------------------------~ c
Conditions after effects of recharge event- Groundwater

discharge to surface water decreases
t
~----------------------------~ n
Direction of groundwater flow- Ught arrows are associated with shallow,

transient groundwater flow systems. Heavy arrows are associated with
more stable groundwater flow systems
_____:sz__ Water level
Figure 1.5.4. Diagrams showing the effect of transient recharge from precipitation on the configuration of a water table and associ- Figur
ated groundwater flow. 33 with 1
:he
)W
Figure 15.5. Local geomorphic features such as streambed topography, streambed roughness, mean-
dering, and heterogeneities in sediment hydraulic conductivities can give rise to localized flow sys-
tems within streambeds and banks. The near-stream subsurface environment with active exchange
between surface water and groundwater is commonly referred to as the hyporheic zone, although the
transition between groundwater and surface water represents a hydrologic continuum, preventing a
precise separation.2
Not to scale
D Glacial deposits - Local groundwater flow path
• Carbonate-rock aquifer •-- Intermediate groundwater flow path
• Basal confining unit - Regional groundwater flow path
- Indicates flow simulated by the regional groundwater
..o<-- flow model constructed for this investigation
Figure 1.5.6. Diagrammatic conceptual model of the Midwestern Basins and Arches aquifer system showing flow paths associated
with local, intermediate, and regional flow systems71 and flow systems simulated by the regional groundwater flow model. 33
Lake
Michigan
Boundary of major surface-water

drainage basin
Figure 1.5.7. Midwestern Basin and Arches aquifer system in Parts of Indiana, Ohio, Michigan, and Ulinois.33
1.6 HYDROLOGIC BUDGET

A hydrologic budget, water budget, or water balance is a measurement of continuity of the
flow of water, which holds true for any time interval and applies to any size area ranging from
local-scale areas to regional-scale areas or from any drainage area to the earth as a whole. The
hydrologists usually must consider an open system, for which the quantification of the hydro-
ie
Stream
inflow qn
Figure 1.6.1. Components of hydrologic

to stream cycle in an open system: the major inflows
and outflows of water from a parcel of land.
Groundwater Source: W.M. Marsh and J. Dozier,
inflow
Landscape: An Introduction to Physical
Geography. Copyright© 1986 by Wiley.
Reprinted by permission of John Wiley &
Sons, Inc.
logic cycle for that system becomes a mass balance equation in which the cbange of storage
of water (dS!dt) with respect to time within that system is equal to the inputs (/) to the system
minus the outputs (0) from the system.
Considering the open system in Figure 1.6.1, the water balance equation can be expressed
for the surface water system and the groundwater system in units of volume per unit time sep-
arately, or for a given time period and area, in depth.
Surface Water System Hydrologic Budget

p + Qin - Qoul + Qg- Es- Ts - I =/lSs, (1.6.1)
where P is the precipitation, Qin is the surface water flow into the system, Q001 is the surface
water flow out of the system, Q8 is the groundwater flow into the stream, Es is the surface evap-
oration, Ts is the transpiration, I is the infiltration, and !1Ss is the change in water storage of the
surface water system.
Groundwater Sysum Hydrologic Budget

I+ Gin- Goul- Qg -Eg- Tg = !lSg, (1.6.2)
where Gin is the groundwater flow into the system, G0111 is the groundwater flow out of the sys-
tem, and ASg is the change in groundwater storage. The evaporation, Eg, and the transpiration,
Tg, can be significant if the water table is near the ground surface.
Sysum Hydrologic Budget

The system hydrologic budget is developed by adding the above two budgets together:
f the (1.6.3)
from
.The Using net mass exchanges, the above system hydrologic budget can be expressed as
tdro- P-Q-G-E-T=AS (1.6.4)
(a)
Oi•charge Figure 1.6.2. Diagrams illustrating water

budgets for a groundwater system for
predeve]opment and development condi-
tions.3 (a) ?redevelopment water-budget
diagram illustrating that inflow equals
outflow. (b) Water-budget diagram show-
ing changes in flow for a groundwater
system being pumped. The sources of
(b) Pumpa!je water for the pumpage are changes in
recharge, discharge, and the amount of
'nCiease in Decrease n
recharge <fiScllargo water stored. The initial predevelopment
values do not directly enter the budget
calculation.
Hydrologic budgets can be used for numerous studies related to groundwater including
• Estimating groundwater exchange with lakes. 32, 39· 46· 60. 63, 64, n
• Estimating surface water and groundwater interaction. 38· 42· 62
• Computing recharge from a well-hydrograph data. 39
A diagram illustrating water budgets for a groundwater system for predevelopment and
development conditions is presented in Figure 1.6.2. A groundwater budget for part of Nassau
and Suffolk Counties, Long Island, New York, is shown in Figure 1.6.3. Both of these water
budgets assume equilibrium conditions with little or no change in storage.
EXAMPLE 1.6.1 During 1996, the water budget terms for Lake Annie in Florida60 included precipitation (P) of 43 inch/yr,
evaporation (E) of 53 inch/yr, surface water inflow (Q;n) of 1 inch/yr, surface outflow (Q001) of 173
inch/yr, and change in lake volume (65') of -2 inch/yr. Determine the net groundwater flow (the ground-
water inflow minus the groundwater outflow).
Fi!
SOLUTION Assuming T8 =0, the water budget equation (1.6.4) to define the net groundwater flow for the lake is Is);
bUI
G = L1S- P + E- Qin + Qou1
;;;: -2- 43 +53 - 1 + 173
=180 inch/yr •
EXAMPLE 1.62 During January 1996, the water-budget terms for Lake Annie in Florida60 included precipitation (P) of 1.9
inch, evaporation (E) of 1.5 inch, surface water inflow (Q;n) of 0 inch, surface outflow (Q0u1) of 17.4 inch,
and change in lake volume (AS) of 0 inch. Determine the net groundwater flow for January 1996 (the
groundwater inflow minus the groundwater outflow).
SOLUTION The water budget equation to define the net groundwater flow for the lake is
G =M- P + E- Qin + Qoul = 0-1.9 + 1.5-0 + 17.4 = 17 inch for January 1996 •
OVERALLPREDEVELOPMENT GROU NO-WATER PREDEVELOPMENT

WATER-BUDGET ANALYSIS WATER-BUDGET ANALYSIS
--,
INFLCNHO LONG ISLAND CUBIC INFlOW TO LONG ISLAND CUBIC
HYDROLOGIC SYSTEM FEET PER GROUND-WATER SYSTEM FEET PEA
water SECOND SECOND
)[
1. Precipitation I 2,475 7. Ground-water recharge ! 1,275
I i
OUTFLOW FROM lONG ISLAND I OUTFLOW FROM LONG ISLAND
>ndi- I
tdget
HYDROlOGIC SYSTEM I GROUND-WATER SYSTEM
I
2. EvapotransplratiQo ol precipjteUon 1.175 8. GIDIJOd-water dlsdwge to slreams 500
tls 3. GrO\Jnd-water dlsd\all1e to saa 725 I 9. Ground-water d"!Scllarge lo sea 725
how- 4. Stteamllow disdlarge to sea 525 I 10. Evapotranspiration ol ground water 25
5. Evapotransplratl011 of ground water 2S 11. Sprlllg now 25
:er 6. Sprill!l How 25 Tolafoutftow 1,275
Jf Total outflow 2,475
m
of
nent
~et
1t and
ass au
water
tch/yr,
lf 173
·ound-
Figure 1.6.3. Groundwater budget for part of Nassau and Suffolk Counties, Long Island, New York. 3 Block diagram of Long
.e is Island, New York, and tables listing the overall water budget and groundwater budget under predevelopment conditions. Both water
budgets assume equilibrium conditions with little or no change in storage.
The components of hydrologic budgets are either measured, calculated, or estimated.

• Evaporation, for example, may be obtained from measured pan evaporation data or calculated
from the energy balance method, the aerodynamic method (such as the Thornthwaite-Holz-
man equation), or a combination of these methods (such as the Priestley-Taylor evaporation
method). Detailed discussions of these methods are presented in References 22, 49, and 52.
of 1.9
Precipitation data are measured or is obtained from recorded data, and in some cases are esti-
~inch,
'6(the
mated based upon recorded data for other time periods of interest. Depending upon the time
period of the budget, average monthly values could also be used. In the case when the hydro-
logic budgets are of lakes, lake volume changes are needed and they can be computed using
actual lake-stage measurements and relationships between lake stage and lake volume. Using
estimates of lake volume changes, evaporation, and precipitation, estimates of net groundwa-
• ter flow to a lake can be made. The individual components of net groundwater flow are the
groundwater inflow to the lake and the leakage from the lake, which can be determined
through groundwater simulation models. Groundwater simulation models (Chapter 9), how-
ever, need to be calibrated, which requires the monitoring of groundwater levels (hydraulic
heads, Chapter 2), lithographic data (Chapter 12), results (hydraulic conductivities) from pump
tests (Chapter 4), or slug tests (Chapter 5). With a calibrated groundwater flow model, the
groundwater flow into a lake and the leakage from a lake can be detennined using simulated 3()
groundwater flow fields.
CASE STUDY Lake Five-0, Florida
Grubbs 39 described the hydrologic budgetary analysis that was tion model was used to determine the groundwater inflows and
performed for Lake Five-0 (a seepage lake), located in Bay seepage from the lake. The simulation model was developed using
County in northwestern Florida (see Figures 1.6.4 and 1.6.5). This lithographic data to define the three geohydrologic units, a net-
hydrologic budget (Table 1.6.1 and Figure 1.6.6) was determined work of monitoring wells to define the hydraulic heads over the
using both collected data and model simulation results (see Chap- time of the hydrologic budget, previously published data, and lim-
ter 9). Lake Five-0 is located in an area characterized by coastal ited slug tests to help determine hydraulic conductivities. The
terrace deposits that have been modified by extensive karst devel- analysis made it possible to develop quantitative estimates of min-
opment. Soils in the area are deep, excessively drained, and con- imum groundwater inflow and leakage rates not only during the
sist of very permeable, Lakeland series sands. Maximum lake study period, but also for long-term average conditions.
depth ranged from 13.5 m to 15.4 m. Surface area ranged from The hydrologic budget for Lake Five-0 is expressed as
10.4 to 11.3 hectares, and the lake volume ranged from 9.09 x 105
to 1.11 x 106 m3.
The net groundwater flow (groundwater inflow minus leak-
age) was determined by the water balance approach utilizing The hydrologic budget by Grubbs39 showed that the ground-
measurements of precipitation, lake evaporation was determined water inflow to the lake and leakage from the lake to the ground-
using the energy budget method, and lake volume changes were water system are the dominant components, respectively, in total
estimated from lake-stage measurements and a relationship inflow (precipitation plus groundwater inflow) and total outflow
between lake stage and lake volume. The analysis was utilized to (evaporation plus leakage) budgets of the lake. The groundwater
make qualitative assessments of the significance of lake-ground- movement, including the head distribution and groundwater flow
water exchanges during the study period. A groundwater simula- near Lake Five-0, is discussed in Section 3.6.6.
30"30'
LOCATION OF 15'
BAY COUNTY :tv
A
:tv
ll
~~
30°00' +/ ll
('0 A
0
I
10 MILES
I
s
I I
10 KILOMETERS
0
0
J\
r::
Figure 1.6.4. Location of Lake Five-0 study area. 39
ned
&5"40'
ow-
ulic
Jmp
the
ated J0025'30"
' w31'
•''
.
"'
p:t2.1
p:t2.2
w26
.
w14',
''
/ttyBitty
Pond &
w6t\ ~r: w15 w29
•
and :ws.1
sing ''
net-
: the
''
''
\
.
w4
.w27
EXPLANATION
'
' w28•
lim- '
The A--A' LINE OF ', w16 Campers
SECTION \ Pond
nin- •"w32
: the
30"25'
+
UNIMPROVED ROAD \
STAFF GAGE
' .
I
I
I
)"
x ELEVATION REFERENCE
..
f '
POINT I
w4 •
.6.
WELL AND IDENTIFIER
CLIMATE STATION
.
'
'
I
'
md- w6.2<> CONTINUOUS RECORDER
''
Jnd- AND IDENTIFIER '
total 0 114 MILE ' ''
'' '
Figure 1.6.5. Location of data-
t1ow ' ''
0 0.25 KILOMETERS
'
'' collection sites in the Lake Five-0
1ater ' ' study area. 39
t1ow
Table 1.6.1 Monthly Net Groundwater Flow to Lake Five-0, 1989-9039
Standard
deviation (percent
Average lake Change in lake Net ground- Standard of net ground-
Month volume volume Precipitation Evaporation water flow deviation waterflow)
1989
Jan. 9891700 -26,100 3,800 3,200 -26,700 2,300 9
Feb. 9641000 -21,900 6,900 4,100 -24,700 2,600 10
Mar. 943,300 -13,900 17,700 7,900 -23,700 3,300 14
Apr. 935,200 -8,700 10,600 11,000 -8,300 2,800 34
May 922,500 -14,700 8,900 13,800 -9,900 2,900 29
June 945,600 69,100 56,800 12,000 24,400 9,300 38
July 1,024,800 66,700 26,300 13,300 53,700 5,400 10
Aug. 1,0681400 26,200 20,900 12,300 17,600 3,700 21
Sept. 1,0911800 22,300 21 ,500 16,800 17,700 3,700 21
Oct. 1,103,400 3,100 15,100 13,000 1,000 2,700 270
Nov. 1,108,200 4,100 13,800 9,400 -300 2,300 770
Dec. 1,102,700 1,000 13,500 7,900 -4,600 21300 50
(continues)
Table 1.6.1 (continued) Monthly Net Groundwater Flow to Lake Five-0, 1989-90
Standard
deviation (percent
Average lake Change in lake Net ground- Standard of net ground-
Month volume volume Precipitation Evaporation water flow deviation water flow)
1990
Jan. 1,107,000 -3,100 3,800 3,300 -3,500 1,500 43
Feb. 1,103,000 -4,800 5,700 4,800 -5,800 1,300 22
Mar. 1,092,000 -15,500 9,600 8,600 -16,500 2,000 12
Apr. 1,076,100 -17,800 12,000 11,100 -18,700 2,400 13
May 1,058,600 -22,000 9,400 15,100 -16,300 2,400 15
June 1,041,300 -9,100 23,700 14,300 -18,500 4,200 23
July 1,046,400 17,900 31,300 15,000 1,600 5,300 330
Aug. 1,055,600 -2,400 16,000 15,600 -2,800 3,600 130
Sept. 1,033,300 -35,600 3,900 17,900 -21,600 2,900 13
Oct. 998,600 -35,100 5,500 15,400 -25,200 2,800 11 1.7
Nov. 963,200 -35,600 800 11,300 -25,100 2,600 10
Dec. 932,000 -28,900 7,000 7,400 -28,500 2,600 9 1.7.
All units are in cubic meters, unless otherwise noted. Standard deviation is the error component of the net groundwater flow estimate. Negative
values of net groundwater flow indicate that leakage exceeded groundwater inflow.
• GROUNDWATER INFLOW
~ 80,000 0 PRECIPITAnON
w
Iii
...•• •• • •
60,000
• • •• •
::E
u 40,000 ~ •• •• ••
iii 0
~
:::l 20,000
o
°0 C De D
o
cc co ooccoc oc d
w 0 ------------ - ----------~
~
:::l
...J
~ ·20,000
-10,000 ~~L...J.....L...J.....~~~~~..........~~
100,000 ,.......,..........,.....,....,,.....,....,,.....,....,,.....,....,......,.....,......,.....,......,.....,......,.....,r-r->......,.....,
e LEAKAGE
encr eo.ooo D EVAPORATION
~ 60,000 :a • • • • • • • • • • • • • • ••••••.
~
u 40,000 •••
m
:::l 20,000
3
w 1.7.;
~
3 ·20,000
~ .
-10,000
100,000
• RESIDUAL
U) 80,000 c LAKE VOLUME CHANGE
cr cc
~
~
60,000
u 40,000
m
::> 00
g 20,000 c
w
::il
:I
.J
0
>
Figure 1.6.6. Monthly hydrologic budget compo-
nents for Lake Five-0, 1989-90.39
1.7 Publication Sources 27
The components of a hydrologic budget, whether they are measured or calculated, have
associated errors. These errors are based on the degree of uncertainty of the measurements,
cent
limitations of methods, and the assumptions made to calculate the values. In many situations
d- information on the rainfall, surface outflow, and withdrawal for supply may be the most reli-
) able. Calculated values of lake evaporation and lake storage may be less reliable because off-
site pan evaporation and estimated pan coefficients are used. Lake stage data may be less
reliable because of the uncertainty in assessing the surface area of the lake. Groundwater flow
is typically the least reliable. When the measured or calculated components are used in the
budget calculations, imbalances between the inflow and outflow components, called residuals,
occur. The residual term in a hydrologic budget is an accumulation of all the errors in the com-
ponents of the budget. The previous hydrologic budget equations do not reflect residual terms;
however, the analysis by Grubbs 39 describes the residuals in detail.
1.7 PUBLICATION SOURCES

1.7.1 Internet Resources
ative
The Internet has changed the availability of sources of information on groundwater hydrology.
Now we can access many types of data and publications very rapidly through the use of a com-
puter. Appendix A provides a list of U.S. government and nongovernmental organizations'
Web sites.
Much of this book has been developed to serve as a portal to the vast resources on ground-
water hydrology that now exist on the Internet. Hopefully this book will help guide the student,
the professional, and the researcher to the general documents, program plans, field project
details, software, and other information found on the Internet. Many of the new figures and
tables in this edition have been taken from various Web sites, in particular the U.S. Geologi-
cal Survey site. These items not only provide specific information but also serve as samples of
graphics and tabulations available on the Internet, making the style of this book more variable
than that of most textbooks. The majority of Web sites referenced in this book are maintained
by government agencies and established organizations, and therefore should be accessible for
many years to come. Several end-of-chapter problems throughout the book are based upon
Internet exercises.
1.7.2 U.S. Geological Survey Publications

In the United States, a majority of the field measurements and investigations of groundwater
have been conducted by the U.S. Geological Survey (USGS). Most work has been on a coop-
erative basis with individual states. Results are published by the Survey as circulars, digital
data series, fact sheets, hydrologic atlases, open-file reports, techniques of water-resources,
water data reports, water-resources investigation reports, professional papers, and water-sup-
ply papers. Table 1.7.1 describes the various types of U.S. Geological Survey reports, and
Table 1.7.2 describes how to find and reference on-line USGS water resources reports. Since
1935 records of groundwater measurements in key observation wells have been published in
water-supply papers under the title Groundwater Levels in the United States. The U.S. Geo-
logical Survey publishes at irregular intervals other papers on the geology and groundwater
0- resources of local areas. Invariably, these intensive investigations concern areas containing
important groundwater problems and are carried out in cooperation with local agencies.
Table 1.7.1 Types of U.S. Geological Survey Reports
Circulars-Purpose: To present to general or scientific audiences short summaries or articles of short-term, popular, or local interest.
Digital Data Series-The Digital Data Series encompasses a broad range of digital data, including computer programs, interpreted
results of investigations, comprehensive reviewed databases, spatial data sets, digital images and animation, and multimedia presen- l
tations that are not intended for printed release. Scientific reports in this series cover a wide variety of subjects and facets of U.S. 0
Geological Survey investigations and research that are of lasting scientific interest and value. Releases in the Digital Data Series
Sl
offer access to scientific information that is available in digital form; the information is primarily for viewing, processing, and (or)
analyzing by computer.
Fact Sheets-Purpose: To describe Water Resources Division (WRD) programs, projects, products, and water-resources topics to
T
either a general or professional audience. Water Fact Sheets are concise and timely publications that increase the understanding and
visibility of WRD activities and accomplishments.
Hydrologic Atlases-Purpose: To present reports on hydrology or geohydrology in map format to a wide range of hydrologically ori- ~
ented audiences. th
Open-File Reports-Purpose: To make available (1) data reports, (2) reports preliminary findings that would be of interest to few
persons other than the cooperating agency, (3) reports and maps pending publication elsewhere but requiring immediate release, and
(4) timely information describing programs, projects, products, and water-resources topics.
Professional Papers-Purpose: To present comprehensive or topical reports on any field in the earth sciences. This series is com-
monly used for summaries of wide popular, scientific, or geographic interest, and for significant scientific contributions-generally
1.
on topics other than hydrology.
Techniques of Water-Resources-Purpose: To present to technically oriented audiences reports on methods and techniques used in
collecting, analyzing, and processing hydrologic data.
Water Data Reports-A series of annual reports that document hydrologic data gathered from the U.S. Geological Survey's and
cooperating agencies' surface and groundwater data-collection networks in each state, Puerto Rico, and Trust Territories. These
records of streamflow, groundwater levels, and water quality provide the hydrologic information needed by state, local, and federal
agencies, and the private sector for developing and managing our Nation's land and water resources.
Water-Resources Investigations Reports-Purpose: To (a) present to interdisciplinary audiences comprehensive or topical interpre-
tive reports and maps that are mainly of local or short-term interest; (b) provide a medium of release for reports and maps that would
not be feasible in any other series or journal or that would be published quickly.
Water-Supply Papers-Purpose: To present significant interpretive results of hydrologic investigations of broader than local interests.
Water Errata Sheets-Changes made to reports after publication.
Source: http://water.usgs.gov/pub.html
Table 1.7.2 How to Find and Reference Online USGS Water Resources Reports
Many USGS reports on water resources topics are now being served online. You can access them by their series and number. For
example, to see Circular 1123, you should enter CIR 1123 in the search box at the following URL: http://water.usgs.gov/pubs.
Constructing a reference
Use one of these prefixes for the report series, followed by the report number. Dashes and underlines are permitted (and ignored), but
blanks are not allowed. Case is insensitive.
Report Series Preferred Prefix Alternate Prefix(es) Allowed
Fact Sheet FAC FS FACTFS-

Open-File Report OFR OF
Water-Resources Investigation WR1 WRWRIR
Professional Paper PRO PROFPP
Water-Data Report WDR WDDATA 1.1
Circular CIR CIRCC
Water-Supply Paper WSP ws
Bulletin BUL BULL
Techniques of Water-Resources Investigations TWRI TWR1
1.8 Data Sources 29
Table 1.7.2 (continued) How to Find and Reference Online USGS Water Resources Reports
rest. Referencing parts of a document
Using only the series and number will reference the home page (index.html) of the document. You also can reference a specific part
~n-
of a document in html format, even if it's not on the home page. For example, to find the section named HRD4 in a file of Circular
1123 called overview.html, use this URL: http://pubs.water.usgs.gov/cir1123/overview.html#HDR4. This technique will work for any
sub-page or figure of the html document.
~)
The "pubs.water.usgs.gov" reference is persistent!
The USGS is committed to supporting this referencing system for the indefinite future. This means you can safely incorporate a
nd "pubs.water.usgs.gov" reference in your Web pages and eve_n in your printed documents and it will still work many years later.
When viewing a document, your browser's location may show you another URL that corresponds to the current physical location of
ori-
the document. Do not use this physical location as a "persistent" reference! As our system grows, these locations will change but the
"pubs.water.usgs.gov" reference will not.
md Source: http://water.usgs.gov/pubs/referencing.html
1. 7.3 Publications
ly
The following journals provide articles on various topics of groundwater:
Environmental Science and Technology, American Chemical Society
Ground Water, National Ground Water Association
Groundwater Management, Water Well Journal Publishing Co.
il Ground Water Monitoring and Remediation, Groundwater Publishing Company
Hydrological Science and Technology, American Institute of Hydrology
:-
uld Journal of the American Water Resources Association, American Water Resources
Association
:sts. Journal of Contaminant Hydrology, Elsevier Scientific Publishers
Journal of Hydraulics, American Society of Civil Engineers
Journal of Hydrology, Elsevier Scientific Publishers
Journal ofWater Resources Planning and Management, American Society of Civil Engineers
There have been many books published on the subject of groundwater. Earlier books include
References 4, 5, 7, 17, 19, 23, 26,29-31,40, 41,43, 56, 58, 59, 61, 67, 70, 73, and 74.
Over the past three decades there are several previous books that have been published on
groundwater including Applied Hydrogeologl 5; Aquifer Hydraulics 8; Contaminant Hydroge-
ologl4; Groundwater36 ; Ground Water Contamination: Transport and Remediation 14 ;
but Groundwater Engineering45 ; Groundwater Hydrology 16 ; Groundwater Hydrologl 9; Ground-
water Hydraulics and Pollutant Transpor?-0 ; Groundwater Mechanics 66 ; Groundwater Sci-
ence36; Groundwater Systems Planning and Management77 ; The Handbook of Groundwater
Engineering 21 ; Hydraulics ofGroundwater10 ; Manual ofApplied Field Hydrogeology76 ; Mod-
eling Groundwater Flow and Pollution 11 ; Principles of Groundwater Engineering 15 ; and
Quantitative Hydrogeology. 51
_ 1.8 DATA SOURCES

Table 1.8.1 lists the principal types of data and data compilations that are required for the
analysis of groundwater systems. The lists are for the physical framework, hydrologic budg-
ets and stresses, and the chemical framework. One of the sources of data used frequently by
groundwater hydrologists is the U.S. Geological Survey NWIS system.
Table 1.8.1 Principal Types of Data and Data Compilations Required for Analysis of Groundwater Systems
Physical framework
Topographic maps showing the stream drainage network, surface-water bodies, landforms, cultural features, and locations of struc-
tures and activities related to water
Geologic maps of surficial deposits and bedrock
Hydrogeologic maps showing extent and boundaries of aquifers and confining units
Maps of tops and bottoms of aquifers and confining units
Saturated-thickness maps of unconfined (water-table) and confined aquifers
Average hydraulic conductivity maps for aquifers and confining units and transmissivity maps for aquifers 1.8.2
Maps showing variations in storage coefficient for aquifers
Estimates of age of groundwater at selected locations in aquifers
Hydrologic budgets and stresses
Precipitation data
Evaporation data
Streamflow data, including measurements of gain and loss of streamflow between gaging stations
Maps of the stream drainage network showing extent of normally perennial flow, normally dry channels, and nonnally seasonal flow
Estimates of total groundwater discharge to streams
Measurements of spring discharge
Measurements of surface-water diversions and return tlows
Quantities and locations of interbasin diversions
History and spatial distribution of pumping rates in aquifers
Amount of groundwater consumed for each type of use and spatial distribution of return flows
Well hydrographs and historical head (water-level) maps for aquifers
Location of recharge areas (areal recharge from precipitation, losing streams, irrigated areas, recharge basins and recharge wells),
and estimates of recharge
Chemical framework
Geochemical characteristics of earth materials and naturally occurring groundwater in aquifers and confining units
Spatial distribution of water quality in aquifers, both areally and with depth
Temporal changes in water quality, particularly for contaminated or potentially vulnerable unconfined aquifers
Sources and types of potential contaminants
Chemical characteristics of artificially introduced waters or waste liquids
Maps of land covernand use at different scales, depending on study needs
Streamflow quality (water-quality sampling in space and time), particularly during periods of low flow
Source: Alley et al. 3
1.8.1 NWISWeb Data for the Nation

The U.S. Geological Survey is the principal federal water data agency in the United States. The
USGS collects and disseminates about 70 percent of the water data currently being used by
numerous state, local, private, and other federal agencies to develop and manage water
resources. The National WATer Data STOrage and REtrieval System (WATSTORE) was estab-
lished in 1972 to provide an effective and efficient means for the processing and maintenance
of water data collected through the USGS and to facilitate release of the data to the public. In
1976, the USGS opened WATSTORE to the public for direct access.
The U.S.G.S. National Water lnfonnation System (NWIS) has replaced WATSTORE and
is referred to as NWISWeb (http://waterdata.usgs.gov/nwis). NWIS is a very large collection of
data and infonnation on the water resources of the United States. This database contains current
and historical water data from more than 1.5 million locations across the nation. The data cate-
1.8 Data Sources 31
gories are real-time data, site information, surface water data, groundwater data, and water qual-
ity. Real-time data includes data transmitted from selected groundwater sites. Site infonnation
includes descriptive site infonnation, including latitude, longitude, weH depth, aquifer, and site
I
IC- use. The groundwater database includes groundwater site inventory, groundwater level data, and
water-quality data. The groundwa1er site inventory consists of more than 850,000 records of
wells, springs, test holes, tunnels, drains, and excavations in the United States. The groundwater
data can be obtained at http://waterdata.usgs.gov/nwis/gw.
1.8.2 Real-Time Data

Real-time groundwater data include data that are automatically collected, transmitted, and made
available to the public at least once a day according to the U.S.G.S. (Fact Sheet 090-01, Decem-
ber 2001). These data can be transmitted by land-lined telephone, cellular telephone, land-
based radio frequency (RF) technology, satellite telemetry, or a combination of these
technologies. Within the U.S.G.S., satellite telemetry is the most common method for real-time
data transmission. Water levels are the most conunon data transmitted in real time by the USGS.
Figure 1.8.1 illustrates a real-time data collection and transmission system. With this method,
flow
water-level data are recorded by a data-collection platform (DCP) (see Figure 1.8.2) and trans-
mitted, often on a four-hour schedule, by satellite telemetry to a U.S.G.S. ground station. The
data are then displayed at http://water.usgs.gov/nwislgw.
Real-time data have many inherent advantages over data collected and distributed by tra-
ditional means, including timeliness, data quality, data availability, and cost. Additional infor-
mation on real-time groundwater data can be obtained at http://water.usgs.gov/nwis/gw or
from the following address: Office of Groundwater, U.S. Geological Survey, 411 National
Center, 12201 Sunrise Valley Drive, Reston, Virginia 20192, 703-648-5001.
),
.The
:d by
Nater
:stab-
lance
ic. In
~and
on of
mnt
cate- Figure 1.8.1. Real-time data collection and transmission system.25
1. Ac
v. 50,
2. AI
storai
3. AI
waler
gov/p
4. Ar
Engrn
5. Ar
ua!M
6. Ba
reg arc
ca/U1
7. Ba
U.S. (
8. Ba
9. Be
York,
10. B
11. B
Reide.
12. B
Geogt
Figure 1.8.2. Multiple sensor data-collection platform (DCP) installation in Kansas.25 13. B·
v. 16,
14. B·
nation
1994.
15. B
348 Pf
16. B(
17. 81
480pi
PROBLEMS 18. a
19. 0
1.6.1 Using the data for LakeFive-0 in Table 1.6.1, show that the 1.7.3 Perform a search of U.S .G.S. publications for the topic Wiley
net groundwater flow for December 1989 is --4,600 cubic meters. "hydrologic budget and water budget." To perform the search, go 20. Cl
1.6.2 Using the data for Lake Five-0 in Table 1.6.1, show that the to http://usgs-georef.cos.com. How many publications are listed? Prenth
net groundwater flow for December 1990 is -28,500 cubic meters. 1.7.4 Perform a search of U.S.G.S . publications for the High 21. Cl
1.6.3 Detennine cumulative net groundwater flow for Lake Five- Plains Aquifer. To perform the search, go to http://usgs-geo NewY
0 during 1989. ref.cos.com. How many publications are listed? 22. Cl
1.7.5 Perform a search of U.S.G.S. publications for the Edwards McGn
1.6.4 Determine cumulative net groundwater flow for Lake Five-
Aquifer. To perform the search, go to http://usgs-georef.cos.com. 23. Cc
0 during 1990. NewY
How many publications are listed?
1.7.1 Perform a search of the U.S. Geological Survey online pub- 24. C1
lications (including circulars, fact sheets, open-file reports, pro- 1.7.6 Perform a search of U.S.G.S. publications for the topic of 27-44,
fessional papers, water-resources investigation reports, and water karst terrains. To perform the search go to http://usgs-geo
25. Ct
supply papers) to detennine what studies, if any, have been per- ref.cos.com. How many publications are listed? u.s.a
formed on the regional aquifer system closest to where you live. 1.7.7 Go to the site http://water.usgs.gov/software and obtain a 26. o~
1. 7.2 Develop an inventory of wells in the county where you live list of the groundwater software that the U.S.G.S. has available. Sons, I
using the USGS data sources for your state. Select a well that has 1.7.8 Write a description of the U.S.G.S. Ground-Water Resources 27. Dt
a time history of water levels and print the hydrograph. Program. Use the site http://water.usgs.gov/ogw/GWRP.htmJ. Press, :
References 33
REFERENCES
1. Adams, F. D., Origin of springs and rivers-an historical review, Fennia, 28. De Villiers, M., Water: The Fate of Our Most Precious Resource,
v. 50, no. 1, 16 pp., 1928. Mariner Books, Houghton Mifflin, Boston, 2000.
2. Alley, W. M., R. W. Healy, J. W. LaBaugh, and T. E. Reilly, Flow and 29. DeWiest, R. J. M., Geohydrology, John Wiley & Sons, New York, 366
storage in groundwater systems, Science, v. 206, June 14, 2002. pp., 1965.
3. Alley, W. M., T. E. Reilly, and 0. L. Franke, Sustainability of Ground- 30. DeWiest, R. J. M. (ed.), Flow through Porous Media, Academic, New
water Resources, U.S. Geological Survey circular 1186, http://water.usgs. York, 530 pp., 1969.
gov/pubs/circ/circ1186, U.S. Geological Survey, Denver, CO, 1999. 31. Domenico, P. A., Concepts and Models in Groundwater Hydrology,
4. Amer. Soc. Civil Engrs. (ASCE), Ground Water Management, Manual McGraw-Hill, New York, 405 pp., 1972.
Engmg. Practice 40, New York, 216 pp., 1972. 32. Duwelius, R. F., Hydrologic data and hydrologic budget for Summit
5. American WaterWorks Assoc. (AWWA), GroundWater,AWWAMan- Lake Reservoir, Henry County, east-central Indiana, water years 1989 and
ual M21, New York, 130 pp., 1973. 1990, 1993.
6. Baker, M. N., and R. E. Horton, Historical development of ideas 33. Eberts, S. M., and L. L. George, Regional Groundwater Flow and
regarding the origin of springs and ground-water, Trans. Amer. Geophysi- Geochemistry in the Midwestern Basins and Arches Aquifer System in
cal Union, v. 17, pp. 395-400, 1936. Parts of Indiana, Ohio, Michigan, and Illinois, U.S. Geological Survey
7. Baldwin, G. V., and C. L. McGuinness, A Primer on Ground Water, Professional Paper 1423-C, 2000.
U.S. Geological Survey, 26 pp., 1963. 34. Fetter, C. W., Contaminant Hydrogeology, 2nd edition, Prentice Hall,
8. Batu, V., Aquifer Hydraulics, Wiley Interscience, New York, 1998. Upper Saddle River, NJ, 1999.
9. Bear, J., Dynamics of Fluids in Porous Media, Amer. Elsevier, New 35. Fetter, C. W., Applied Hydrogeology, Prentice Hall, Upper Saddle
York, 1972. River, NJ, 2001.
10. Bear, J., Hydraulics of Groundwater, McGraw-Hill, New York, 1979. 36. Fitts, C. R., Groundwater Science, Academic Press, San Diego, 2002.
11. Bear, J. and A. Verruijt, Modeling Groundwater Flow and Pollution, 37. Freeze, R. A., and J. A. Cherry, Groundwater, Prentice Hall, Engle-
Reidel, Dordrecht, The Netherlands, 1987. wood Cliffs, NJ, 604 pp., 1979.
12. Beaumont, P., Qanats on the Varamin Plain, Iran, Trans. Inst. British 38. Gronberg, J. A.M., and K. R. Belitz, Estimation of a Water Budget for
Geographers, Publ. no. 45, pp. 169-179, 1968. the Central Part of the Western San Joaquin Valley, California, U.S. Geo-
13. Beaumont, P., Qanat systems in Iran, Bull. Inti. Assoc. Sci. Hydrology, logical Survey Water-Resources Investigation, WRI-91-4192, 1992.
v. 16,pp. 39-50,1971. 39. Grubbs, J. W., Evaluation of Groundwater Flow and Hydrologic Bud-
get for Lake Five-0, A Seepage Lake in Northwestern Florida, U.S. Geo-
14. Bedient, P. B., H. S. Rifai, and C. J. Newell, Ground Water Contami-
logical Survey Water-Resources Investigations Report 94-4145, 1995.
nation: Transport and Remediation, Prentice Hall, Englewood Cliffs, NJ,
1994. 40. Harr, M. E., Groundwater and Seepage, McGraw-Hill, New York,
315 pp., 1962.
15. Biswas, A. K., History of Hydrology, Amer. Elsevier, New York,
348 pp., 1970. 41. Heath, R. C., and F. W. Trainer, Introduction to Groundwater Hydrol-
ogy, John Wiley & Sons, New York, 284 pp., 1968.
16. Bouwer, H., Groundwater Hydrology, McGraw-Hill, New York, 1978.
42. Hedman, E. R., and Jorgenson, Suiface- and Ground-water Interac-
17. Bureau of Reclamation, Ground Water Manual, U.S. Dept. Interior,
tion and Hydrologic Budget of the Missouri River Valley Aquifer Between
480 pp., 1977.
Yankton, South Dakota and St. Louis, Missouri, U.S. Geological Survey
18. Carson, R., The Silent Spring, Houghton Mifflin, Boston, 1962. Hydrologic Investigations Atlas, HA-0721, 1990.
19. Cedergren, H. R., Seepage, Drainage, and Flow Nets, 2nd ed., John 43. Huisman, L., Groundwater Recovery, Winchester Press, New York,
opic Wiley & Sons, New York, 534 pp., 1977.
1, go
336 pp., 1972.
20. Charbeneau, R. J., Groundwater Hydraulics and Pollutant Transport, 44. Johnston, R. H., Hydrologic Budgets of Regional Aquifer Systems of
ted? Prentice Hall, Upper Saddle River, NJ, 2000. the United States-Predevelopment and Development Conditions, U.S.
-:ligh 21. Chow, V. T. (ed.), Handbook of Applied Hydrology, McGraw-Hill, Geological Survey Professional Paper No. 1425, 1997.
-geo New York, 1453 pp., 1964. 45. Kashef, A. A. I., Groundwater Engineering, McGraw-Hill, New York,
22. Chow, V. T., D. R. Maidment, and L. W. Mays, Applied Hydrology, 1986.
·ards McGraw-Hill, New York, 1988. 46. Lee, T. M., and A. Swancar, The Influence of Evaporation, Ground-
;om. 23. Collins, R. E., Flow of Fluids through Porous Materials, Reinhold, water and Uncertainty in the Hydrologic Budget of Lake Lucerne, A Seep-
New York, 270 pp., 1961. age Lake in Polk County, Florida, U.S. Geological Survey Water-Supply
24. Cressey, G. B., Qanats, karez and foggaras, Geogr. Review, v. 48, pp. Paper 2439, 1997.
ic of
27-44, 1958. 47. Lightfoot, D. R., The origin and diffusion of qanats in Arabia: New
-geo
25. Cunningham, W.L., Real-Time Ground-Water Data for the Nation, evidence from the northern and southern peninsula, Geogr. Jour., v. 166,
U.S. Geological Fact Sheet 090-01, Dec. 2001. pp.215-226,2000.
rin a 26. Davis, S. N., and R. J. M. DeWiest, Hydrogeology, John Wiley & 48. MacK.ichan, K. A., Estimated use of water in the United States, 1955,
)le. Sons, New York, 463 pp., 1966. Jour. Amer. Water Works Assoc., v. 49, pp. 369-391, 1957.
rrces 27. Delleur, J. W., ed., The Handbook of Groundwater Engineering, CRC 49. Maidment, D. R., ed., Handbook of Hydrology, McGraw-Hill, New
Press, Boca Raton, FL, 1999. York, 19~3.
50. Marsh, W. M., and J. Dozier, Landscape: An Introduction to Physi-

cal Geography, John Wiley & Sons, New York, 1986.
66. Strack, 0. D. L., Groundwater Mechanics, Prentice Hall, Engle-
woods Cliffs, NJ, 1989.
C.
51. Marsily, de G., Quantitative Hydrogeology, Academic Press, New 67. Thomas, H. E., The Conservation of Ground Water, McGraw-Hill,
York, 1986. New York, 327 pp., 195 I.
52. Mays, L. W. (ed.), Water Resources Handbook, McGraw-Hill, New 68. Thurner, A., Hydrogeologie, Springer, Vienna, 350 pp., 1967.
York, 1996. 69. Todd, D. K., Groundwater Hydrology, 2nd ed, John Wiley & Sons,
53. Mays, L. W., Water Resources Engineering, John Wiley & Sons, New York, 1980.
New York, 2001. 70. Tolman, C. F., Ground Water, McGraw-Hill, New York, 593 pp., 1937.
54. Meinzer, 0. E., The history and development of groundwater hydrol- 71. Toth, J., A theoretical analysis of groundwater in small drainage
ogy, loUJ: WashingtonAcad. Sci., v. 24, pp. 6-32, 1934. basins, Jour. Geophys. Res., v. 68, pp. 4795-4812, 1963.
55. Murray, C. R., and E. B. Reeves, Estimated Use ofWater in the United 72. Trommer, J. T., M. J. De!Charco, and B. R. Lewelling, Water Budget
States in 1975, U.S. Geological Survey Circular 765, 39 pp., 1977. and Water Quality of Ward Lake, Flow and Water-Quality Characteristics
56. Muskat, M., The Flow of Homogeneous Fluids through Porous . of the Braden River Estuary, and the Effects ofWard Lake on the Hydro-
Media, McGraw-Hill, New York, 763 pp., 1937. logic System,· West-Central Florida, U.S. Geological Survey Water-
57. Perrault, P., On the Origin of Springs, trans. by A. LaRocque, Hafner, Resources Investigations Report 98-4251, 1999.
New York, 209 pp., 1957. 73. Verruijt, A., Theory of Groundwater Flow, Gordon and Breach, New
58. Polubarinova-Kochina, P. Y., Theory of Groundwater Movement, York, 190 pp., 1970.
Princeton Univ. Press, Princeton, NJ, 613 pp., 1962. 74. Walton, W. C., Groundwater Resource Evaluation, McGraw-Hill,
59. Raudkivi, A. 1., and R. A. Callander, Analysis of Groundwater Flow, New York, 664 pp., 1970.
John Wiley & Sons, New York, 214 pp., 1976. 75. Walton, W. C., Principles of Groundwater Engineering, Lewis Pub-
60. Sacks, L.A., A. Swancar, and T. M. Lee, Estimating Groundwater lishers, Chelsea, MI, 1991.
Exchange with Lakes Using Water-Budget and Chemical Mass-Balance 76. Weight, W. D., and J. L. Sonderegger, Manual of Applied Field
Approaches for Ten Lakes in Ridge Areas of Polk and Highlands Coun- Hydrogeology, McGraw-Hill, New York, 2001.
ties, Florida, U.S. Geological Survey Water-Resources Investigations 77. Willis, R., and W. W-G. Yeh, Groundwater Systems Planning and
Report, WRI-98-4133, 1998. Management, Prentice Hall, Englewood Cliffs, NJ, 1987.
61. Scheidegger, A. E., The Physics of Flow through Porous Media, 3rd 78. Winter, T. C., Uncertainties in estimating the water balance of lakes,
ed., Univ. ofToronto, Toronto, 353 pp., 1974. Water Resources Bull., v. 17, pp. 82-115, 1981.
62. Shade, P. J., Water Budget for the Lahaina District, Island of Maui, 79. Winter, T. C., The interactions of lakes with variably saturated 2.1
Hawaii, U.S. Geological Survey Water-Investigations Report, WRI-96- porous media, Water Resources Research, v. 19, pp. 1203-1218, 1985.
4238, 1996.
80. Winter, T. C., J. W. Harvey, 0. L. Franke, and W. M. Alley, Ground-
63. Skrobialowski, S. C. and M. J. Focazio, Hydrologic Characteristics water and Surface Water: A Single Source, U.S. Geological Survey Cir-
and Water Budgets for Swift Creek Reservoir, Virginia, 1996, U.S. Geo- cular 1139, http://water.usgs.gov/pubs/circ/circl139, Denver, CO, 1998.
logical Survey Open-File Report 97-0229, 1997. 81. Wulff, H. E., The qanats oflran, Sci. Amer., v. 218, pp. 94-100, 105,
64. Skrobialowski, S.C., Hydrologic Characteristics and Water Budget 1968.
for Swift Creek Reservoir, Virginia, 1997, U.S. Geological Survey Water-
Resources Investigations, WRI-98-4122, 1998.
65. Solley, W. B., Preliminary Estimates of Water Use in the United
States, I995, U.S. Geological Survey Open-File Report 97-0645, Reston,
VA, 1997.
EXAMPLE PUBLICATIONS OF ORGANIZATIONS AND GOVERNMENT AGENCIES

American Water Works Association, AWWA Standard for Disinfection of U.S. Environmental Protection Agency, Office of Drinking Water, Citi-
Water Wells, ANSJJAWWA C654, Denver, CO. zen's Guide to Ground-Water Protection, Washington, DC, 1990.
American Water Works Association, AWWA Standard for Water Wells, U.S. Environmental Protection Agency, Office of Drinking Water, Guide
ANSIIAWWAAlOO, Denver, CO. to Ground-Water Supply Contingency Planning for Local and State
American Water Works Association, Manual 21, Groundwater, Denver, Governments, Washington, DC, 1991.
CO, 1989. U.S. Environmental Protection Agency, Office of Drinking Water, Pro-
Borch, M.A., S. A. Smith, and L. N. Noble, Evaluation and Restoration tecting Local Ground-Water Supplies Through Wellhead Protection,
of Water Supply Wells, American Water Works Association and Ameri- Washington, DC, 1991.
can Water Works Association Research Foundation, Denver, CO, 1993.
U.S. Environmental Protection Agency, Office of Drinking Water, Local
Financing for Wellhead Protection, Washington, D. C., 1989.
:ngle- Chapter2
-Hill,
Sons,
1937.
inage Occurrence of Groundwater
udget
·istics
"ydro-
later-
.New To describe the occurrence of groundwater necessitates a review of where and how ground-
water exists; subsurface distribution, in both vertical and areal extents, needs to be considered.
-Hill, The geologic zones important to groundwater must be identified, as well as their structure in
terms of water-holding and water-yielding capabilities. If hydrologic conditions furnish water
Pub-
to the underground zone, the subsurface strata govern its distribution and movement; hence the
important role of geology in groundwater hydrology cannot be overemphasized. Springs,
Field
hydrothermal phenomena, and water in permanently frozen ground constitute special ground-
? and water occurrences.
lakes,
rated 2.1 ORIGIN AND AGE OF GROUNDWATER

>85.
Almost all groundwater can be thought of as a part of the hydrologic cycle, including surface
'Jund-
and atmospheric (meteoric) waters. Relatively minor amounts of groundwater may enter this
1 Cir-
1998. cycle from other origins.

• 105, Water that has been out of contact with the atmosphere for at least an appreciable part of
a geologic period is termed connate water; essentially, it consists of fossil interstitial water that
has migrated from its original burial location. 62 This water may have been derived from
oceanic or freshwater sources and, typically, is highly mineralized. 20 Magmatic water is water
derived from magma; where the separation is deep, the term plutonic water is applied, while
volcanic water designates water from relatively shallow depths (perhaps 3 to 5 km). 61 New
water of magmatic or cosmic origin that has not previously been a part of the hydrosphere is
referred to as juvenile water. And finally, metamorphic water is water that is or has been asso-
ciated with rocks during their metamorphism. The diagram in Figure 2.1.1 illustrates the inter-
Citi-
relations of these genetic types of groundwater.
The residence time of water underground has always been a topic of considerable specu-
"Juide lation. But with the advent of radioisotopes, determinations of the age of groundwater have
State become possible. Hydrogen-3 (tritium) and carbon-14 are the two isotopes that have proved
most useful. Tritium with a half-life of 12.33 years is produced in the upper atmosphere by
Pro- cosmic radiation; carried to earth by rainfall and hence underground, this natural level of tri-
~tion,
tium begins to decay as a function of time, such that
(2.1.1)
where A is the observed radioactivity, A0 is the activity at the time the water entered the aquifer,
'A is the decay constant, and tis the age of the water. Carbon-14 has a half-life of 5,730 years
and is also produced at an established constant level in the atmosphere. This isotope is present
in groundwater as dissolved bicarbonate originating from the biologically active layers of the
35
36 Chapter 2 Occurrence of Groundwater
Juvenile water
("new" water)
Magmatic water
(in or from magma;
plutonic if deep;
volcanic if shallow)
2.2
Metamorphic water
Meteoric water
(is or has been
Oceanic water (in or recently from
with rocks during
the atmosphere)
metamorphism)
Connate water
("fossil" water
generally in unmeta-
morphosed rocks)
Figure 2.1.1. Diagram illustrating relationships of genetic types of water62 (courtesy The Geological
Society of America, 1957).
soil where C0 2 is generated by root respiration and the decay of humus. 59• 66 Tritium i~ appli-
cable for estimating groundwater residence times of up to 50 year~, while carbon-14 spans the
age bracket of a several hundred to about 50,000 years.
Applications of the age-dating techniques have revealed groundwaters ranging in age
from a few years or less to many thousand years.* Measurements of water samples taken from
deep wells in deserts of the United Arab Republic and Saudi Arabia indicate ages of 20,000 to
30,000 years. 35 This period is compatible with the Wisconsin ice age, when these desert areas
last possessed a high rainfall capable of recharging the underlying major aquifers.
2.2 ROCK PROPERTIES AFFECTING GROUNDWATER

2.2.1 Aquifers
Groundwater occurs in many types of geologic formations; those known as aquifers are of
most importance. An aquifer may be defined as a fonnation that contains sufficient saturated
penneable material to yield significant quantities of water to wells and springs. 35 This implies
an ability to store and to transmit water; unconsolidated sands and gravels are a typical exam-
ple. Furthennore, it is generally understood that an aquifer includes the unsaturated portion of
the penneable unit. Synonyms frequently employed include groundwater reservoir and water-
bearing formation. Aquifers are generally areally extensive and may be overlain or underlain
*The fallout of bomb tritium and C-14 in precipitation since the advent of nuclear weapon testing in 1952 has
greatly complicated much of the dating of groundwater because recent levels greatly exceed the prebomb level.
by a confining bed, which may be defined as a relatively impermeable material stratigraphi-

cally adjacent to one or more aquifers. Clearly, there are various types of confining beds; the
following types are well established in the literature:
1. Aquiclude-A saturated but relatively impenneable material that does not yield apprecia-
ble quantities of water to wells; clay is an example.
2. Aquifuge-A relatively impenneable fonnation neither containing nor transmitting water;
solid granite belongs in this category.
3. Aquitard-A saturated but poorly permeable stratum that impedes groundwater movement
and does not yield water freely to wells, that may transmit appreciable water to or from
adjacent aquifers and, where sufficiently thick, may constitute an important groundwater
storage zone; sandy clay is an example. 45 *
2.2.2 Porosity
Those portions of a rock or soil not occupied by solid mineral matter can be occupied by
groundwater. These spaces are known as voids, interstices, pores, or pore space. Because inter-
stices serve as water conduits, they are of fundamental importance to the study of groundwater.
Typically, they are characterized by their size, shape, irregularity, and distribution. Original
interstices were created by geologic processes governing the origin of the geologic formation
and are found in sedimentary and igneous rocks. Secondary interstices developed after the rock
was formed; examples include joints, fractures, solution openings, and openings formed by
plants and animals. With respect to size, interstices may be classed as capillary, supercapillary,
and subcapillary. Capillary interstices are sufficiently small that surface tension forces will hold
water within them; supercapillary interstices are those larger than capillary ones; and subcapil-
lary interstices are so small that water is held primarily by adhesive forces. Depending on the
;al
connection of interstices with others, they may be classed as communicating or isolated.
The porosity of a rock or soil is a measure of the contained interstices or voids expressed
as the ratio of the volume of interstices to the total volume. If a is the porosity, then
pli-
the
a=-=--
vv ~ -~
(2.2.1)
V ~
age
~om
where Vv is the volume of interstices (voids), Vs is the volume of solids, and Vis the total vol-
0 to ume (bulk volume). Porosity may also be expressed by
~eas
a= Pm -pd = 1_& (2.2.2)
Pm Pm
where Pm is the density of mineral particles (grain density) t and Pd is the bulk density.
The term effective porosity refers to the amount of interconnected pore space available for
fluid flow and is expressed as a ratio of interconnected interstices to total volume. For uncon-
solidated porous media and for many consolidated rocks, the two porosities are identical.
~of
Porosity may also be expressed as a percentage by multiplying the right-hand side of Equa-
1ted
tions 2.2.1 or 2.2.2 by 100. The terms primary and secondary porosity are associated with orig-
lies
inal and secondary interstices, respectively.
:un-
n of
*The word aquifer can be traced to its Latin origin. Aqui- is a combining fonn of aqua ("water") and -fer comes
ter- fromferre ("to bear"). Hence, an aquifer is a water-bearer. The suffix -elude of aquiclude is derived from the Latin
lain claudere ("to shut or close"). Similarly, the suffix -fuge of aquifuge comes fromfugere ("to drive away"), while the
suffix -tard of aquitard follows from the Latin tardus ("slow").
tThe density of solid rock varies with the type of mineral. For alluvium where quartz is the predominant mineral, a
value of 2.65 glcm3 is typical; limestone and granite fall in the range 2.7-2.8 g/cm3, and basalt can approach 3.0 glcm3.
EXAl
(a) (b) (c)
SOLV
(d) (e) (f)
Figure 2.2.1. Examples of rock interstices and the relation of rock texture to porosity. (a} Well-sorted
sedimentary deposit having high porosity. (b) Poorly sorted sedimentary deposit having low porosity.
(c) Well-sorted sedimentary deposit consisting of pebbles that are themselves porous, so that the
deposit as a whole has a very high porosity. (d) Well-sorted sedimentary deposit whose porosity has
been diminished by the deposition of mineral matter in the interstices. (e) Rock rendered porous by
solution. (j) Rock rendered porous by fracturing.42
Figure 2.2.1 shows several types of interstices and their relation to porosity. In tenns of
groundwater supply, granular sedimentary deposits are of major importance. Porosities in
these deposits depend on the shape and arrangement of individual particles, distribution by
size, and degree of cementation and compaction. In consolidated fonnations, removal of min-
eral matter by solution and degree of fracture are also important. Porosities range from near
zero to more than 50 percent, depending on the above factors and the type of material. Repre-
sentative porosity values for various geologic materials are listed in Table 2.2.1. It should be
recognized that porosities for a particular soil or rock can vary considerably from these values.
In sedimentary rocks subject to compaction, measurements show that porosity decreases
with depth of burial.37 Thus, a typical relation has the fonn
(2.2.3)
Table 2.2.1 Representative Values of Porosity (after Morris and Johnson45)
Material Porosity, percent Material Porosity, percent

Gravel, coarse 28" Loess 49
Gravel, medium 32a Peat 92
Gravel, fine 34a Schist 38
Sand, coarse 39 Siltstone 35
Sand, medium 39 Claystone 43
Sand, fine 43 Shale 6
Silt 46 Till, predominantly silt 34
Clay 42 Till, predominantly sand 31
Sandstone, fine grained 33 Tuff 41
Sandstone, medium grained 37 Basalt 17
Limestone 30 Gabbro, weathered 43
Dolomite 26 Granite, weathered 45
Dune sand 45
"These values are for repacked samples; all others are undisturbed.
where az is the porosity at depth z, a0 is the porosity at the surface, a is a constant, and e is the
base of Naperian logarithms.
An undisturbed sample of a medium sand weighs 484.68 g. The core of the undisturbed sample is 6 em
in diameter and.l 0.6 ~ em high. The sample is oven-dried for 24 hr at 11 ooc to remove the water content.
At the end of the 24 hr, the core sample weighs 447.32 g. Determine the bulk density, void ratio, water
content, porosity, and saturation percentage of the sample.
SOLUTION: The dry weight of the sample is Wd =447.32 g and the total weight is Wr = 484.68 g. The total volume
of the undisturbed sample is
rted
The bulk density is defined as the density of solids and voids together, after drying. Thus,
[ty.
_ wd _ 447. 32 g _
lS Pd- - 1 3
3 - 1.491 gem
~ 300cm
y
Assuming quartz is the predominant mineral in the sample, then Pm = 2.65 g/cm3
Thus, the volume V5 of the solid phase of the sample is
s of
s in v = wd = 447.32 g 168.8 cm3
1 by s Pm 2.65 g/cm 3
nin-
Thus, the total volume of voids in the sample is
rlear
pre-
j be
ues. With this information, we can calculate the void ratio e of the sample is
ases
e- Vv - 131.2 cm3 0.777
- ~ - 168.8 cm 3
2.3)
The volumetric water content of a sample is the volume of the water divided by the volume of the sample
484 68 447 32
· g- 3 · g llg/cm 3 =0.1245g/cm 3 =0.125
300cm
;ent
where Ww is the total weight of the undisturbed sample before drying.
The gravimetric water content of the sample is
8 w = Wr- Wd X 100 = 484.68 g-447.32 g X 100 = 8.3 5%

wd 447.32 g
The porosity of the sample is
a=~-~ x 100 = 300 cm3 -168.8 cm3 x 100 = 43.73%

3
~ 300cm
Finally, the saturation percentage of a sample is defined as the percentage of the pore space that is filled
by water,
8 (0.1245)
~x100= ( ) x100=28.47%
a 0.4373
•
-EXAMPLE 2.2.2 The void ratio of an unconsolidated clay sample is L 19. Detennine the porosity of the sample.
SOLUTION Using the definition of the void ratio of an undisturbed sample, e = Vv, and substituting Vv = V1 - V5,
h the vm'd ratio
ten . ts ~ - ~ ~ 1~ -~= 1+e.
. e=--=-- vs
~ vs vs
Substituting this into the porosity equation, we obtain
v.-v
a=-
~
5
~
v]1+e 1+e
[ 1]
- - xlOO= [1-_£ xlOO= 1 - - xlOO=-xlOO
1 e
Thus, the porosity of the sample is

e 1.19
a = - X 100 = - - X 100 = 54.34%
1+e 1+1.19
•
EXAMPL;E 2.-2.1 The porosity of a quartz sand sample is 38.41%. Detennine the bulk density of the sample.
, c. ~"'"'· , .,""~· ..Ji •... ~
SOLUTION The bulk density and porosity of an undisturbed sample are defined as pd = Wd and a = ~ - Vs x 100 ,
~~~ ~ ~
Substituting the dry weight of a sample Wd = Pm V 5
into the bulk density expression, we have
Pd =Wd- =Pm~ - - andtheporostty. .ts
~ ~
a= ~ - ~ X100 = [1- ~]X 100 ~ ~ = 1- ~
~ VI ~ 100
Using the bulk density expression then yields Pd = Pm ~ = Pm[1-~].

~ 100
For quartz sand, Pm = 2.65 g/cm3, the bulk density is
2.65 g/cm l[1- ~~ ] =1.63 g/cm

~
10 ] 38 1
1
3 3
Pd =p,.[1- =(
•
2.2.3 Soil Classification
Unconsolidated geologic materials are normally classified according to their size and distri-
bution. A commonly employed system based on particle, or grain, size is listed in Table 2.2.2.
Evaluation of the distribution of sizes is accomplished by mechanical analysis. This involves
sieving particles coarser than 0.05 mm and measuring rates of settlement for smaller particles
in suspension. Results are plotted on a particle-size distribution graph such as that shown in
Figure 2.2.2. The percentage finer scale on the ordinate shows the percentage of material
smaller than that of a given size particle on a dry-weight basis. 2.2.4
The effective particle size is the 10 percent finer than value (d10). The distribution of par-
ticles is characterized by the uniformity coefficient Uc as
(2.2.4)
where d60 is the 60 percent finer than value. A uniform material has a low unifonnity coeffi-
cient (the dune sand in Figure 2.2.2), while a well-graded material has a high unifonnity coef-
ficient (the alluvium).
Table 2.2.2 Soil Classification Based on Particle

Size (after Morris and Johnson 45 )
Material Particle size, mm
Clay <0.004
Silt 0.004- 0.062
Very fine sand 0.062-0.125
Fine sand 0.125-0.25
Medium sand 0.25-0.5
Coarse sand 0.5- 1.0
Very coarse sand 1.0-2.0
Very fine gravel 2.0-4.0
Fine gravel 4.0-8.0
• Medium gravel
Coarse gravel
8.0-16.0
16.0-32.0
Very coarse gravel 32.0-64.0
x100, 80
e have
~ 60
~
cQ)
eQ) 40
a..
20
0~~----~--~--~--~--~--~----~--~--~--~
.001 .002 .005 .010 .020 .050 .1 00 .200 .500 1.00 2.00 5.00
• Particle size, mm
Figure 2.2.2. Particle-size distribution graph for three geologic samples (data from U.S. Geological
Survey).
distri- The texture of a soil is defined by the relative proportions of sand, silt, and clay present in
2.2.2. the particle-size analysis. This can be expressed by the soil-textural triangle in Figure 2.2.3.
volves Note, for example, that a soil composed of 30 percent clay, 60 percent silt, and 10 percent sand
rticles constitutes a silty clay loam.
1wn in
:tterial
2.2.4 Porosity and Representative Elementary Volume
,f par- We can define porosity in a spatial field as a function, a(x), at any point x using spatial averag-
ing over a representative elementary volume (REV). Consider a porous medium with different
2.2.4) sizes of averaging volumes, V*, centered at point i. The point value of porosity (volumetric
porosity) is associated with an averaging volume centered at that point, expressed as 14
:oeffi-
coef- 1
a(x)=- J x(x-.Y)d.Y
V * V*(x)
(2.2.4)
60 ~ '0
Percent sand
Figure 2.2.3. Triangle of soil textures for describing various combinations of sand, silt, and clay
(after Soil Survey Staff50).
where X(i) is an indicator function for void space: X(i) =0 if point i is located in the solid
space or X(i) = 1 if point x is located in the void space. V* is the averaging volume centered
on the point .X, with the estimated porosity value as a function of averaging volume size. If the
volume is too small, the average value is not well defined and the limit approaches either 0 or
1. There may be a range of volumes surrounding pointi for which the average is well defined,
and if the averaging volume is too large, then soils with different textures may be included,
resulting in a deviation from the average.
SOL[
2.2.5 Specific Surface
The water retentive property of a soil or rock is markedly influenced by its surface area. This area
depends on particle. size and shape and on the type of clay minerals present. The term specific
surface refers to the area per unit weight of the material, usually expressed as m2/g. Relative
methods for measuring specific surface are based on retention of a polar organic molecule such
as ethylene glycol; these have been related to absolute values derived from statistical calculations
of surface area. 6 Clay particles contribute the greatest amount of surface area in unconsolidated
formations. Nonswelling clays such as kaolinite have only an external surface and exhibit spe-
cific surfaces in the range of 10-30 m2/g; however, swelling clays such as montmorillonite and
vermiculite have internal and external surfaces that yield specific surface values near 800m2/g.
An illustration of the importance of particle size to specific surface is presented in Table
2.2.3. Here, considering only uniform spheres, it can be seen that when a given volume is
transformed into 100 small spheres totaling the same volume, the specific surface increases by
a factor of 100. Furthermore, it can be shown that when the volume is deformed into rod, disk,
or plate shapes, specific surface increases even more. 4
Table 2.2.3 Relation of Surface Area to Particle Size for Uniform Spheres
Diameter of Soil Number of Total surface

particle, mm classification particles per cm3 area, cm 2
10 Medium gravel 3.14
3
1 Coarse sand 1 X 10 31.4
0.1 Very fine sand 1 X 106 314
0.02 Silt 1.25 X 108 1,570
0.002 Clay 1.25 X 10 11 15,700
Note: Rectangular packing is assumed in a cubic container 1 em on a side so that
the total volume, and weight, of spheres remains constant at rt/6 cm3.
Using the tabulated results of a grain size distribution test on a field sample, perform the following tasks:
(a) Prepare a grain size distribution curve for this sample.
(b) Is this a well-graded or poorly graded sample?
(c) Classify the sample using Table 2.2.2.
(d) What would be reasonable porosity values for this sample?
U.S. Standard Sieve Number Mass retained (g)

3/8 49.95
4 26.70
8 25.29
16 50.58
30 72.57
e solid
40 25.50
~ntered
100 33.60
:. If the
ter 0 or 200 7.53
ei1ned, Pan (passes through #200 sieve) 8.28
;luded, Total sample weight 300.00
SOLUTION (a) The given data are analyzed as shown in the table below. Note that the partic1e size (sieve opening)
corresponding to each U.S. Standard Sieve number is given in the table. The results yield the grain-
tis area size distribution curve shown in Figure 2.2.4.
pecific
elative Sieve Grain size (mm) Mass retained (g) Percent finer by mass
.e such 3/8 9.5 49.95 83.35
lations 4 4.75 26.70 74.45
lidated 8 2.36 25.29 66.02
·it spe- 16 1.18 50.58 49.16
[te and 30 0.6 72.57 24.97
m2/g. 0.425 25.50 16.47
40
Table 33.60 5.27
100 0.15
tme is
200 0.075 7.53 2.76
ses by
Pan <0.075 8.28
, disk,
Total sample weight 300
90
80
..cf-""'
/
70
r/)
Cll
m
E 60 I v
>-
.c
...Q) 50 /
c ll'
-= 40
E
Q)
...a> 30
0
a.. 20
......... ~
10 Figure 2.2.4.
..cr"' Grain-size distribu-
0
0.01 0.1 10 tion curve for
Particle size, mm Example 2.2.4.
(b) From the grain-size distribution curve:
t4o= 1.6 mm and d 10 :0.23 rom

From Equation 2.2.4, the uniformity coefficient is
U = dfiJ ::;: 1.6 mm ,. 7

c d 10 0.23 mm
Since Uc > 6, the sample can be described as well graded (i.e., low unifonnity).
{c) The percentage of clay and silt in the sample is approximately 2-3 percent, while about 60 percent
of the sample is sand. The remaining 37- 38 percent is composed of very fine to coarse gravel.
{d) The porosity of the sample could be somewhere between 20 and 35 percent based on our classifica-
tion in part (c). •
The following data are obtained from a hydrometer test to determine the gradation of a sample of fine 2.3
sediments.
{a) Prepare a grain size distribution curve for this sample.
(b) Is this a well-graded or poorly graded sample?
(c) Classify the sample using Table 2.2.2.
(d) What would be reasonable porosity values for tills sample?
Particle grain size (mm) Percent finer than

1.000 100.00
0.700 98.90
0.500 88.50
0.350 49.40
0.250 10.30
0.175 0.90
0.125 0.00
0.088 0.00
0.0625 0.00
SOLUTION (a) Using the given data. the grain size distribution curve is plotted for this sample in Figure 2.2.5.
2.3 Vertical Distribution of Groundwater 45
100 ~
90 I /
80 I
70 I
I
~
Q; J
c 60
~
c 50 I I
Q)
e 40 I
I
Q)
a..
30
20 L
10 /
stribu-
0 ,__ ~ Figure 2.2.5. Grain-
0.100 1.000 size distlibution curve
4. Particle size, mm for Example 2.2.5.
(b) From Figure 2.2.5, d60 =0.36 mm and d =0.25 mm, the uniformity coefficient is given by Equa-
10
tion 2.2.4:
Uc = d6o = 0.36 mm "" 1.44

d10 0.25 mm
Since Uc < 4, the sample can be classified as poorly graded (i.e., highly uniform).
(c) From the given data and the soil classifications of Table 2.2.2, the sample consists of 11.5 percent
coarse sand, 78.2 percent medium sand, and 10.3 percent fine sand. Thus, the sample can be classi-
~ercent
fied as a medium sand with some proportions of coarse and fine sand.
1.
(d) Based on the classification in part (c), the porosity of the sample would be in the range of 35 to 45
ssifica-
• percent. •
of fine 2.3 VERTICAL DISTRIBUTION OF GROUNDWATER

The subsurface occurrence of groundwater may be divided into zones of aeration and satura-
tion. The zone of aeration consists of interstices occupied partially by water and partially by
air. In the zone of saturation, all interstices are filled with water under hydrostatic pressure. On
most of the land masses of the earth, a single zone of aeration overlies a single zone of satu-
ration and extends upward to the ground surface, as shown in Figure 2.3.1.
In the zone of aeration, vadose water* occurs. This general zone may be further subdi-
vided into the soil water zone, the intermediate vadose zone, and the capillary zone (Fig-
ure 2.3.1). 14
The saturated zone extends from the upper surface of saturation down to underlying
impermeable rock. In the absence of overlying impermeable strata, the water table, or phreatic
surface, t forms the upper surface of the zone of saturation. This is defined as the surface of
atmospheric pressure and appears as the level at which water stands in a well penetrating the
aquifer. Actually, saturation extends slightly above the water table due to capillary attraction;
however, water is held there at less than atmospheric pressure. Water occurring in the zone of
saturation is commonly referred to simply as groundwater, but the term phreatic water is also
employed.
*vadose is derived from the Latin vadosus ("shallow").

I,
tPhreatic is derived from the Greekphrear, -atos ("a well").
Ground surface
t
Soil-water
zone
+
c
0
-~
Q)
ctl
15
T
Immediate
vadose
Q;
~
~
Q)
(/)
0
Q) "'0
c zone ctl
0 >
j_
N
Capillary
zone
Water table
'"} .r-D-·ll-:' ~~..15

Impermeable rock
2.4.3
Figure 2.3.1. Divisions of subsurface water.
2.4 ZONE OF AERATION

2.4.1 Soil-Water Zone
Water in the soil-water zone exists at less than saturation except temporarily when excessive
water reaches the ground surlace as from rainfall or irrigation. The zone extends from the
ground surface down through the major root zone. Its thickness varies with soil type and vege-
tation. Because of the agricultural importance of soil water in supplying moisture to roots, agri-
culturists and soil scientists have studied soil moisture distribution and movement extensively.
The amount of water present in the soil-water zone depends primarily on the recent expo-
sure of the soil to moisture. Under hot, arid conditions, a water-vapor equilibrium tends to
become established between the ambient air and the surlaces of fine-grained soil particles. As
a result, only thin films of moisture-known as hygroscopic water-remain adsorbed on the
surfaces. For coarse-grained materials and where additional moisture is available, water also
forms liquid rings surrounding contacts between grains, as sketched in Figure 2.4.1. This
water is held by surface tension forces and is sometimes referred to as capillary water. Tem-
porarily, the soil-water zone may contain water in excess of capillary water from rainfall or
irrigation; this gravitational water drains through the soil under the influence of gravity.
2.4.2 Intermediate Vadose Zone

The intermediate vadose zone extends from the lower edge of the soil-water zone to the upper
limit of the capillary zone (Figure 2.3.1). The thickness may vary from zero, where the bound-
Figure 2.4.1. Illustration of vadose water held at

contact points of particles in the unsaturated
zone.
ing zones merge with a high water table approaching ground surface, to more than I 00 m
under deep water table conditions. The zone serves primarily as a region connecting the zone
near ground surface with that near the water table through which water moving vertically
downward must pass. Nonmoving vadose water is held in place by hygroscopic and capillary
forces. Temporary excesses of water migrate downward as gravitational water.
2.4.3 Capillary Zone

The capillary zone (or capillary fringe) extends from the water table up to the limit of capil-
lary rise of water. If a pore space could be idealized to represent a capillary tube, the capillary
rise he (Figure 2.4.2) can be derived from an equilibrium between surface tension of water and
the weight of water raised. Thus,
2't
;essive he= -cos A (2.4.1)
ry
lm the
l vege- where 't is surface tension, yis the specific weight of water, r is the tube radius, and Ais the angle
i, agri- of contact between the meniscus and the wall of the tube. For pure water in a clean glass, A=0,
sively. and at 20° C, 't =0.074 g/cm and r= 1 g/cm3, so that for r in em, the capillary rise (in em) is
:expo-
nds to h = 0.15
e (2.4.2)
les.As r
on the
!r also It follows from Equation 2.4.2 that the thickness of the capillary zone will vary inversely
. This with the pore size of a soil or rock. Measurements of capillary rise in unconsolidated materi-
Tern- als shown in Table 2.4.1 bear out this relationship. Furthennore, for a material containing
fall or innumerable pores of a wide range in size, the upper boundary of the zone will fonn a jagged
limit when studied microscopically. Taken macroscopically, however, a gradual decrease in
'· water content results with height. That is, just above the water table almost all pores contain
capillary water; higher, only the smaller connected pores contain water; and still higher, only
the few smallest connected pores contain water lifted above the water table. This distribution
upper of water above the water table is shown in Figure 2.4.3 from a drainage test on a sand. The
ound- visual capillary rise is invariably less than the actual capillary zone as defined in Figure 2.4.3.
Intermediate
vadose
zone
j
I
Porosity -_j
I
I
I
I
I
Figure 2.4.2. Rise of I 2.4.4
I
water in a capillary tube.
Figure 2.4.3. Distribution
0 10 20 30 40 of water in a coarse sand
Percent moisture above the water table after
content by volume drainage (after Pril!47).
Table 2.4.1 Capillary Rise in Samples of Unconsolidated Materials (after Lohman34)
Material Grain size (rnm) Capillary rise (em)
Fine gravel 5-2 2.5

Very coarse sand 2-1 6.5
Coarse sand 1-0.5 13.5
Medium sand 0.5-0.2 24.6
Fine sand 0.2-0.1 42.8
Silt 0.1-0.05 105.5
Silt 0.05-0.02 20<f
Note: Capillary rise measured after 72 days; all samples have virtually the same porosity
of 41 percent
"Still rising after 72 days.
Assuming uniform spherical grains of cubic packing, approximate the capillary rise in a soil as a function
of the grain diameter. Using this relationship, estimate the capillary rise for each material in Table 2.4.1.
SOLUTION The accompanying figure shows a typical arrangement of cubical packing with spherical grains of equal
diameter. It can be shown, using geometry, that the radius of the pore space between the grains is 0.2
times the grain diameter.
Assuming that this pore space could be idealized to represent a capillary tube, r = 0.20D can be
. d mto
substltute . Equauon . lding hc O.l 5 = O.lS
. 2.4.2 yte =-,-_ D
3
D =
0 20 4
Material Grain size (mm) Estimated capillary Actual capillary

rise (em) rise (em)
Fine gravel 5-2 1.5-3.75 2.5
Very coarse sand 2-1 3.75-7.5 6.5
Coarse sand 1-0.5 7.5-15 13.5
Medium sand 0.5-0.2 15-37.5 24.6
Fine sand 0.2-0.1 37.5-75 42.8
Silt 0.1-0.05 75-150 105.5
Silt 0.05-0.02 150-375 200
•
2.4.4 Measurement of Water Content
mtion Determination of the water content of soils can be accomplished by various direct methods
;and based on removal of the water from a sample by evaporation, leaching, or chemical reaction,
! after followed by measurement of the amount removed. 1• 24 Thus, the gravimetric method involves
7). weighing a wet soil sample, removing the water by oven-drying it, and reweighing the sample.
Indirect methods consist of measuring some soil property affected by soil-water content.
Specifically, electrical and thermal conductivity and electrical capacitance of porous materials
vary with water content.
Another useful instrument for measuring soil moisture is the neutron probe. When low-
ered in a small-diameter tube in the ground, determination of soil moisture can be made as a
function of depth. The instrument contains a radium-beryllium source of fast neutrons and a
detector for slow neutrons. The fast neutrons are slowed by collisions with hydrogen, and
because most of the hydrogen in soil is associated with water, the intensity of slow neutrons
measured yields, after calibration, the local soil moisture content. 6
Within the vadose zone a negative-pressure head of water exists, often referred to as suc-
tion, or tension in a positive sense. This tension can be measured by a tensiometer; Figure 2.4.4
shows a tensiometer installed in a soil column. The depression M in water level measures the
local soil tension. Such instruments function in the range from atmospheric pressure (near
1,000 em of water) to about 200 em of water (800 em water tension). Calibration data for soil
suction and water content reveal that the relation between the two variables is not single valued;
Jnction instead, soil structure and compaction, as well as effects of wetting or drying, influence the
2.4.1. results. 6
.f equal
; is 0.2
Porous
cup ~::-:--~~~
Suction head,
ll:lh_.)(or tension
j
Unsaturated
Figure 2.4.4. Illustration of a
can be porous medium tensiometer for measuring water
Water tension in unsaturated porous
media.
2.4.5 Available Water

Soils absorb and retain water, which may be withdrawn by plants during periods between rain-
fall or irrigations. This water-holding capacity is defined by the available water, which is the
range of plant-available water, the moist end being the field capacity and the dry end the wilting
point. Field capacity can be defined as the amount of water held in a soil after wetting and after
subsequent drainage has become negligibly small. The negligible drainage rate is often assumed
after two days; however, different soils possess varying drainage rates so that quantitative values
may not be comparable. The wilting point defines the water content of soils when plants grow-
ing in that soil are reduced to a permanent wilted condition. Because factors such as soil type,
volume, plant type, and age influence wilting point, this moisture content can also be variable.
2.5 ZONE OF SATURATION

In the zone of saturation, groundwater fills all of the interstices; hence, the (effective) poros-
ity provides a direct measure of the water contained per unit volume. A portion of the water can
be removed from subsurface strata by drainage or by pumping of a well; however, molecular
and surface tension forces hold the remainder of the water in place.
2.5.1 Specific Retention

The specific retentionS, of a soil or rock is the ratio of the volume of water it will retain after
saturation against the force of gravity to its own volume. Thus,
s = w, (2.5.1)
r V:
t
where w, is the volume occupied by retained water* and Vt is the bulk volume of the soil or rock.
2.5.2 Specific Yield

The specific yield SY of a soil or rock is the ratio of the volume of water that, after saturation,
can be drained by gravity to its own volume. 17 Therefore,
w
sy =__1_
v. (2.5.2)
t
where wy is the volume of water drained. Values of S, and SY can also be expressed as percent-
ages. Because w, and wu constitute the total water volume in a saturated material, it is appar-
ent that SOLI
(2.5.3)
or
(2.5.4)
where all pores are interconnecting.
Values of specific yield depend on grain size, shape and distribution of pores, compaction
of the stratum, and time of drainage. 40 Representative specific yields for various geologic
materials are listed in Table 2.5.1; individual values for a soil or rock can vary considerably
from these values. It should be noted that fine-grained materials yield little water, whereas
coarse-grained materials pennit a substantial release of water and hence serve as aquifers. In
*It should be noted that the terms .field capacity and retained water refer to the same water content but differ by the
zone in which they occur.
2.5 Zone of Saturation 51
Table 2.5.1 Representative Values of Specific Yield (after Johnson 25 )

:n rain- Material Specific yield (percent)
1 is the
wilting Gravel, coarse 23
td after Gravel, medium 24
:sumed Gravel, fine 25
values Sand, coarse 27
grow- Sand, medi urn 28
ll type, Sand, fine 23
iable. Silt 8
Clay 3
Sandstone, fine grained 21
Sandstone, medium grained 27
poros- Limestone 14
:ercan Dune sand 38
ecular Loess 18
Peat 44
Schist 26
Siltstone 12
Till, predominantly silt 6
1 after
Till, predominantly sand 16
Till, predominantly gravel 16
Tuff 21
2.5.1)
general, specific yields for thick unconsolidated formations tend to fall in the range of 7 to 15
·rock.
percent, because of the mixture of grain sizes present in the various strata; furthermore, they
normally decrease with depth due to compaction.
Specific yield can be measured by a variety of techniques involving laboratory, field, and
estimating techniques. 6• 24• 26 Methods based on well-pumping tests, described in Chapter 4,
:ttion,
generally give the most reliable results for field measurements.
~.5.2) Estimate the average drawdown over an area where 25 million m3 of water has been pumped through a
number of uniformly distributed wells. The area is 150 km2 and the specific yield of the unconfined
cent- aquifer is 25 percent.
ppar-
SOLUTION The volume of water drained is wy =25 x 106 m3• Eq. 2.5.2 is used to determine the bulk volume, V, of
:.5.3) the aquifer to extract this volume of water:
w
s =-y
.5.4) y v;
0.25 = 25 x 106 m3 ~ V, =1x 1os m3

;tion v;
ogic
ably V. 1x108 m 3
Thus, the average water level drop over the area is Mz =....!... = =0.67 m.
:reas A 150xl06 m2
s. In
1 The storage coefficient for confined aquifers can be approximated by (Jacob, 1950)
the
-~ S = ya~b + Yinl
4
where"( is the unit weight of water, o. is the porosity of the aquifer, ~ is the reciprocal of the bulk modu- Table
lus of elasticity of water, 11 is the reciprocal of the modulus of elasticity of the skeleton of the aquifer, and
b is the saturated thickness of the aquifer.
The first and the second tenns in the above equation for S represent the fractions of storage attribut- Type of
able to the expansibility of water and the compressibility of the aquifer skeleton, respectively. Assuming porosity
reasonable ranges of values, develop an expression to estimate the storage coefficient of a confined
aquifer considering the expansion of water only. Intergrru
SOLUTION Since we consider only the expansion of water, we must deal with the first term in the given equation.
Reasonable ranges of values for the given properties, assuming a temperature range of 5-25°C, are
Intergra1
"( =9806 to 9779 N/m 3 andfra
~ = 4.85 X 10-IO to 4.5 X 10-IO m 2;N
o. =0.10 to 0.50 Fracture

Substituting these values into the expression for the storage coefficient attributable to the expansibility
of water yields
s = yo.~b = (9779 N/rn 3)(0.10)(4.5 X w-IO m21N)(b) = (4.4 X w-7)b
or
S = yo.pb = (9806 N/m3)(0.50)(4.85 X 10-IO m2/N)(b) = (2.4 X 10-6)b
Thus, in most cases the storage coefficient attributable to the expansion of water would be in the range
of 4.4 X 10-7b to 2.4 X lQ~b, where b is the saturated thickness of the aquifer.
Determine the volume of water released by lowering the piezometric surface of a confined aquifer by 5
mover an area of A = 1 knl. The aquifer is 35m thick and has a storage coefficient of 8.3 x 10- 3•
SOLUTION The released volume can be determined utilizing the definition of the storage coefficient, V =
(A)(M)(S) = (1 x 106 rrl)(5 m)(8.3 x w-3) = 41,500 m3.
2.6 GEOLOGIC FORMATIONS AS AQUIFERS

2.6.2 :
A geologic formation that will yield significant quantities of water has been defined as an
aquifer. Many types of fonnations serve as aquifers. 7· 38 A key requirement is their ability to
store water in the rock pores. Porosity may be derived from intergranular spaces or from frac-
tures. Table 2.6.1 summarizes the geologic origin of aquifers in terms of type of porosity and
rock type. The roles of various geologic fonnations as aquifers are briefly described in the fol-
lowing subsections.
2.6.1 Alluvial Deposits

Probably 90 percent of all developed aquifers consist of unconsolidated rocks, chiefly gravel
and sand. These aquifers may be divided into four categories, based on manner of occurrence:
water courses, abandoned or buried valleys, plains, and intermontane valleys. Water courses
consist of the alluvium that fonns and underlies stream channels, as well as fanning the adja-
cent floodplains. Wells located in highly penneable strata bordering streams produce large
quantities of water, as infiltration from the streams augments groundwater supplies. Abandoned
or buried valleys are valleys no longer occupied by the streams that fanned them. Although such
valleys may resemble water courses in permeability and quantity of groundwater storage, their
2.6 Geologic Formations as Aquifers 53
k modu- Table 2.6.1 Geologic Origin of Aquifers Based on Type of Porosity and Rock Type (after Dept. of Economic and Social Affairs 16)
lifer, and
Igneous and
attribut- Type of Sedimentary metamorphic Volcanic
porosity
ssuming Consolidated Unconsolidated Carbonates Consolidated Unconsolidated
~onfined
Intergranular Gravelly sand Weathered zone Weathered zone Volcanic ejecta,
Clayey sand of granite-gneiss of basalt blocks, and
Sandy clay fragments
quation. Ash
are
Intergranular Breccia Zoogenic Volcanic tuff
and fracture Conglomerate limestone Cinder
Sandstone Oolitic Volcanic
Slate limestone breccia
Calcareous grit Pumice
Fracture Limestone Granite Basalt
llSibility Dolomite Gneiss Andesite
Dolomitic Gabbro Rhyolite
limestone Quartzite
Diorite
Schist
Mica schist
e range recharge and perennial yield are usually less. Extensive plains underlain by unconsolidated sed-
iments exist in the United States. In some places gravel and sandbeds form important aquifers
under these plains; in other places they are relatively thin and have limited productivity. These
plains flank highlands or other features that served as the source of the sedimentary deposits.
:er by 5 The aquifers are recharged chiefly in areas accessible to downward percolation of water from
3
precipitation and from occasional streams. Intermontane valleys are underlain by tremendous
volumes of unconsolidated rock materials derived by erosion of bordering mountains. Many of
lt, v= these more or less individual basins, separated by mountain ranges, occur in the western United
States. The sand and gravel beds of these aquifers produce large quantities of water, most of
which is replenished by seepage from streams into alluvial fans at mouths of mountain canyons.
as an 2.6.2 Limestone
[lity to Limestone varies widely in density, porosity, and permeability depending on degree of con-
afrac- solidation and development of permeable zones after deposition. Those most important as
ty and aquifers contain sizable proportions of the original rock that have been dissolved and
~e fol- removed. 64 Openings in limestone may range from microscopic original pores to large solution
caverns forming subterranean channels sufficiently large to carry the entire flow of a
stream. 9• 28 The term lost river has been applied to a stream that disappears completely under-
ground in a limestone terrain.* Large springs are frequently found in limestone areas.
The dissolution of calcium carbonate by water causes prevailingly hard groundwater to be
gravel found in limestone aquifers; also, by dissolving the rock, water tends to increase the pore space
renee: and permeability with time. Solution development of limestone forms a karst terrain,t charac-
)Urses terized by solution channels, closed depressions, subterranean drainage through sinkholes, and
: adja-
large *The growing interest of venturesome scuba divers in springs, sinkholes, and caves is yielding new infonnation
joned regarding the hydrogeology of limestone aquifers.
1 such 1The term karst is derived from the German form of the Slavic word kras or krs, meaning a black waterless place.
, their Also, it is the German name for a district east of Trieste having such a terrain. 44
I
.c 50
a.
II)
0
Figure 2.6.1. Diagram showing factors influenc-

ing cavity distribution in carbonate rocks (after
100 Lauman and Parizek 32).
caves (see Figure 2.6.1). Such regions normally contain large quantities of groundwater.33
Major limestone aquifers occur in the southeastern United States and in the Mediterranean
area. 11 · l2, 31 ' A national karst map is being developed by the U.S. Geological Survey. 18 2.6.3
The Edwards Aquifer (Figure 2.6.2) in south-central Texas is an example of karst terrain.
The limestone and dolomites are exposed at the land surface, (referred to as·the outcrop area)
where numerous solution openings along vertical joints and sinkholes provide an efficient link
°
between the land surface and water table. 7 Figure 2.6.3 shows a sample of the limestone
bedrock from the Edwards Aquifer, illustrating the porousness of the karst formation.
0 30 miles
2.6.4
0 .
0
~ Edwards ~~~,.
,....t!il
f J .. - -.__ .·
(j .• \
.:::;
,...
. v
-~ . ,
2.6.5
Figure 2.6.2. A large area ofkarst ter-

rain is associated with the Edwards 2.6.6
Aquifer in south-central Texas. Large
streams lose a considerable amount of
EXPLANATION
water to ground water as they traverse
• Outcrop area of Edwards Aquifer - - Une separating freshwater zone to the
the outcrop area of the Edwards Aquifer
O Artesian area within freshwater zone north from saltwater zone to the south
(modified from Brown and Patton8).
• Awareness of the extensive solution development of limestone along the southern perimeter of Europe continues to
increase. Thus, at Gibraltar, St. Michael's Cave was discovered only during fortification work in 1942; it is large
enough to serve today as an underground amphitheater for concerts.
2.6 Geologic Fonnations as Aquifers 55
influenc-
s (after Figure 2.6.3. Limestone bedrock from the Edwards
Aquifer, a karst fonnation, is extremely porous and
l provides for very fast movement of ground water. 21
iwater.33
erranean
18 2.6.3 Volcanic Rock
t terrain. Volcanic rock can form highly permeable aquifers; basalt flows in particular often display such
·op area) characteristics.* The types of openings contributing to the permeability of basalt aquifers
ientlink include, in order of importance: interstitial spaces in clinkery lava at the tops of flows, cavities
mestone between adjacent lava beds, shrinkage cracks, lava tubes, gas vesicles, fissures resulting from
faulting and cracking after rocks have cooled, and holes left by the burning of trees over-
whelmed by lava.38 Many of the largest springs in the United States are associated with basalt
deposits. Rhyolites are less permeable than basalt, whereas shallow intrusive rocks can be
practically impermeable.
2.6.4 Sandstone
Sandstone and conglomerate are cemented forms of sand and gravel. As such, their porosity
and yield have been reduced by the cement. The best sandstone aquifers yield water through
their joints. Conglomerates have limited distribution and are unimportant as aquifers.
2.6.5 Igneous and Metamorphic Rocks

In solid forms, igneous and metamorphic rocks are relatively impermeable and hence, serve as
poor aquifers. Where such rocks occur near the surface under weathered conditions, however,
they have been developed by small wells for domestic water supply.
li'St ter- t
rds 1 2.6.6 Clay
"
..arge •'t Clay and coarser materials mixed with clay are generally porous, but their pores are so small
unt of l
1verse
'¥· that they may be regarded as relatively impermeable. Clayey soils can provide small domestic
Aquifer JI water supplies from shallow, large-diameter wells.
~
,ns).
l'
;
*An excellent example of a highly permeable volcanic aquifer is in Managua, Nicaragua. Here a circular Jake con-
1tinues to tained in an extinct volcanic crater serves as the major municipal water source and yields some 75,000 m3/day.
large There is no surface inflow, and evaporation exceeds precipitation; hence, the lake, fed entirely by groundwater, acts
as a large natural well.
2.7 TYPES OF AQUIFERS

Most aquifers are of large areal extent and may be visualized as underground storage reservoirs,
such as the Edwards Aquifer shown in Figure 2.6.2. Water enters a reservoir from natural or arti-
ficial recharge; it primarily flows out under the action of gravity or is extracted by wells (see
Section 1.6.2). Ordinarily, the annual volume of water removed or replaced represents only a
small fraction of the total storage capacity. Aquifers may be classed as unconfined or confined,
depending on the presence or absence of a water table, while a leaky aquifer represents a com-
bination of the two types.
2.7.1 Unconfined Aquifer

An unconfined aquifer is one in which a water table varies in undulating form and in slope,
depending on areas of recharge and discharge, pumpage from wells, and permeability. Rises
and falls in the water table correspond to changes in the volume of water in storage within an
aquifer. Figure 2.3.1 is an idealized section through an unconfined aquifer; the upper aquifer
in Figure 2.7.1 is also unconfined. The outcrop area of the Edwards Aquifer (Figure 2.6.2) is
the unconfined portion of an aquifer. Contour maps and profiles of the water table can be pre-
pared from elevations of water in wells that tap the aquifer to determine the quantities of water
available and their distribution and movement.
A special case of an unconfined aquifer involves perched water bodies, as illustrated by
Figure 2.7.2. This occurs wherever a groundwater body is separated from the main ground-
water by a relatively impermeable stratum of small areal extent and by the zone of aeration
above the main body of groundwater. Clay lenses in sedimentary deposits often have shallow
perched water bodies overlying them. Wells tapping these sources yield only temporary or
small quantities of water.
2.7.2 Confined Aquifers

Confined aquifers, also known as artesian· or pressure aquifers, occur where groundwater is
confined under pressure greater than atmospheric by overlying relatively impermeable strata.
Recharge
2.7.3
Impermeable
strata
Figure 2.7.1. Schematic cross section iUus· 2.7.4

trating unconfined and confined aquifers.
'The word artesian has an interesting origin. It is derived from the French artesien, meaning "of or pertaining to
Artois," the northernmost province of France. Here the first deep wells to tap confined aquifers were drilled and
investigated, from about 1750. Originally the tenn referred to a well with freely flowing water, but at present it is
applied to any well penetrating a confined aquifer or simply the aquifer itself.
2.7 Types of Aquifers 57
Ground water
;ervoirs,
1or arti-
:lls (see
~wat"tablos
s only a
)nfined,
:a com-
Impermeable strata
.·... - ...
1 slope, Water table

y. Rises
Figure 2.7.2. Sketch of
ithin an Unconfined aquifer perched water tables.
aquifer
~.6.2) is
be pre- In a well penetrating such an aquifer, the water level will rise above the bottom of the confin-
,fwater ing bed, as shown by the artesian and flowing wells of Figure 2.7.1. Water enters a confined
aquifer in an area where the confining bed rises to the surface (see Figure 2.6.2); where the
ated by confining bed ends underground, the aquifer becomes unconfined. A region supplying water
7"0und- to a confined aquifer is known as a recharge area; water may also enter by leakage through a
.eration confining bed (see Figure 2. 7.1 ). Rises and falls of water in wells penetrating confined aquifers
:hallow result primarily from changes in pressure rather than changes in storage volumes. Hence, con-
rary or fined aquifers display only small changes in storage and serve primarily as conduits for con-
veying water from recharge areas to locations of natural or artificial discharge.
The piezometric surface, or potentiometric surface, of a confined aquifer is an imaginary
surface coinciding with the hydrostatic pressure level of the water in the aquifer (Figure 2.7.1).
tater is The water level in a well penetrating a confined aquifer defines the elevation of the piezomet-
strata. ric surface at that point Should the piezometric surface lie above ground surface, a flowing
well results. Contour maps and profiles of the piezometric surface can be prepared from well
data similar to those for the water table in an unconfined aquifer. It should be noted that a con-
fined aquifer becomes an unconfined aquifer when the piezometric surface falls below the bot-
tom of the upper confining bed. Also, quite commonly an unconfined aquifer exists above a
confined one, as shown in Figure 2.7 .1.
2.7.3 Leaky Aquifer

Aquifers that are completely confined or unconfined occur less frequently than do leaky, or
semiconfined, aquifers. These are a common feature in alluvial valleys, plains, or fanner lake
basins where a penneable stratum is overlain or underlain by a semipervious aquitard, or
semiconfining layer (see Figure 2.7.3). Pumping from a well in a leaky aquifer removes water
in two ways: by horizontal flow within the aquifer and by vertical flow through the aquitard
into the aquifer.
Ius- 2.7.4 Idealized Aquifer

For mathematical calculations of the storage and flow of groundwater, aquifers are frequently
; to
assumed to be homogeneous and isotropic. A homogeneous aquifer possesses hydrologic
nd properties that are everywhere identical. An isotropic aquifer's properties are independent of
it is direction. Such idealized aquifers do not exist; however, good quantitative approximations can
be obtained by these assumptions, particularly where average aquifer conditions are employed
Ground surface
- Unit 1
piezom1
Water table
____ sz __ Piezometric surface

=--------------------
Aquitard
-----=~---------------------------
Leaky aquifer
~ .
f _ _ _. . . . . . . _ . . _ _ . . . . . _ " - " - - - - - _ _........._.._......._.
Figure 2.7.3. Sketch of a leaky, or

Impermeable strata semiconfined, aquifer.
on a large scale. Anisotropic aquifers, which possess directional characteristics, are discussed
in Chapter 3.
2.8 STORAGE COEFFICIENT

Water recharged to, or discharged from, an aquifer represents a change in the storage volume
within the aquifer. For unconfined aquifers this is simply expressed by the product of the vol-
ume of aquifer lying between the water table at the beginning and at the end of a period of time
and the average specific yield of the formation. In confined aquifers, however, if the aquifer
remains saturated, changes in pressure produce only small changes in storage volume. Thus, 2.9 {
the hydrostatic pressure within an aquifer partially supports the weight of the overburden GROl
while the solid structure of the aquifer provides the remaining support. When the hydrostatic
pressure is reduced, such as by pumping water from a well penetrating the aquifer, the aquifer
load increases. A compression of the aquifer results, forcing some water from it. In addition,
lowering of the pressure causes a small expansion and subsequent release of water. The water-
yielding capacity of an aquifer can be expressed in terms of its storage coefficient.
A storage coefficient (or storativity) is defined as the volume of water that an aquifer
releases from or takes into storage per unit surface area of aquifer per unit change in the com-
ponent of head normal to that surface. For a vertical column of unit area extending through a
confined aquifer, as in Figure 2.8.l(a), the storage coefficientS equals the volume of water
released from the aquifer when the piezometric surface declines a unit distance. The coefficient
is a dimensionless quantity involving a volume of water per volume of aquifer. In most con-
fined aquifers, values fall in the range 0.00005 < S < 0.005, indicating that large pressure
changes over extensive areas are required to produce substantial water yields. Storage coeffi-
cients can best be determined from pumping tests of wells (Chapter 4), or from groundwater
2.9.1
fluctuations in response to atmospheric pressure or ocean tide variations (see Chapter 6).
The fact that S normally varies directly with aquifer thickness enables the rule-of-thumb
relationship 34 for a confined aquifer
(2.8.1)
where b is the saturated aquifer thickness in meters to be applied for estimating purposes.
The storage coefficient for an unconfined aquifer corresponds to its specific yield, as
shown in Figure 2.8.l(b).
2.9 Groundwater Basins/Regional Groundwater Flow Systems 59
Unit cross-sectional area

Unit decline of Piezometric
pieioi~etric:sirtace~::r=ti;liice;-:
Unit cross-sectional area
Water table
......
: ·.·.·:: .· .. ·.·.·l······.·i.············
· ... · ..
.-_: :; :·.· .· .·_.
..... . ...
.· ·. . . .. :.· · .· .. ··
. . •. . ·.: .· .. . .
or . ·.. : ·-·.·: ·.· ..
..
..
··.· : :.·. ·
..
: .· ._· Aquifer;~:·>· •
0
• 0
.
•
• ••••
• • •
'•
•.....
• ••
:__.:_·:Aquifer .·· ·
. ·. .. ·
cussed . ..
•,
. .... ,.
:: ·.. ·.. .:~ -~ -~ .. ..
·. · .'·
. . .. ·. .... : . ~. ·. ..
. . ..
.. . · ... ··.·.·.·.: .. . ...
. ·· ·.· · .·._.: . .. ..... .•'
·:·:·.·· .. · .··.· .. ·.··

..
.
-
~ . . .. ~ .~ ./ . . ·~.
·olume Figure 2.8.1. Illustrative sketches for

Impermeable lmpem1eable defining the storage coefficient of (a)
1e vol-
)ftime (a) (b) confined and (b) unconfined aquifers.
tquifer
Thus, 2.9 GROUNDWATER BASINS/REGIONAL
Jurden GROUNDWATER FLOW SYSTEMS
ostatic A groundwater basin may be defined as a hydrogeologic unit containing one large aquifer or
tquifer several connected and interrelated aquifers.* Such a basin may or may not coincide with a
dition, physiographic unit. In a valley between mountain ranges, the groundwater basin may occupy
water- only the central portion of the stream drainage basin. In limestone and sandhill areas, drainage
and groundwater basins may have entirely different configurations. The concept of a ground-
tquifer water basin becomes important because of the hydraulic continuity that exists for the contained
!COffi-
groundwater resource. In order to ensure continued availability of subsurface water, basin-
mgh a wide management of groundwater, which is described in Chapter 10, becomes essential.
water A regional groundwater flow system can be comprised of subsystems at different scales and
fi.cient a complex hydrogeologic framework, illustrated in Figure 2.9.1. The Edwards Aquifer, illustrated
:t con- in Figure 2.6.2, is a regional groundwater flow system. Two regional groundwater flow systems,
essure the High Plains Aquifer and the Gulf Coastal Plain Aquifer system, are illustrated in Figure 2. 9.2.
~oeffi
lwater
).
2.9.1 High Plains Aquifer
:humb An example of a groundwater basin is the High Plains Aquifer (Figures 2.9.2 and 2.9.3), which
underlies an area of about 174,000 square miles extending through parts of Colorado, Kansas,
Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming. Approximately 20
2.8.1)
percent of the irrigated agricultural land in the United States overlies the High Plains aquifer,
~s.
:ld, as *In practice the term groundwater basin is loosely defined; however, it implies an area containing a groundwater
reservoir capable of furnishing a substantial water supply.
®
l
!
I
" ----
Figun
CD Local groundwater subsystem
Low hydraulic-conductivity confining unit
0 Very low hydraulic-conductivity bedrock
® Sub regional groundwater subsystem
- Direction of groundwater flow @) Regional groundwater subsystem
Figure 2.9.1. A regional groundwater flow system that comprises subsystems at different scales and a complex hydrogeologic
framework (modified from Sun5\
Significant features of this depiction of part of a regional groundwater flow system include (I) local groundwater subsystems in
the upper water-table aquifer that discharge to the nearest surface-water bodies (lakes or streams) and are separated by groundwater
divides beneath topographically high areas; (2) a subregional groundwater subsystem in the water-table aquifer in which flow paths
originating at the water table do not discharge into the nearest surface-water body but into a more distant one; and (3) a deep,
regional groundwater flow subsystem that lies beneath the water-table subsystems and is hydraulically connected to them. The
hydrogeologic framework of the flow system exhibits a complicated spatial arrangement of high hydraulic-conductivity aquifer
units and low hydraulic-conductivity confining units. The horizontal scale of the figure could range from tens to hundreds of miles?
and about 30 percent of the groundwater used for irrigation is withdrawn from the aquifer (U.S. 2.9.2
Geological Survey5\ The aquifer boundary approximates the boundary of the Great Plains
Physiographic Province, which is characterized by a flat to gently rolling land surface and
moderate precipitation. The region is underlain by sedimentary rocks that dip gently to the east
and are upturned with their contact with the Rocky Mountains and other dome mountains, such
as the Black Hills. Figure 2.9.3 shows the regional subdivisions of the High Plains Aquifer: the
Northern High Plains, the Central High Plains, and the Southern High Plains.
The altitude and configuration of the water table in the High Plains Aquifer is affected
mostly by the altitude and configuration of the underlying bedrock surface, the transmissivity
of the aquifer, and the rate and distribution of recharge and discharge. Large areas of the
aquifer are not continuously saturated or are saturated only in isolated channels in the bedrock
surface. The water table in other parts of the aquifer is continuous and slopes eastward at gra-
High Plains
aquifer
Gulf Coastal Plain

aquifer system
SilO MILES
l
I';
••
SilO KILOM£TUIS
Figure 2.9.2. Location of the High Plains Aquifer and the Gulf Coastal Plain Aquifer system.55
clients of about 10 to 40 ftlmi. In most locations, the water levels have declined since irrigation
withdrawal became widespread.
:ic
Irrigation with groundwater began in the 1800s using windmill-powered pumps, with
large-scale withdrawals beginning after the 1930s drought. Because of the large water-level
ems in declines in the aquifer, the saturated thickness has substantially decreased, significantly
tdwater increasing the cost to obtain the water. The result has been to use deeper wells and larger
v paths pumps, leading to higher energy costs. Figure 2.9.3 shows the location of selected wells whose
hydrographs are plotted in Figure 2.9.4. These hydrographs clearly indicate historical water-
te level declines. The effects of the groundwater withdrawals on the southern High Plains Aquifer
fer are illustrated in Figure 2.9.5. Figure 2.9.6 shows the changes in groundwater levels and the
miles. 3 percent change in saturated thlck.ness of the aquifer from predeveloproent to 1997.
:r(U.S.
Plains
2.9.2 Gulf Coastal Plain Aquifer System
ce and The Gulf Coastal Aquifer system (Figure 2.9.2) consists of a large, complex system of aquifers
he east and confining units that underlie about 290,000 square miles.55 This system extends from
s, such Texas to westernmost Florida, including offshore areas to the edge of the Continental Shelf.
'er: the The aquifer system has been coarsely subdivided into 17 regional aquifers and confining units,
most of which are shown in Figure 2.9.7. Groundwater withdrawals have significantly altered
ffected ; the groundwater conditions in the aquifer system, resulting in large-scale regional changes in
ssivity ~ direction of horizontal flows, changes in vertical direction of flow between aquifers, increases
of the i
l
in regional recharge to aquifers, and decreases in regional discharge from aquifers. 55 The sim-
~rock 1 ulated widespread reversal of vertical-flow directions from predevelopment to 1987 for the
at gra- upper part of the Gulf Coastal Plain Aquifer is shown in Figure 2.9.8.
Figure 2.9.3.
Regional subdivid-
ions of the High
Plains Aquifer, and
location of selected
wells.39
2,975 ::J
ti~
2,955 ~~
~L:i
w{l)
2,935 g~
!::o
Altitude of base of aquifer: 2,600 ft. ~m
2,915 <{ <{
UJ B.
...J
I ~ 20
Well number: 411238104070801
Laramie County, Wyoming
5,062 1- w
w:>i
ww
luw~
u. {I) 40 5,042 ~~
u.-'
~C ~CJ)
:r:z
r- :5 60 5,022 ~~
Hi~ ~CD
og so 5,002
<<.
w Altitude of base of aquifer: 4,800 feet
m
100 ':L:'-L..I.....I..."'--'--'--:'-:'-:'-::'--'-'-..J.....W.-'-:'-:-'~'-'-'-L.J.....J.....o..,.u-c'-'-'--'-L...J..-'--'-::lc:.'":'-"...L...I...J....L...W':...L.:J-L..I....L-L.J.....J.....L...J.-:l-:"":'-'-..L..J....L•I-L.':·J 4,982
1930 1940 1950 1960 1970 1980 1990 2000
o~~~~~~~~~~~~~~~~~~~~~~~~~2.~1
w
0
M. 2,941 ...J
~ 50
1/kll number: 380201101030101
Finney County, Kansas
2,921 ti ~
~~
~~ 2,901
~~
u.{l)
~a
:r:z 100
t::5
2,881
2,861
2,841
g;
!:: 0
w~ 2,821 ~m
ag 150
2,801
<{ <{
w
m
2,781
200 L....L.LJL...I....WL...I....WL...L..L..>....L..L..I....L..L..L...L..I....L...L..I....L.J....l...I...J....l...I....L..L..L..L..L.J.....L...L..L...L...L.J....L...L.L.L..L..I....L....L.LJL....L.LJL....L.LJ..................L..L..I...L..I....L............ 2..~761
1930 1940 1950 1960 1970 1980 200u
300 4,134
w
0 0. ...J
4,114
tu~
320
~ Well number: 342736103203701
tuO:: Curry County, New Mexico w...J
~~ 340 4,094 ~~
~a
z ~CJ)
~ :5 360 4,074 ~~
-o
lh~ ~CD
ogw 380 Altitude of base of aquifer. 4,000 feet 4,054
<{c()
m
400 4,034
1930 1940 1950 1960 1970 1980 1990 2000
120
w
0 ...J
140 P. 3,2901- ~
~ \Nell number: 363033101440701 Ww
lijO::
Texas County, Oklahoma ~...J
w=>
u.{l) 160 3,270 ~i1i
',3. ~a ~{I)
ubdivid- J:z
t-:5 180 3,250 ~~
High fh~ ~m
1ifer, and co...J 3,230 <{ <{
w 200 Figure 2.9.4. 1998 Hydrographs
selected ID Altitude of base of aquifer: 2,990 feet
for selected wells in the High Plains
220
1930 1940 Aquifer. 39
3,335
(a)
3,325
-
...I
a. 3,305
tli~
Weft number: 360750101462800
Sherman County, Texas 3,285 ~~
-U1
3,265 ~(IJ
3,245 ::>~
t:o
3,225 ~~
AUude of base or aquifer: 3,000 feet
3,205
1990 ~d165
iii ...I
Iii~ 80
3,224 Iii~
~~
w::>
u..(IJ
100 3,204 ~U1
~a ~(IJ
z
~~ 120
s. 3,164 ~~
~§ Will number: 333455101583200
Altitude of base of aquifer: 3, 140 feet
~~.
uJ140 Lubbock County, Texas 3,164
Ill Figure 2.9.4 (continued). 1998
Hydrographs for selected wells is
the High Plains Aquifer.39
West East
(a) vertical scale greatly exaggerated

Syatem before devel<l$)nlent Syatem during dewlopment Figure 2.9.5. The effects of groundwater with- Figur
drawals on the southern High Plains Aquifer. nessi
Recharge
Natural and lrrigalon (a) Schematic cross section of the southern EJ~
recharge return flow Pumpege High Plains Aquifer illustrating that groundwater Plains
24 510 830
withdrawal in the middle of the southern High thickn
Plains Aquifer has a negligible short-tenn effect cause
on the discharge at the boundaries of the ed in ·
33
Natural Natural aquifer. (b) Water budgets of the southern
cfiSCharge discharge High Plains Aquifer (all flows in million cubic
24 10
feet per day) before development and during
(b) development.3• 26• 36
2.9 Groundwater Basins!Regioual Groundwater Flow Systems 65
(b)
l98
1s is
Decreases
• More than 50
.._......_._....o....:..:.lo.iiC::i:!~' ~ 25-50
D 1o-24
Rises Increases
• More than 10 • More than 10
D Less than 10-foot change 0 Less than 10-foot change

- Boundary of the High - Boundary of the High
Plains Aquifer Plains Aquifer
vith- Figure 2.9.6. (a) Changes in groundwater levels in the High Plains Aquifer from predevelopment to 1997 and(b) saturated thick-
:r. ness in the High Plains Aquifer from predevelopment to 1997. (V L. McGuire, U.S. Geological Survey, written cornmun_, 1998.)
1em Extensive pumping of ground water for irrigation has led to groundwater-level declines in excess of 100 feet in parts of the High
Iwater Plains Aquifer in Kansas, New Mexico, Oklahoma. and Texas. These large water-level decliiles have led to reductions in saturated
igh thickness of the aquifer exceeding 50 percent of the predeveloprnent saturated thickness in some areas. Lower groundwater levels
.ffect cause increases in pumping lifts. Decreases in saturated thickness result in declining well yields. Surface-water irrigation has result-
ed in water-level rises in some parts of the aquifer system, such as along the Platte River in Nebraska.3
bic
g
N s
fEET
2.
s.ooo MJSSlSSIPPI EM8AYMENT COASTAl LOWLANDS
AQUIFER SYSTEM AQUIFER SYSTEM 2.
SEA
LEVEL
5.000
10.000
-
15.000
AqJfer •nd ~ • nurnllet-

Pottemed area eon~<~lns saline wator
rn Conlnntl unit- rnodol 1.,.,

"""'bor
. • • • • • • • • • Unit tnnaled It odgoo o! Continomal
Shelf
69
Figure 2.9.7. Aquifers and confining units and designation of layers in a regional model of the Gulf Coastal Plain Aquifer system (as
presented in Taylor and Aller\
ILLINOIS
• Area where simulated vertical flow was reve1sed

from upward under predevelopment cond~ions
to downwa1d by 1987
- Updip boundary of uppermost aquifer layer
(Downdip boundary is edge o1 Continental Shelf)
Figure 2.9.8. Areas

where vertical flow
between uppermost
Gull ol
,, _, aquifer layers reversed Fig~
from upward under prede·
___ _., 'I
Mexico
(d)~
; _____________ ____ '\ ___ _ _,_
velopment conditions to
downward by 1987, as
,' Edge of Conlinental Shelf simulated in regional
I
model of Gulf Coastal
' Plain Aquifer system22 (as
' 100
presented in Taylor and
Allers).
2.!0 Springs 67
j 2.10 SPRINGS
2.10.1 What Are Springs?
A spring is a concentrated discharge of groundwater appearing at the ground surface as a
current of flowing water. To be distinguished from springs are seepage areas,* which indicate
a slower movement of groundwater to the ground surface. Water in seepage areas may pond
and evaporate or flow, depending on the magnitude of the seepage, the climate, and the
topography.
Springs occur in many forms and have been classified as to cause, rock structure, dis-
charge, temperature, and variability. Bryan 10 divided all springs into (1) those resulting from
nongravitational forces and (2) those resulting from gravitational forces. Under the former cat-
egory are included volcanic springs, associated with volcanic rocks, and fissure springs, result-
ing from fractures extending to great depths in the earth's crust. Such springs are usually
thermal (see following section).
Gravity springs result from water flowing under hydrostatic pressure; the following gen-
eral types are recognized (see Figure 2.10.1 ): 10
l. Depression Springs-Formed where the ground surface intersects the water table.
2. Contact Springs-Created by a permeable water-bearing formation overlying a less per-
meable formation that intersects the ground surface.
3. Artesian Springs-Resulting from releases of water under pressure from confined aquifers
either at an outcrop of the aquifer or through an opening in the confi.o.ing bed.
4. Impervious Rock Springs-Occurring in tubular channels or fractures of impervious rock.
5. Tubular or Fracture Springs-Issuing from rounded channels, such as lava tubes or solu-
tion channels, or fractures in impermeable rock connecting with groundwater.
(c)
(a)
Spring Spring
fmpeiViOUS
(b) (d)
sed
Figure 2.10.1. Diagrams illustrating types of gravity springs. (a) Depression spring. (b) Contact spring. {c) Fracture artesian spring.
prede-
(d) Solution tubular spring (after Bryan,10 copyright © !9!9 by the University of Chicago Press).
ns to
. as
al
;ta)
22 *Seepage is a general term describing the movement of water through the ground or other porous media to the
n (as
ground surface or surface water bodies. The term is well established in the engineering literat\lre in connection with
and
groundwater movement from and to surface water bodies, particularly where associated with stroctures such as
dams, canals, and levees.
To define magnitudes of springs, Meinzer proposed a classification by discharge shown in

Table 2.10.1. Large-magnitude springs occur primarily in volcanic and limestone terrains. 43
The discharge of a spring depends on the area contributing recharge to the aquifer and the
rate of recharge. Adequate spring water to supply the needs of a single family can be obtained
from a few hectares, whereas large areas with high rainfalls are necessary to produce a first-
magnitude spring. Figure 2.10.2 shows the contributing recharge area (795krn2 , 310 square rni)
for the Maramec Spring, near Rolla, Missouri. The spring discharges about 1,555 ft 3/sec
(4.4m3/s) into the nearby Maramec River, more than doubling the flow in the river most of the
time. 58
Most springs fluctuate in their rate of discharge. Fluctuations are in response to variations
in rate of recharge with periods ranging from minutes to years, depending on geologic and
hydrologic conditions. Figure 2.10.3 shows the relationship among precipitation, discharge,
and dissolved calcium,65 indicating the mixing of differing water sources emerging from the
spring. The groundwater and surface-water basins coincide with karst features such as caves,
sinkholes, losing stream reaches, springs, and conduits in the sedimentary features. Perennial
springs drain extensive permeable aquifers and discharge throughout the year, whereas inter-
mittent springs discharge only during portions of the year when sufficient groundwater is
recharged to maintain flow. Areas of volcanic rock and sandhills are noted for their perennial
springs of nearly constant discharge. Springs that exhibit more or less regular discharge fluc-
tuations not associated with rainfall or seasonal effects are periodic springs. Such fluctuations
may be caused by variations in transpiration, by atmospheric pressure changes, by tides affect-
ing confined aquifers, and by natural siphons acting in underground storage basins.
In coastal areas containing limestone or volcanic rock aquifers, large subsurface channels
often discharge groundwater through openings to the sea. Such submarine springs can be
found along the borders of the Mediterranean Sea and also in Hawaii. Where the discharge is
sufficiently large, potable water can be lifted directly from the sea surface.*
Table 2.10.1 Classification of Springs by Discharge

(after Meinzer41 ) 8
Magnitude Mean discharge ..

I IIIII
First >10 m3/s

Second 1-10 m3/s
Third 0.1-1 m3/s
Fourth W-100 lis
Fifth 1-10 Vs
Sixth 0.1-11/s
Seventh 10-100 rnVs
Eighth <10 rnVs
•Another discharge classification of springs, also pro-
posed by Meinzer and based on English units, has been in
use for many years in the United States.
...
*Lucretius, a Roman poet and philosopher of the fl!st century B.C., described a submarine spring in the Mediter-
ranean Sea in his epic poem De Rerum Natura: "In the sea at Arados is a fountain of this kind, which wells up with
fresh water and keeps off the salt waters all around it . . . a seasonable help in need to thirsting sailors, vomiting
forth fresh waters amid the salt."
2.10 Springs 69
~
own in f Figure 2.10.2. Gaining and losing slreams,
1 dye traces, and the recharge area for Maramec
ins.43 ~ St. James
Spring. 58
md the - J Steelvi'le .
>tained
a flfSt-
Gaining stream
are mi)
' Losing stream
ft3/sec
"" Dye trace
t of the
Maramec Spring
Recharge Area
iations - County line
;ic and
:barge,
:>m the
caves,
rennial
; inter-
·ater is
rennial
·e flue-
1ations
s s 0 u
affect-
.annels
~an be
arge is
1.0~--~----r---,----r----r---or---r----r---~--~----~---r~
I
t' '
I ''
I
I
'
I'
10
111
0 --------------------------------------------------------
.~ J - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - ---------
c t
Figure 2.10.3. The rela-
I
tionship between precipi-
fiter-
lpwith tation, discharge, and dis·
ing solved calcium, Maramec
Spring, 1985-1986.58
2.10.2 Edwards Aquifer-Discharge of Springs

The recharge to the Edwards Aquifer (Figure 2.6.2) is derived mainly from seepage into the
aquifer from streams that cross the outcrop of the aquifer and direct infiltration of precipita-
tion on the outcrop. The watershed areas providing recharge to the aquifer are shown in Fig-
ure 2.10.4. Net recharge into the aquifer from the Guadalupe River Basin is negligible48 . The
estimated annual recharge to the aquifer by basin for 1980 to 2001 is listed in Table 2. I 0.2. The
estimated average annual recharge for the 1980 to 2001 time period is 684,700 ac-ft.
Discharge from the Edwards Aquifer is from wells and springs (refer to Figure 2.10.5).
The major discharge from wells is in Bexar, Medina, and Uvalde Counties. The City of San ...
.-..
v
~I
Antonio, Texas, obtains all of its water supply from the aquifer. Table 2.10.3 lists the estimated
!::!
annual discharge from the Edwards Aquifer by county and lists the portion from wells and 0
from springs. Note the relation of discharge from springs as compared to well discharge as a ....."'
0
function of time. Discharge from the Comal Springs (Figure 2.1 0.6) and San Marcos Springs "'
-o
~
for 200 I was 414,800 ac-ft, accounting for 78 percent of the discharge for that year. "'0
;:l
£
.._.
0
0
d !=:
~
.s-
()()
0\
tv
.c
~
Ill
>-
]
.&l
...
~0'
<:
]"'
~
-o
~
u
f3
.....,0
~
..:
(J
4J
~
~t::
~
-o
" Streamflow-gaging station .J

~
0 10 20 30 40 miles
~
EXPLANATION ""'
N
Study area C.J Gaged basins - Ungaged areas
Gl
:c
G) Nueces-West Nueces River Basin ® Area between Medina River Basin and ~
® Frio-Dry Frio River Basin Cibolo-Dry Comal Creek Basins
tJ'I Cibolo Creek and Dry Comal Creek Basins
®3 Sabinal River Basin
f.\
\!..J
@ Guadalupe River Basin
~ Area between Sanlnal and Medina River Basins
®
5 Medina River Basin ® Blanco River Basin
Figure 2.10.4. Map showing gaged basins and ungaged areas (modified from Pueote,48 Figure 1, as presented by Slattery and
Thomas49).
en::;·
en
L""'
o
~. g
<0
en Clf
Ci)
"'"';
.g' 8~
,..
.g>~(J)
V> r '\'- >•' --o-
::I
l]q ::::= - . 0 ; -
<'D en S ""+> ~--" ~ .~s·
~~~..a· 0
n ;::J
:::l~ ::::::!. ~ DJ :::s· ~ ~ ~ en 9
s· e-
(J)
m c:r
en ::I cg cg ~ ::::~'·'<· oqtn:::S('D~Vl
g g~·
cn~Q..Q..:::S':--" • ('D
-~ ... .,....~-
Table 2.10.2 Estimated Annual Recharge to the Edwards Aquifer by Basin, 1980-2001 (thousands of acre-feet)
Area between
Area between Medina
Nueces- Sabinal River Basin Cibolo
West Frio-Dry Sabinal and Medina Medina and Cibolo- Creek and Blanco
Calendar Nueces Frio River River River Dry Comal Dry Comal River
year River Basin River Basin 1 Basin 1 Basins 1 Basin2 Creek Basin 1 Creek Basins Basin 1 Total 3
1980 58.6 85.6 42.6 25.3 88.3 18.8 55.4 31.8 406.4
1981 205.0 365.2 105.6 252.1 91.3 165.0 196.8 67.3 1,448.4
1982 19.4 123.4 21.0 90.9 76.8 22.6 44.8 23.5 422.4
1983 79.2 85.9 20.1 42.9 74.4 31.9 62.5 23.2 420.1
1984 32.4 40.4 8.8 18.1 43.9 11.3 16.9 25.9 197.9
1985 105.9 186.9 50.7 148.5 64.7 136.7 259.2 50.7 1,003.3
1986 188.4 192.8 42.2 173.6 74.7 170.2 267.4 44.5 1,153.7
1987 308.5 473.3 110.7 405.5 90.4 229.3 270.9 114.9 2,003.6
1988 59.2 117.9 17.0 24.9 69.9 12.6 28.5 25.5 355.5
1989 52.6 52.6 8.4 13.5 46.9 4.6 12.3 23.6 214.4
1990 479.3 255.0 54.6 131.2 54.0 35.9 71.8 41.3 1,123.2
1991 325.2 421.0 103.1 315.2 52.8 84.5 109.7 96.9 1,508.4
1992 234.1 586.9 201.1 566.1 91.4 290.6 286.6 228.9 2,485.7
1993 32.6 78.5 29.6 60.8 78.5 38.9 90.9 37.8 447.6
1994 124.6 151.5 29.5 45.1 61.1 34.1 55.6 36.6 538.1
1995 107.1 147.6 34.7 62.4 61.7 36.2 51.1 30.6 531.3
1996 130.0 92.0 11.4 9.4 42.3 10.6 14.7 13.9 324.3
1997 176.9 209.1 57.0 208.4 63.3 193.4 144.2 82.3 1,134.6
1998 141.5 214.8 72.5 201.4 80.3 86.2 240.9 104.7 1,142.3
1999 101.4 136.8 30.8 57.2 77.1 21.2 27.9 21.0 473.5
2000 238.4 123.0 33.1 55.2 53.4 28.6 48.6 34.1 614.5
2001 297.5 126.7 66.2 124.1 90.0 101.5 173.7 89.7 1,069.4
Average 121.2 134.1 42.2 107.2 61.9 69.9 105.0 43.1 684.7
1
Includes recharge from ungaged areas (Figure 2.10.4).
2
Recharge to Edwards aquifer from the Medina River Basin consists entirely of losses from Medina Lake (Puente 48 ).
3
Total might not equal sum of basin values due to rounding.
Source: Slattery and Thomas49 ~
0
C/.l
""0
::1.
::l
(JQ
C/J
-l
Jo-ol.
-
Outcrop of the Edwards Aquifer
Boundary of geographic subarea
I Tabte2
Calend<
- Model boundary 98' 00 year
I
-4 Minor spring 1980
-<!> Major spring 99'00'
I 1981
I Well and number G1LLESf'lE 1982
I 1983
.... 100' 30. 100000
1- 1984
'
I I
, . . ,IU'f'S /1. I '
EDWARDS KERR 1985
I
EDWARDS PLATEA
I
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
4
1999
2000
2001
zs 50 Milos 1
Tc
2S 50 Kllomolers 2 D:
the Leom
Jo,
4o,
Figure 2.10.5. Edwards Aquifer with springs.30
~ 0.
Sou
Figure 2.10.6. Comal Springs, Edwards

Aquifer. 2.11
The highly productive Edwards Aquifer,
the first aquifer to be designated as a sole 2.11.1
source aquifer under the Safe Drinking Water
Act, is the source of water for more than one
million people in San Antonio, Texas, some
military bases and small towns, and for south-
central Texas farmers and ranchers. The
aquifer also supplies water to sustain threat-
ened and endangered species habitat associat-
ed with natural springs in the region and sup-
plies surface water to users downstream from
the major springs. These various uses are in
direct competition with gcoundwater develop-
ment and have created challenging issues of
groundwater management in the region (pho-
tograph by Robert Morris, U.S. Geological
Survey).
Table 2.10.3 Estimated Annual Discharge from the Edwards Aquifer by County, 1986-2001 (thousands ofacre-feet)
Calendar Kinney-Uvalde Medina Bexar Comal Hays Well Spring

year Counties County County County County Total 1 discharge discharge
1980 151.0 39.9 300.3 220.3 107.9 819.4 491.1 328.3

1981 104.2 26.1 280.7 241.8 141.6 794.4 387.1 407.3
1982 129.2 33.4 305.1 213.2 105.5 786.4 453.1 333.3
1983 107.7 29.7 277.6 186.6 118.5 720.1 418.5 301.6
1984 151.1 46.9 309.7 108.9 85.7 702.3 529.8 172.5
1985 156.9 59.2 295.5 200.0 144.9 856.5 522.5 334.0
291.7 2 2
1986 41.9 294.0 229.3 16Q.4 817.3 429.3 388.1
1987 Zg5.1 15.9 326.6 286.2 198.4 2922.0 364.1 2
558.0
2
156.7 82.2 317.4 236.5 116.9 2909.7 540.0 2
369.8
1988
1989 156.9 70.5 305.6 147.9 85.6 766.6 542.4 224.1
1990 118.1 69.7 276.8 171.3 94.1 730.0 489.4 240.6
1991 76.6 25.6 315.5 221.9 151.0 790.6 436.3 354.3
1992 76.5 9.3 370.5 412.4 261.3 1,130.2 327.3 802.8
1993 107.5 17.8 371.0 349.5 151.0 996.7 407.3 589.4
1994 95.5 41.1 297.7 269.8 110.6 814.8 424.6 390.2
/.
{ 1995 90.8 35.2 272.1 235.0 127.8 761.0 399.6 361.3
,~·
1996 117.6 66.3 286.8 150.2 84.7 705.6 493.6 212.0
3 37.0 3 3 3
1997 29.9 255.3 243.3 149.2 684.7 300.7 384.0
1998 113.1 51.3 312.8 271.4 169.2 915.9 451.7 464.1
4 5
4%.3 142.3 427.8 456.2
1999 99.8 298.3 295.2 884.0
2000 89.1 45.1 283.6 226.1 108.4 752.3 414.8 337.5
2001 68.7 33.9 291.6 327.4 175.3 896.9 367.7 529.1
1
Total might not equal sum of county values due to rounding.
2
Differs from value in Edwards Underground Water District Bulletins 46--48, table 3, due to correction of an error in the method of computing
the Leona Fonnation underflow.
3 Does not include irrigation discharge (Bexar, Medina, and Uvalde Counties).
4
Does not include discharges for domestic supply, stock, and miscellaneous use.
5 Does not include discharge from Kinney County.
Source: Slattery and Thomas49
2.11 HYDROTHERMAL PHENOMENA

fer,
2.11.1 Thermal Springs
rater
one Thennal springs discharge water having a temperature in excess of the nonnallocal ground-
ne water. The relative tenns wann springs and hor springs are common. Waters of thennal springs
JUth- are usually highJy mineralized and consist for the most part of meteoric water that has been
modified in quality by its passage underground. 60
.at- Hydrothennal phenomena involving the release of water and steam are nearly always
ciat- associated with volcanic rocks and tend to be concentrated in regions where large geothermal
sup- gradients occur. Also, by implication, aquifers must be present that permit water to percolate
rom to great depths-often 1,500 to 3,000 m. This water, heated from below, forms a large con-
In
vective current that rises to supply hydrothennal areas (see Figure 2.11.1 ).
:lop-
A geyser is a periodic thermal spring resulting from the expansive force of superheated
of
lhO-
steam within constricted subsurface channels (see Figure 2.11.2). Water from surface sources
i.l
'The word geyser is derived from the Icelandic word geysir, meaning to gush or rage.
Descending cool surtace water
Water begins to /+'-

boil near ground
surface because o
2 Hot spring
;:::
of greatly reduced O>
t
~ or geyser "" Water enters at
ground surface
and sinks in con-
duit formed by
Surface (meteoric) pressures t :00: fault or fracture
water sinks to levels J'- ~· ......_ Permeable zone

perhaps as much as-
Tl
3,000 meters below
above
ground.itsHeated
normaltar
boil·
ing point, it begins to
rise toward the sur- \
HEA
--·~ - . ··-Lt allows water to
flow throughout
face /
;r~ ;..._;-=l-.,.-==Coo~h-ng""'m-eg.-~m...a chamber ''(~ Figure 2.1 1.1. Schematic diagram of a

hydrothennal system (after Keefer9).
Figure
Figure 2.11.2. Eruptions of

the Midway Geysers in
Yellowstone National Park, 2.11.2
Wyoming (after Keefer9)
and/or shallow aquifers drains downward into a deep vertical tube where it is heated to above
the boiling point. With increasing pressure the steam pushes upward; this releases some water
at the surface, which reduces the hydrostatic pressure and causes the deeper superheated water
to accelerate upward and to flash into steam. The geyser then surges into full eruption for a
short interval until the pressure is dissipated; thereafter, the filling begins again and the cycle
is repeated.
Another kind of hot spring, known as a mudpot, results when only a limited supply of
water is available. Here water mixes with clay and undissolved particles brought to the surface,
forming a muddy suspension by the small amount of water and steam continuing to bubble to
the surface. Afumarole" is an opening through which only steam and other gases such as car-
'The word fumarole stems from the Latinfumus, meaning smoke.

'": ....
i I
\_
\.
Figure 2.11.3. Thennal springs in the United States (after Waring 60).
bon dioxide and hydrogen sulfide discharge. These features are nonnally found on hillsides
above the level of flowing thennal springs; water can often be heard boiling underground. 46
Thennal springs of various kinds are found throughout the world59 notable areas exist in
Iceland, New Zealand, and the Kamchatka Peninsula of Russia. Figure 2.11.3 shows the
regional distribution of thennal springs in the United States. Yellowstone National Park in
Wyoming, which contains literally thousands of hydrothennal features, possesses the greatest
concentration of thennal springs in the world. 29 This area marks the site of an enonnous vol-
canic eruption 600,000 years ago. Today, temperatures of 240°C exist only 300 m below
ms of ground surface.
2.11.2 Geothermal Energy Resources

Heat within the earth flows outward at an average rate of 1.5 x 10--6 callcm2/s- 1 and creates an
average geothennal gradient of 1°C/50 m. But in areas of volcanic and tectonic activity, heat
above flows several orders of magnitude larger than nonnal have been found. These tremendous
water reservoirs of heat close to ground surface have been recognized as invaluable sources of
. water energy. 18•52 But most important from a hydrologic standpoint is the fact that utilization of these
1 for a geothennal resources invariably involves water as the mechanism for extracting heat. Four
:cycle types of sources are generally recognized:
ply of 1. Dry Steam Field-Penneability is so low that groundwater reaching the heat source is lim-
rrface, ited and is fully vaporized.
•ble to 2. Wet Steam Field-Sufficient groundwater reaches the heat source so that a mixture of
ts car- water and steam is produced.
3. Low- Temperature Fields-Temperatures in the range of 50-80°C are found at shallow
depths, enabling the groundwater to be employed directly for heating purposes.
4. Dry Rock-In some areas at depths exceeding 3,000 m, temperatures are found in the
range of 200-500°C without the presence of groundwater; by injecting water into frac-
tured rock zones through one dritl hole, steam can be produced from an adjacent hole.
Geothermal resources exist through the world, but they have been developed in only a few
localities.* Most homes in Reykjavik, Iceland, are heated by naturally occurring hot water.
Notable power plants generating energy from geothermal resources are located in Italy, New
Zealand, and the western United States.
2.12 GROUNDWATER IN PERMAFROST REGIONS

Permafrost, or perennially frozen ground, is defined as unconsolidated deposits or bedrock that
continuously have had a temperature below ooc for two years to thousands of years.67· 68 Fig-
ure 2.12.1 illustrates the upper and lower limits of permafrost in terms of the depths at which
a 0°C ground temperature occurs. Regions of permafrost in the Northern Hemisphere are
shown in Figure 2.12.2. In the continuous-permafrost zone, permafrost is present everywhere
to depths of 150-400 m; while in the more southerly discontinuous-permafrost zone, per-
mafrost is perforated by unfrozen zones that depend on local conditions.
Frozen ground creates an impermeable layer that restricts the movement of groundwater,
acts as a confining layer, and limits the volume in which liquid water can be stored. In many
areas of frozen ground, shallow aquifers are entirely eliminated, thereby requiring that wells
be drilled deeper than in similar geologic environments without permafrost. Groundwater can
occur above, below, and locally within permafrost. 13 In the continuous-permafrost zone, the
Mean annual ground-surface temperature
-20 -10 /o +10 +20°C So
,....._ _ _..:....._ Maximum annual

Minimum annual
temperature temperature
Level of zero
Depth below annual amplitude
ground surface
l
·t
2.13
Base of permafrost Figure 2.12.1. Location of per-

mafrost below ground surface in
~ relation to ground temperature

(after Williams67).
'It has been estimated that the total stored heat in the earth to a depth of 3 km amounts to 2 x l o2 1 cal and that one
percent, or 2 x l0 19 cal, of this can be commercially recovered. 63
I in the
to frac-
10le.
y a few
·water.
y, New
ck that
68 Fig-
which
:re are
where
~. per-
water,
many
wells
ercan
te, the
Figure 2.12.2. Distribution of pennafrost in

Southern limit of continuous permafrost Southern limit of discontinuous permafrost the Northern Hemisphere (after Williams 67 ).
best sources of water are in unfrozen alluvium beneath large lakes, in major valleys, and adja-
cent to riverbeds. In the discontinuous-permafrost zone, groundwater can be produced locally
from shallow aquifers; however, because of potential pollution from ground surface, sources
beneath the frozen layer are preferable.
2.13 GROUNDWATER IN THE UNITED STATES

Productive aquifers and withdrawals from the wells in the United States are shown in Figures
2.13.1 and 2.13.2. These maps show regions in which moderate to large supplies of water can
be obtained from wells. Unconsolidated and consolidated aquifers are indicated in addition to
Jer- primary geologic fonnations. Blank areas delineate generally those regions not known to pro-
·e in duce yields of more than 3 lis (48 gpm) to a well.
re Areal distribution of groundwater in the United States can best be described by dividing
the contenninous United States into the groundwater regions shown in Figures 2.13.1 and
2.13.2. Table 2.13.1 is a summary of the principal physical and hydrologic characteristics of
the groundwater regions. Brief geologic and groundwater summaries for each of the 15 regions
are provided on the following pages.
'
A TLANTIC
O CEAN
of
Mexico
..
0 100 2QO 300 ......
0 50 Miles """
o 250 Kilometers f--T-J I I

1
0 100 2Q0 300
1 1
1
Figure 2
0 50 Kilomelets
Figure 2.13.1. Groundwater regions in the United States. [The Alluvial Valleys region (region 12) is shown on Figure 2.13.2.] 23
1. Western Mountain Ranges
This mountainous region serves as the principal source of water in the western United States
because the bulk of precipitation falls here and thereafter supplies streams and aquifers by its
runoff. Rocks are mainly hard and dense; they shed water rather than absorbing it, although
weathered surficial rock may locally yield limited groundwater. Some aquifers are to be found
in alluvium contained in small intermontane valleys. Because of the thinness and rapid drain-
ing of the mantle rock, it is not easy to obtain groundwater from wells. Small springs, wells in
valleys, and small surface reservoirs meet most domestic water supply needs.
2. Alluvial Basins
The basins in this region consist of vast depressed areas bounded by adjacent highlands. They
are partly filled by erosional debris in the form of alluvium and serve as the storage volumes
for water flowing from nearby highlands. The alluvial fill functions as an ideal aquifer and cre-
ates the opportunity for development of high-yielding wells. Because of the prevailing arid cli-
mate, groundwater development for irrigation is much in demand. Replenishment rates usually
are far less than withdrawal rates, so groundwater levels decline as storage is depleted. Locally,
artificial recharge (see Chapter 13) has helped to alleviate this problem of overproduction.
•d
region
.ciated
I region
- v
HAWAII
o::s
~
~ I '=?•• ·' ~PUERTO
.o
~~~ Qo ~
RICO AND
., - VIRGIN ISLANDS
Figure 2.13.2. Alluvial Valleys groundwater region of the United States.23
3. Columhill Lava Plateau

This area is fonned principally by extrusive volcanic rocks, mainly lava flows, interbedded
with or overlain by alluvium and lake sediments. Water originates chiefly from mountains on
the perimeter of the region. The lava rocks tend to be highly permeable as a result of tubes and
shrinkage cracks and thus fonn highly productive aquifers. The large volumes of groundwater
States .1. discharge as major springs or as streams with sustained base flows. Because of the great thick-
; by its •
.,~ ness of the lava flows, groundwater is most readily available in valley bottoms; however, in the
though higher plateau areas, deep wells are required to extract groundwater for irrigation.
:found
drain- •
1ells in 4. Colorado Plauau and Wyoming Basin
This region consists of sedimentary strata, chiefly interbedded sandstone and shale; these are
generally horizontal but in places are folded, tilted, or broken by faults. The plateaus are rather
high, dry, and deeply dissected by streams. Prospects for large-scale groundwater development
are poor; nevertheless, small water supplies for domestic and livestock purposes are widely
:. They available. Most aquifers are sandstone beds, although limestone and alluvium yield water in a
>lumes few places.
1d ere-
rid cli-
tsually 5. High Plains
ocally, Here alluvium forms a vast plain extending eastward from the Rocky Mountains. The bulk of
.on. it is classified as a single stratigraphic unit, the Ogallala Formation, which covers older rocks
QC)
Table 2.13.1 Summary of the Principal Physical and Hydrologic Characteristics of the Groundwater Regions of the United States
=
Characteristics of the dominant aquifers
Components of the system Water-bearing Composi- Storage and transmission Recharge and
openings tion properties discharge conditions
Unco~fined Confining Con~ined Presence and Primary Secon- Degre~ of Porosity Transmissivity Recharge Discharge
aqu1fer beds aqu1fers arrangement dary solubility
~ -
~ ~ ~ C/) c:a C/)
0 c:a ~ ..0 ~ ~ > ~ "0
-5 ·:;
Region
No
Name _
c
e
a.
_
c
~
~
~
_
c
~
~
g
-
~
C)
-~
c
c
~
cr-
ar. -og. . .
-o
~
-~ C/)
~
~
..c
1>-
.{g
N ~
~
c
~ ~
ara. ~
-~
• c:a ~ c:a >
C/)
.._ c:a ·- g? ~ ·:; 'E ~ -o ~
C/)
...l<: ·- ..::1 ~ E ~ 1>-
C/)
E ·c: a> C/)
~ g? ~ 0 ·~ -~ ~ 5- U -~ g 8 ~ ~ ~ ~ [ g - I~ 0 I~ .{g ~ 'E ~ -~
~ - §, 00 0' §, _:?; C:0 ::J C:0 "0 .._: ~ "0 0 O Cll 0 -~ C\! "0 ~ "0 N t) 0 ~ C:0
·c;; o ..... ·c;; c:a ·c;; n g? '8 -o ~ ~ :g o c:::n -s -o -o o E c o c:a ..c ~
~
c =
~ 3
~ -~
~
g
~
::J
£
~
-~
~
-5
e
t5
::J
C..
~
~
'E
~ ·:; Q; ~
c g~ a;
Cll
§ ~
C/)
§ ~
g
~
-o
~
~ ~
~
_
,.!.
q _
N
g
E
g
"ff
- N
E
~
o.o
~ 1
a> Cll
E
-§,
~
'§
C/)
-g
c:a
~
3
c:a ~ g c:a § c:a '8 c:a 0 § -og ~ g ~ (.) ~ c "! £ q Lll ~ "ff - ~ ~ ..c -g § g....
·s -
"0
~
0.
c.. -~ g ~ ~ c:a ·- ::J -g
..0
~ ~ ~ ro ..c _
£I
.2 .2 u .2 >- en a> o o N o.o - c:a -..:::::
.a
C/)
~ g ~ ~ g?, ~ :g, ~ g :J ~ ~ ~ ~ -~ -~ c:a ~ § ~ g ~ ~ ~ ::!.- ~ ~ ~ 55 ~ ~ ~ ~ ~

~>-
o
-~
·E
o
~ .5
>- ..c
~
-
~
>-
!:o ~::1
~ ~
..c -~
·c;:;:: 8c ~ o~ o~ ~
o
~::1 n
~
~o -§ -g
-~
~
o
~ ~
c:a o
ro
E
~ ~c:a o
ro ~ .§ -~ ~ .§
E ~ o. a> a.
~
>
~
-
0 C/) 0
:r: ::2: 0 :r: 1- .E :r: Z ::2: 1- (f) 1- ::::> u 0 a.. a.. 1- u. en £ ::2: (f) ...J ::2: (f) 1 ...J ::2: en > =:) ...J ...J (f) w E
1 Western Mountain Ranges X X X X X X X X X X X X
2 Alluvial Basins X X X X X X X X X X X
3 Columbia Lava Plateau X X X X X X X X X X X X
4 Colora_do Plat~au and X X X X X X X X X X X X

Wyommg Bas1n
1
5 High Plains X X X X X X X X X X
6 Non~laciated Central X X X X X X X X X X X X X X I
Reg1on J
7 Gla~iated Central
Reg1on
X X X X X X X X X X X X X J
8 Piedmont and Blue Ridge X X X X X X X X X X X
g Northeast and Superior X X X X X X X X X X X

Uplands
1O Atlantic and_ Gulf

Coastal Plam
X X X X X X X X X X X X X J
11 Southeast Coastal Plain X X X X X X X X X X X X
12 Alluvial Valleys X X X X X X X X X X
13 Hawaii X X X X X X X X X X X
14 Alaska X X X X X X X X X X X
15 Puerto Rico and Virgin X X X X X X X X X X X X X

Islands
Source: Heath 23
X
to thicknesses exceeding 150m. The sand and the gravel of the formation consitute an aquifer
X
yielding 10 1/s (151 gpm) to more than 60 1/s (451 gpm)of water from individual wells. The
X region is generally semiarid so that groundwater recharge from precipitation is extremely
small. The productiveness of wells has encouraged pumping of groundwater, especially for
irrigation in Texas. This water is derived from storage; as a result, water tables have declined
substantially for many years.
6. Nonglaciated Central Region

This is a large, complex area characterized by plains and plateaus underlain by consolidated
sedimentary rocks. Alluvial deposits of substantial width and thickness fonn good aquifers but
only along major streams. Aquifers in most of the region are dolomitic limestone and sand-
stone with low to moderate yields. The region includes some of the least productive aquifers
in the United States because of low yield, saline water, or both. Wells in some of the karst areas
penetrate caverns and exhibit highly variable yields.
7. Glaciated Central Region

Although hydrogeologically similar to the previous region, this area differs fundamentally by
the presence of glacial drift deposited by the ice and meltwaters of the continental glaciers. The
drift consists mostly of fine-grained rock debris together with beds of water-sorted sand and
gravel. In portions of the area, the drift is more than 300 m thick and forms an important
aquifer. In this glaciated region, large-diameter wells will yield sufficient water to meet domes-
tic needs of a family. Excellent aquifers can be found along watercourses where rapidly flow-
ing meltwaters removed fine materials and left behind permeable deposits of sand and gravel.
8. Piedmont and Blue Ridge

This mountainous area consists of varying parallel highlands with differing geologic struc-
tures, including crystalline rocks, limestone, dolomite, sandstone, and shale. Groundwater
productivity ranges from small to moderate or is erratic. Shallow wells can usually obtain
small yields for domestic supplies from the weathered rock. Wells of highest average yield
occur in the valleys because rocks are more intensely fractured and because of groundwater
drainage from the surrounding hills. In the limestone areas, much of the drainage is under-
ground so the chances of obtaining moderately productive wells are good.
9. Northeast and Superior Uplands

Because this region is considerably more rugged than the Glaciated Central Region, there is
greater local variation in thickness of the glacial drift. Usually, hilltops are covered by a thin
mantle of till; slopes contain even less with bedrock outcrops, and valleys are underlain by
r:: thicker drift and till. Near the sea, glacial material is overlain or interbedded with marine
'2>
> deposits. In most of the area, aquifers are not productive. Except for a few areas where bedrock
'0
r::
111
yields moderate quantities of groundwater, the principal groundwater sources are sand and
0
0
gravel deposits occurring as outwash plains or as channel fillings in the stratified drift.
a:ocn
t:'O gJ
CDr::
:J~
Q.,!!! ~CD I 0. Atlantic and Gulf Coastal Plain
J:
....
I.C) Cli
~
This extensive coastal plain consists of a huge seaward-thickening wedge of generally uncon-
=::!
~ solidated sedimentary rocks. The sediments are mainly stratified layers of clay, silt, sand,
gravel, marl, and limestone. The maximum thickness of deposits beneath the coastline varies
from 100 m in the northeast to more than 10,000 m in the south. Essentially all of the area PROI
contains extensive and productive aquifers. The principal aquifers are found in sand and gravel 2.2.1 A
beds, while limestone aquifers are most important in the southeastern portion of the region. 51 sizes a:
Problems of seawater intrusion and land subsidence are significant in concentrated pumping spheres.
localities. packing
11. Southeast Coastal Plain

The Southeast Coastal Plain region consists of layers of sand clay over semiconsolidated and
consolidated carbonate rock. Surface areas are underlain by unconsolidated deposits of clay,
sand, gravel, and shell beds, and deeper layers consist of alternating layers of serniconsolidated
and consolidated limestones and dolomites. One of the most prolific aquifers in the world is Cubic P
the Floridian Aquifer, which is located in carbonate units. The surficial sand and gravel 2.2.2 c
deposits also have water. in Table
between
ofProbl
12. Alluvial Valleys Region
2.2.3 Tl
The Alluvial Valleys Region (see Figure 2.13.2) consists of thick sand and gravel deposits these da
beneath floodplains and stream terrace deposits. The many river systems in this region have and asse
deposited thick sequences of highly porous and permeable sand and gravel-glacial outwash What WI
by streams that carried water from melting ice during the Pleistocene era. The resulting long, u.~
narrow aquifer systems were the foundations for the modem river systems. Sie
13. Hawaii
Basaltic lava flows make up the bulk of the islands and constitute the most important aquifers.
Openings within or between the flows are responsible for the generally high permeability. Near
eruptive centers of volcanoes vertical dikes of dense impermeable rock may separate high-level
groundwater in inland areas from basal groundwater in the coastal low lands. The basal water ,
forms a lens of fresh groundwater floating on underlying sea water (see Chapter 14). Ground-
water is extensively developed for irrigation and municipal purposes on Maui and Oahu.
Totals:
14. Alaska
2.2.4 TI
The principal aquifers in this region are bodies of water-sorted sand and gravel within both gla- distributi
cial drift that covers the uplands and glacial outwash and other alluvial deposits extending from Plot the l
the uplands into the lowlands. The most productive aquifers are found in the vast central plateau the samp
and on the slopes of the southern mountain system. Pennafrost predominates in the penneable Pc:
deposits of the arctic slope, so groundwater is available only locally near large bodies of surface
water. Along the southeastern coastal area, productive alluvial deposits are scarce.
15. Puerto Rico and Virgin Islands

This island region consists of alluvium and limestone overlying and bordering fractured
igneous rocks. It is underlain by both limestone and volcanic and intrusive igneous rocks. The
islands receive large amounts of rainfall; Puerto Rico has almost 6 ft (2 m) of annual ground-
water recharge. The alluvium with sand and gravel in the stream valleys and along the coast is
an effective aquifer. There are also limestone areas that are aquifers.
Problems 83
[ne varies PROBLEMS

f the area
nd gravel 2.2.1 A cubical pattern of grain packing is shown below; grain 2.5.1 Determine the water level rise in an unconfined aquifer pro-
region. 51 sizes are uniform and each sphere touches all neighboring duced by a seasonal precipitation of four inches. The aquifer's
pumping spheres. Determine the porosity of a sample with this type of grain porosity is 20 percent and its specific retention is 9 percent.
packing. Does the porosity depend on the grain size? 2.5.2 The leakage from the artificially constructed Tempe Town
Lake in Tempe, AZ, can be as low as 0.5 ft/day or as high as 3
ft/day. The lake covers 222 surface acres. If the specific yield of
the subsurface formation is 20 percent, estimate the average
lated and regional groundwater level rise assuming that the aerial extent of
the effect of leakage is
~of clay,
(a) 222 acres
solidated
world is Cubic Packing (b) I mile 2
d gravel 2.2.2 Considering the ranges of values for sedimentary materials (c) 5 mile2
in Tables 2.2.1 and 2.2.2, comment on a possible relationship (d) 25 mile2
between porosity versus grain size. Compare your answer to that 2.5.3 How much water can be produced by lowering the water
of Problem 2.2.1. What conclusion(s) can you draw? table of an unconfined aquifer 7 ft over an area of 1 mile2? The
2.2.3 The results of a sieving analysis are tabulated herein. Using aquifer's porosity and specific retention are 0.38 and 0.15,
deposits these data, prepare a grain size distribution curve for this sample respectively.
[on have and assess whether the sample is well graded or poorly graded. 2.8.1 The coefficient of storage of a confined aquifer is found to
What would be reasonable porosity values for this sample? be 6.8 x 10-4 as a result of a pumping test. The thickness of the
outwash
ng long, U.S. Standard aquifer is 50 m and the porosity of the aquifer is 0.25. Determine
Sieve Number Grain size (mm) Mass retained (g) the fractions of the expansibility of water and compressibility of
4 4.750 11.92 the aquifer skeleton in making up the storage coefficient of the
aquifer. (Refer to Example 2.5.2.)
8 2.360 2.66
16 1.180 37.30 2.8.2 The specific storage of a 45-m thick confined aquifer is
20 0.840 87.53 3.0 x w-5 m- 1• How much water would the aquifer produce if the
tquifers.
piezometric surface is lowered by 10m over an area of 1 km2?
ty. Near 25 0.710 78.07
30 0.600 72.95 2.13.1 Prepare a summary (with illustrative figures) addressing
gh-level
40 0.425 98.82 the following issues for one or more of the regions listed below:
al water
]round- 100 0.150 50.85 Geographical area
u. 200 0.075 8.10 Climate
Pan <0.075 1.80 Geology
Total sample weight 450.00 Surface water (short description of major surface waters)
2.2.4 The results of a hydrometer test to determine the grain size Groundwater (types and locations of aquifers and other con-
oth gla- distribution of a sample of fine sediments are tabulated herein. ditions that have a bearing on the groundwater of the region)
lg from Plot the grain size distribution and comment on the uniformity of Groundwater facts (recharge, uses, pollution, salt water intru-
plateau the sample. sion, etc.)
meable Particle grain The regions are to be selected from the following:
surface size (mm) Percent finer than Africa
1.000 98.900
Artie and Antarctic
0.700 95.700
Alaska, Canada, and Greenland
0.500 82.800
Australia and Oceana
0.350 65.600
tctured 0.250 32.300 China
<s. The 0.175 6.750 Central Europe
round- 0.125 0.600 Eastern Europe
;oast is 0.088 0.100 Great Britain
0.0625 0.000 India
(continued on next page)
Indochina Russia 38. Ma

Mexico and Central America Scandinavian Countries cepts in
Middle Eastern Countries South America 39. Me
Plains 1
http://w
1998
REFERENCES
40. Me
I. Agricultural Research Service, Field Manual for Research in Agricul- 20. Feth, J. H., Selected References on Saline Ground-Water Resources of nations,
tural Hydrology, Agric. Handbk. no. 224, U.S. Dept. of Agric., 215 pp., the United States, U.S. Geological Survey Circular 499, 30 pp., 1965.
41. Me
1962. 21. Groundwater Management Districts Association, Newsletter, v. 2, tions, l.J
2. Alley, W. M., R. W. Healy, J. W. LaBaugh, and T. E. Reilly, Flow and no. 2, http://www.gmdausa.org/newsletter.htm, Spring 2002. 42. Me
storage in groundwater systems, Science, v. 206, June 14, 2002. 22. Grubb, H. F., Summary of Hydrology of the Regional Aquifer Sys- States, i
3. Alley, W.M., T. E. Reilly, and 0. L. Franke, Sustainability of Ground- tems, Gulf Coastal Plain, South Central United States, U.S. Geological 43. Me
Water Resources, U.S. Geological Survey Circular 1186, http://water. Survey Professional Paper 1416-A, 1998. Survey'
usgs.gov/pubs/circ/circ1186, U.S. Geological Survey, Denver, CO, 1999. 23. Heath, R. C., Groundwater Regions of the United States, U.S. Geo- 44. Mo
4. Baver, L. D., et al., Soil Physics, 4th ed., John Wiley & Sons, New logical Survey Water-Supply Paper 2241, Washington, DC, 1984. Survey'
York, 498 pp., 1972. 24. Johnson, A. 1., Methods of Measuring Soil Moisture in the Field, U.S. 45. Mo
5. Bear, 1., Dynamics of Fluids in Porous Media, Dover Publications, Geological Survey Water-Supply Paper 1619-U, 25 pp., 1962. ical Pn
Mineola, NY, 1972. 25. Johnson, A. 1., Specific Yield-Compilation of Specific Yields for Var- logic L
6. Black, C. A. (ed.), Methods of Soil Analysis, Part 1, Agronomy Mono- ious Materials, U.S. Geological Survey Water-Supply Paper 1662-D, 74 Geolog1
graph no. 9, Amer. Soc. Agronomy, Madison, WI, 770 pp., 1965. pp., 1967. 46. Pol
7. Blank, N. R., and M. C. Schroeder, Geologic classification of aquifers, 26. Johnston, R. H., The hydrologic responses to development in regional the Mec
Ground Water, v. 11, no. 2, pp. 3-5, 1973. sedimentary aquifers, Ground Water, v. 27, no. 3, pp. 316-322, 1989. Witlulrc.
8. Brown, D. S. and J. T. Patton, Recharge to and from the Edwards 27. Jones, 0. R., and A. D. Schneider, Determining specific yield of the 1972.
Aquifer in the San Antonio Area, Texas, U.S. Geological Survey Open-File Ogallala aquifer by the neutron method, Water Resources Research, v. 5, 47. Pril
Report 96-181, 1995. pp. 1267-1272, 1969. Effect~
9. Brucker, R. W., et al., Role of vertical shafts in the movement of ground 28. Karst in China, Institute of Hydrogeology and Engineering Geology, ply Pap.
water in carbonate aquifers, Ground Water, v. 10, no. 6., pp. 5-13, 1972. Chinese Academy of Geological Sciences, Shanghai People's Publishing 48. Put:
10. Bryan, K., Classification of springs, Jour. Geology, v. 27, pp. House, 1976. Aquifer
522-561, 1919. 29. Keefer, W. R., The Geologic Story of Yellowstone National Park, U.S. Investig
11. Burdon, D. J., and N. Papakis, Handbook of Karst Hydrogeology with Geological Survey Bull. 1347,92 pp., 1971. 49. Sla
§pecial Reference to the Carbonate Aquifers of the Mediterranean Region, 30. Kuniansky, E. L., L. Fahlquist, and A. F. Ardis, Travel Times along the Edr1
Inst. Geology and Subsfc. Research, United Nations Spec. Fund, Karst reports/
Selected Flow Paths of the Edwards Aquifer, Central Texas, in U.S. Geo-
Groundwater Inv., Athens, 276 pp., 1963. logical Survey Karst Interest Group Proceedings, ed. E. L. Kuniansky, 50. Soi
12. Burdon, D. J., and C. Safadi, The karst groundwaters of Syria, Jour. U.S. Geological Survey Water-Resources Investigations Report 01-4011, book, n•
Hydrology, v. 2. pp. 324-347, 1964. 2001. 51. Stri
31. LaMoreaux, P. E., eta!., Hydrology of limestone terranes, annotated terranes
13. Cederstrom, D. J., et al., Occurrence and Development of Ground
bibliography of carbonate rocks, Geological Survey Alabama Bull. 94(A), v. 8, pp.
Water in Pennafrost Regions, U.S. Geological Survey Circular 275, 30 pp.,
1953. 242 pp., 1970. 52. Sur
32. Lattman, L. H., and R. R. Parizek, Relationship between fracture mal ph
14. Charbeneau, R. J., Groundwater Hydraulics and Pollutant Transport,
traces and the occurrence of ground water in carbonate rocks, Jour. Socorro
Prentice Hall, Upper Saddle River, NJ, 2000.
Hydrology, v. 2, pp. 73-91, 1964. 53. Sur
15. Childs, E. C., An Introduction to the Physical Basis of Soil Water Phe-
33. LeGrand, H. E., and V. T. Stringfield, Karst hydrology-A Review, U.S. Ge.
nomena, John Wiley & Sons, London, 493 pp., 1969.
Jour. Hydrology, v. 20, pp. 97-120, 1973. Survey 1
16. Dept. of Economic and Social Affairs, Ground-Water Storage and
34. Lohman, S. W., Ground-Water Hydraulics, U.S. Geological Survey 54. Sur
Artificial Recharge, Natural Resources, Water Ser. no. 2, United Nations,
Professional Paper 708, 70 pp., 1972. gram of
New York, 270 pp., 1975.
Circula.
17. Dos Santos, A. G., Jr., and E. G. Youngs, A study of the specific yield 35. Lohman, S. W., et al., Definitions of Selected Ground-Water Terms-
55. Tay
in land-drainage situations, Jour. Hydrology, v. 8, pp. 59-81, 1969. Revisions and Conceptual Refinements, U.S. Geological Survey Water-
the Imp.
18. Dutcher, L. C., Preliminary Appraisal of Ground Water in Storage Supply Paper 1988, 21 pp., 1972.
Circula
with Reference to Geothermal Resources in the Imperial Valley Area, Cal- 36. Luckey, R. R., E. D. Gutentag, F. J. Heimes, and J. B. Weeks, Digital 2001.
ifornia, U.S. Geological Survey Circular649, 57 pp., 1972. Simulation of the Ground-Water Flow in the High Plains Aquifer in Parts of
56. The:
19. Epstein, J. B., D. J. Weary, R. C. Orndorff, Z. C. Bailey, and R. C. Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota,
3472, PJ
Kerbo, National Karst Map Project: An Update, in U.S. Geological Sur- Texas, and "Yoming, U.S. Geological Survey Professional Paper 1400-D,
vey Karst Interest Group Proceedings, Shepherdstown, West Virginia, 1986.
August 20-22, 2002, ed. E. L. Kuniansky, U.S. Geological Survey Water- 37. Manger, G. E., Porosity and Bulk Density of Sedimentary Rocks, U.S.
Resources Investigations Report 02-417 4, 2002. Geological Survey Bull. 1144-E, 55 pp., 1963.
References 85
38. Maxey, G. B., and J. E. Hackett, Applications of geohydrologic con- 57. U.S. Geological Survey, Ground Water Atlas of the United States,
cepts in geology, Jour. Hydrology, v. 1, pp. 35-46, 1963. U.S. Geological Survey Hydrologic Investiga~ions Atlases,
39. McGuire, V. L., and B. C. Fisher, Water-Level Changes in the High http://capp.water.usgs.gov/gwa!gwa.html
Plains Aquifer-1980-1998, U.S. Geological Sun,ey Web Report HP 98, HA 730-B California, Nevada 1995
http://www-ne .cr. usgs. gov /high plai nslh p98 _web_reportlh p9 8fs. htm, HA 730-C Arizona, Colorado, New Mexico, Utah 1995
1998. HA 730-D Kansas, Missouri, Nebraska 1997
40. McQueen, I. S., Evaluating the reliability of specific-yield determi-
HA 730-E Oklahoma, Texas 1996
sources of nations, low: Research U.S. Geological Survey, v. I, pp. 371-376, 1973.
HA 730-F Arkansas, Louisiana, Mississippi 1998
1965. 41. Meinzer, 0. E., Outline of Ground- Water Hydrology with Defini-
HA 730-G Alabama, Florida, Georgia, South Carolina 1990
Her, v. 2, tions, U.S. Geological Survey Water-Supply Paper 494, 71 pp., 1923.
HA 730-H Idaho, Oregon, Washington 1994
42. Meinzer, O.E., The Occurrence of Ground Water in the United
iJ.ifer Sys- States, U.S. Geological Survey Water-Supply Paper 489, 321 pp., 1923. HA 730-I Montana, North Dakota, South Dakota, Wyoming 1996
:eological 43. Meinzer, 0. E., Large Springs in the United States, U.S. Geological HA 730-J Iowa, Michigan, Minnesota, Wisconsin 1992
Survey Water-Supply Paper 557, 94 pp., 1927. HA 730-K Illinois, Indiana, Kentucky, Ohio, Tennessee 1995
JS. Geo- 44. Monroe, W. H., A Glossary of Karst Terminology, U.S. Geological HA 730-L Delaware, Maryland, New Jersey, North Carolina, Pennsyl-
k Survey Water-Supply Paper 1899-K, 26 pp., 1970. vania, Virginia, West Virginia 1997
ield, U.S. 45. Morris, D. A., and A. I. Johnson, Summary of Hydrologic and Phys- HA 730-M Connecticut, Maine, Massachusetts, New Hampshire, New
ical Properties of Rock and Soil Materials, as Analyzed by the Hydro- York, Rhode Island, Vermont 1995
rfor Var- logic Laboratory of the U.S. Geological Survey 1948-1960, U.S. HA 730-N Alaska, Hawaii, Puerto Rico, and U.S. Virgin Islands 1999
62-D, 74 Geological Survey Water-Supply Paper 1839-D, 42 pp., 1967.
58. Vandike, J. E., The Hydrology of Maramec Spring, Brochure, Water
46. Poland, J. F., et a\., Glossary of Selected Terms Useful in Studies of Resources Report no. 55, Missouri Department of Natural Resources,
regional the Mechanics of Aquifer Systems and Land Subsidence Due to Fluid Division of Geology and Land Survey, Rolla, Missouri, 1996.
989. Withdrawal, U.S. Geological Survey Water-Supply Paper 2025, 9 pp., 59. VogeL J. C., and D. Ehhalt, The use of carbon isotopes in groundwa-
ld of the 1972.
ter studies, Radioisotopes in Hydrology, Inti. Atomic Energy Agency,
rch, v. 5, 47. Prill, R. C., Specific Yield-Laboratory Experiments Showing the Vienna, pp. 383-395, 1963.
Effect of Time on Column Drainage, U.S. Geological Survey Water-Sup-
60. Waring, G. A., Thermal Springs of the United States and Other
}eology, ply Paper 1662-B, 55 pp., 1965. Countries of the World-A Summary, U.S. Geological Survey Profes-
blishing 48. Puente, C., Method of Estimating Natural Recharge to the Edwards sional Paper 492, 383 pp., 1965.
Aquifer in the San Antonio Area, Texas, U.S. Geological Survey Water-
61. White, D. E., Thermal waters of volcanic origin, Bull. Geological
zrk, U.S. Investigations Report 78-10, 1978. Soc. Amer., v. 68, pp. 1637-1658, 1957.
49. Slattery, R. N., and D. E. Thomas, Recharge to and Discharge from
62. White, D. E., Magmatic, connate, and metamorphic waters, Bull.
~salong the Edwards Aquifer in the San Antonio Area, 2001, http://tx.usgs.gov/
Geological Soc. Amer., v. 68, pp. 1659-1682, 1957.
S. Geo- reports/dist/dist-2002-01, San Antonio, TX, October 2002.
63. White, D. E., Geothermal Energy, U.S. Geological Survey Circular.
1iansky, 50. Soil Survey Staff, Soil Survey Manual, U.S. Dept. Agriculture Hand-
519, 17 pp., 1965.
1-4011, book, no. 18, 503 pp., 1951.
64. White, W. B., Conceptual models for carbonate aquifers, Ground
51. Stringfield, V. T., and H. E. LeGrand, Hydrology of carbonate rock
Water, v. 7, no. 3, pp. 15-21, 1969.
notated terranes-A review with reference to the United States, Jour. Hydrology,
65. Wicks, C. M. and J.A. Hoke, Prediction of quality and quantity of
94(A), v.8,pp. 349-417,1969.
Maramec Spring Water, Ground Water, vol. 38, no. 2, pp. 218-225,2000.
52. Summers, W. K., Annotated and Indexed Bibliography of Geother-
66. Wigley, T. M. L., Carbon 14 dating of groundwater from closed and
racture mal phenomena, New Mexico Bur. Mines and Mineral Resources,
open systems, Water Resources Research, v. 11, pp. 324-328, 1975.
;, Jour. Socorro, 665 pp., 1972.
67. Williams, J. R., Ground Water in Permafrost Regions-An Annotated
53. Sun, R. J. (ed.), Regional Aquifer-System Analysis Program of the
Bibliography, U.S. Geological Survey Water-Supply Paper 1792, 294 pp.,
teview, U.S. Geological Survey-Summary of Projects, 1978-84, U.S. Geological
1965.
Survey Circular 1002, 1986.
68. Williams, J. R., Ground Water in the Permafrost Regions of Alaska,
Survey 54. Sun, R. J., and R. H. Johnson, RegionalAquifer-SystemAnalysis Pro-
U.S. Geological Survey Professional Paper 696, 83 pp., 1970.
gram of the U.S. Geological Survey, 1978-1992, U.S. Geological Survey
Circular 1099, 1994. 69. Williamson, A. K., and H. F. Grubb, Groundwater Flow in Gulf
~rms
Coast Aquifer Systems, South Central United States, U.S. Geological Sur-
Water- 55. Taylor, C. J., and W. M. Alley, Ground- Water-Level Monitoring and
vey Professional Paper 1416-F, 2001.
the Importance of Long-Term Water-Level Data, U.S. Geological Survey
Circular 1217, http://water.usgs.gov/pubs/circ/circ 1217, Denver, CO, 70. Winter, T. C., J. W. Harvey, 0. L. Franke, and W. M. Alley, Ground-
Digital
2001. water and Surface Water: A Single Source, U.S. Geological Survey Cir-
~rtsof
cular 1139, http://water.usgs.gov/pubs/circ/circl139, Denver, CO, 1998.
·akota, 56. Thatcher, L., et al., Dating desert ground water, Science, v. 134, no.
fOO-D, 3472,pp. 105-106,1961.
s, u.s.
Chapter 3
Groundwater Movement
Groundwater in its natural state is invariably moving. This movement is governed by estab-
lished hydraulic principles. The flow through aquifers, most of which are natural porous
media, can be expressed by what is known as Darcy's law. Hydraulic conductivity, which is a
measure of the permeability of the media, is an important constant in the flow equation. Deter-
mination of hydraulic conductivity can be made by several laboratory or field techniques.
Applications of Darcy's law enable groundwater flow rates and directions to be evaluated. The
dispersion, or mixing, resulting from flows through porous media produces irregularities of
flow that can be studied by tracers. In the zone of aeration, the presence of air adds a compli-
cating factor to the flow of water.
3.1 DARCY'S LAW

More than a century ago Henry Darcy,* a French hydraulic engineer, investigated the flow of
water through horizontal beds of sand to be used for water filtration. He reported 24 in 1856:
, I have attempted by precise experiments to detennine the law of the flow of water through
filters .... The experiments demonstrate positively that the volume of water which passes
through a bed of sand of a given nature is proportional to the pressure and inversely proportional
to the thickness of the bed traversed; thus, in calling s the surface area of a filter, k a coefficient
depending on the nature of the sand, e the thickness of the sand bed, P - H0 the pressure below
the filtering bed, P + H the atmospheric pressure added to the depth of water on the filter; one
has for the flow of this last condition Q = (k&le)(H + e + H0), which reduces to Q = (k&le)(H +e)
when H0 = 0, or when the pressure below the filter is equal to the weight of the atmosphere.
This statement, that the flow rate through porous media is proportional to the head loss
and inversely proportional to the length of the flow path, is known universally as Darcy's law.
It, more than any other contribution, serves as the basis for present-day knowledge of ground-
water flow. Analysis and solution of problems relating to groundwater movement and well
hydraulics began after Darcy's work.
3.1.1 Experimental Verification

The experimental verification of Darcy's law can be performed with water flowing at a rate Q
through a cylinder of cross-sectional area A packed with sand and having piezometers a
*An interesting summary of the life and accomplishments of Henry Darcy was prepared by Fancher. 31
86
3.1 Darcy's Law 87
distance L apart, as shown in Figure 3.1. 138 · 52 Total energy heads, or fluid potentials, above a
datum plane may be expressed by the energy equation
p v2 2 p v2
_L+-1-+z = - +-2- +z +h (3. 1.1)
r 2g ' r 2g 2 L
where p is pressure. y is the specific weight of water, vis the velocity of flow, g is the acceler-
ation of gravity, z is elevation, and hL is head loss. Subscripts refer to points of measurement
identified in Figure 3.1.1. Because velocities in porous media are usually low, velocity heads
may be neglected without appreciable error. Hence, by rewriting, the head loss becomes
(3.1.2)
· estab- Therefore, the resulting head loss is defined as the potential loss within the sand cylinder, this
porous energy being lost by frictional resistance dissipated as heat energy. It follows that the head loss
ich is a is independent of the inclination of the cylinder.
Deter- Now, Darcy's measurements showed that the proportionalities Q- hL and Q- 1/L exist.
rtiques. Introducing a proportionality constant K leads to the equation
xi.The
ities of Q=-KAhL (3.1.3)
ompli- L
Expressed in general tenns

dh (3. 1.4)
Q=-KA-
dl
or simply
low of
1856: Q dh (3.1.5)
v=-=-K-
A dl
·n.al
It
w
. e)
.d loss
'slaw.
uund-
i well
:ateQ
ters a
Datum plane /o Figure 3.1.1. Pressure
distribution and head loss in
flow through a sand column.
88 Chapter 3 Groundwater Movement
where vis the Darcy velocity or specific discharge; K is the hydraulic conductivity, a constant
that serves as a measure of the penneability of the porous medium; and dh/dl is the hydraulic
gradient. The negative sign indicates that the flow of water is in the direction of decreasing
head. Equation 3.1.5 states Darcy's law in its simplest form, that the flow velocity v equals the
product of the hydraulic conductivity and the hydraulic gradient.
EXAMPLE 3.1.1 A field sample of an unconfined aquifer is packed in a test cylinder (see Figure 3.1.1). The length and the
diameter of the cy Iinder are 50 em and 6 em, respectively. The field sample is tested for a period of 3 min
under a constant head difference of 16.3 em. As a result, 45.2 cm3 of water is collected at the outlet.
Determine the hydraulic conductivity of the aquifer sample.
SOLUTION The cross-sectional area of the sample is
A= 1tD2 = 7t(0.06m)2 =0.00283m2

4 4
The hydraulic gradient, dh/dl, is given by
dh {-16.3cm) -0.3
26
dl 50cm
and the average flow rate is

3
45.2 cm . /d
Q= . =1 5.07cm 31mm= 0.0217m 3 ay
3 rrun
Apply Darcy's law, Equation 3.1.4, to obtain the hydraulic conductivity as

3
Q=-KA dh ~K= Q 0.0217 m /day 23 .5 m/da
dl A(dhldl) (0.00283 m2)( -0.326) y
•
E.fAMPLE 3.1.2 A confined aquifer with a horizontal bed has a varying thickness as shown in Figure 3.1.2. The aquifer
is inhomogeneous with K = 12 + 0.006x, wherex =0 at section (1), and the piezometric heads at sections
(1) and (2) are 14.2 m and 18.8 m, respectively measured above the upper confining layer. Assuming the
flow in the aquifer is essentially horizontal, determine the flow rate per unit width.
3.1.2
SOLUTION Darcy's law for a constant thickness aquifer is given by Equation 3.1.4,
Q=-KA dh
dl
K = 12 + 0.006x
30m
t
~+--------3,600 m-------+~ Figure 3.1.2. Aquifer for Example 3.1.2
3.1 Darcy's Law 89
:onstant Since the aquifer thickness is variable in this problem, we must also write the cross-sectional area and
1draulic
the hydraulic gradient as a function of the distance x. Assuming a unit width, A = b1 + (bz - bl )x , where
:reasing
b1 =30m, b2 = 75 m, and L = 3,600 m, then we have L
uals the
A=30+ (? 5 - 30)x 30+0.0125x
3,600
1 andthe Substituting the expressions for A and K into Darcy's equation yields the expression for Q in follow-
of3 min ing form:
e outlet.
dh
Q= -(12 + 0.006x )(30 + 0.0125x ) -
dx
Rearranging this equation and integrating from section (1) to section (2) yields
3600 18.8
I
1
(12+0.006x)(30+0.0125x)
dx= I _J._dh
Q
0 14.2
This equation is integrated using partial fraction decomposition to obtain
36r[ 0.2 0.416 ]dx = 'Y _J._dh

0 (12+0.006x) (30+0.0125x) 14 .2 Q
x=3,600 1 ]~=18 8
[ 33.333ln ( 12+0.006x ) -33.28ln ( 30+0.0125x )] x=O =-Qh h,= 14.·2
1
-26.54- (-30.36) = --(18.8-14.2)
Q
• Q =-1.20 (m3/day/m)
!aquifer
sections
The minus sign implies that the flow is from section (2) to (1).
•
ning the
3.1.2 Darcy Velocity

The velocity v in Equation 3.1.5 is referred to as the Darcy velocity because it assumes that
flow occurs through the entire cross section of the material without regard to solids and pores.
Actually, the flow is limited only to the pore space so that the average interstitial velocity
v
a
=iL
a.A
(3.1.6)
where a is the (effective) porosity. This indicates that for a sand with a porosity of 33 percent,
va = 3v. To define the actual flow velocity, one must consider the microstructure of the rock
material. In water flowing through a sand, for example, the pore spaces vary continuously with
location within the medium. This means that the actual velocity is nonunifonn, involving end-
less accelerations, decelerations, and changes in direction. Thus, the actual velocity depends
on specifying a precise point location within the medium. For naturally occurring geologic
materials, the microstructure cannot be specified three-dimensionally; hence, actual velocities
can only be quantified statistically.
3.1.3 Validity of Darcy's Law

In applying Darcy's law it is important to know the range of validity within which it is appli-
cable. Because velocity in laminar flow, such as water flowing in a capillary tube, is propor-
tional to the fll"st power of the hydraulic gradient (Poiseuille's law), it seems reasonable to
believe that Darcy's law applies to laminar flow in porous media. For flow in pipes and other
large sections, the Reynolds number, which expresses the dimensionless ratio of inertial to vis-
cous forces, serves as a criterion to distinguish between laminar and turbulent flow. Hence, by
analogy, the Reynolds number has been employed to establish the limit of flows described by
Darcy's law, corresponding to the value where the linear relationship is no longer valid.
Reynolds number is expressed as 3.2 f
3.2.1
(3.1.7)
where pis the fluid density, v the velocity, D the diameter (of a pipe), and J.l the (dynamic) vis-
cosity of the fluid. To adapt this criterion to flow in porous media, the Darcy velocity is
employed for v and an effective grain size (d10) is substituted for D. Certainly a grain diame-
ter represents only an approximation of the critical flow dimension for which it is intended;
however, measuring pore size distribution is a complex research task.
Experiments show that Darcy's law is valid for NR < 1 and does not depart seriously up to
NR = 10. 1 This, then, represents an upper limit to the validity of Darcy's law. A range of val-
ues rather than a unique limit must be stated because as inertial forces increase, turbulence
occurs gradually. 85• 92• 106 The irregular flow paths of eddies and swirls associated with turbu-
lence occur first in the larger pore spaces; with increasing velocity they spread to the smaller
pores. For fully developed turbulence the head loss varies approximately with the second
power of the velocity rather than linearly.
Fortunately, most natural underground flow occurs with NR < 1, so Darcy's law is appli-
cable. Deviations from Darcy's law can occur where steep hydraulic gradients exist, such as
near pumped wells; also, turbulent flow can be found in rocks such as basalt and limestone 17
that contain large underground openings.*
The following additional infonnation is given for the aquifer sample in Example 3.1.1. The sample has
a median grain size of 0.037 em and a porosity of 0.30. The test is conducted using pure water at 20°C.
Determine the Darcy velocity, average interstitial velocity, and assess the validity of Darcy's law.
SOLUTION Darcy velocity is computed using Equation 3.1.5:

dh
v =-K di =-(23.54 rnlday)(-0.326) = 7.67 rnlday
The average linear velocity is computed using Equation 3.1.6:
v = _Q_ = ~ = 7·67 mlday =25.6 rnlday

a <1A a 0.30
In order to assess the validity of Darcy's Law we must detennine the greatest velocity for which 3.2.2 1
Darcy's law is valid using Equation 3.1.7, NR = pvD , knowing Darcy's law is valid for NR<l. For water at
)l
20°C, )l = 1.005 X 10-3 N/m2 and p = 998.2 kg/m3, so that for NR =1,
!t should also be noted that investigations have shown that Darcy's law may not be valid for very slow water flow
through dense clay. Here the effects of electrically charged clay particles on water in the minute pores produce non-
linearities between flow rate and hydraulic gradienL53
3.2 Permeability 91
Jl 1.005 X 10-3 kg/ms

V max =- =( ) 0.00272 m/s =235 m/day
is appli- pD 998.2 kg/m 3 (0.00037 m)
. propor-
nable to Then Darcy's law will be valid for Darcy velocities equal to or less than 235 rnlday for this sample. Thus,
nd other the answer we have found in Example 3.1.1 is valid since v =7.67 rnlday < 235 m/day. •
:tl to vis-
ence, by
ribed ~y
id.
3.2 PERMEABILITY
3.2.1 Intrinsic Permeability
(3.1.7)
The penneability of a rock or soil defines its ability to transmit a fluid. This is a property only
of the medium and is independent of fluid properties. To avoid confusion with hydraulic con-
1ic) vis- ductivity, which includes the properties of groundwater, an intrinsic permeability k may be
ocity is expressed as
diame-
tended; k= K~ (3.2.1)
pg
ly up to
of val- where K is hydraulic conductivity,~ is dynamic viscosity, pis fluid density, and g is acceler-
ation of gravity. Inserting this in Equation 3.1.5 yields
Julence
1 turbu-
k=- ~v (3.2.2)
smaller pg(dhldl)
second
which has units of
; appli-
mch as (3.2.3)
stone 17
Thus, intrinsic penneability possesses units of area. Because values of kin Equation 3.2.3 are
so small, the U.S. Geological Survey expresses kin square micrometers (~m) 2 = 10-12m2.
tple has In the petroleum industry the value of k is measured by a unit termed the darcy, defined as
lt 20°C.
(1 centipoise)(1 em
3
Is)
2
1 cm
ldarcy=----~~~---- (3.2.4)
1 atmosphere/em
By substitution of appropriate units it can be shown that 66

1 darcy = 0.987 {I.Lm)2 (3.2.5)
so the darcy corresponds closely to the intrinsic permeability unit adopted by the U.S.
Geological Survey.
which
3.2.2 Hydraulic Conductivity
1ater at
For practical work in groundwater hydrology, where water is the prevailing fluid, hydraulic
conductivity K is employed. A medium has a unit hydraulic conductivity if it will transmit in
unit time a unit volume of groundwater at the prevailing kinematic viscosity* through a cross
flow
e non-
*Kinematic viscosity equals dynamic viscosity divided by fluid density.
section of unit area, measured at right angles to the direction of flow, under a unit hydraulic
gradient. The units are
v mlday
K = -~- =- - - =m/day (3.2.6)
dh/dL mlmo:
indicating that hydraulic conductivity has units of velocity.
3.2.3 Transmissivity
The term transmissivity Tis widely employed in groundwater hydraulics. It may be defined as
the rate at which water of prevailing kinematic viscosity is transmitted through a unit width of
aquifer under a unit hydraulic gradient. It follows that
T = Kb =(mlday)(m) =m2/day (3.2.7)
where b is the saturated thickness of the aquifer.
E.'fAMPLE 3.2.1 A leaky confined aquifer is overlain by an aquitard that is also overlain by an unconfined aquifer. The
estimated recharge rate from the unconfined aquifer into the confined aquifer is 0.085 m/year. Piezo-
metric head measurements in the confined aquifer show that the average piezometric head in the confined
aquifer is 6.8 m below the water table of the unconfined aquifer. If the average thickness of the aquitard
is 4.30 m, find the vertical hydraulic conductivity, K"' of the aquitard. What type of material could this
possibly be?
SOLUTION Given v = 0.085 m/year = 2.329 x 10-4m/day, Equation 3.2.6 is used to compute the vertical hydraulic
conductivity of the aquitard:
K = __v_ = 2.329 x 10-4m/day= l.473x 10-4m/day

dhJdl (6.8 m/4.30 m)
From Table 3.2.1, the aquitard is composed of clay.

•
3.2.4 Hydraulic Conductivity of Geologic Materials
The hydraulic conductivity of a soil or rock depends on a variety of physical factors, includ-
ing porosity, particle size and distribution, shape of particles, arrangement of particles, and
other factors. 63· 79 In general, for unconsolidated porous media, hydraulic conductivity varies
with particle size; clayey materials exhibit low values of hydraulic conductivity, whereas sands
and gravels display high values.
An interesting illustration of the variation of hydraulic conductivity with particle size is
shown by data in Figure 3.2.1. Here conductivities were measured for two uniform sieved
sands. These two sands were then mixed in varying proportions, and the corresponding
hydraulic conductivities were again determined. Results show that any mixture of the two
sands displays a conductivity less than a linearly interpolated vaJue. The physical explanation
lies in the fact that the smaller grains occupy a larger fraction of the space around larger grains
than do uniform grains of either size.
Table 3.2.1 contains representative hydraulic conductivities for a variety of geologic mate-
rials. It should be noted that these are averages of many measurements; clearly, a range of val-
ues exists for each rock type depending on factors such as weathering, fracturing, solution
channels, and depth of burial.
3.2 Permeability 93
1draulic Percentage of coarser sand

100 75 50 25 0
BOOr-------,--------,------~~------,
(3.2.6) 100% of sand retained

on sieve with openings
~ / of 1.00 mm
32 600
(/)
Q;
Q)
E
~ 100% of sand retained
fined as c on sieve with openings
vidth of ~400
(.) of0.71 mm ~
:J
!'\
"0
c::
0
(.)
(3.2.7) .~
:; 1 75% of coarser sand
~200 i and 25% of finer sand
>- 1 has K = 367 m/day
I I
I
I Figure 3.2.1. Hydraulic
I conductivity of various
fer. The I
·. Piezo- proportions of two uniform
00 25 50 75 100 sands (courtesy Illinois
:onfined
aquitard Percentage of finer sand State Water Survey).
mld this
Table 3.2.1 Representative Values of Hydraulic Conductivity (after Morris and Johnson75 )
ydraulic
Material Hydraulic conductivity (rnlday) Type of measuremene
Gravel, coarse 150 R
Gravel, medium 270 R
Gravel, fine 450 R
Sand, coarse 45 R
• Sand, medium 12 R
Sand, fine 2.5 R
Silt 0.08 H
Clay 0.0002 H
includ- Sandstone, fine-grained 0.2 v
es, and Sandstone, medium-grained 3.1 v
·varies Limestone 0.94 v
s sands Dolomite 0.001 v
Dune sand 20 v
size is Loess 0.08 v
sieved Peat 5.7 v
onding Schist 0.2 v
he two Slate 0.00008 v
mation Till, predominantly sand 0.49 R
·grains Till, predominantly gravel 30 R
Tuff 0.2 v
; mate- Basalt 0.01 v
of val- Gabbro, weathered 0.2 v
Jlution Granite, weathered 1.4 v
aH is horizontal hydraulic conductivity, R is a repacked sample, and V is vertical hydraulic conductivity.
~
Unconsolidated
deposits k k K K K
(darcy) (cm 2) (cm/s) (m/s) (gal/day/ft2)
105 w-3 1o 2
1o 6
104 w-4 10 1 o- 1
105
I 103 10-5 10 - 2
Q)
c I Q5 104
0 >
10 2
"0
1i) I c ~ 1o- 6 10-1 10-3
Q) ...... (!)
(ti
ell
(/) 103
~ (/) "0 c
...... ell
.0
c ell 10 10-7 10-2 10- 4
~ ell(/) Q)
ell Q) (/)~
"'0
(3 I 1o2
~ :0 :::::Jo c
ell oo 2 10-8 10-3 1o- 5 3.3.2
Q) Q)"-
co
.E ell
(/)
I E
(»
e>·-
·-.1::. -
0
~ I 10
a.. "0
Q) 0
Q. 0
"0 (/)
(/)
U5 10-1 10-9 w-4 10- 6
,_ E -o >. Q)
I .ao .....ell c
ell 0
ell
.Q
~ Q) Q) Q) :!:! I 10-2 10-10 10-5 10- 7
LL E c c
U5
·~ 10-1
~E
:;:;
I Q)
c E (ti
·o
10-3 10-11 w-6 10-8
0
1i) I "0
ell
I
:J ~ 10-2
"'0
c
0
:.ca. Q)
.r:.
a 10-4 10-12 10-7 10-9
ell .....
(J)
0 ell
Q)
I E(/J !: 10- 3
-
ell.:.::
o c I 10-5 10-13 10-8 1 o-10
I
Q)o
E._ I ::)
"10- 4
"0(/J Q)
10-14 10-9 10-11
Q):::::l
.... o
:::::JQ)
(ti
.r:. I 10-6
..... c (J)
10-5
00>
ell.- 10-7 10-15 10-10 1o-12
~"0
cc
:)ell I 10-6
10-8 10-16 10-11 1o-13
I
10-7
Figure 3.2.2. Range of values of hydraulic conductivity and permeability (Freeze, R. A. and Cherry, J. A., Groundwater, Prentice
Hall, Englewood Cliffs, NJ, 1979.)
Magnitudes of hydraulic conductivity for various classes of unconsolidated and consoli-

dated rocks are shown in Figure 3.2.2.
3.3 DETERMINATION OF HYDRAULIC CONDUCTIVITY

Hydraulic conductivity in saturated zones can be determined by a variety of techniques,
including calculation from formulas, laboratory methods, tracer tests, auger hole tests, and
pumping tests of wells.
a==
3.3.1 Formulas
Numerous investigators have studied the relationship of permeability or hydraulic conductiv-
ity to the properties of porous media. Several formulas have resulted based on analytic or
Ea·
~-
experimental work. Most pem1eability formulas have the general form
k =cd2 (3.3.1)
3.3 Detennination of Hydraulic Conductivity 95
where cis a dimensionless coefficient, or
(3.3.2)
where.fs is a grain (or pore) shape factor,/0 is a porosity factor, and dis characteristic grain
diameter. 30· 62· 68 Few formulas give reliable estimates of results because of the difficulty of
including all possible variables in porous media. For an ideal medium, such as an assemblage
of spheres of uniform diameter, hydraulic conductivity can be evaluated accurately from
known porosity and packing conditions.
Because of the problems inherent in fonnulas, other techniques for determining hydraulic
conductivity are preferable.
3.3.2 Laboratory Methods

In the laboratory, hydraulic conductivity can be determined by a permeameter, in which flow
is maintained through a small sample of material while measurements of flow rate and head
loss are madel07 The constant-head and falling-head types of permeameters are simple to
operate and widely employed.
The constant-head permeameter shown in Figure 3.3.la can measure hydraulic conduc-
tivities of consolidated or unconsolidated fonnations under low heads. Water enters the
medium cylinder from the bottom and is collected as overflow after passing upward through
the material. From Darcy's law it follows that the hydraulic conductivity can be obtained from
VL
K=- (3.3.3)
Ath
where Vis the flow volume in timet, and the other dimensions, A, L, and h, are shown in Fig-
ure 3.3Ja. It is important that the medium be thoroughly saturated to remove entrapped air.
Several different heads in a series of tests provide a reliable measurement.
rentice
r,:::::===:J - Continuous
Constant supply
water level
;on soli-
Overflow
h
niques,
;ts, and
Horizontal area 2r,
of sample, A
t T T
1i~~olume Vln timet

L
ductiv- i Figure 3.3.1. Perrneameters
ytic or for measuring hydraulic
conductivity of geologic
samples. (a) Constant head.
(3.3.1) (a) (b) (b) Falling head.
A second procedure utilizes the falling-head permeameter illustrated in Figure 3.3.1 b.

Here water is added to the tall tube; it flows upward through the cylindrical sample and is col-
lected as overflow.
The test consists of measuring the rate of fall of the water level in the tube. The hydraulic
conductivity can be obtained by noting that the flow rate Q in the tube
SOLUTII
Q=1tr12dh/dt (3.3.4)
must equal that through the sample, which by Darcy's law is

Q::: nr/K h/L (3.3.5)
After equating and integrating,
(3.3.6)
where L, r~' and rc are shown in Figure 3.3.lb, and tis the time interval for the water level in
the tube to fall from ht to h2.
Permeameter resuLts may bear little relation to actual field hydraulic conductivities.
Undisturbed samples of unconsolidated material are difficult to obtain, while disturbed sam-
ples experience changes in porosity, packing, and grain orientation, which modify hydraulic
conductivities. Then, too, one or even several samples from an aquifer may not represent the
overall hydraulic conductivity of an aquifer. Variations of several orders of magnitude fre-
quently occur for different depths and locations in an aquifer. Furthermore, directional prop-
erties of hydraulic conductivity may not be recognized.
SOLUT[(
EXAMPLE 3.3.1 A field sample of medium sand with a median grain size of 0.84 mrn will be tested to determine the
hydraulic conductivity using a constant-head permeameter. The sample has a length of 30 em and a diam-
eter of 5 em. For pure water at 20°C, estimate the range of piezometric head differences to be used in the
test.
SOLUTION The maximum allowable Darcy velocity (assuming NR =I) ford= 0.84 mrn is
1-1. 1.005 X Io-3 kglrns

Ymax =-= ( ) = 0.0012 rn/s =103.6 rn/day
pD 998.2 kg/m 3 (0.00084 m)
Thus, the Darcy velocity in the test must be equal to or Jess than 103.6 rn/day so that Darcy's law will
be valid, so that
v
dh
=-K- I I (103.6 m/day)(0.30 m)
~ 103.6 rn/day -4 dh ~ _,____ _......:....:,..,___....:.
dl K
For the representative value of hydraulic conductivity for medium sand given in Table 3.2.1,
(103.6 rn/day)(0.30 m)
K =12m/day, then ldhl ~
12 rn/day
=2.6 m = 260 ern
It should be noted that the K value for clean sand ranges approximately from 0.1 rn/day to 4,320
rn/day. See Figure 3.2.1. Therefore, the early series of tests must be conducted with relatively low piezo-
metric head differences if possible. After analyzing the results of early test data, a better estimate of the ·
maximum allowable piezometric head difference can be made using the above inequality. •
3.3 Determination of Hydraulic Conductivity 97
3.3.1 b. If the field sample in Example 3.3.1 is tested with a head difference of 5.0 em and 200 ml of water is col-
EXAMPLE 3.3.2
lis collected at the outlet in 15 min, determine the hydraulic conductivity of the sample. What should the max-
imum allowable piezometric head difference be for a series of tests?
·draulic
SOLUTION Equation 3.3.3 is used to compute the hydraulic conductivity in a constant-head penneameter test:
(3.3.4)
VL (200 cm 3)(30 ern)
K =-= [ 2] = 0.0679 crnls = 58.7 rnlday
Ath n(5 em) ( s }
15rrlin x60-. (5.0 em)
(3.3.5) 4 mm
Based upon this estimate and referring to Example 3.3.1 , the maximum allowable piezometric head
difference for tests should be approximately
(3.3.6)
i
)
(103.6 rnlday)(0.30 m)
IdhI~ -0.53m=53cm
level in 58.7 m/day
•
ivities.
d sam-
draulic
i
ent the ~' A 20-cm long field sample of silty, fine sand with a diameter of 10 em is tested using a falling-head
~-
de fre- permeameter. The falling-head tube has a diameter of 3.0 em and the initial head is 8.0 em. Over a period
lprop- l of 8 hr, the head in the tube falls to 1.0 ern. Estimate the hydraulic conductivity of the sample.
) SOLUTION Equation 3.3.6 is used to compute the hydraulic conductivity in a falling-head permeameter test:
{
t
line the -~ 2
r: 2 L
K=-1 -ln_!_=
h (1.5 cm) (20 em)
8 0 em
ln-·- -=1.3X10-4cm/s=0.112m/day
adiam- l r/t ~
2
din the ·~
'
··~
(5.0 em) (8x3600sec) 1.0 em
•
fi
~
~:
:1
-~
-~
, I Add Sample
I
~
-~
f tracer t fortracer
aw will 7
Ground surface >--Hole B
>--Hole A
l'ii' _l
I'V h
- / - t
) 4,320'
- -
-
-
-
-
-
Water table
-----.. ~
-
-
-
-
-
-
- Figure 3.3.2. Cross section of an
•piezo- - - unconfined aquifer illustrating a
- - tracer test for deterrrlining hydraulic
! of the
• k----------L----------~ conductivity.
3.3.3 Tracer Tests

Field detenninations of hydraulic conductivity can be made by measuring the time interval for
a water tracer to travel between two observation wells or test holes. 18 · 25 · 40 · 57 · 72 · 73 · 109 For the
tracer, a dye, such as sodium fluorescein, or a salt, such as calcium chloride, is convenient,
inexpensive, easy to detect, and safe. Figure 3.3.2 shows the cross section of a portion of an
unconfined·aquifer with groundwater flowing from hole A toward hole B. The tracer is injected
as a slug in hole A after which samples of water are taken from hole B to determine the time 3.3.4 A
of passage of the tracer. Because the tracer ±lows through the aquifer with the average inter-
stitial velocity va' then
Kh
v =-- (3.3.7)
a a L
where K is hydraulic conductivity, a is porosity, and hand L are shown in Figure 3.3.2. But va
also is given by
(3.3.8)
where tis the travel time interval of the tracer between the holes. Equating these and solving
forK yields
aL2
K=- (3.3.9)
ht
Although this procedure is simple in principle, results are only approximations because of
serious limitations in the field.
1. The holes need to be close together; otherwise, the travel time interval can be excessively Table 3.3.1
long.
2. Unless the flow direction is accurately known, the tracer may miss the downstream hole
entirely. Multiple sampling holes can help, but these add to the cost and complexity of
conducting the test. y
3. If the aquifer is stratified with layers with differing hydraulic conductivities, the fust arrival
of the tracer will result in a conductivity considerably larger than the average for the aquifer.
An alternative tracer technique, which has been successfully applied under field condi-
2
tions; is the point dilution method. 28 • 45 · 53 Here a tracer is introduced into an observation well
and thoroughly mixed with the contained water. Thereafter, as water flows into and from the
well, repeated measurements of tracer concentration are made.
5
Analysis of the resulting dilution curve defines the groundwater velocity; this, together with
the measured water table gradient and Darcy's law, yields a localized estimate of the hydraulic
(
conductivity and also the direction of groundwater flow. 85 Additional references on tracer tests
10 1
include Barth et al. 7 on heterogeneous porous media; Mullet al.1 6 on carbonate aquifer systems;
and Novakowski77 on divergent radial flow fields. A new Efficient Hydrologic Tracer-test
c
0
Design (EHTD) has been developed by the U.S. Environmental Protection Agency (EPA). 33- 36
20 1
0
EXAMfLE-3.3.4 A tracer test is conducted to determine the hydraulic conductivity of an unconfined aquifer. The water 0
levels in the two observation wells 20 m apart are 18.4 m and 17.1 m. The tracer injected in the first well 50
arrives at the second observation well in 167 hours. Compute the hydraulic conductivity of the uncon- 0
fined aquifer given that the porosity of the formation is 0.25. 0
100
0
SOLUTION Given a= 0.25, L =20m, h = 18.4 m- 17.1 m = 1.3 m, t = 167 hours= 6.96 days, Equation 3.3.9 is used
0.
to compute the hydraulic conductivity of the aquifer:
3.3 Determination of Hydraulic Conductivity 99
2
aL2 (0.25)(20 m)
K = ht = (1.3 m)(6.96 days)
11. 1 rnfday
~rval for
'For the
venient,
•
m of an
injected
the time 3.3.4 Auger Hole Tests
~e inter- The auger hole method involves the measurement of the change in water level after the rapid
removal of a volume of water from an unlined cylindrical hole. If the soil is loose, a screen may
be necessary to maintain the hole. The method is relatively simple and is most adaptable to
(3.3.7) shallow water table conditions. The value of K obtained is essentially that for a horizontal
direction in the immediate vicinity of the hole.
.. But va Figure 3.3.3 illustrates an auger hole (also refer to Figure 5.2.4) and the dimensions
required for the calculation. It can be shown 10 that hydraulic conductivity is given by
(3.3.8)
K=_£ dy (3.3.10)
solving 864 dt
where dy!dt is the measured rate of rise in cm/s and the factor 864 yields K values in m/day.
The factor Cis a dimensionless constant listed in Table 3.3.1 and governed by the variables
(3.3.9)
shown in Figure 3.3.3.
ause of
:ssively Table 3.3.1 Values of the Factor C for the Auger Hole Test to Determine Hydraulic Conductivity (after Boast and Kirkham 10)
(H- Lw)!Lw for (H- Lw)/Lw for infinitely
m hole
impermeable layer H-LIV permeable layer
:xity of
Lw/rw y/Lw 0 0.05 0.1 0.2 0.5 2 5 00 5 2 0.5
arrival 1 447 423 404 375 323 286 264 255 254 252 241 213 166
1quifer. 0.75 469 450 434 408 -360 324 303 292 291 289 278 248 198
0.5 555 537 522 497 449 411 386 380 379 377 359 324 264
condi-
2 186 176 167 154 134 123 118 116 115 115 113 106 91
m well
0.75 196 187 180 168 149 138 133 131 131 130 128 121 106
om the
0.5 234 225 218 207 188 175 169 167 167 166 164 156 139
5 51.9 48.6 46.2 42.8 38.7 36.9 36.1 35.8 35.5 34.6 32.4
er with
0.75 54.8 52.0 49.9 46.8 42.8 41.0 40.2 40.0 39.6 38.6 36.3
draulic
0.5 66.1 63.4 61.3 58.1 53.9 51.9 51.0 50.7 50.3 49.2 46.6
~r tests
10 18.1 16.9 16.1 15.1 14.1 13.6 13.4 13.4 13.3 13.1 12.6
·stems;
0.75 19.1 18.1 17.4 16.5 15.5 15.0 14.8 14.8 14.7 14.5 14.0
:er-test
~). 33-36 H 0.5 23.3 22.3 21.5 20.6 19.5 19.0 18.8 18.7 18.6 18.4 17.8
~ 20 5.53 5.30 5.06 4.81 4.70 4.66 4.64 4.62 4.58 4.46
~· 5.91
'1 0.75 6.27 5.94 5.73 5.50 5.25 5.15 5.10 5.08 5.07 5.02 4.89
~water 0.5 7.67 7.34 7.12 6.88 6.60 6.48 6.43 6.41 6.39 6.34 6.19
rst well ;( 50 1.25 1.18 1.14 1.11 1.07 1.05 1.04 1.03 1.02
i
uncon- 1
'Y
0.75 1.33 1.27 1.23 1.20 1.16 1.14 1.13 1.12 1.11
:l
~· 0.5 1.64 1.57 1.54 1.50 1.46 1.44 1.43 1.42 1.39
100 0.37 0.35 0.34 0.34 0.33 0.32 0.32 0.32 0.31
is used 0.75 0.40 0.38 0.37 0.36 0.35 0.35 0.35 0.34 0.34
0.5 0.49 0.47 0.46 0.45 0.44 0.44 . 0.44 0.43 0.43
Ground surface
I
-·--- - ·--,
Impermeable, or highly Figure 3.3.3. Diagram of an auger hole and dimen-

permeable, layer sions for detennining hydraulic conductivity.
Note that the tabulated values cover the following conditions below the hole: a shallow
impenneable layer, an infinite homogeneous stratum, and a shallow, highly penneable (gravel)
layer. The value y should correspond to that when dyldt is measured.
Several other techniques similar to the auger hole test have been developed in which
water level changes are measured after an essentially instantaneous removal or addition of a
volume of water. With a small-diameter pipe driven into the ground, K can be found by the
piezometer, or tube, method. 103 For wells in confined aquifers, the slug method can be
employed. 22·66 Here a known volume of water is suddenly injected or removed from a well
after which the decline or recovery of the water level is measured in the ensuing minutes.
Where a pump is not available to conduct a pumping test on a well, the slug method serves as
an alternative approach.
3.3.5 Pumping Tests of Wells

The most reliable method for estimating aquifer hydraulic conductivity is by pumping tests of
wells. Based on observations of water levels near pumping wells, an integrated K value over a
sizable aquifer section can be obtained. Then, too, because the aquifer is not disturbed, the reli-
ability of such determinations is superior to laboratory methods. Pump test methods and com-
putations are described in Chapter 4.
3.4 ANISOTROPIC AQUIFERS

The discussion of hydraulic conductivity up to this point assumed that the geologic material
was homogeneous and isotropic, implying that the value of K was the same in all directions. ,
However, this is rarely the case, particularly for undisturbed unconsolidated alluvial materials.·
3.4 Anisotropic Aquifers 101
t Figure 3.4.1. Diagram of two horizontal strata, each

isotropic, with different thicknesses and hydraulic
conductivities.
Instead, anisotropy is the rule where directional properties of hydraulic conductivity exist. ln
alluvium this results from two conditions. One is that individual particles are seldom spheri-
cal so that when deposited under water they tend to rest with their flat sides down. The second
is that alluvium typically consists of layers of different materials, each possessing a unique
value of K. ff the layers are horizontal, any single layer with a relatively low hydraulic con-
ten- ductivity causes vertical flow to be retarded, but horizontal flow can occur easily through any
stratum of relatively high hydraulic conductivity. Thus, the typical field situation in alluvial
deposits is to find a hydraulic conductivity Kx in the horizontal direction that will be greater
than a value Kz in a vertical direction.
illow Consider an aquifer consisting of two horizontal layers, each individually isotropic, with
avel) different thicknesses and hydraulic conductivities, as shown in Figure 3.4.1. For horizontal
flow parallel to the layers, the flow q 1 in the upper layer per unit width is
'ruch
(3.4.1) .
of a
y the
.n be
where i is the hydraulic gradient and K 1 and z1 are as indicated in Figure 3.4.1. Because i must
well be the same in each layer for horizontal flow, it follows that the total horizontal flow qx is
utes.
(3.4.2)
es as
For a homogeneous system this would be expressed as
(3.4.3)
;ts of. where Kx is the horizontal hydraulic conductivity for the entire system. Equating these and
·vera solving for Kx yield
: reli-
com- (3.4.4)
which can be generalized for n layer's as
(3.4.5)
terial
jons.
rials. This defines the equivalent horizontal hydraulic conductivity for a stratified material.
Now, for vertical flow through the two layers in Figure 3.4.1, the flow q2 per unit hori-
zontal area in the upper layer is
-K1
q- dhl- (3.4.6)
z ZJ
where dh 1 is the head loss within the first layer. Solving for the head loss, we find
dh = _s_q (3.4.7)
1 K z
I
From continuity, qz must be the same for the other layer so that the total head loss is
dh + dh == [_s_ + _2_]q (3.4.8)

I 2 K K z
1 2
In a homogeneous system
(3.4.9)
where Kz is the vertical hydraulic conductivity for the entire system. Rearranging yields
(3.4.10)
SOLV
and equating with Equation 3.4.8, we have
K == + z2
Zt
z Ztz2 (3.4.11)
-+-
Kl K2
which can be generalized for n layers as
(3.4.12)
This defines the equivalent vertical hydraulic conductivity for a stratified material.
As mentioned earlier, the horizontal hydraulic conductivity in alluvium is normally
greater than that in the vertical direction. This observation also follows from the above deri-
vations; thus, if
Kx>Kz (3.4.13)
then for the two-layer case from Equations 3.4.4 and 3.4.11
3.5
K1z1 + K2z2 > Zt + z2
----.!...~-----.!:::.......::... ____:._-=.__
z1 + z2 _s_ + _2_ (3.4.14)

Kl K2
which reduces to 67
(3.4.15)
3.5 Groundwater Flow Rates 103
1it hori- Because the left side is always positive, it must be greater than zero, thereby confirming that
Kx ~ 1 (3.4.16)
(3.4.6) Kz
Ratios of K/Kz usually fall in the range of2 to 10 for alluvium, 75 but values up to 100 or
more occur where clay layers are present. For consolidated geologic materials, anisotropic
conditions are governed by the orientation of strata, fractures, solution openings, or other
(3.4.7) structural conditions, which do not necessarily possess a horizontal alignment.
In applying Darcy's law to two-dimensional flow in anisotropic media, the appropriate
value of K must be selected for the direction of flow. For directions other than horizontal (Kx)
and vertical (K2), the K value can be obtained from
2
(3.4.8) 1 cos 2 ~ sin p
-=--+-- (3.4.17)
K~ Kx Kz
where K~ is the hydraulic conductivity in the direction making an angle pwith the horizontal.
(3.4.9)
fiX;iMPLE 3.4.1 An unconfined aquifer consists of three horizontal layers, each individually isotropic. The top layer has
a thickness of 10 m and a hydraulic conductivity of 11.6 m/day. The middle layqr has a thickness of 4.4
Is m and a hydraulic conductivity of 4.5 m/day. The bottom layer has a thickness qf 6.2 m and a hydraulic
conductivity of 2.2 m/day. Compute the equivalent horizontal and vertical hyddulic conductivities.
3.4.10)
SOLUTION Equation 3.4.5 is used to compute the equivalent horizontal hydraulic conductivity:
K = K1z1 + K2 z2 + K3z3
x z1 +z2 + z3
3.4.11) _ (11.6 m/day)(lO m)+(4.5 m/day)(4.4 m)+(2.2 m/day)(6.2 m)
7.25 m/day
- (10m+4.4m+6.2m)
The equivalent vertical hydraulic conductivity is computed using Equation 3.4.12:
K = z1 +z 2 +z3
3.4.12) z z z z
-1+ -2 + -
3
K, K2 K3
10 m + 4.4 m + 6.2 m
= 10 m 4.4 m 6.2 m =4.42 m/day
rmally ----+ +---
11.6 m/day 4.5 m/day 2.2 m/day
e deri-
Note that the equivalent hydraulic conductivities above are computed based on the assumption that
each layer is individually isotropic, that is, Kx =Kz in each layer. •
L4.13)
3.5 GROUNDWATER FLOW RATES

~.4.14) From Darcy's law it follows that the rate of groundwater movement is governed by the
hydraulic conductivity of an aquifer and the hydraulic gradient. To obtain an idea of the order
of magnitude of natural velocities, assume a productive alluvial aquifer with K =75 m/day and
a hydraulic gradient dh/dl =-10 m/1000 m =-0.01. Then from Equation 3).5,
dh
~.4.15) v = K dz =75( -0.01) =0.75 m/day (3.5.1)
Stream Ground surface

.~----~l--~_L~------
Figure 3.5.1. Cross sec-
Water table
tion of a typical alluvial ·EXAMI
floodplain containing an
unconfined aquifer with
groundwater flowing per- SOLUTI'
f + - - - - - - - - - -1000 rn - - - - - - - - - - + 1 pendicular to the section
Impermeable strata (not to scale).
This is approximately equivalent to 0.5 mm/min, which demonstrates the sluggish nature of
natural groundwater movement.
If the above flow occurs within and perpendicular to the large alluvial cross section
sketched in Figure 3.5.1, then the total flow rate is
Q = Av = (50)(1000)(0.75) = 37,500 m3/day (3.5 .2)
which, when converted to usual streamflow units, amounts to only 0.43 m3/s. Thus, ground-
water typically can be conceived of as a massive, slow-moving body of water.
Groundwater velocities vary widely depending on local hydrogeologic conditions; values 3.6 GF
from 2m/year to 2m/day are nonnal. Usually, velocities tend to decrease with depth as porosi-
ties and penneabilities also decrease. Velocities can range from negligible• to those of turbu-
3.6.1 F
lent streams in underground openings within basalt and limestone. Mechanisms such as wells
and drains act to accelerate flows.
An illustration of one-dimensional vertical flow is shown in Figure 3.5.2. Here an aquitard
separates an overlying unconfined aquifer from an underlying leaky aquifer. The water table
Ground surface
Unconfined
aquifer
K =10m/day
Aqultard
Sm K= 0.2m/day
i Leaky
Figure 3.5.2. Diagram illustrating applica-
tion of Darcy's law for vertically downward
aquifer flow.
'All groundwater within the hydrologic cycle should be regarded as in continuous motion, although, it must be
granted, some of it flows at extremely small rates.
stands above the piezometric surface so that water moves vertically downward from the uncon-
fined aquifer, through the aquitard, and into the confined aquifer.
sec-
vial EXAMPLE 3.5.1 Determine h8 and the vertical velocity for the situation shown in Figure 3.5.2.
g an
~ith
~per
SOLUTION Assume steady-state conditions. Writing Darcy's law from point A to B with the dimensions indicated in
:tion Figure 3.5.2, we have
v = K dh = 10 27- h8
dl 27 (3.5.3)
ture of and from point B to C,
v = K dh = 0. 2 h8 + 5-30
~ection dl 5
(3.5.2)
Solving these yields, h8 = 26.8 m* and v = 0.07 rnlday.
•
;round-
values 3.6 GROUNDWATER FLOW DIRECTIONS

porosi-
3.6.1 Flow Nets
~ turbu-
s wells For specified boundary conditions flow lines t and equipotential lines can be mapped in two
dimensions to form a flow net. The two sets of lines form an orthogonal pattern of small
quitard squares. In a few simplified cases, the differential equation governing flow can be solved to
~r table obtain the flow net. Flow net analysis techniques have been applied in a number of ways for
groundwater studies. Hollett48 and Fenemore 32 used flow net analysis to provide initial
transmissivities prior to trial-and-error numerical model calibration. Rice and Gorelick82
applied flow net analysis to three field problems and demonstrated the value of the method for
the conceptualization of flow systems. Other applications include Scott and Thorpe 87 and
vanTonder. 104
Consider the portion of a flow net shown in Figure 3.6.1. The hydraulic gradient i is given by
dh
t=- (3.6.1)
ds
and the constant flow q between two adjacent flow lines by
dh
q=K-dm (3.6.2)
ds
for unit thickness. But for the squares of the flow net, the approximation
ds=:dm (3.6.3)
can be made so that Equation 3.6.2 reduces to
lplica-
mward q=Kdh (3.6.4)
*It should be noted that the piezometer open at the bottom of the unconfined aquifer (B) displays a water level
be below the water table (A) because of the head loss associated with vertical flow through the aquifer.
t A flow line is defined here as a line such that the macroscopic velocity vector is everywhere tangent to it.
h-dh
~+--Equipotential
/ line
Flow line
Figure 3.6.1. Portion of an

orthogonal flow net. formed
by flow and equipotential
lines.
Applying this to an entire flow net, where the total head loss h is divided into n squares
between any two adjacent flow lines, then
dh =!!_ (3.6.5) Ku /
/
/
n 1---t--
If the flow is divided into m channels by flow lines, then the total flow KL :
I
..,_ I
Kmh
Q=mq=-
n
(3.6.6) ""'--
:
1
10
Thus, the geometry of the flow net, together with the hydraulic conductivity and head loss, ""'--
I
enables the total flow in the section to be computed directly. ;
In anisotropic media, flow lines and equipotential lines are not orthogonal except when I
the flow is parallel to one of the principal directions. 8 In order to calculate flows for this situ- :
ation, the boundaries of a flow section must be transfonned so that an isotropic medium is
obtained. For the typical alluvial case of Kx > Kz, all horizontal dimensions are reduced by the
ratio ~ Kz I Kx. This creates a transformed section with an isotropic medium having an equiv-
alent hydraulic conductivity
K' =~KXKZ (3.6.7)
With this transfonned section, the flow net can be drawn and flow rate determined.
After the flow net has been defined, it can be converted back to the true anisotropic sec-
Ku___ _
-KL
tion by multiplying all horizontal dimensions by~ Kx I Kz . Figure 3.6.2 illustrates the proce-
dure for an earth dam as well as demonstrates the distortion created by anisotropy in an /
/
element of the flow net. The technique can also be extended to anisotropic two-layer systems. 67 I
I
Figure 3.6.3 shows contrasting flow nets for channel seepage through layered anisotropic I
I
media. I
I
Computer-assisted tlow net analysis techniques have been developed and evaluated by I
I
Scott. 86 I
I
1.0
I
3~6.2 Flow in Relation to Groundwater Contours I
I
Because no flow crosses an impenneable boundary, flow lines must parallel it. Similarly, if no
flow crosses the water table of an unconfined aquifer, it becomes a bounding flow surface. The
Flow net
element forms a
parallelogram
z
+X
Figure 3.6.2. Illustration of flow net analysis
•f an 1------- L -------+-1 I· ·I for anisotropic hydraulic conductivity in an
med
L~;;Lfi:=!:_ earth dam. (a) True anisotropic section with
.a! K1 9 3 Kx =9Kl (b) Transformed isotropic section
(a) {b) with Kx::;; Kz
:3.6.5)
3.6.6)
10 20
I loss,
when
; situ-
um is
JY the (a)
:quiv-
Impermeable
3.6.7) dike
Channel
:sec-
Ku
roce- - KL-- --: . .-------------,..,
.
in an / /
ms.67
I
I
/
//
/
/
I
ropic
I
I
1
I
I
\
I
I I
:d by I
I
I
I
I
I 99.5 \
\ , I
I I I \ 90.,j
I I I r \
Figure 3.6.3. Row nets for seepage from
I f I I \
1.0 5 20 40 80
'
I one side of a channel through two different
I I
I anisotropic two-layer systems. (a) Ku!KL =
r
r
1
, I I 1/50 (b) Ku!KL::;; 50 The anisotropy ratio
if no for all layers is K/Kt::;; 10 (after Todd and
. The (b) B~s) .
energy head hE, or fluid potential, from Equation 3.1.2 at any point on the water table can be
approximated by
(3.6.8)
\
so that by letting the atmospheric pressure reference be zero, p =0, and hE = z. Therefore,
under steady-state conditions, the elevation at any point on the water table equals the energy
head and, as a consequence, flow lines lie perpendicular to water table contours. Similarly,
flow lines within a confined aquifer are orthogonal to contours of the potentiometric surface.
With only three groundwater elevations known from wells, estimates of local groundwa- \
\
ter contours and flow directions can be determined as demonstrated by Figure 3.6.4. From field I
I
measurements of static water levels in wells within a basin, a water level contour map can be I
I
constructed. Flow lines, sketched perpendicular to contours, show directions of movement. An I
example appears in Figure 3.6.5. I
I
Contour maps of groundwater levels, together with flow lines, are useful data for locating I
new wells. Convex contours indicate regions of groundwater recharge, while concave contours
are associated with groundwater discharge. Furthermore, areas of favorable hydraulic con-
ductivity can be ascertained from the spacing of contours. The procedure can be illustrated by
treating two adjacent flow lines as impermeable boundaries because there can be no flow
across a flow line. If the aquifer is uniformly thick, the flow at sections 1 and 2 in Figure 3.6.5
equals
(3.6.9)
where v is velocity and W is the width of the flow section perpendicular to the flow. From
Darcy's law
(3.6.10)
which can be rewritten

Kt W2it_
- =-- (3.6.11)
Kz ~il
where K is hydraulic conductivity and i is hydraulic gradient. The ratios W2/W1 and i 2/i 1 can
be estimated from the water level contour map (see Figure 3.6.5). For the special case of
nearly parallel flow lines, Equation 3.6.11 reduces to
Water table
elevation
\
61.2 Groundwater
contours
Figure 3.6.4.
Estimate of ground-
water contours and
flow direction from
water table eleva-
tions in three wells.
can be 348 346
(3.6.8)
I
I
/
/
:refore, /
/
/
energy /
/
/
nilarly,
1rface. \
\
\
mdwa- \
\
\
mfield I \
I \
contours of
can be I \ groundwater
I \ suriace
~nt. An I \
\
I
I \ ____ ,.,
>eating I '' ' .......
mtours
--------
500 m Figure 3.6.5. Contour map of a groundwater
c con- surface showing flow lines.
tted by
o flow
e 3.6.5
(3.6.12)
(3.6.9)
From which may be interpreted as indicating that in an area of uniform groundwater flow, areas with
wide contour spacings (flat gradients) possess higher hydraulic conductivities than those with
narrow spacings (steep gradients). Therefore, in Figure 3.6.5, prospects for a productive well
L6.10) are better near section 2 than 1.
Where a contour map of groundwater levels contains closed contours around a group of
wells of known total discharge Q, the transmissivity of the regional aquifer can be calculated.
Figure 3.6.6 illustrates such a situation resulting from heavy groundwater pumping in and near
1.6.11) Savannah, Georgia in 1957. If a flow net can be constructed, Equation 3.6.6 in the form
li 1 can T= nQ
ase of mh (3.6.13)
can be applied where h represents the difference in elevation between any two selected closed
contour lines. The typical irregularity of groundwater contours often makes construction of an
accurate flow net difficult. As a convenient alternative involving contours but no flow net,
Lohman66 suggested the equation
T= 2Q
(Lt + ~)M111r (3.6.14)
where L 1 and L2 are the lengths of any two concentric closed contours, !l.h is the contour inter-
val, and 11r is the average distance between the two closed contours.
Natural permeable boundaries of aquifers include surface water bodies and the ground
surface. In a surface water body, the energy head is constant everywhere within the water body
mnd-
and equals the elevation of the water surface; consequently, aquifer flow lines must intersect
and
normal to such a bounding surface. For the ground surface, this does not apply, as only atmos-
from
va- pheric pressure exists at the ground surface. Hence, in Equation 3.6.8, by lettingp = 0, hE= z,
{ells. which is identical to the case for a water table boundary.
Oire1
groun
mov
lines of equal head

meters Figure 3.6.6. Contour map of the piezometric
S 0 TOKM
surface near Savannah, Georgia, 1957, showing
lull! I closed contours resulting from heavy local
groundwater pumping (after U.S.G.S. Water-Supply SOLi
81°15' 81°00' so•4s' so•Jo' Paper 1611).
Three observation wells are installed to determine the direction of groundwater movement and the
hydraulic gradient in a regional aquifer. The distance between the wells and the total head at each well
are shown in Figure 3.6.7a.
Well1
(32.55 m)
Wsll2
(32.41 m) SOLi
Figure 3.6.7a.
0 25 50 100 Configuration of three
Well3
L------'----l...-----'' (32.66 m) observation wells io
Scale, meters (a) Example 3.6.1
Well1
(32.55 m)
Well3
Scale, meters (32.66 m)
Figure 3.6.7b. Illustration of graphical proce-
(b) dure in Example 3.6. L.
pply SOLUTION Step 1: Identify the well with the intermediate water level-Well I in this case.
Step 2: Along the straight line between the wells with the highest head and the lowest head, iden-
tify the location of the same head of the well from Step 1. Note that this is accomplished by locating
the elevation of 32.55 m between Well2 and Well 3 in the graphical solution.
r1d the
Step 3: Draw a straight line between the intermediate well from Step 1 and the point identified in
Step 2. This is a segment of the equipotential line along which the total head is the same as that in
h well
the intermediate well (i.e, equipotential line of 32.55 m head in this case).
Step 4: Draw a line perpendicular to the equipotential line passing through the well with the lowest
head. The hydraulic gradient is the slope of that perpendicular line. Also, the direction of the line indi-
cates the direction of groundwater movement. The graphical procedure above is illustrated in Figure
3.6.7b. The hydraulic gradient is then computed as
i =32.55 m- 32.41 m ::; 0_0012

115.93 rn
•
The average daily discharge from the Patuxent Fonnation (see Figure 3.6.8) in the Sparrows Point dis-
trict of Baltimore, Maryland, in 1945 was estimated as 1 x 106 ft3/day. A flow net of the region is con-
structed using the available contour lines as shown in Figure 3.6.8. (This example is adapted from
Lohman.66) Compute the transmissivity of the regional aquifer.
SOLUTION As shown in the flow net, there are 15 flow channels, hence m = 15. There are four equipotential drops
from the 60-ft contour line to the 20-ft contour line, soh= 40ft and n::; 4. Then the overall transmissiv-
ity of the district can be computed using Equation 3.6.13:
hree 6
nQ {4)(lx10 ft /day)
3
in T::; - ::; ( X ) =6700 ft 2/day
mh 15 40ft
•
N -- Bad.. Rf,49r •..

·.
NIICA ·. I
.-.
-~
·. ~ _J. .• t
Figure 3.6.8. Map of

Baltimore industrial area.
-30- Potentiometric contour:
Maryland, showing poten-
2 Hiles
shows altWde of potentiometric 0
I' I I 11
1
II I I
tiometric surface in 1945
surface of Patuxent Formation I I
In 1945. Countour interval tO 0 1 2 3 4 KilometaJs and generalized flow lines
leet. Datum is mean sea level. in the Patuxent Formation.
----- Row line From Bennett and Meye.-9
• Well for which altitude (as presented in
of head was obtained Lohman66) .
3.6.3 Flow Across a Water Table

As long as no flow crosses a water table, it serves as a groundwater boundary; however, if flow,
such as percolating water, reaches the water table, flow lines no longer parallel the surfa~ as
an impermeable boundary.52 To illustrate this refraction effect, let vu represent the unsaturated
vertical velocity approaching the water table and vs the saturated velocity below the water table
(Figure 3.6.9). The head loss dh for flow along the left flow line below the water table occurs
in a distance of bs tan(8 +e), as defined in Figure 3.6.9. Thus
. K dh
v
s
=Kt=
bs tan(O+E)
(3.6.15)
but SOLUTl
(3.6.16)
hence
(3.6.17) .
From continuity
(3.6.l8)
Unsaturated
flow
Figure 3.6.9.
Refraction of flow
lines across a water
table (after Jacob56).
of where bu and bs are as shown in Figure 3.6.7 so that

I area.
poten- (3.6.19)
1945
w lines
nation. which, when solved for£, gives
l<ieyer9
(3.6.20)
This states that for v11 > 0, flow lines will form an angle of (90 degrees- o- E) below the water
if flow, o,
table. For the case of no percolating flow, v11 = 0 and E =90 degrees - so that vs parallels the
fa~e as water table.
~
turated
!r table
occurs Consider a case where 1.6 mlyear of percolating water reaches the water table of an unconfined aquifer.
'EXAMPLE J.6,J
If the hydraulic conductivity and the regional hydraulic gradient of the aquifer are 9.3 m/day and 0.01,
respectively, find the deflection angle of the flow lines as they cross the water table of the unconfined
3.6.15) aquifer.
·•~'
·~ SOLUTION Given i =O.oJ, Vu =1.6 m/year::: 4.38 X w-3 m/day, K =9.3 m/day, and tan 0 = i = O.Ql so <'i::: 0.573°.
~ The deflection angle, e, is computed using Equation 3.6.20:
3.6.16) f.
..~
''
t
,.
e =tan -t~ tano)-o
3.6.17)
~
1
:::tan -t( 4.38xl0-
93
· m/day
3
m/day
tan(0.573• ))- 0.573· =86.7"
:~
The angle between the groundwater flow direction and the refracted flow lines is 90° - (0 + e) = 90° -
3.6.18) (0.573° + 86.7°) = 2.7°. •
3.6.4 Flow Across a Hydraulic Conductivity Boundary

Similar to the above analysis, where flow passes from a region of hydraulic conductivity K1 to
one of K2, a change in flow direction results. The change of direction can be derived from con-
tinuity considerations and expressed in terms of the two K values. Visualizing a flow field as
shown in Figure 3.6.10, we see that the normal components of flow approaching and leaving
the boundary must be equal; hence, the normal velocities vn must be such that
(3.6.21)
or
(3.6.22)
where 8 1 and 82 are angles with the normal shown in Figure 3.6.1 0. Also, the distance b along
the boundary between two adjacent flow lines must be the same on each side of the boundary.
From Figure 3.6.10, the distance b can be given as
b = dJ4 = dLz (3.6.23)

sin8 1 sin8 2
which can be arranged

(3.6.24)
Dividing this equation by Equation 3.6.22 and noting that dh 1 = dh 2 between two equipoten-
tial lines, gives
K1 = tan8 1
K2 tan8 2 (3.6.25)
Thus, for saturated flow passing from a medium of one hydraulic conductivity to that of
another, a refraction in flow lines occurs such that the ratio of the Ks equals the ratio of the tan-
gents of the angles the flow lines make with the nonnal to the boundary. Consequences of the
relation are illustrated in Figure 3.6.11.
SOLUT.
v, v,
Hydraulic
conductivity K1
Boundary 3.6.5
Hydraulic
conductivity K2
Figure 3.6.10. Refraction of flow lines

across a boundary between media of differ-
ent hydraulic conductivities.
;tivity K1 to
d from con-
low field as K2 = 10
K,
Figure 3.6.11. Refraction across layers of
and leaving coarse and fine sand with a hydraulic con-
ductivity ratio of 10 (after Hubbert51 copy-
right © 1940 by the University of Chicago
(3.6.21) Press).
(3.6.22)
nee b along Unconfined

aquifer
e boundary.
(3.6.23)
Aquitard
(3.6.24) Confined
aquifer
> equipoten- Figure 3.6.12. Example
3.6.4
(3.6.25)
Consider a case where a leaky confined aquifer with 4.5 m/day horizontal hydraulic conductivity is over-
y to that of lain by an aquitard with 0.052 m/day vertical hydraulic conductivity. If the flow in the aquitard is in the
.o of the tan- downward direction and makes an angle of so with the vertical (see Figure 3.6.12), determine 92.
.ences of the
SOLUTION Given K1 =0.052 m/day, K2 = 4.5 rnlday, and 9 1 =5°, Equation 3.6.25 is used to compute Elz:
~ tan(s ) ~ = _ •
=tan9 1 ~ 0.052 rnlday =__
0
92 82 5
K2 tanEl2 · 4.5m/day tan92
The flow lines become nearly horizontal as they enter into the confined aquifer. This is a typical case for
regional flow systems, as the hydraulic conductivity of a confined aquifer is generally a few orders of
magnitude larger than that of the confining layers. •
3.6.5 Regional Flow Patterns

Although most groundwater movement in shallow aquifers tends to be nearly horizontal,
regional flow patterns can become quite complex. Reasons for this stem from the diversity of
field parameters: areas and magnitudes of recharge and discharge, topography, stratigraphy,
97
ow lines and anisotropy. 88 Analytic solutions for specified aquifer cross sections by Totb •98 demon-
iia of differ- strated that the variability of a water table could produce a variety of flow patterns. Subsequent
work by Freeze37 extended this approach to other subsurface boundary conditions. From these
Potential distribution on the surface

of the theoretical flow region A
METERS
28
24
20
16
12
SEA
LEVEl
.....
:~: .. ~ . . :. Figure 3.6.13 Vertical cross section showing
8
theoretical flow pattern of groundwater
llF"1 Region of local system r-:-1 Region of intermediate D Region of regional system
through an aquifer with a sloping sinusoidal 12
~ ol groundwater flow ~ system o1 groundwater flow of groundwater flow
- - - Line of force water table (after Toth98). 18
20
_l
contributions it is clear that accurate evaluation of groundwater flows is contingent on a .v
detailed knowledge of hydrogeologic conditions.

An illustration of a regional groundwater flow pattern is shown in Figure 3.6.13 for a vertical
cross section through an unconfined aquifer. Flow lines were obtained by computer assuming:
l. A homogeneous and isotropic aquifer with impermeable boundaries along the sides and
the bottom. 11
2. A rectangular flow region 6,100 m wide by 3,050 m deep.

3. A sinusoidal potential distribution (equivalent to the water table) with an amplitude of .J 1D
15 m and a mean slope of 2 percent on the surface of the theoretical flow region. ~
iS 17
Local, intermediate, and regional systems of flow are indicated in Figure 3.6.13, as well as II)
flows counter to the mean water table gradient. Although the aquifer portrayed is highly ide- ~
0
Ill
alized, a similar multiformity in flow pattern can be anticipated for actual aquifers where < 18
irregularities of topography, stratigraphy, and anisotropy are introduced.

~
:l 15
!
~
a:
14
CJ\SR STIIDY Lake Five-0, Florida

~
3:
13
The purp<Jse of this case study is to examine the head distribution components in the total inflow (precipitation plus groundwater
and the groundwater flow near Lake Five-0, which was intro- inflow) and the total outflow (evaporation plus leakage) budgets
duced in Chapter 1. Lake Five-0 is a softwater seepage lake in of the lake. 12
northwestern Florida. The Lake Five-0 study area (see Figures As reported by Grubbs,44 head fluctuations for the two-year
1.6.4 and 1.6.5) has three hydrogeologic units (surficial aquifer, (1989-1990) study period were generally consistent with typical
11
intermediate confining unit, and Upper Floridan aquifer) that seasonal patterns of precipitation. Seasonal precipitation in north· J I
influence the hydrology. Figure 3.6.14 illustrates these three units western Florida causes wet conditions during the summer and
for the hydrogeologic section A-A', location as shown in Figure dryer conditions during the fall and spring, so that as shown in
1.6.5. Figure 3.6.15, heads increased during the wet sununer seasons of Theh•
Grubbs44 evaluated the temporal and spatial distributions of 1989 and decreased during the dryer spring months of 1989 and potential
groundwater flow into and the leakage from Lake Five-0 using 1990 and the fall of 1990. Head data from a monitoring network Floridan,
hydrologic data and simulation models of the shallow ground- indicated a consistent pattern of groundwater flow toward the stage oft
water system adjacent to the lake. As shown by the study (hydro- lake, so that water table elevations (a) increased with distance eating ab
logic budget by Grubbs44) groundwater inflow to the lake and from the lake (Figures 3.6.16 and 3.6.17) and (b) were consis- 1989 and
leakage from the lake to the groundwater system are the dominant tently higher than the stage of the lake.
3.6 Groundwater Aow Directions 117
A A'
METERS
Figure 3.6.14. Hydrogeologic section

1 showing
tter A-A' through Lake Five-0 showing mid-
points of screened intervals of observation
nusoidal
wells used to construct the section.
(Location of section is shown in figure
Vonk:al oxaggen~tion IS klO
EXPLANATION 1.6.5.) Hydrogeologic section A-A' revised
0 250 500 FEET
..1. WE SURFACE OR WATER TABLE
I I I I
1
1
from Andrews and others4 (as presented in
ent on a 0 w2.2 ~EN~U.ItmRVAI. 0 50 100 150 METERS Grubbs44) .
:1 vertical
20
Ulllng: WELL W1.1 0
0 0 0
.
(SURFICIAL 0
I 0 00
• • • • • ••
ides and 0 AQUIFER) 0 0
19 0 00 0
~
• •• •
I Oo 0
• 0 •
0 •• • 0
!itude of (8 •
• • ••
0
0 0 0 WELL W1.3
(INTERMEDIATE
. o
...J • CONFINING UNIT)
• 0
0
well as ~ 11
••
WELL W6.2
~
(SURFICIAL AQUIFER)
:hly ide-
s where
m
<
Cl)
II:
16 /_,..r-~..........
"-..,
\
, -\.
~
LAKE FIVE-<> ""--~-I ,',',
~ 15
\ '\
'\
~14
\
',
'',,.
mdwater
budgets
a:
w
J::
13 ...
...
...
.
... ... ...
•
.........
.·r···
WELL W1.5
......
.........
... ...
(UPPER FLORIDAN
AQUIFER)
...
12 ...
:wo-year ...... Figure 3.6.15. Water levels of Lake Five-
'I typical 0 and selected groundwater monitoring
in north- 11
J A $ 0 N D J F MA M J J " S 0 N 0 J F M A M J J A S 0 N 0 J F
wells, July 1988 to February 1991
mer and 1988 1969 1990 1991 (Grubbs44).
hown in
asons of The head data shown in Figures 3.6.15-3.6.18 indicate a strong of 1990. Leakage from the lake decreased during the summer of
989 and potential for downward leakage from Lake Five-0 to the Upper 1989 and increased in the summer of 1990. Grubbs44 notes that the
network Floridan Aquifer. Figure 3.6.18 shows the differences between the changes in head difference (and the consequent leakage from the
~ard the
;
stage of the lake and the head in the Upper Floridan aquifer, indi- lake) were probably due to regional variations in recharge to the
distance ~ cating abrupt declines in head difference during the summer of Upper Floridan Aquifer during the two-year time period.
consis- 1< 1989 and abrupt increases in head differences during the summer
(continues)
A A'
METERS
28
24 SURFICIAL AQUIFER
20
16
12
4
SEA
LEVEL
4
8 Figure 3;6.16.
12 Hydrogeologic sec-
tion A-A' through
16
Lake Five-0, show-
20 ing vertical head dis-
Vttbl e~ II x10 EXPLANATION
0 250 500 Fl:ET
DIRECTION Of GROU!+[).WATER FLOW tribution near the lake
I I I
I II 13.74 LAKE OR WATER TABLE SURFACE for Aprill3, 1989.
--1-
50 100 150 METERS UNE OF EQUAL tM>RAUUC HEAil-41 melell above SN !owl. (Location of section
- \6- CcnlcoJ< ln\oMJ Is ve<toble
13.74 L..AKE-LEVEL~ meters tbow ... !ovol is shown in Figure
e15.88 HYOIW.IUC HEAD VALUE-In - - SM love! 1.6.5.) (Grubbs44)
A A'
METERS
28
enc
24 fici
oft
ute.
16
oft
12 is t
hea
4
SEA
LEVa 3.i
4
3.i
12 Figure 3.6.17.
Hydrogeologic sec-
18
tion A-A' through
20 Vertical ~ b xtO Lake Five-0, show-
EXPLANATION
0 250 500 FE£T ing vertical head dis-
DIRECTION OF GROUND-WATER FLOW
I I
I
I
I
I tribution near the Jake
_.L.. LAKE OR WATER-TABlE SURFACE
0 50 100 150 METERS
for January 17, 1990.
-16- LINE OF EQUAL HYDRAULIC HEAI)-In mele~
above sae kM!L Concour interval Is variable (Location of section
13.74 LAKE LEVEL-In metem aboVe sea level is shown in Figure
•15.86 HYDRAULIC HEAD VALUE-ln melers abo~ sea level 1.6.5.) (Grubbs44)
3.7 Dispersion 119
15.4 2.25
152 / 2.20 en
I a:
w
15.0 2.15
I l;j
LAKE FIVE.() STAGE I
"'I
14.8 2.10 ~
14.6 2.05 3;:

...J
w 14.4 I 2.00
ci
>
w '\ I ;;)
...J 142 1.95 J:
\. I HEAD IN z
;;)
en
w
14.0
13.8
\.\ I L
UPPER FLORIDAN AQUIFER
NEAR lAKE FIVE.()
1.90
1.85
<{
0
ii:
> ' , .._-- I 0
0
a:l 13.6
0_.8- -e----e-...__&---0.. 1.80
...J
u.
<{ I -o.."-G.. a:
en 13.4 ,l ....()~ 1.75 w
a: a.
w / 1.70
a.
132 ~
Iii en
~ 13.0 1.65 ~
~
z
12.8 1.60 ;E
c sec- \
)Ugh
~
:::;) 12.6 \(fl.. w
1.55 (.!)
~
f=
\
1.50 ~
show- 12.4
'&,, en
~d dis-
<t. 12.2 1.45 0I
w
the lake 1.40 > Figure 3.6.18
12.0
HEAD DIFFERENCE BElWEEN u:
989. 11.8
LAKE FIVE.() AND THE w
1.35 :!!::: Comparison of Lake Five-
UPPER FlORIDAN AQUIFER
ection ~ 0 stage and head in the
11.6 1.30
gure Upper Floridan Aquifer,
s44) 1.25
JJASONO December 1988 to
1988 1990 January 1991 (Grubbs44).
CASE STUDY Lake Five-0, Florida (continued)
Head data indicate significant hydraulic conductivity differ- larger than the head differences between wells 1.3 and 1.4 (refer to
ences within the intermediate confining unit and between the sur- Figure 3.6.14 for locations). An abrupt decrease in hydraulic con-
ficial aquifer and intennediate confining unit. Confining properties ductivity occurs at contact of the surficial aquiferlintennediate con"
of the intermediate confining unit are probably not evenly distrib- fming units. This is indicated in Figures 3.6.16-3.6.18 by the
uted within the unit. A dense, shelly clay layer is located at the base vertical head losses for the group of wells 1.1 through 1.5. The
of the unit and is the most effective confining unit in the unit. This lower hydraulic conductivity of the intermediate confining unit in
is based upon Figures 3.6.16 and 3.6.17, which indicate the large relation to the hydraulic conductivity of the surficial aquifer unit
head loss that occurs across the basal clay unit, which is much limits the downward movement of water.
3.7 DISPERSION
3.7.1 Concept
In saturated flow through porous media, velocities vary widely across any single pore, just as
sec- in a capillary tube where the velocity distribution in laminar flow is parabolic. In addition, the
tgh
pores possess different sizes, shapes, and orientations. As a result, when a labeled miscible liq-
lOW-
d dis-
uid, referred to as a tracer, is introduced into a flow system, it spreads gradually to occupy an
le lake
increasing portion of the flow region. This phenomenon is known as dispersion and constitutes
1990. a nonsteady, irreversible mixing process by which the tracer disperses within the surrounding
~tion water. 20· 84
ure In a column packed with sand as in Figure 3.7.la and supplied continuously after a time
14) t0 with water containing a tracer of concentration c0 , dispersion in the longitudinal direction of
tOverflow
I
Tracer curve ---1
without dispersion I Tracer curve
I
with dispersion
Screen
Figure 3.7.1. Longitudinal
dispersion of a tracer pass-
Outflow containing Time after introduction of tracer --+-
ing through a sand column.
tracer concentration c
(a) Sand column
(a) (b) (b) Dispersion curve
flow can be measured. 42 · 89 From samples ofwater emerging from the column, a tracer
concentration cis found. In Figure 3.7.lb, the solid line shows a typical S-shaped dispersion
curve. Besides longitudinal dispersion, lateral dispersion also occurs because water is contin-
ually dividing and reuniting as it follows tortuous flow paths around grains of a medium, as
illustrated in Figure 3.7 .2. 26
The equation for dispersion in homogeneous and isotropic media for the two-dimensional
case has the form
(3.7.1)
where cis the relative tracer concentration (0 ~ c ~ 1), DL and Dr are longitudinal and trans-
verse dispersion coefficients, v is fluid velocity, x is the coordinate in the direction of flow, y is
the coordinate normal to flow, and tis time. Dimensions of the dispersion coefficients are L2/T.
Dispersion is essentially a microscopic phenomenon caused by a combination of molec-
ular diffusion and hydrodynamic mixing occurring with laminar flow through porous media.
The net result produces a conic form downstream from a continuous point-source tracer (Fig-
ure 3.7.3a), and an expanding ellipsoid for a single tracer injection (Figure 3.7.3b).* Many
theoretical and experimental studies have provided a better understanding of the phenome-
non.5· 29· 33· 34· 35· 36· 72• 73· 76· 77 Most mathematical descriptions of dispersion are based on sta-
tistical concepts because of the difficulties in defining the microstructure of porous media as 3.7.2 l
well as the relative roles of molecular diffusion and mechanical dispersion. 20· 65 · 96 Experi-
mental studies have attempted to establish relations between factors such as velocity, media
structural properties, and permeability with dispersion coefficients. 35· 47 · 49· 76· 92· 99 The pattern
of a point tracer as it moves downstream from its source tends to a normal (Gaussian) distri-
bution both longitudinally and transversely, as sketched in Figure 3.7 Jb. Furthermore, the lon-
gitudinal component is larger than that of the transverse so that the major axis of mixing
occurs in the direction of flow.
*one may visualize these two fonns, by analogy, as a continuous plume and as a single puff of smoke drifting
downwind from a smokestack, respectively, even though the flow in porous media is laminar whereas that in the
atmosphere is turbulent.
3.7 Dispersion 121
Tracer
entrance
'~
'
I Direction
of flow
itudinal
:r pass-
:olumn.
Figure 3.7.2. Lateral dispersion of a tracer origi-
nating from a point source in a porous medium.
a tracer
;persion ~
Flow
contin- direction
lium, as
:nsional
(a)
Tracer boundaries
(3).1)
d --------t=s--
-----
<=====14[_---=-:::::__ ~-
~
j trans- Flow
direction
ow, yis
reL2/T.
molec- Longitudinal tracer
media. distribution at two Figure 3.7.3. Sketches of tracer distribution result-
~r (Fig- distances
ing from dispersion by flow in porous media. (a)
' Many (b) Continuous tracer (b) Single slug of tracer
:nome-
on sta-
~aas 3.7.2 Dispersion and Groundwater Hydrology
~xperi
In groundwater hydrology, dispersion may be encountered whenever two fluids with different
media characteristics come into contact. Prime examples of this include tracers for evaluating direc-
mttem tions and velocities of groundwater flow, introduction of pollutants into the ground (Chapter
distri- 8), artificial recharge of water with one quality into an aquifer containing groundwater of
le Ion- another quality (Chapter 13), and intrusion of saUne water into freshwater aquifers
nixing (Chapter 14). In general, the magnitude of dispersion for uniform sands can be measured in
terms of only a few meters over a travel distance of 103 meters.
In addition, the heterogeneity of most geologic materials introduces irregularities of flow
g with the consequent mixing of a tracer; these effects often far overshadow the effects of micro-
the scopic dispersion. The rapid nonsymmetric dispersal of two continuous dye streams in the lab-
oratory demonstration shown in Figure 3.7.4 amply illustrates the important role of
Injection
points
Figure 3.7.4. Dispersion of two dye streams in a heterogeneous porous medium. Dotted areas indicate bands of much higher per-
meability (after Skibitzke and Robinson90).
heterogeneity. In essence, a macroscopic dispersion is superimposed on the flow system. Fur-

thermore, the irregular geometries of groundwater recharge and discharge zones, together with
the lack of specific data on aquifer characteristics, preclude quantitative evaluation of disper-
sion coefficients for most natural groundwater flow situations.
3.8 GROUNDWATER TRACERS

A variety of tracers have been employed for studying dispersion and also for evaluating direc-
tions and rates of groundwater flow under field conditions (see Section 3.3.3). An ideal tracer
should (1) be susceptible to quantitative determination in minute concentrations, (2) be absent
or nearly so from the natural water, (3) not react chemically with the natural water or be
absorbed by the porous media, (4) be safe in terms of human health, and (5) be inexpensive and
readily available. No tracer completely meets all these requirements, but a reasonably satis-
factory tracer can be selected to fit the needs of a particular situation. Possibilities include
water-soluble dyes (such as sodium fluorescein), which can be detected by colorimetry; solu-
ble chloride and sulfate salts and sugars, which can be detected chemically; and strong elec-
trolytes, which can be detected by electrical conductivity.
Radioisotopes such as H3 (tritium), CofiJ, Rb86, ! 131 have also served as tracers.41 · 54•58•59· 94• 108
Naturally occurring radioisotopes, H3 and C 14, not only have enabled residence times of
groundwater to be estimated (see Section 2.1) but also have provided a means for evaluating
regional groundwater flow. 6• 46• 78· 105 Refer to the work by Field33- 36.
3.9 GE~ERAL FLOW EQUATIONS

The control volume (CV) for a saturated flow is shown in Figure 3.9.1. The sides, which
define the control suiface (CS), have lengths dx, dy, and dz in the coordinate directions. The
total volume of the control volume is dxdyd.z and the volume of water flowing into or out of
the control volume is Odxdydz, where e is the moisture content. The use of the control-volume
approach is described in detail in Chow et al.;21 Crowe et al.;23 and Mays69. The general con-
trol volume equation for continuity is applicable:
0 :-: ~
dt cv
I I
p dV + pv. dA
cs
(3.9.la)
or
JpV ·dA =-~ Jp·d\f (3.9.lb)

cs dt cv
which states that the rate of outflow of groundwater through the control surface (CS) is equal
to the rate of change of groundwater stored in the control volume (CV). The elemental volume
3.9 General Flow Equations 123
z aq
q+-dz
az
1 c
0
i
~
w
er per- q
X
Figure 3.9.1. Control volume

Darcy flux for development of the continu-
~m. Fur- y ity equation in porous medium
her with
(from Chow et al21 ).
f disper-
is tN =dxdydz, the velocity vector of flow is V, and the volume of flowrate past a given area
dA is V · d.A, where A is tlle area vector. Equation 3.9.1b also states that the net rate of out-
flow of mass (groundwater) from the control volume is equal to the time rate of change of mass
1g direc- (groundwater) stored within the control volume.
a1 tracer The time rate of change of mass stored in the control volume is defined as the time rate of
e absent change of fluid mass in storage, expressed as
...
~r or be
sive and
'
i_
dt
Jp dV =pSs 'dhar (dxdydz) +pW(dxdydz) (3.9.2)
ly satis- cv
include where Ss is the specific storage and W is the flow into or out of the control volume, W =
y; solu-
Q/(dxdydz). The term pSs dh (dxdydz) includes two parts: (1) the mass rate of water produced
ng elec- ar
by an expansion of the water under change in density and (2) the mass rate of water produced
'59, 94. 108 by the compaction of the porous medium due to change in porosity.
imes of The inflow of water through the control surface at the bottom of the control volume is
aluating qdxdy and the outflow at the top is [q + (dq!dz)dz]dxdy so that the net out flow in the vertical
direction z is
Jpv.dA =p(q + oq dz)dxdy- qdxdy
cs oz (3.9.3)
, which
=pdxdydz oq
•ns. The oz
rout of
where q (in the z direction) in "dq is qt.
volume
raJ con-
oz
Considering all three directions, we have
JPv.dA =pdxdydz oqax + pdxdydz dyaq + pctxdydz aaqz

cs
(3.9.4)
(3.9.la)
Substituting 3.9.2 and 3.9.4 into 3.9.1 results in
0 =pS5 ~~ dxdydz + pW(dxdydz) +pdxdydz ~! +pdxdydz ~; + pdxdydz ~~ (3.9.5)
Dividing through by p dxdydz gives

is equal s ()h + dq + iJq + iJq + w=0 (3.9.6)
volume s dt dx dy dZ
Using Darcy's law, the Darcy flux in each direction is 3.10

oh
qx = -Kx ox (3.9.7a)
qy =-K-
oh (3.9.7b)
y ()y
oh (3.9.7c)
qz = -Kl ()z
Subsitituting these definitions of qx, qy, and qz into 3.9.6 and rearranging results in
(3.9.8)
This is the equation for three-dimensional transient flow through a saturated anisotropic
porous medium. For a homogenous, isotropic medium (Kx = KY = Kz) then Equation 3.9.8
becomes
cPh cPh o2h s
-+-+-=_1._-+-
ah w (3.9.9)
Clx2 CJi dz 2 K dt K
For a steady-state flow, ah =0, Equation 3.9.9 becomes Ground

at .2 2 2
CJ h Cl h Cl h w
-+-+-=- (3.9.10)
Clx2 a/ ()z 2 K
For a horizontal confined aquifer of thickness b, S =S)J and the transmissivity T =Kb, so
the two-dimensional form of 3.9.9 with W =0 becomes
(3.9.11)
The governing equation for radial flow can be derived using the control volume approach.
Alternatively, Equation 3.9.11 can be converted into radial coordinates using the relation
r =~ x2 + y2 • This is known as the diffusion equation, expressed as
~i(r
r ar
ahJ
or
= ah +~ ah
2
ar r ar 2
= §_ ah
T at
(3.9.12)
where r is the radial distance from a pumped well and t is the time since the beginning of
pumping. For steady-state conditions, ()hJ()t =0, so Equation 3.9.12 reduces to t
~~(,ah)=o (3.9.13)
c
r dr ar 0
~
QJ
Application ro Aquifers. The equations derived above will be applied subsequently to obtain Ui
analytic solutions to particular groundwater flow problems. For solution of any problem, ide-
alization of the aquifer and of the boundary conditions of the flow system is necessary. Results
may only approximate field conditions; nevertheless, known deviations from assumptions fre-
quently allow analytic solutions to be modified to obtain an answer that otherwise would not
have been possible. A common assumption regarding the aquifer is that it is homogeneous and
isotropic (see Chapter 2). Often aquifers can be assumed to be infinite in areal extent; if not,
boundaries are assumed to be (1) impermeable, such as underlying or overlying rock or clay 0
layers, dikes, faults, or valley walls; or (2) permeable, including surface water bodies in con-
tact with the aquifer, ground surfaces where water emerges from underground, and wells.
3.!0 Unsaturated Flow 125
3.10 UNSATURATED FLOW

In groundwater hydrology, unsaturated flow is important for downward vertical flow (natural
3.9.7a)
and artificial recharge), upward vertical flow (evaporation and transpiration), movement of
pollutants from ground surface, and horizontal flow in the capillary zone above the water
3.9.7b) table.91 A large amount of literature exists on unsaturated flow, most of it contributed by soil
scientists. Significant summaries of the subject are available in several references.61 • 93• 103
In order to discuss infiltration, we must first consider the various subsurface tlow processes
3.9.7c)
shown in Figure 3.10.1. These processes are infiltration of water to become soil moisture, sub-
surface flow (unsaturated flow) through the soil, and groundwater flow (unconfined saturated
flow). Unsaturated flow refers to flow through porous medium when some of the voids are
occupied by air. Saturated flow occurs when the voids are filled with water. The water table is
(3.9.8) the interface between the saturated and unsaturated flow where atmospheric pressure prevails.
Saturated flow occurs below the water table and unsaturated flow occurs above the water table.
otropic A typical distribution of water content above a water table is sketched in Figure 3.10.2a.
n 3.9.8 If the water table is lowered llh, then the capillary zone shifts downward. The shaded area
serves as a measure of the volume of water drained from above the water table. This water is
released by vertical percolation; consequently, specific yield becomes an asymptotic function
(3.9.9) of time (Figure 3.10.2b).
'f.
3.9.10) .~
···~
··,
Kb, so
i.
~
;
3.9.11) Figure 3.10.1. Subsurface water zones and

l processes (from Chow et al.21).
~.
)roach.
J
elation
'
3.9.12)
Initial water
., content
ting of
t
3.9.13) .,
\
c:
0 t - - ----------=-~-;-~-=------
> ~
[jj
CJ)"
u
obtain i Final water -a;
I
content ·:;.
m, ide- ~
~esults .
~
~
Initial water table
0
!8
Q>
a.
ms fre- }
(f)
uld not
IUS and
'
~ Final water table
Time
(b)
~
if not,
" 0 50 100
or clay Figure 3.10.2. Movement of water in the zone of aeration by lowering the
in con- Percent saturation water table. (a) Water content above the water table. (b) Specific yield as a
Us. (a) function of the time of drainage.
3.10.1 Flow Through Unsaturated Soils

Consider the control volume (element) in Figure (3.10.3b) for an unsaturated soil which has
sides of lengths d.x, dy, and dz with a volume of dx dy dz. The volume of water contained in the ·~
control volume is edx dy dz [where e is the moisture content]. Flow through the control vol-
ume is defined by the Darcy flux, q =Q!A, which is the volumetric flow rate per unit of soil
area. For this derivation, the horizontal fluxes are ignored and only the vertical (z) direction is
considered, with z positive upward.
Using the general control volume equation for continuity (3.9.lb)
fpV · dA = ~~ fpd\f (3.10.1)

cs
The time rate of change of mass stored in the element (control volume) is
- dJ d
- pa'V =-(p9dxdydz) ~
= pOd.xdyd~:)" (3.10.2)
dt dt ot
where the density is assumed constant. The net outflow of water is the difference between
the volumetric inflow at the bottom (q dx dy) and the volumetric outflow at the top [q +
(oqldz)dz]dx dy, so
• Solid particles
Water
0 Air-filled voids
(a)
c:
0
~(I)
[jj
q Figure 3.10.3. (a) Cross-section

X tluough an unsaturated porous
medium (b) Control volume for
Darcy flux development of the continuity
equation in an unsaturated
porous medium (from Chow
(b) et al.21 )
1ich has
IpV. dA =p(q + aq~ dzJdxdy -pqdxdy =pdxdydz aq~
a
(3.10.3)
:din the
rol vol- Dividing through by p dx dy dz and rearranging results in the following continuity equation for
t of soil one-dimensional unsteady unsaturated flow in a porous medium in the z direction:
:ction is
(3.10.4)
Darcy's law relates the Darcy flux, q, to the rate of headloss per unit length of medium. For
3.10.1) now in the vertical direction, the headloss per unit length is the change in total head ah over
a distance, az, that is, - ah' where the negative sign indicates that total head decreases (as a
az
result of friction) in the direction of flow. Darcy's law for unsaturated flow can be expressed as
3.10.2) ah
q =-Ku(e) az (3.10.5)
etween
where Ku(e) is the hydraulic conductivity, as a function of the moisture content, e. This law
lp [q +
applies to areas that are large compared with the crosssection of individual pores and grains of
the medium. Darcy's law describes a steady uniform flow of constant velocity with a net force
of zero in a fluid element. In unconfined saturated flow, the forces are gravity and friction. For
unsaturated flow, the forces are gravity, friction, and the suction force that binds water to soil
particles through surface tension.
In unsaturated flow the void spaces are only partially filled with water so that water is
attracted to the particle surfaces through electrostatic forces between the water molecule polar
bonds and the particle surfaces. This in tum draws water up around the particle surfaces, leav-
ing air in the center of the voids. The energy due to the soil suction forces is referred to as the
suction head (or matric potential) 'I' in unsaturated flow, which varies with moisture content
(see subsection 3.11.2). Total head is then the sum of the suction and gravity heads:
h='Jf+Z (3.10.6)
Note that the velocities are so small that there is no term for velocity head in this expression
for total head.
Darcy's law can now be expressed as
q =-K(e) a( 'I'+ z) (3.10.7)

az
Darcy's law was originally conceived for saturated flow and was extended by Richards 83 to
unsaturated flow with the provision that the hydraulic conductivity is a function of the suction
head, that is, K = K('Jf). Also, the hydraulic conductivity can be related more easily to the
degree of saturation so that K =K(8). Because the soil suction head varies with moisture con-
tent and moisture content varies with elevation, the suction gradient can be expanded by using
the chain rule to obtain
a"' = d'Jf ae (3.10.8)
ection az de az
ous
e for in which aetaz is the wetness gradient and the reciprocal of d'Jf/de, namely de/d'Jf, is the spe-
jty cific water capacity. Now Equation 3.10.7 can be modified to
q= -K(e)(a"' + azJ =-K(e)(d"' ae + 1J =-(K(e) d'Jf ae + K(e)) (3.10.9)

az az de az de az
The soil water diffusivity D[L2ff] is defined as
D = K(9) d\jf (3.10.10) '~

de
The diffusivity is considered as the ratio of hydraulic conductivity to the water capacity of the
soil, d8fdljl. Substituting the above expression forD into Equation 3.10.9 results in
q=-( D~: +K(e)) (3.10.11)
Using the continuity equation (3.10.4) for one-dimensional, unsteady, unsaturated flow in a
porous medium, we obtain
(3.10.12)
which is a one-dimensional form of Richard's equation. This equation is the governing equa-
tion for unsteady unsaturated flow in a porous medium (Richards83). For a homogeneous soil,
oKioz =0, so o9/ot =oldz(Do9/dz).
Detennine the flux for a soil in which the unsaturated hydraulic conductivity is expressed as a function
of the suction head asK= 250(-'l'f2· 11 in cm/d at depth z1 = 80 em, h1 = -145 em, and 'I' I= -65 em at
depth z2 = 100 em, hz. = -160 em, and '1'2 = -60 em.
The flux is detennined using Equation 3.10.5. Hydraulic conductivity is computed using an average value
SOLUTION
of 'I'= -65 + (60)]12::;; -62.5 ern. Then Ku:;;; 250(-'lff2· 11 =250 (62.5f2·ll:. 0.041 ern/d. The flux is then
q::;; -Ku(hl -~) = -0.041[-145- (-160)]=-0.03 crnfd

Zt -z2 -80-(-100)
The flux is negative because the moisture is flowing downward in the soil.
•
Detennine the soil water diffusivity for a soil in which e=0.1 and K(9) :;;; 3 X ro-ll mrnls from a rela-
liXAMPL'E J.lQ.i tionship of '!f(9) at 9 = 0.1, ~'I'= 107 rnm, and ~e = 0.35.
SOLUTION Using Equation 3.10.10, we compute the soil water d.iffusivity as
D= K(e) ~~ =(3x 10- 11 nun/s)(107 mrn/0.35) =8.57 x 10-4nun 2/s

•
3.10.2 Unsaturated Hydraulic Conductivity
Unsaturated flow in the zone of aeration can be analyzed by Darcy's law; however, the unsat-
urated hydraulic conductivity Ku is a function of the water content as well as the negative pres-
sure head (tension).3• 14• 38 · 100 Because part of the pore space is filled with air, the available
cross-sectional area available for water flow is reduced; consequently, Ku is always less than
the saturated value K.
Although there are hysteresis effects present in the relations of Ku with water content and
negative pressure, approximations based on empirical evidence can be stated. Water content
data fit the fonn 31
3
Ku = ( Ss -So ) (3.10.13)
K ll - So
W.lO) :::.::
~ 0.8
1 of the 0
-~
.c 0.6
:~
10.11) tl
;::J
"C
c
8 0.4
-~
;::J
~
Threshold
'0 saturation
f 0.2
5 0 = 0.20
10.12)
~ Figure 3.10.4. Ratio of unsaturated to sat-
~equa- urated hydraulic conductivity as a function
lS soil, Degree of saturation, 55 of saturation (after Irmay 55).
unction
5 em at where Ss is the degree of saturation and S0 is the threshold saturation-that part of the voids
filled with nonmoving water held primarily by capillary forces. Equation 3.10.13 is plotted in
Figure 3.10.4; note that Ku ranges from zero at Ss = S0 to Ku =Kat Ss = 1, which is saturation.
For hydraulic conductivity and negative pressure, S-shaped relations as indicated in Fig-
:e value ure 3.1 0.5 are generally applicable. 101• 103 These can be approximated by a step function or by
. is then
Ku =----
a
K ~(-hf +a (3.10.14)
b
• where a, b, and n are constants that vary with particle sizes of unconsolidated material and h
is the pressure head measured in centimeters. It can be seen that when h =0, ~hlch occurs at
t a rela- atmospheric pressure, Ku = K. Orders of magnitude of the constants in Equation 3.10.14 for
different soils are as follows: 11
1.0
• ~
':6:.:0
.,.6
~
unsat- if!'
·:;:
e pres- ·;:::;
u 0.5
ailable ;::J
"C
c
;s than 0
u
u
"5
mtand ~
"C
>.
Figure 3.10.5. Median
:ontent :c relationships between
hydraulic conductivity
0 and soil-water pressure
10.13) -200 -150 -100 -50 0 (tension) (after
Pressure (em ot water) Bouwer 11 ) .
Material a b n
Medium sands 5 X 10 9 107 5
Fine sands, sandy loams 5 X 10 6 105 3
3 3
Loams and clays 5 X 10 5 X 10 2
Hydraulic conductivity is a measure of the ability of soils to transmit water and depends upon
both soil properties (total porosity, pore-size distribution, and pore continuity) and fluid prop-
erties (viscosity and density).
Table 3.10.1 describes three mathematical relationships for soil water retention and
hydraulic conductivity proposed by Brooks and Corey, 13 Campbell, 16 and Van Genuchten. 102
The relationships by Brooks and Corey and Campbell describe only a portion of the soil water-
Table 3.10.1 Soil Water Retention and Hydraulic Conductivity Relationships

Hydraulic soil characteristic Parameters Parameter correspondence
13
Brooks and Corey
Soil water retention A = pore-size index A=A
hb = bubbling capillary pressure hb= hb
e-er =(hb )"
9, = residual water content
<j>-er h 9,= 9,
<1> =porosity =
Hydraulic conductivity Ks = fully saturated conductivity (9 = <j>) Ks= Ks
h = matric potential
Se = effective saturation
h ="'
2
n=3+-
A
Campbe11 16
Soil water retention =porosity
<1> = 
~=r~r
Hb= scaling parameter with dimension of length Hb= hb
b =constant
b = _!_
Hydraulic conductivity n =3 +2b A.
Van Genuchten 102

Soil water retention <1>=porosity =
m 9r = residual water 9,= 9,.
e-e 1
a =constant a= (hbt 1
<1>-e: = [ l+(aht ]
n =constant n =A+ 1
A.
m =constant m=--
A-+1
h = absolute value of matric potential h = 1'1'1
Hydraulic conductivity
r rrr
Ki~) =(:=:: {!-[!-(:=::
e= water content; 'I' = capillary suction, em; K(O) = hydraulic conductivity for given water content, cmlh
Source: Rawls et ai. 80 Figure 3
retention curve for matric potentials less than the bubbling pressure (pressure at which air will
enter the soil); whereas the Van Genuchten 102 relationship describes the total soil water-retention
curve. The water retention relationship for two contrasting soil textures is shown in Figure
3.1 0.6a. It is interesting to note that the sandy loam soil retains Jess water than the clayey soil.
Figure 3. I0.6b illustrates that the hydraulic conductivity is a nonlinear function of volumetric
is upon soil water content, and varies with soil texture. The points for saturated water content-bl (on
d prop- the sandy loam topsoil curve) and b2 (on the clayey subsoil curve)-show that the hydraulic
conductivity for the sandy loam is higher than for the clay soil, even though the porosity is higher
on and in the clay soil. As shown in the figure, the hydraulic conductivity of both soils decreases rap-
hten. 102 idly as the soil water content decreases. Figure 3.10.6c illustrates that the hydraulic conductiv-
l water- ity of the sandy loam decreases more rapidly with a decrease in the matric potential head than
that of the clay subsoil. At lower values of the matric potential head, which refers to higher suc-
tions, the hydraulic conductivity of the clay soil is higher. The rate of change of the matric poten-
tial head in the sandy soil also decreases much more rapidly than that of the clay soil.
The absorbing (wetting) and desorbing (draining) have differing relationships of the
e matric potential head as a function of soil water content This difference is called hysteresis and
is illustrated in Figure 3.10.7. Hysteresis is caused by the entrapment of air in pockets con-
necting different size pores during wetting. Cycles of wetting and drying exhibit secondary
hysteretic loops (see Figure 3.10.7), which causes considerable difficulty in modeling soil
water movement. The K(8 ) has a much smaller hysteresis effect than '1'(8) which is why the
hydraulic conductivity KJ9) is used in the Darcy equation as opposed to K14('1').
-10,000 I 100.--------------------,
'0
E
00
'E ~ Sandy loam
•t: ~ -1,000
Cii~
§: topsoil~
Eci '::t:.,
... t'i5
(I) (I)
(ii.c.. -100 i-
S:] >
~c
u:J
(/)~ "0
0 -10 c:
c. 8 0.1
.Q
:i
0 ~
-g, 0.01 L_...._ ___.,___ _ _ _ _ __u__ ____J_ ____J
0.05 0.1 0.2 0.3 0.5 :r: 0.1 0.2 0.3 0.5
Volumetric soil water Volumetric soil water
content, 0 (cm 3 em..,) content, 9 (cm3 em..,)
(a) (b)
Matric potential head, 1/1 (em)

(c)
Figure 3.10.6. (a) The 'lf(9), (b) Ku(9), and (c) KJ'!I) relationships of sandy loam and clayey horizons (Rawls et a180).
-1000 ..-----...- - -- -- - - ----, 3.10.3
Figure 3.10.7. The effect of hysteresis

on the '1'(9) function of sandy loam
0.1 0.2 0.3 0.4 topsoil during wetting and drying
Soil water content, 8 (cm3 cm-3) (Rawls et al80).
3.11 }
Pressure head (em of water)

-60 -50 -40 -30 -20 -10 0 +1 0
'E
~
~
li
~ - 100
I
I
I
I
I
(a) (b)
Figure 3.10.8. Profiles of water content and pressure head in a sand column following infiltration of water. Numbers on the curves
represent time in minutes after start of infiltration (a) Water content (b) Pressure head. Data are computer-based numerical results
(after Watson in van Schllfgaarde 103 ; reproduced from Drainage for Agriculture, ASA Monograph No. 17, pp. 368-369, 1974 by
pennission of the American Society of Agronomy).
r
3.11 Kinematic Wave 133
3.10.3 Vertical and Horizontal Flows

Illustrative of the vertical drainage of water in unsaturated soils are the data shown in Figure
3.10.8. An initially saturated sand 180 em in depth was allowed to drain for one hour and was
then subjected to infiltration for four min under a surface head of 3 em. The resulting depth pro-
files of water content and pressure head are indicated for various time intervals after the end of
the infiltration interval. Note that at the start of the subsurface redistribution of water (t = 0), sat-
uration (water content= 0.29 cm3/cm 3) extends to a depth of about 30 em and corresponds to a
linear pressure gradient. For field conditions where stratifications in soil texture can be
expected, the complexity of such downward water migration, as from rainfall or irrigation, can
be appreciable39• 103 .
Lateral flow occurs above the water table in the capillary zone. The flow rate decreases
with the degree of saturation, and hence the hydraulic conductivity (see Figure 3.10.4). The
fraction of flow above the water table can be calculated from an equivalent saturated thick-
ysteresis ness. 71 For aquifers of substantial depth, the flow component above the water table is negligi-
loam ble, but it can be significant in shallow unconfined aquifers.
ng
3.11 KINEMATIC WAVE

Figure 3.11.1 shows an individual soil water wave propagating downward through the soil
under gravity drainage. The kinematic wave model is a redistribution model based upon grav-
+10 ity drainage (Charbeneau 19• 20 and Morel-Seytoux74). Figure 3.11.1 a shows the wave immedi-
~ ately after the end of the period of infiltration, consisting of a rectangular portion added to the
profile for the antecedent water content. The horizontal line is the wetting front, indicating a
-20 drop in water content from the wave above to the unwetted soil below. After a short period of
drainage, the wave is shown in Figure 3.11.1 b, consisting of three parts: (1) a draining part
with increasing water content (curved portion of the wave) with depth, (2) a constant or verti-
-40 cal plateau part, and (3) the wetting front. The draining part has not reached the wetting front
and is separated from it by the vertical plateau. In Figure 3.11.1c, the drainage profile has
caught up with the wetting front and the vertical plateau region no longer exists.
-60
The basic assumption in the kinematic wave theory is that the pressure gradients are neg-
ligible. With negligible capillary pressure gradients and with 'JI constant within the profile, the
-80 soil water diffusivity is zero and Equation 3.10.11 simplifies to
'E
~
.J::. q = -CKuCe)) (3.11.1)
c.
Q)
-100 0 so that Darcy's law reduces to Equation 3.11.1 for the kinematic wave theory. Equation 3.10.12
then reduces to
-120 (3.11.2)
Water content
-140
o,c
I () ()
I
-160 I
I
1\1
I 1\1 1\1
I
=- .c .J::.
-100 0.
Q)
c.
Q)
c.
Q)
0 0 0
curves
esults Figure 3.11.1. Soil moisture profiles for
'4 by individual kinematic wave
(a) (b) (c) (Charbeneau 19).
so that flow is downward under a unit gradient. Equation 3.11.2 can be expressed as
ae + dK( e) ae = 0 (3.11.3)
dt de dz
Charbeneau 19· 20 shows the analytical solution of the kinematic wave model, which can be used
to model soil water redistribution and recharge.
3.12 INFILTRATION: THE GREEN-AMPT METHOD

The process of water penetrating into the soil is infiltration. The rate of infiltration is influ-
enced by the condition of the soil surface, vegetative cover, and soil properties including
porosity, hydraulic conductivity, and moisture constant. Figure 3.12.1 illustrates the distribu-
tion of soil moisture within a soil profile during downward movement. These moisture zones
are the saturated zone, the transmission zone, a wetting zone, and a wetting front. This profile
changes as a function of time as shown in Figure 3.12.2.
The infiltration rate, f, is the rate at which water enters the soil surface, expressed in inlhr
or em/hr. The potential infiltration rate is the rate when water is ponded on the soil surface, so
if no ponding occurs, the actual rate is less than the potential rate. Most infiltration equations
describe a potential infiltration rate. Cumulative infiltration, F, is the accumulated depth of
water infiltrated, defined mathematically as
l
F(t) = Jf(t)dt (3.12.1)

0
0 Moisture content -----+
Transmission
zone
' Transition zone
..c
15.
Q)
Cl Wetting
zone
0
t
Saturation
"' Wetting front
100%
Figure 3.12.1. Moisture zones during
infiltration (from Chow et al. 21 ).
Ground surface
..c
15.
Q)
0 Figure 3.12.2. Moisture profile as a
t function of time for water added to the

soil surface (Chow et al. 21 ).
Rainfall
hyetograph
Figure 3.12.3. Rainfall infiltra·
tion rate and cumulative infi.l-
(3.11.3) tration. The rainfall hyetograph
ll..
c0 illustrates the rainfall pattern as
e~ be used a function of time. The cumula-
~
~a);
--.::: tive infiltration at time 1 is F, or
·(ij£ F F(t) and at time t + 111 is F, +I!.!
!!~c
~ c.Q or F(t +ill). The increase in
=·Q(ij cumulative infiltration from
~~"3
EEE
co -.= ::J time t to t + 111 is F,... 61 - F1 or
is influ- a: .Eo F(t + !1t)- F(t) as shown in the
i.
ncluding ·~
figure. Rainfall excess is rain-
f= dF
distribu- dt fall that is neither retained on
1re zones t r+M the land surface nor infiltrated
is profile Time into the soil (Chow et al. 21 ).
j in inlhr and the infiltration rate is the time derivative of the cumulative infiltration given as
rface, so
quations f(t) =dF(t)
depth of dt (3.12.2)
Figure 3.12.3 illustrates a rainfall hyetograph with the infiltration rate and cumulative
infiltration curves.
(3.12.1) Green and Ampt43 ( 1911) proposed the simplified picture of infiltration shown in Figure
3.12.4. The wetting front is a sharp boundary dividing initial soil moisture content 9; below
from saturated soil with moisture content TJ above. The wetting front has penetrated to a depth
L in time t since infiltration began. Water is ponded to a small depth ho on the soil surface.
Consider a vertical column of soil of unit horizontal cross-sectional area (Figure 3.12.5)
and let the control volume be defined around the wet soil between the surface and depth L. If
the soil was initially of moisture content 9i throughout its entire depth, the moisture content
will increase from 9i to TJ (the porosity) as the wetting front passes. The moisture content e is
the ratio of the volume of water to the total volume within the control surface, so the increase
'ii'
0
luring
Wetted zone
(conductivity K) L
Wetting front
9r-....,
- TO;
I
I ''3 Oe
11
Figure 3.12.5. Infiltration into a column
Figure 3.12.4. Variables in the Green-Ampt infil- of soil of unit cross-sectional area for the
as a tration modeL The vertical axis is the distance from Green-Ampl model (from Chow et ai.21 ).
to the the soil surface, the horizontal axis is the moisture
rontent of the soil (from Chow et al.2 1).
in the water stored within the control volume as a result of infiltration is L(ll - 8) for a unit
cross section. By definition this quantity is equal to F, the cumulative depth of water infiltrated
into the soil. Hence
F(t) = L(ll - 8) =L/18 (3.12.3)
where /18 = (11- 8J
Darcy's law may be expressed as
dh 11h
q=-K-=-K- (3.12.4)
dz Llz
In this case the Darcy flux q is constant throughout the depth and is equal to -f, because q
is positive upward while f is positive downward. If points 1 and 2 are located respectively at
the ground surface and just on the dry side of the wetting front, Equation 3.12.4 can be approx-
imated by
f=K[~]
z2 Zt-
(3.12.5)
The head h1 at the surface is equal to the ponded depth h0. The head h2, in the dry soil below
the wetting front, equals -'V- L. Darcy's law for this system is written as
(3.12.6a)
and if the ponded depth h0 is negligible compared to 'V and L,
(3.12.6b)
This assumption (h 0 =0) is usually appropriate for surface water hydrology problems because
it is assumed that ponded water becomes surface runoff.
From Equation 3.12.3, the wetting front depth is L = F/118, and assuming h0 = 0, substi-
tution into Equation 3.12.6 gives
(3.12.7)
Since f =dF/dt, Equation 3.12.7 can be expressed as a differential equation in the one unknown
Fas
dF
dt
=K['V 118 +
F
F]
To solve for F, cross-multiply to obtain
[ F +F
'V 118
]dF= Kdt
then divide the left-hand side into two parts
F +'V 118 118

[ F + 'V 118 'V ]dF = Kdt
F +'V 118
and integrate
~or a unit F(t)[ 8 ] t

!lfiltrated fo 1- 'V ~
F + 'V ~e
f
dF = Kdt
o
(3.12.3) to obtain
F(t)- 'V~8{ln[F(t) + o/~8] -ln(o/~8) = Kt (3.12.8a)
or
(3.12.4)
F(t)- 'V ~e ln(1 + F(t)) =Kt (3.12.8b)
'V ~8
ecause q
;tively at Equation 3.12.8 is the Green-Ampt equation for cumulative infiltration. Once F is com-
: approx- puted using Equation 3.12.8, the infiltration ratefcan be obtained from Equation 3.12.7 or
(3.12.9)
(3.12.5)
When the ponded depth h0 is not negligible, the value of 'V- h0 is substituted for 'V in Equa-
1il below tions 3.12.8 and 3.12.9.
Equation 3.12.8 is a nonlinear equation in F, which can be solved by the method of suc-
cessive substitution by rearranging,
:3.I2.6a)
F(t) = Kt + 'V ~8 ln(1 + F(t)) (3.12.10)
'V~e
Given K, t, 'V and ~e, a trial value F is substituted on the right-hand side (a good trial value is
3.12.6b) F = Kt), and a new value ofF calculated on the left-hand side, whlch is substituted as a trial
value on the right-hand side, and so on until the calculated values ofF converge to a constant.
because The final value of cumulative infiltration F is substituted into Equation 3.12.9 to determine the
corresponding infiltration rate f
I, substi- Equation 3.12.8 can also be solved by Newton's method, whlch is more complicated than
the method of successive substitution but converges in fewer iterations. Referring to Equation
3.12.10, application of the Green-Ampt model requires estimates of the hydraulic conductiv-
(3.12.7) ity K, the wetting front soil suction head 'V (see Table 13.12.1), and ~e.
The residual moisture content of the soil, denoted by e,, is the moisture content after it has
tnknown been thoroughly drained. The effective saturation is the ratio of the available moisture (8- 8,)
to the maximum possible available moisture content (11- 8,), given as
· e-8
se =, - -'
_ e, (3.12.11)
where,_ 8, is called the effective porosity ee.

The effective saturation has the range 0 s; se s; 1.0, provided 8, s; es; 11· For the initial con-
dition, when 8 =8;, cross-multiplying Equation 3.12.11 gives 8;- 8, =seSe, and the change in
the moisture content when the wetting front passes is
~8 =,_ ei =11- (seee + 8,);

(3.12.12)
~8 =(1- se)Se
A logarithmic relationship between the effective saturation se and the soil suction head 'V
can be expressed by the Brooks-Corey equation (Brooks and Corey 13)
CumuJa
(3.12.13)
Green-
in which 'Jib and 'A are constants obtained by draining a soil in stages, measuring the values of
se and \jf at each stage, and fitting Equation 3.12.13 to the resulting data.
Brakensiek et al. 12 presented a method for determining the Green-Ampt parameters using
the Brooks-Corey equation. Rawls et al. 81 used this method to analyze approximately 5,000
soil horizons across the United States and determined average values of the Green-Ampt
parameters 11, ee, 'Jf, and K for different soil classes, as listed in Table 3.12.1. As the soil
becomes finer, moving from sand to clay, the wetting front soil suction head increases while Horton's
~
the hydraulic conductivity decreases. Table 3.12.1 also lists typical ranges for 11, 8e, and 'l'· The
ranges are not large for 11 and 8e, but \jf can vary over a wide range for a given soil. K varies
scs :;1
along with 'lf, so the values given in Table 3.12.1 for both 'V and K should be considered typi-
cal values that may show a considerable degree of variability in application. 2· 27
Table 3.12.2 compares other infiltration methods. Refer to Chow et al. 21 for detailed dis- F.= s(
cussions of each of these methods. t ~
Table 3.12.1 Green-Ampt Infiltration Parameters for Various Soil Classes*

Effective Wetting front Hydraulic
Porosity porosity soil suction conductivity
Soil class 11 ee head 'I' (em) K(cm!h)
Sand 0.437 0.417 4.95 11.78

(0.374-0.500) (0.354-0.480) (0.97-25.36)
Loamy sand 0.437 0.401 6.13 2.99 SOLUT11
(0.363-0.506) (0.329-0.473) (1.35-27.94)
Sandy loam 0.453 0.412 11.01 1.09
(0.351-0.555) (0.283-0.541) (2.67-45.47)
Loam 0.463 0.434 8.89 0.34
(0.375-0.551) (0.334-0.534) (1.33-59.38)
Silt loam 0.501 0.486 16.68 0.65
(0.420-0.582) (0.394-0.578) (2.92-95.39)
Sandy clay 0.398 0.330 21.85 0.15
loam (0.332-0.464)
0.464
(0.235-0.425)
0.309
(4.42-108.0)
20.88 0.10
I ~
Clay loam
(0.409-0.519) (0.279-0.501) (4.79-91.10) '&.
,i
.'1.
Silty clay 0.471 0.432 27.30 0.10 '&

:~
,7,
loam (0.418-0.524) (0.347-0.517) (5.67-131.50) .1
Sandy clay 0.430
(0.370-0.490)
0.321
(0.207-0.435)
23.90
(4.08-140.2)
0.06
I Timet (hr
Silty clay 0.479

(0.425-0.533)
0.423
(0.334-0.512)
29.22
(6.13-139.4)
0.05
i Infiltratior
Infiltratior
I'
Clay 0.475 0.385 31.63 0.03 Timet (hr:
(0.427-0.523) (0.269-0.50 1) (6.39-156.5) l
Infiltration
*The numbers in parentheses below each parameter are one standard deviation around the parameter value given.
Infiltration
Source: Rawls, Ahuja, Brakensiek, and Shirmohammadi 80 and Rawls, Brakensiek, and Miller81
3.12 Infiltration: The Green-Amp! Method 139
Table 3.12.2 Infiltration Equations
Cumulative infiltration (F,) Infiltration rate (j,) Couunents

12.13)
Green-Arnpt equation • Hydraulic conductivity (K)
ues of
~
• Wetting front soil suction head ('JI)
F, -'I' 68tn(1+
"'~8
J= Kt J, = K( "';6 + tJ • Change in moisture content (66)
using ~6=11-6;
5,000 • Porosity l1
Ampt • Initial moisture content (8;)
1e soil
wbile Horton's equation • fo = initial infiltration rate
F, =fct + (/o - fc)( I - e-k~lk fr = fc + (/o- fJekl • Constant infiltration rate (/c)
v. The • Decay constant (k)
vanes
I typi- SCS method • Potential maximum retention (S)
s2 dP, S = 1000 _ 10
:d dis- S(P,-10 ) J,1 = dt
F,:= 2
CN
P,-/0 +S (P,-/a+S) • Dimensionless curve number (CN)
O ~ CN$; 100
p~Io. • Initial abstraction 10

10 =0.2S
• Total rainfall to time t, P,
Philip's equation • Sorptivity (S)

,-
F- St 112 + kt J, =JI2Sr112 + K • Hydraulic conductivity (K)
11lic
ivity Use the Green-Amp! method to evaluate the infiltration rate and cumulative infiltration depth for a silty
/h) clay soil at 0.1-hour increments up to 6 hours from the beginning of infiltration. Assume an irtitial
effective saturation of 20 percent and continuous ponding.
'8
J9 SOLUTION From Table 3.12.1, for a silty clay soil, 6l = 0.423, 'I' = 29.22 ern, and K = 0.05 em/hr. The initial effec-
tive saturation iss,= 0.2, so 68 =(I-s.) 8e = (1- 0.20)0.423 = 0.338, and 'I' 68 = (29.22) (0.338) = 9.89
em. Assuming continuous pending, the cumulative infiltration F is found by successive substitution in
)9
Equation 3.12.10:
14 F = Kt +'I' 68ln[l + F I ('I'M)] = 0.05 t + 9.891n(l + F 19.89]

For example, at timet= 0.1 hr, the cumulative infiltration converges to a final value F = 0.29 ern. The
iS inftltration rate fis then computed using Equation (3.12.9)
f = K (1 +'I' 68/F) = 0.05 (1 + 9.89/F)
l5
As an example, at timet:=: 0.1 hr,f = 0.05(1 + 9.89/0.29) = 1.78 em/hr. The infiltration rate and the
lO cumulative infiltration are computed in the same manner between 0 and 6 hours at 0.1 hr or 0.5 hr inter-
vals, as shown in Table 3.12.3.
lO Table 3.12.3 Infiltration Computations Using the Green- Ampt Method
Time 1 (hr) 0.0 0.1 0.2 0.3 0.4 0.5 1.0 1.5 2.0
Infiltration ratef(cm!hr) 00 1.78 1.20 0.97 0.84 0.75 0.54 0.44 0.39
)5 Infiltration depth F (ern) 0.00 0.29 0.43 0.54 0.63 0.71 1.02 1.26 1.47
)3 Timet (hr) 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Infiltration ratef(cmlbr) 0.35 0.32 0.30 0.28 0.27 0.26 0.25 0.24
siven.
Infiltration depth F (ern) 1.65 1.82 1.97 2.12 2.26 2.39 2.51 2.64
•
;::
/ 3.1.5 c
I Cumulative rainfall a field s
i I
E' /
/ "'f......_
........
1~ Potential infiltration rate
{)
/ a medi;;
E
{)
0 / Cumulative infiltration
c / 3.1.6 If
0 \
\
~
/
/
.c
........ \
(1)
> ~ conduc1
§_ --- -------- Rainfall rate ~ ure 3.1.
(1) Actual infiltration "3
~ E able he
::J
a: () median
3.1.7 R,
Time Time
ductivit:
Figure 3.12.6. Ponding time. This figure illustrates the concept of ponding time for a constant intensity rainfall. Ponding time is the 3.2.1 Th
elapsed time between the time rainfall begins and the time water begins to pond on the soil surface. 69 as 11.2· J
the intri
Ponding time, tP' is the elapsed time between the time rainfall begins and the time water begins to pond (b) Wha
EXAA1PLE 3.12.2
on the soil surface. Develop an equation for ponding time under a constant rainfall intensity, i, using the ural gro1
Green-Ampt infiltration equation (refer to Figure 3.12.6). 3.2.2 Th
the hyciJ
from 5°(
SOLUTION The infiltration rate f and the cumulative infiltration Fare related in Equation 3.12.8. The cumulative
infiltration at ponding time tP is FP = i tP' in which i is the constant rainfall intensity (see Figure 3.12.6). 3.3.1 A,
Substituting FP = i tP and the infiltration ratef= i into Equation 3.12.9, we have 175 cm2
field san
i= K[~ztP~S +1] 320 sec<

hydrauli1
the intrir
whose solution is is used f1
ditions (1
t = -~e
- K"' conducti·
p i(i-K)
3.3.2 Ift
which is the ponding time for a constant rainfall intensity.
• pled and
3.3.4, ho
em?
PROBLEMS 3.3.3 A~
meamete
3.1.1 A confined aquifer with a porosity of 0.15 is 30m thick. The 3.1.3 A confined aquifer with a horizontal bed has a varying thick- tube is lC
potentiometric surface elevations at two observation wells 1,000 ness as shown in the figure. Assuming the flow in the aquifer is let level ;
m apart are 52.35 m and 56.90 m. If the horizontal hydraulic con- essentially horizontal, determine the flow rate if the piezometric em after:
ductivity of the aquifer is 25 m/day, detennine the flow rate per heads at sections (1) and (2) are 23.7 m and 27.1 m, respectively. the samp:
unit width of the aquifer, specific discharge, and average linear 3.3.4 If tl
velocity of the flow assuming steady unidirectional flow. How the same
long would it take for a tracer to travel the distance between the lect 200 r
obervation wells? 3.3.5The
K= 16m/day to be 0.6~
3.1.2 A field sample of an aquifer is packed in a test cylinder (see 20m
Figure 3.1.1). The cylinder has a length of 120 em and a diameter
of 12 em. The field sample with a porosity of 0.24 is tested under
t the piezm
materiall
~+--------1200 m-------+1
a constant head difference of 160 em with water at 10°C. If the tivity oft]
estimated hydraulic conductivity of the sample is 30 m/day, 3.1.4 What should be the range of water temperature in Problem 3.4.1 The
calculate (a) the expected total discharge, (b) the specific dis- 3.1.2 for Darcy's law to be applicable if the median diameter of shown be
charge, (c) the average flow velocity, and (d) the hydraulic gradi- the field sample is 2.2 mm? Is the water temperature given in the hydraulic
ent along the cy Iinder. problem appropriate? degree of
Problems 141
3.1.5 Could the same apparatus in Problem 3.1.2 be used to test 1
a field sample having a hydraulic conductivity of 86.4 m/day and

a median grain diameter of 5 rrun?
K z 2.3 mlday, b ~ 15 m K= 2.3 mlday. b= 10m
K = 2.3 mlday, D= 20 m
II
3.1.6 If a 50-em-long field sample with an estimated hydraulic K = 12.8 mlday, b =20m
conductivity of 8 m/day is to be tested in the laboratory (see Fig-

K= 12.8mlday. b= 15m
K= l2.8mlday,b= 10m I
ure 3.1.1) using water at 20°C, what would be maximum allow-
I I I
able head difference across the column? The sample has a
(a) (b) (c)
median grain diameter of 0.5 mm.
3.1.7 Rework Probl~m 3.1.6 with an estimated hydraulic con-
ductivity of 40 rn/day and a median grain diameter of 1.5 rom.
3.4.2 The stratification of a confined aquifer with a horizontal
me is the 3.2.1 The hydraulic conductivity of a medium sand was measured
bed varies as follows. Calculate the equivalent horizontal and
as 11.2 rn!day in a laboratory test with water at 25"C. (a) What is
vertical hydraulic conductivities in each case. What happens to
the intrinsic permeability of the sample in square micrometers?
the degree of anisotropy?
s to pond (b) What is the expected in-situ hydraulic conductivity if the nat-
using the ural groundwater temperature is 10°C? l '-------<
JJ-- I
3.2.2 The intrinsic permeability of a silty sand is 0.36 darcy. Plot I K = 1.4 mlday, b =21 m
K= 1.4 mlday, b= 5 m
the hydraulic conductivity as a function of water temperature

I K= 1.4 m/day, b= 5 m J
from SOC to 30°C. 1------___j
1mulative
! 3.12.6).
3.3.1 A constant head permeameter has a cross-sectional area of K= 11.7 mlday, b= 5 m l K= 11.7 mlday, b= 7m K= 11.7m/day,b=24ml
175 cm2 and is used to measure the hydraulic conductivity of a

I ' I
field sample 40 em long. If the permeameter discharges I 00 ml in
(a) (b) (c)
320 seconds under a head difference of 60 em: (a) what is the
hydraulic conductivity as measured in the laboratory? (b) what is
the intrinsic permeability of the sample given that water at 20°C
is used for the test? (c) is Darcy's law valid under the given con- 3.4.3 A confined aquifer has a constant thickness of 34 m and is
ditions (D = 0.5 rrun)? and (d) what would be the in-situ hydraulic stratified between two observation wells as shown in the follow-
conductivity if the regional groundwater temperature is l2°C? ing figure. Given the piezometric surface elevations in the figure,
3.3.2 If the medium sand in Examples 3.3.2 and 3.3.3 were sam- determine the flow per unit width of the aquifer. What is the
• pled and tested exactly the same way as described in Example

3.3.4, how long would it take for the head to drop from 8 em to 1
equivalent horizontal hydraulic conductivity of that section of
the confined aquifer between the observation wells?
em?
3.3.3 A 30-cm-long field sample is tested in a falling head per-
meameter with a 10-cm diameter cylinder. The diameter of the £+2.8m
ing thick- tube is 10 rom. The water level in the tube is 35 em above the out-
>quifer is let level after one hour of operation, while the level drops to 22
~zometric . ern after 12 hours of operation. Find the hydraulic conductivity of
'ectively. the sample.
3.3.41f the silty, fine sand sample in Example 3.3.4 were tested
- the same way as in Example 3.3.3, how long would it take to col-
lect 200 mJ of water at the outlet?
3.3.5 The average flow velocity at a point in an aquifer was found
to be 0.68 rnlday by means of a point dilution test. If the slope of
the piezometric surface at the same point is O.Ql8 and the aquifer
material has a porosity of 0.23, detennine the hydraulic conduc-
tivity of the aquifer at that point.
Problem 3.4.1 The stratification of the same aquifer at three locations is
>meter of shown below. Calculate the equivalent horizontal and vertical 3.5.1 The groundwater temperatures in the United States vary from
ten in the hydraulic conductivities for each location. What happens to the about 4oc in the northern part to approximately 20°C in the south-
degree of anisotropy? em part Assuming a productive alluvial aquifer with an intrinsic
penneability of 100 darcys and a hydraulic gradient of 0.01 (see 1957 was 30,000 m3 per day. Estimate the transmissivity of the II.Bou~
Figure 3.5.1), how much would the production of the same aquifer regional aquifer. Hydraulic
change between a northern state and a southern state? 1964.
3.10.1 Determine the soil water diffusivity for a soil in which 9 =
12. Brake
3.5.2 A confined aquifer is recharged from an unconfined aquifer 0. t and K = 3 X 1o-Il mm/s from a relationship of '1'(9) at e =0. 1' textured
through an aquitard as shown in the figure. If the water levels in 6'1' = 107 nun, and 69 =0.35. no. 2, pp.
piezometers A, B, and Care 4.31 m, 4.47 m, and 6.64 m below 13. Brook
3.12.1 Determine the infiltration rate and cumulative infiltration
the ground surface, respectively, detemtine the permeability of Hydrolog:
curves (0 to 5 h) at 1-hr increments for a clay loam soil. Assume an
the aquitard. 1964.
initial effective saturation of 40 percent and continuous ponding.
14. Brook
A B C Ground surface 3.12.2 Rework Problem 3.12.1 using an initial effective satura- fluid now,
tion of 20 percent. pp. 61--88
3.12.3 Rework Example 3. 12.1 for a sandy loam soil. 15. Burea•
480 pp., i·
3.12.4 Compute the ponding time and cumulative infiltration at 16. Campi
ponding for a sandy clay loam soil with a 30 percent initial effec- ductivity f
tive saturation, subject to a rainfall intensity of 2 ern/h. 1974.
25m 17. Castill
3.12.5 Rework Problem 3.12.4 for a silty clay soil.
Resources
3.12.6 Determine the cumulative infiltration and the infiltration 18. Cederg
rate on a sandy clay loam after one hour of rainfall at 2 cm!hr if Wiley &S
the initial effective saturation is 25 percent. Assume the pending 19. Charb<
depth is negligible in the calculations. transport, I
20. Charbe
3.12.7 Rework Problem 3.12.6, assuming that any ponded water
PrenticeH
remains stationary over the soil so that the ponded depth must be
21. Chow,
accounted for in the calculations. McGraw-H
3.6.1 The coordinates and the total heads of three observation 3.12.8 Use the Green-Ampt method to compute the infiltration 22. Cooper
wells in a regional aquifer are tabulated below. Determine the rate and cumulative infillration for a silty clay soil (11 = 0.479, instantane(
hydraulic gradient and direction of groundwater flow. 263-269, I
'I' =29.22 em, K = 0.05 cmlhr) at 0.25 hour increments up to
hours from the beginning of infiltration. Assume an initial effec- 23. Crowe,
Well# .x (m) Y (m) Total head (m)
Mechanics,
tive saturation of 30 percent and continuous ponding.
0 0 57.79 24. Darcy, I
3.12.9 Compute the ponding time and cumulative infiltration at Paris, 647 p
2 200 340 55.11
pending for a sandy clay loam soil of 25 percent initial yffective 25. Davis, S
3 190 -150 52.80
saturation for a rainfall intensity of (a) 2 crn/h, (b) 3 cmlh, and (c) Water Trace
5 em/h. 1985.
3.6.2 Suppose that the average discharge due to the local ground-
26. de Josse.
water pumping near Savannah, Georgia (see Figure 3.6.6), in 3.12.10 Rework Problem 3.12.9 considering a silt loam soil.
ular deposit:
27. Devaurs
infiltration e
REFERENCES pp. 19-27, I
1. Ahmed, N., and D. K. Sunada, Nonlinear flow in porous media, Jour. 6. Back, W., et al., Carbon-14 ages related to ocurrence of salt water, Jour. 28.Drost, W
Hydroulics Div.,Amer. Soc. Civil Engrs., v. 95, no. HY6, pp. 1847-1857, Hydroulics Div., Amer. Soc. Civil Engrs., v. 96, no. HY! l, pp. 2325-2336, flow by me:
1969. 1970. 125-146, 19
2. American Society of Agricultural Engineers, Advances in infiltration, 7. Barth, G. R., T. H. Illaogasekare, M. C. Hill, and H. Rajaram, A new 29. Eldor, N.
Proc. National Conf. on Advances in Infiltration, Chicago, IT..., ASAE Pub- tracer-density criterion for heterogeneous porous media, Water Resources porous medi
lication 11-83, St. Joseph, Ml, 1983. Research, v. 37, no. I, pp. 2!-3!, 2001. 30. Fair, G.
3. Amer. Soc. Testing Matis (ASTM), Permeability and capillarity of soils, 8. Bear, J., and G. Dagan, The relationship between solutions of flow stteamline f].
ASTM SpL Tech. Publ. no. 417, Philadelphia, 210 pp.. 1967. problems in isotropic and anisotropic soils, Jour. Hydrology, v. 3, pp. 25, pp. 1551·
4. Andrews, W. J., J. P Oliveros, and J. J. Collins, Preliminary Repon on 88-96. 1%5. 3!. Fancher, ·
rhe Hydrogeology of Lake Five-0 and Vicinity, Bay County, Florida, U.S. 9. Bennett, R. R. and R. R. Meyer, Geology and ground-water resources of Petr. Tech. , v.
Geological Survey Warer-Resource /nvesrigarions Repon 90-4148, 1990. the Baltimore area (Maryland), Maryland Department of Geology, Mines 32. Fenemor,
5. Bachmat, Y., and J. Bear, The general equations of hydrodynamic dis- and Warer Resources BuU. 4, 1952. Waimea Pia
persion in homogeneous, isotropic porous mediums, Jour. Geophysical 10. Boast, C. W., and D. Kirkham, Auger hole seepage theory, Soil Sci. Research, H}
Research, v. 69, pp. 2561-2567, 1964. Soc. Amer. Proc., v. 35, pp. 365-373, 1971. Zealand, 198
References 143
1 of the 11. Bouwer, H., Unsaturated flow in ground-water hydraulics, Jou1: 33. Field, M. S., Ground· Water Tracing and Drainage Basin Delineation
Hydraulics Div., Amer. Soc. Civil Engrs., v. 90, no. HY5, pp. 121-144, for Risk Assessment Mapping for Spring Protection in Clarke County,
!964. Virginia, U.S. Environmental Protection Agency NCEA· W-0936, 2000.
tich e= 12. Brakensiek, D. L., R. L. Engleman, and W.J. Rawls, Variation within 34. Field, M. S., Efficient hydrologic tracer-test design for tracer-mass
8 =0.1, texture classes of soil water parameters. Trans. Am Soc. Agric Eng., v. 24, estimation and sample-collection frequency, I. Method development,
no. 2, pp. 335-339, 1981. Environmental Geology, v. 42. no. 7, pp. 827-838,2001.
13. Brooks. R.H., and A. T. Corey, Hydraulic properties of porous media, 35. Field, M. S., Efficient hydrologic tracer-test design for tracer-mass
iltration
Hydrology Papers, no. 3. Colorado State University, Fort Collins, CO, estimation and sample-collection frequency, 2. Experimental results,
;umean Environmental Geology, v. 42, no. 7, pp. 839-850,2001.
1964.
nding. 36. Field, M., Tracer-test planning using the efficient hydrologic tracer-
14. Brooks, R. H., and A. T. Corey, Properties of porous media affecting
satura- fluid flow, Jour. b·rig. Drain. Div., Amer. Soc. Civil Engrs., v. 92, no.IR2, test/design program, EPA/600/R-03/034, National Center for Environ-
pp. 61-88, 1966. mental Assessment-Washington Office, U.S. Environmental Protection
!5. Bureau of Reclamation, Ground Water Manual, U.S. Dept. Interior, Agency, Washington, DC, 2003.
480 pp., 1977. 37. Freeze, R. A., Theoretical analysis ofregional groundwater flow, Sci-
ation at 16. Campbell, G. S., A simple method for determining unsaturated con- entific Ser. no. 3, Inland Waters Branch, Dept. Energy, Mines and
ductivity from moisture retention data, Soil Science, v. 117, pp. 311-314, Resources, Ottawa, Canada, 202 pp., 1969.
1! effec-
1974. 38. Freeze, R. A., Three-dimensional, transient, saturated-unsaturated
17. Castillo, E., et a!., Unconfined flow through jointed rock, Water flow in a groundwater basin, Water Resources Research, v. 7, pp.
Resources Bull., v. 8, pp. 266-281, 1972. 347-366, 1971.
iltration 18. Cedergren, H. R., Seepage, Drainage, and Flow Nets, 2nd ed., John 39. Freeze, R. A., and J. Banner, The mechanism of natural ground-water
Wiley & Sons, New York, 534 pp., 1977. recharge and discharge, Water Resources Research, v. 5, pp. 153-171,
cmfhr if
1969; v. 6, pp. 138-155, 1970.
ponding 19. Charbeneau, R. J., Kinematic models for soil moisture and solute
40. Gaspar, E., Modem Trends in Tracer Hydrology, CRC Press, Boca
transport, Water Resources Research, 17(3), pp. 699-706, 1984.
Raton, FL, !987.
20. Charbeneau, R. J., Groundwater Hydraulics and Pollutant Transpon,
~d water 41. Gaspar, E., and M. Oncescu.,Radioactive Tracers in Hydrology, Else-
Prentice Hall, Upper Saddle River, NJ, 2000.
must be vier, Amsterdam, 342 pp., 1972.
21. Chow, V. T., D. R. Maidment, and L. W. Mays, Applied Hydrology,
42. Gelhar, L. W., and M.A. Collins, General analysis of longitudinal dis-
McGraw-Hill, New York. 1988.
persion in nonuniform flow, Water Resources Research, v. 7, pp.
iltration 22. Cooper, H. H., Jr., et al., Response of a finite-diameter well to an 1511-1521, 1971.
=0.479, instantaneous charge of water, Water Resources Research, v. 3, pp.
43. Green, W.H., and G. A. Ampt, Studies on soil physics, Part I, The
263-269, 1967.
tts up to flow of air and water through soils, J. Agric. Sci., v. 4, no. 1, pp. 1-24,
23. Crowe, C. T., D. F. Elger, and J. A. Roberson, Engineering Fluid 1911.
a! effec-
Mechanics, 7th edition, John Wiley & Sons, New York, 2001.
44. Grubbs, J. W., Evaluation of Ground-Water Flow and Hydrologic
24. Darcy, H., Les fontaines publiques de Ia ville de Dijon, V. Dalmont, Budget for Lake Five-0, A Seepage Lake in Northwestern Florida, U.S.
ration at Paris, 647 pp., 1856. Geological Survey Water-Resources Investigations Report 94-4145,1995.
~ffective 25. Davis, S. N., D. J. Campbell, H. W., Bentley, and T. J. Flynn, Ground 45. Halevy, E., et al., Borehole dilution techniques: A critical review, Iso-
t, and (c) Water Tracers, National Ground Water Association, Dublin, OH, 200 pp., topes in Hydrology, Intl. Atomic Energy Agency, Vienna, pp. 531-564,
1985. 1967.
26. de Josselin de Jong, G., Longitudinal and transverse diffusion in gran- 46. Hanshaw, B. B., et al., Carbonate Equilibria and Radiocarbon Dis-
soil. ular deposits, Trans. Amer. Geophysical Union, v. 39, pp. 67-74, 1958. tribution Related to Groundwater Flow in the Floridan Limestone
27. Devaurs, M., and G. F. Gifford, Applicability of the Green and Ampt Aquifer, U.S.A., Inti. Assoc. Sci. Hydrology Pub!. 74, pp. 601-{)14, 1965.
infiltration equation to rangelands, Wa1er Resource Bulletin, v. 22, no.1, 47. Harleman, D. R. F., et al., Dispersion-permeability correlation in
pp. 19-27. 1986. porous media, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 89, no.
ater,Jour. 28. Drost, W., et al., Point dilution methods of investigating ground water HY2, pp. 67-85, 1963.
i25-2336, flow by means of radioisotopes, Water Resources Research, v. 4, pp. 48. Hollett, K. J., Geohydrology and Water Resources of the Papago
125-146, 1968. Fanns--Great Plain Area, Papago lnd.icll Reservation, Arizona and the
m,A new 29. Eldor, M., and G. Dagan, Solutions of hydrodynamic dispersion in Upper Rio Sonoyta Area, Sonoram Mexico, U.S. Geological Survey Water
Resources porous media, Water Resources Research, v. 8, pp. 1316-1331, 1972. Supply Paper 2258, 1985.
30. Fair, G. M., and L. P. Hatch, Fundamental factors governing the 49. Hoopes, J. A., and D. R. F. Harleman, Dispersion in radial flow from
ts of flow streamline flow of water through sand, Jour. Amer. Water Works Assoc., v. a recharge well, Jour. Geophysical Research, v. 72, pp. 3595-3607, 1967.
'v. 3, pp. 25,pp. 1551-1565, 1933. 50. Horton, R. E., The role of infiltration in the hydrologic cycle, Trans.
31. Fancher, G., Henry Darcy: Engineer and benefactor of mankind, Jour. Am Geophysical Union, v. 14, pp. 446-460, 1933.
sources of Petr. Tech. , v. 8, pp. 12-14, Oct. 1956. 51. Hubbert, M. K.• The theory of ground-water motion, Jour. Geol. , v.
>gy, Mines 32. Fenemor, A. D., A three-dimensional model for management of the 48,pp. 785-944, 1940.
Waimea Plains Aquifer, Nelson. Dept. of Scientific and Industrial 52. Hubbert, M. K., Darcy's Jaw and the field equations of the flow of
'• Soil Sci. Research, Hydrology Centre Publication No. 18, 33. Christchurch, New underground fluids, Trans. Amer. Inst. Min. and MetaL Engrs., v. 207, pp.
Zealand, 1988. 222-239, 1956.
53. Intl. Assoc. Hydraulic Research (IAHR), Fundamentals ofTransport 74. Morel-Seytoux, H. J., Multiphase flow in porous media, in Develop- 93. Stalin
Phenomena in Porous Media, Elsevier, Amsterdam, 392 pp., 1972. ments in Hydraulic Engineering 4, P. Novak, ed., Ch. 3, Elsevier, New science, \
54. Inti. Atomic Energy Agency, Working Group on Nuclear Techniques York, 1987. York, 196
in Hydrology, Guidebook on Nuclear Techniques in Hydrology, Tech. 75. Morris, D. A., and A. I. Johnson, Summary of Hydrologic and Physical 94. Stout,
Rept. Ser. no. 91, Vienna, 214 pp., 1968. Properties of Rock and Soil Materials, as Analyzed by the Hydrologic Lab- physical l
55. Irmay, S., On the hydraulic conductivity of unsaturated soils, Trans. oratory of the U.S. Geological Survey 1948--60, U.S. Geological Survey 1967.
Amer. Geophysical Union, v. 35, pp. 463-467, 1954. Water-Supply Paper 1839-D, 42 pp., 1967.
95. Todd,
56. Jacob, C. E., Flow of ground water, Engineering Hydraulics, H. 76. Mull, D. S., J. L. Smoot, and T. D. Liebermann, Dye Tracing Tech- media,!OJ
Rouse, ed., John Wiley and Sons, New York, pp. 321-386, 1950. niques Used to Determine Ground-Water Flow in a Carbonate Aquifer 31-57, 19
57. Kass, W., Tracing Technique in Geohydrology, A. A. Balkema, Rot- System Near Elizabethtown, Kentucky, U.S. Geological Survey Water- 96. Todor1
terdam, Netherlands, 1998. Resource Investigations Report WRI87-4174, Washington, DC, 1988. media, We
58. Kaufman, W. J., and D. K. Todd, Application of tritium tracer to canal 77. Novakowski, K. S., The analysis of tracer experiments conducted in 97. Toth,.
seepage measurements, Tritium in the Physical and Biological Sciences, divergent radial flow fields, Water Resources Research, v. 28, no. 12, pp. Central f
Inti. Atomic Energy Agency, Vienna, pp. 83-94, 1962. 3215-3225, 1992. 4375-438
59. Keeley, J. W., and M. R. Scalf, Aquifer storage determination by 78. Pearson, F. J., Jr., and D. E. White, Carbon 14 ages and flow rates of 98. Toth, J
radio-tracer techniques, GroundWater, v. 7, pp. 17-22, 1969. water in Carrizo Sand, Atascosa County, Texas, Water Resources basins, Jo;
Research, v. 3,pp.251-261, 1967.
60. Kilpatrick, F. A., and E. D. Cobb, Measurement of discharge using 99. U.S.<
tracers, U.S. Geological Survey Techniques ofWater Resources Investi- 79. Rasmussen, W. C., Permeability and storage of heterogeneous GeologictA
gations, Book 3, Ch. A16, 1985. aquifers in the United States, Inti. Assoc. Sci. Hydrology Pub/. 64, pp.
100. Yael
317-325, 1964.
61. Kirkham, D., and W. L. Powers, Advanced Soil Physics, John Wiley unsaturate
& Sons, New York, 534 pp., 1972. 80. Rawls, W. J., L. R. Ahuja, D. L. Brakensiek, and A. Shirmohammadi, Research,
Infiltration and soil water movement, Handbook of Hydrology, D. R.
62. Krumbein, W. C., and G. D. Monk, Permeability as a function of the 101. Vach
Maidment, ed., Ch. 5, McGraw-Hill, New York, 1993.
size parameters of unconsolidated sand, Trans. Amer. Inst. Min. and Met. redistribu
Engrs., v. 151, p. 153-163, 1943. 81. Rawls, W. J., D. L. Brakensiek, and N. Miller, Green-Ampt infiltra- Resources
tion parameters from soils data, J. Hydraulic Div., ASCE, v. 109, No.I,
63. LeGrand, H. E., and V. T. Stringfield, Development and distribution
pp. 62-70, 1983.
of permeability in carbonate aquifers, Water Resources Research, v. 7, pp.
1284-1294, 1971. 82. Rice, W. A., and S.M. Gorelick, Geological inference from "flow
net" transmissivity determination: Three case studies, Water Resources
64. Lenhard, R. J., J. C. Parker, and S. Mishra, On the correspondence
Bulletin, v. 21, no. 6, pp. 919-930, 1985.
between Brooks-Corey and Van Genuchten Models, Journal Irrigation
Drainage Engineering, 15(4), pp. 744-751, 1989. 83. Richards, L.A., Capillary conduction of liquids through porous medi~
urns, Physics, v. I, pp. 318-333, 1931.
65. Li, W. H., and G. T. Yeh, Dispersion at the interface of miscible
liquids in a soil, Water Resources Research, v. 4, pp. 369-378, 1968. 84. Rumer, R. R., Longitudinal dispersion in steady and unsteady flow,
Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 88, no. HY 4, pp.
66. Lohman, S. W., Ground-Water Hydraulics, U.S. Geological Survey
147-172, 1962.
Prof Paper 708, 70 pp., 1972.
67. Luthin, J. N., ed., Drainage of agricultural lands, Agronomy Mono- 85. Rumer, R. R., and P. A. Drinker, Resistance to laminar flow through
graph no. 7, Amer. Soc. Agronomy, Madison, WI, 620 pp., 1957. porous media, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 92, no.
HY5, pp. 155-163, 1966.
68. Masch, F. D., and K. J. Denny, Grain size distribution and its effect on
the permeability of unconsolidated sands, Water Resources Research, v. 86. Scott, D. M., An evaluation of flow net analysis for aquifer identifi-
2,pp.665-677, 1966. cation, Ground Water, v. 30, no. 5, pp. 755-764, 1992.
69. Mays, L. W., Water Resources Engineering, John Wiley & Sons, New 87. Scott, D. M. and H. R. Thorpe, Ground-water resources between the
York, 2001. Rakaia and Ashburton Rivers, Ministry of Works and Development for
the National Water and Soil Conservation Authority, Hydrology Centre
70. Meigs, L. C., and R. L. Beauheim, Tracer tests in a fractured
Publication No.6, Christchurch, New Zealand, 1986.
dolomite, 1. Experimental design and observed tracer recoveries, Water
Resources Research, v. 37, no. 5, pp. 1113-1128, 2001. 88. Shahbazi, M., et al., Effect of topography on ground water flow, Inti.
Assoc. Sci. Hydrology Publ. 77, pp. 314-319, 1968. ~.
71. Mobasheri, F., and M. Shahbazi, Steady-state lateral movement of
water through the unsaturated zone of an unconfined aquifer, Ground 89. Shamir, U. Y., and D. R. F. Harleman, Numerical solutions for dis- }
Water, v. 7, no. 6, pp. 28-34, 1969. persion in porous mediums, Water Resources Research, v. 3, pp.
557-581, 1967.
72. Molz, F. J., 0. Guven, J. G. Melville, R. D. Crocker, and K. T.
Matteson, Performance, analysis, and simulation of a two-well tracer test 90. Skibitzke, H. E., and G. M. Robinson, Dispersion in Ground Water
at the Mobile site, Water Resources Research, v. 22, no. 7, pp. Flowing through Heterogeneous Materials, U.S. Geological Survey Prof
1031-1037, 1986. Paper 386-B, 3 pp., 1963.
73. Molz, F. J., 0. Guven, J. G. Melville, R. D. Crocker, and K. T. Matte- 91. Smith, W. 0., Infiltration in sands and its relation to groundwater
son, Peifonnance and Analysis of Aquifer Tracer Tests with Implications recharge, Water Resources Research, v. 3, pp. 539-555, 1967.
for Contaminant Transport Modeling, Technical Report EPA/600/2- 92. Smith, W. 0., and A. N. Sayre, Turbulence in Ground-Water Flow,
86/062, U.S. Environmental Protection Agency, Washington, DC, 1986. U.S. Geological Survey Prof Paper402-E, 9 pp., 1964.
References 145
1 Develop- 93. Stallman, R. W., Flow in the zone of aeration, Advances in Hydro- 102. Van Genuchten, M. Th., A closed-form equation for predicting the
:vier, New science, v. 4, in Chow, V. T., ed., pp. 151-195, Academic Press, New hydraulic conductivity of unsaturated soils, Soil Science Society Am. 1., v.
York, !967. 44, pp 892-898, 1980.
dPhysical 94. Stout, G. E., ed., Isotope Techniques in the Hydrologic Cycle, Geo- 103. van Schilfgaarde, J. ed., Drainage for Agriculture, Agronomy
•logic Lab- physical Monograph Ser. no. 11, Amer. Geophysical Union, 199 pp., Monograph no. 17, Amer. Soc. Agronomy, Madison, Wisconsin, 700 pp.,
cal Survey 1967. 1974.
95. Todd, D. K., and J. Bear, Seepage through layered anisotropic porous 104. van Tonder, G. J., A computer code for the calculation of the relative
cing Tech- media, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 87, no. HY3, pp. transmissivity distribution in an aquifer for steady state ground-water lev-
!te Aquifer 31-57, 1961. els, Water SA, v. 15, no. 3, pp. 147-152, 1989.
1ey Water- 96. Todorovic, P., A stochastic model of longitudinal diffusion in porous 105. Vogel, J. C., Carbon-14 dating of groundwater, Isotope Hydrology
:, 1988. media, Water Resources Research, v. 6, pp. 211-222, 1970. 1970, Inti. Atomic Energy Agency, Vienna, pp. 225-239, 1970.
nducted in 97. Toth, J., A theory of groundwater motion in small drainage basins in 106. Ward, J., Turbulent flow in porous media, Jour. Hydraulics Div.,
no. 12, pp. Central Alberta, Canada, Jour. Geophysical Research, v. 67, pp. Amer. Soc. Civil Engrs., v. 90, no. HY5, pp. 1-12, 1964.
4375-4387, 1962. 107. Wenzel, L. K., Methods for Determining Permeability of Water-
>w rates of 98. Toth, J., A theoretical analysis of groundwater flow in small drainage Bearing Materials with Special Reference to Discharging- Well Methods,
Resources basins, Jour. Geophysical Research, v. 68, pp. 4795-4812, 1963. U.S. Geological Survey Water-Supply Paper 887, 192 pp., 1942.
99. U.S. Geological Survey, Fluid Movement in Earth Materials, U.S. 108. Wiebenga, W. A., et aJ., Radioisotopes as groundwater tracers, Jour.
rogeneous Geological Survey Professional Paper 411, Chaps. A to I, 1961-1970. Geophysical Research, v. 72, pp. 4081-4091, 1967.
bl. 64, pp. 100. Vachaud, G., Determination of the hydraulic conductivity of 109. Zaluski, M., Tracer Test, Manual of Applied Field Hydrogeology,
unsaturated soils from an analysis of transient flow data, Water Resources W. D. Weight and J. L. Sonderegger, eds., Ch. 13, McGraw-Hill, New
>hammadi, Research, v. 3,pp.697-705, 1967. York, 2001.
?gy, D. R. 101. Vachaud, G., and J. L. Thony, Hysteresis during infiltration and
redistribution in a soil column at different water contents, Water
tpt infiltra- Resources Research, v. 7, pp. 111-127, 1971.
109, No.1,
·rom "flow
Resources
rous medi-
:eady flow,
HY4, pp.
>w through
, v. 92, no.
er identifi-
etween the
)pment for
)gy Centre
r flow, lntl.
ms for dis-
v. 3, pp.
JUnd Water
urvey Prof.
:oundwater
vater Flow,
Chapter 4
Groundwater and Well Hydraulics
Darcy's law and the fundamental equations governing groundwater movement can now be
applied to particular situations. Solutions of groundwater flow to wells rank highest in impor-
tance. From pumping tests of wells, storage coefficients and transmissivities of aquifers can be
detennined; furthermore, with these aquifer characteristics known, future declines in ground-
EXAM
water levels associated with pumpage can be calculated. Well flow equations have been devel-
oped for steady and unsteady flows, for various types of aquifers, and for several special
boundary conditions. For practical application, most solutions have been reduced to conven-
ient graphic or mathematical form, or computer programs.
SOLUTj
4.1 STEADY UNIDIRECTIONAL FLOW

Steady flow implies that no change occurs with time. Flow conditions differ for confined and
unconfined~quifers and hence need to be considered separately, beginning with flow in one
direction.
4.1.1 Confined Aquifer 4.1.2 l

Let groundwater flow with a velocity v in the x-direction of a confined aquifer of uniform
thickness. Then for one-dimensional, steady flow, Equation 3,9.10 reduces to
o2h
-=0 (4.1.1)
ox
2
which has for its solution

(4.1.2)
where his the head above a given datum and C1 and C2 are constants of integration. Assum-
=
ing h 0 when X= 0 and oh/Ox =-( v!K) from Darcy's law, then we have
h=- vx (4.1.3)
K
This states that the head decreases linearly, as sketched in Figure 4.1.1, with flow in the
x-direction.
146
Ground surface
I
-
------
~ --- -- --- Is<:.
Piezometric
surface
-- --f~ -- l2
- - -- -
cs h
GC::
Impermeable
~1~
I~
!J
=
t:-...-......
-·
Confined
~
=
v
aquifer I
"" Figure 4.1.1. Steady unidirectional
t now be ---x flow in a confined aquifer of uni-
n impor- Impermeable form thickness.
rs can be
1 ground- Referring to Figure 4.U, if the distance and the observed piezometric surface drop between two adja-
EXAMPLE 4.1.1
~n devel- cent wells are 1,000 m and 3 m, respectively, find an estimate of the time it takes for a molecule of water
.1 special to move from one well to the other. Assume steady unidirectional flow in a homogeneous silty sand con-
, conven- fined aquifer with a hydraulic conductivity K = 3.5 m/day and an effective porosity of 0.35.
SOLUTION First compute the discharge velocity:

hK {-3m)(3 .5rnld)
v=--=- =0.ol05m/d
x (lOOOm)
The pore (seepage) velocity is computed using the velocity:
fined and vp =vine= (0.0105 m/d)/(0.35) = 0.03 mid
•W m one
It would take 1000 m/(0.03 mid x 365) => 91.3 years.
•
'uniform For the similar flow situation in an unconfined aquifer, direct analytic solution of the Laplace
equation is not possible. The difficulty arises from the fact that the water table in the two-
dimensional case represents a flow line. The shape of the water table determines the flow dis-
tribution, but at the same time the flow distribution governs the water table shape. To obtain a
(4.1.1)
solution, Dupuit 13 assumed (1) the velocity of the flow to be proportional to the tangent of the
hydraulic gradient instead of the sine as defined in Darcy's law and (2) the flow to be hori-
zontal and uniform everywhere in a vertical section. These assumptions, although permitting
(4.1.2) a solution to be obtained, limit the application of the results. For unidirectional flow, as
sketched in Figure 4.1.2, the discharge per unit width q at any vertica1 section can be given as
1. Assum- dh
q= -Kh dx (4.1.4)
(4.1.3) where K is hydraulic conductivity, h is tbe height of the water table above an impervious base,
and xis the direction of flow. Integrating yields
>Win the K z
qx=--h +C (4.1.5)
2
148 Chapter 4 Groundwater and Well Hydraulics
and, if h =~ where x = 0, then the Dupuit equation
(4.1.6)
SOLU)
results, which indicates that the water table is parabolic in form.
For flow between two fixed bodies of water of constant heads h0 and h1 as in Figure 4.1.2,
the water table slope at the upstream boundary of the aquifer (neglecting the capillary zone) is
dh q
-=--- (4.1.7)
dx Kho
But the boundary h =h0 is an equipotential line because the fluid potential in a water body is 4.1.3
constant; consequently, the water table must be horizontal at th.is section, which is inconsistent T
with Equation 4.1. 7. In the direction of the flow, the parabolic water table described by Equa-
tion 4.1.6 increases in slope. By so doing, the two Dupuit assumptions, previously stated,
become increasingly poor approximations to the actual flow; therefore, the actual water table
deviates more and more from the computed position in the direction of flow, as indicated in Fig-
ure 4.l.2. The fact that the actual water table lies above the computed one can be explained by
the fact that the Dupuit flows are all assumed horizontal, whereas the actual velocities of the ·~
same magnitude have a downward vertical component so that a greater saturated thickness is 1
required for the same discharge. At the downstream boundary a discontinuity in flow forms •
because no consistent flow pattern can connect a water table directly to a downstream freewater
surface. The water table actually approaches the boundary tangentially above the water body
surface and fonns a seepage face.
The above discrepancies indicate that the water table does not follow the parabolic form
of Equation 4.1.6; nevertheless, for flat slopes, where the sine and tangent are nearly equal, it
closely predicts the water table position except near the outflow. The equation, however, accu-
rately determines q or K for given boundary heads.41
A stratum of dean '"'d aod gravel between two oh'""'~

(soe Figure 4.1.2) bas • hydrnulio oonduotivity
K == w-1 em/sec, and is supplied with water from a ditch (h0 = 6.5 m deep) that penetrates to the bottom 1Ij
of the stratum. If the water surface in the second channel is 4 m above the bottom of the stratum and its
Ground surface
Actual water table

Computed
water table
---=:.:
________..
--
. . -~-- --........
.........._ .... ...........
"''IF:....,---l'--
1 Assumed . - . --...
velocity --
distribution - h Actual
--
- - - velocity
distribution
Figure 4.1.2. Steady flow in an unconfined

---•x aquifer between two water bodies with vertical
Impermeable boundaries.
distance to the ditch is x = !50 m (which is also the thickness of the stratum), estimate the unit flow rate
into the gallery.
(4.1.6)
SOLUTION The flow is computed using the Dupuit equation (4.1.6) for unit flow, where
K = 10- 1 em/sec= 86.4 m/day
1re 4.1.2,
• zone) is q =!£(ho2 -h2) =86t m/d)y (6.52 -42)m2 = 7.56 m2/day
(4.1.7)
2x 2150 m
•
r body is 4.1.3 Base Flow to a Stream
onsistent
Estimates of the base flow to streams (see Chapter 6) or average groundwater recharge can be
by Equa-
computed by applying the above analysis of one-directional flow in an unconfined aquifer. For
ly stated,
example, picture the idealized boundaries shown in Figure 4.1.3 of two long parallel streams
ater table
completely penetrating an unconfined aquifer with a continuous recharge rate Woccurring uni-
~dinFig
formly over the aquifer. With the Dupuit assumptions, the flow per unit thickness is
lained by
ies of the dh
q:-Kh dx (4.1.8)
ck:ness is
JW fonns
and by continuity
freewater
q=Wx (4.1.9)
ater body
Combining these equations and integrating leads to the result
olic form
r equal, it h
2
=h/+; (a 2 -.x 2 ) (4.1.10)
ver, accu-
where h, ha, a, and x are as defined in Figure 4.1.3, and K is the hydraulic conductivity. From
synunetry and continuity
'nductivity (4.1.11)
the bottom
where Qb is the base flow entering each stream per unit length of stream channel. If h is
urn and its
known at any point, Qb or W can be computed provided K is known.
Extensions of this analysis have been applied to design the spacing of parallel drains on
agricultural land for specified soil, crop, and irrigation conditions.
Recharge rate W
Unconfined
aquifer Figure 4.1.3. Steady flow
to two parallel streams
fined from a uniformly
vertical recharged unconfined
aquifer.
Figure 4.1.4. Unconfined aquifer between two

rivers (x;:::: 0, h "' h1 and x = L, h = h2).
EXAA
Consider the unconfined aquifer between two rivers as shown in Figure 4.1.4 with
recharge rate of W. The flow is only in one direction so that the x-axis is aligned parallel to the
flow. The flow is then determined by
!( :)=-W
Kh (4.1.12)
SOLUi
or
(4.1.13)
Integration of Equation 4.1.13 yields
(4.1.14)
where c 1 and c2 are constants of integration. Boundary conditions (h =h1 at x =0 and h =h2 at
x =L) are applied to obtain
(4.1.15)
The discharge qx per unit width at any distance from the origin (see Figure 4.1.4) can be com·
puted using Equation 4.1.8 with dh/dx computed by differentiating Equation 4.1.15:
q =K(ht2 - hz2) - w( L - x) (4.1.16)

X 2L 2
where the units are ft2/day or m2/day for qx, ft/day or rnlday forK and W, and ft or m for x, h1, .
hz, andL. ·
Figure 4.1.4 shows the location where h =hmax (a crest in the water table), for the case of
infiltration, which is essentially a water divide where qx =0. The distanced from the origin to
the water divi& is computed using Equation 4.1.16 with qx = 0 and x = d to obtain
(4.1.17)
At x = d , h = hmax which can then be substituted into Equation 4.1.15 to obtain the following
expression for ~ax:
(h12 ,_ 2)d
h2 =h12 - -,'2 + W(L-d)d (4.1.18)
max L K
W =1.6 m/year
Figure 4.1.5.
Impermeable
1 two Example 4.1.3
[EXAMPLE4.1.3 An unconfined aquifer of clean sand and gravel is located between two fully penetrating rivers (see Fig-
ure 4.1.5) and has a hydraulic conductivity of K = 10-2 em/sec. The aquifer is subject to a unifonn
1.4 with
recharge of 1.6 m/year. The water surface elevations in rivers A and Bare 8.5 m and 10m, respectively,
lel to the
above the bottom. Estimate (a) the maximum elevation of the water table and the location of groundwater
divide, (b) the travel times from groundwater divide to both rivers (n, = 0.35), and (c) the daily discharge
per kilometer from the aquifer into both rivers.
(4.1.12)
SOLUTION (a) The maximum elevation of the water table occurs at the location of the groundwater divide computed
using Equation 4.1.17 with W = 1.6 m/year = 0.0044 m/day and K = 1o-2 cm/s = 8.64 m/day:
(4.1.13) d"' ~-,! (hl2 -~ 2) =460 m- 8.64 m/d (102 -8.52 )m2 = 17l m from river B
2 W 2L 2 0.0044 mid 2(460 m)
The maximum head at the divide is computed using Equation 4.1.18:

(4.1.14)
~ax=
(4.1.15) (b) The average pore velocity is computed using Darcy's law with the Dupuit assumptions:
n be com- VA= (.!J(M)

n, I:Jx
= (8.64 m/d )( 10.7 -8.5] m =O.l 90 m/day
0.35 460-171 m
So the travel time from the groundwater divide to river A is

(4.1.16)
L 460m-171 m
t =.....!!.. = = 1524 days= 4.18 years
vA 0.190 m/day
n for x, h1,
Similarly, the travel time from the groundwater divide to river B is computed as
the case of
1e origin to vo =(K)(M) = (8.64 mid J(l0.7 -lOJ =O.lOl m/day
n ~ 0.35 171
L 171m
t = ~= = 1692 days= 4.64 years
vA 0.104 rnlday
(4.1.17)
(c) From Equation 4.1.16, for x = 0:
~following
(4.1.18)
The minus sign occurs due to opposite flow direction to the x-axis (see Figure 4.1.5). So, (0.751 x
1000 m) =751 m3/day is the daily discharge from the aquifer per kilometer into river B.
Similarly, for x =460 m:
The daily discharge from the aquifer per kilometer into river A is (1.27 x 1000 m) =1270 m3/day.
•
4.2 STEADY RADIAL FLOW TO A WELL
When a well is pumped, water is removed from the aquifer surrounding the well, and the water
table or piezometric surface, depending on the type of aquifer, is lowered. The drawdown at a
given point is the distance the water level is lowered. A drawdown curve (or cone) shows the
variation of drawdown with distance from the well (see Figure 4.2.1). In three dimensions, the
draw down curve describes a conic shape known as the cone of depression, as shown in Figure
4.2.2. Also, the outer limit of the cone of depression (zero draw down) defines the area of influ-
ence of the well.
4.2.1 Confined Aquifer

To derive the radial flow equation (which relates the well discharge to drawdown) for a well ·~
completely penetrating a confined aquifer, referring to Figure 4.2.1 will prove helpful. The ~j:
flow is assumed two-dimensional to a well centered on a circular island and penetrating a ~
homogeneous and isotropic aquifer. Because the flow is everywhere horizontal, the Dupuit I
1:
I
assumptions apply without error. Using plane polar coordinates with the well as the origin, we
obtain the well discharge Q at any distance r as I
dh
i
I
Q = Av = -2rtrbK- (4.2.1)
dr
Ground surface Q
llllii:.G;;~ZI:i~• t Original piezometric

--L ~uif~c~-- -·~1--"'i=--------''
r~
curve
I ll
Figure 4.2.1. Steady
radial flow to a well
penetrating a confined
Impermeable aquifer on an island.
.0.75 1 X for steady radial flow to the well. Rearranging and integrating 4.2.1 for the boundary condi-
tions at the well, h = h,. and r =rw• and at the edge of the island, h =h0 and r =r0 , yield
ho -h Q
=--ln-
ro
(4.2.2)
w 2rtKb rw
or
~-h
Q =2rtKb "' (4.2.3)
ln(r0 /rw)
with the negative sign neglected.

In the more general case of a well penetrating an extensive confined aquifer, as in Figure
4.2.2, there is no external limit for r. From the above derivation at any given value of r,
Q=2rtKb h-~ (4.2.4)

1n(rlrw)
1e water
>wn at a which shows that h increases indefinitely with increasing r. Yet, the maximum his the initial
ows the uniform head h0 . Thus, from a theoretical aspect, steady radial flow in an extensive aquifer
ons, the does not exist because the cone of depression must expand indefinitely with time. However,
1 Figure from a practical standpoint, h approaches h0 with distance from the well, and the drawdown
ofinflu- varies with the logarithm of the distance from the well.
The flow net in Figure 4.2.3 illustrates the distribution of flow in a confined aquifer for a
fully penetrating well and a lOO percent open hole. Figure 4.2.4 illustrates the flow distribution
to a discharging well in a confined aquifer. The well is a fully penetrating, 100 percent open
hole. Figure 4.2.5 illustrates the flow net for a well that penetrates 50 percent of the confined
•r a well aquifer with an open hole. The flow net in Figure 4.2.6 illustrates the distribution of flow in a
ful. The confined aquifer for a well that penetrates through the upper confining bed but not into the
rating a artesian aquifer.
:Dupuit Equation 4.2.4, known as the equilibrium, or Thiem, 60 equation, enables the hydraulic
i.gin, we conductivity or the transmissivity of a confined aquifer to be determined from a pumped well
Ground surface
(4.2.1) 0
/· Original piezometric surface
I
t j ____
------- r--------L__ 51
_1)2_----------
f..
T\ PI
'--orawdown
curve (cone)
~ h
1- ~ /
-~ Observation
wells
I
hw '1
II
h1
I
h2
I '2 I
1- I
4- 2rw l Impermeable
--
; -- 3
l
- -
Confined
Steady aquifer
a well
confined
li5i ! Figure 4.2.2. Radial flow to a well penetrating
island. Impermeable an extensive confined aquifer.
Static piezometric surface

------------------
---- ---
Drawdown cone (curve)
----
-----
r
Upper confining bed
--~ ~--
Artesian
--~ ~-- aquifer
Artesian -- ~
aquifer --~
~--
~--
L
l
--~ ~--
--~ ~-- Figure 4.2.3. Distribution of flow to a discharging

well in an artesian aquifer-a fully penetrating and
100 percent open hole (from the U.S. Bureau of
Lower confining bed Reclamation 62 ).
that fully penetrates the aquifer. Because any two points define the logarithmic drawdown
curve, the method consists of measuring drawdowns in two observation wells at different
distances from a well pumped at a constant rate. Theoretically, hw at the pumped well can serve
as one measurement point; however, well losses caused by flow through the well screen and
inside the well introduce errors so that hw should be avoided. The transmissivity is given by
T= Kb = Q ln ' 2 (4.2.5)
2n(~- ht) r1
r:_ r w
Drawdown
!static piezometric surface
b = Thickness of aquifer
h 0 = Undisturbed artesian head
h 1 , h2 =Undisturbed artesian heads at
r2 respectively when well
is discharging
s 1 ,s 2 = Drawdown at r 1 and
r2 respectively when well
is discharging
Ow= 01 = 02
Aw = 27Trwb
b A1 = 21rr1 b
A2 =21rr2 b
Vw = OwiAw
T --
Artesian
V1 = 01/A1 aquifer --
V2 = 02IA2 Figure 4.2.4. Flow
distribution to a
discharging well in an
artesian aquifer-a
l--•
Lo
fully penetrating and
100 percent open hole
(from the U.S. Bureau
of Reclamation62 ).

--- ~~--=--~-....:.._. _.:...._-.- .- ·
·-·--..
r- Figure 4.2.5.
Distribution of flow to a
well in an artesian
aquifer-a 50-percent
Lower confining bed penetrating and open
:harging - - +-- Flow lines hole (from U.S. Bureau
1ti.ng and - - - Equipotential lines of Reclarnation62).
au of
where r 1 and r 2 are the distances and h1 and h2 are the heads of the respective observation
wells.
rawdown
From a practical standpoint, the drawdown s rather than the head h is measured so that
different
Equation 4.2.5 can be rewritten
:an serve
:reen and
iven by (4.2.6)
(4.2.5) where s 1 and s2 are shown in Figure 4.2.2. In order to apply Equation 4.2.6, pumping must con-
tinue at a uniform rate for a sufficient time to approach a steady-state condition-that is, one
in which the drawdown changes negligibly with time.• The observation wells should be located
close enough to the pumping well so that their drawdowns are appreciable and can be readily
measured. The derivation assumes that the aquifer is homogeneous and isotropic, is of uniform
thickness, and is of infinite areal extent; that the well penetrates the entire aquifer; and that ini-
tially the piezometric surface is nearly horizontal.
-----
- - -t~ ~ ~ : -·--·--- -
/.,
·
.- . --:::..: . ___:..,-:-=--
Drawdown cone
Upper confining bed
T ---
Artesian Figure 4.2.6. Distribution of flow to a discharg-
aquifer -- -- ing well-just penetrating to the top of an arte-
Flow sian aquifer. A strong vertical component of
I in an
-a
l lllllllillli_.....................
Lower confining bed
flow is established out to a distance approxi-
mately equal to the thickness of the aquifer
(from U.S. Bureau of Reclamation62).
~and
nhole
Jureau "In fact, the difference in drawdowns (s 1- s,) becomes essentially constant while both values are still increasing so
·2).
that Equation 4.2.6 generally gives good results after only a few days of pumping.
EXAMPLE 4.2.1 A well fully penetrates a 25-m thick confined aquifer. After a long period of pumping at a constant rate
of 0.05 m3/s, the drawdowns at distances of 50 and 150m from the well were observed to be 3 and 1.2 m,
respectively. Determine the hydraulic conductivity and the transmissivity. What type of unconsolidated
deposit would you expect this to be?
SOLUTION Use Equation 4.2.5 to compute the hydraulic conductivity with Q = 0.05 m3/s, r 1 =50 m, r 2 = !50 m, s 1 =
l~o- h1, and s2 === ho- ~.so s 1 - s2 = ~- h 1 = 3- 1.2 = 1.8 m. Q = 0.05 m3/s = 4320 m3/day, and
3
K= Q ln( r2 J = 4320 m /day In( 150J =16.8 m/day
271b(~ -h 1) r1 2rr(25 m)(1.8 m) 50
The transimissivity is T = Kb = (16.8 rn/day)(25 m) =420m2/day. Referring to Figure 3.2.2 and Table
3.2.1 with K = 1.94 x 10-4 m/s shows that this aquifer is probably a medium clean sand. •
EXAMPLE" 4.2.2 A 1-m diameter well penetrates vertically through a confined aquifer 30 m thick. When the well is
pumped at 113 m3/hr, the drawdown in a welll5 m away is 1.8 m; in another well 50 m away, it is 0.5 m.
What is the approximate head in the pumped well for steady-state conditions and what is the approximate
drawdown in the well? Also compute the transmissivity of the aquifer and the radius of influence of the
pumping well. Take the initial piezometric level as 40 m above the datum.
SOLUTION First determine the hydraulic conductivity using Equation 4.2.5: Q = 113 m3/hr = 2712 m3/day. Then
3
K= Q J
ln('2 = 2712 m /day In( 50)= l3.3 m/day
27tb(s 1 -s2 ) r1 2tt(30m)(1.8m-0.5m) 15
The transmissivity is T =Kb = 13.3 m/day x 30m= 400m2/day.
To compute the approximate head, hw, in the pumped well, rearrange Equation 4.2.5 and use h2 = ho- s2 =
40-0.5 = 39.5 m
~ _ __lL_ln(.2_J==
3
In( m J=34.5 m
2712 50
h,. == 39.5 m- m /day
2rtKb rw 2rt{1Bm/day)(30m) 0.5m
Drawdown is then
sw = ho- h..,= 40 m- 34.5 m = 5.5 m
The radius of influence (R) of pumping well can be found by rearranging Equation 4.2.5 and solving for
r0 which is R: .J
~
I
2rtKb(ho-h1)] (
R = (r1)exp = 15 m ) exp[2rt(l3.3m/day)(30m)(40m-38.2m)] = 79 m
[ 2712 m3/day
Q ·
•
An equation for steady radial flow to a well in an unconfined aquifer also can be derived with
the help of the Dupuit assumptions. As shown in Figure 4.2.7, the well completely penetrates
the aquifer to the horizontal base and a concentric boundary of constant bead surrounds the
well. The well discharge is
dh
Q=-21trKh- (4.2.7)
dr
which, when integrated between the limits h =hw at r =r wand h =h0 at r =r 0, yields
tant rate Ground sulface

d 1.2 m, Original water table
olidated
Drawdown
curve
nd Table ho h h1 h2
• (1 I
I (2 I
Unconfined
aquifer
e well is
is0.5 m.
roximate
tee of the Figure 4.2. 7. Radial flow to a well penetrating
Impermeable an unconfined aquifer.
·.Then
(4.2.8)
Converting to heads and radii at two observation wells (see Figure 4.2.7),
1.2 h2
Q=rtK " 2- I (4.2.9)
Inh/r1)
and rearranging to solve for the hydraulic conductivity
K= Q
rt(hi -hn ln'2
'i
(4.2.10)
This equation fails to accurately describe the drawdown curve near the well because the large
olving for vertical flow components contradict the Dupuit assumptions; however, estimates of K for
given heads are good. In practice, drawdowns should be small in relation to the saturated thick-
ness of the unconfined aquifer.
The transmissivity can be approximated from Equation 4.2.10 by
11
• (4.2.11 )
Where drawdowns are appreciable, the heads h1 and h-z in Equation 4.2JO can be replaced by
(h0 - s1) and (h0 - s?J, respectively, as shown in Figure 4.2.7. Then the transmissivity for the
ived with full thickness becomes3• 37
enetrates
T = Kho = Q 1n '2
mnds the
(4.2.7)
2{[s 1- ;LJ+- ~)] r,
(4.2.12)
Figure 4.2.8 illustrates the flow distribution in an unconfined (free) aquifer, for a fully
s penetrating well that has the openings in the lower one-third of the aquifer.
Static water tabla

-·\-f-·~·--,-· -----~·-·r·;,-
\ 1
SOLU
(a) Initial stage in pumping a free aquifer. Most water follows

a path with a high vertical component from the water
table to the screen.
{b) Intermediate stage in pumping a free aquifer. Radial

component of flow becomes more pronounced but
contribution from drawdown cone in immediate vicinity
of well is still important
----·-·-·-·-·
4.2.3
Figure 4.2.8. Development of flow

(c) Approximate steady-state stage in pumping a free aquifer. distribution about a discharging well
Profile of cone of depression is established. Nearly all in a free aquifer-a fully penetrating
water originating near outer edge of area of influence and 33 percent open hole (from U.S.
and stable primarily radial flow pattern established. Bureau of Reclamation62).
A well penetrates an unconfined aquifer. Prior to pumping the water level (head) is h0 = 25 m. After a
UA'/tfPLE 4.2.3
long period of pumping at a constant rate of 0.05 m3/s, the drawdowns at distances of 50 and 150m from
the well were observed to be 3 and 1.2 m, respectively. Compute the hydraulic conductivity of the aquifer
and the radius of influence of pumping well. What type of deposit is the aquifer material?
SOLUTION Use Equation 4.2. 10 to compute K with Q =0.05 m3/s= 4320 m3/day, r1 =50 m, r2 =150m, h1 = 25-3
=22m, and h2 :;: 25-1.2 =23.8 m.
Jn(' J= 4320m /day ln(150mJ=lS.3m/day

3
K=
n(hi
Q
-hn 2
'i n(23.82 -222 ) 50 m
The deposit is probably a medium clean sand. Equation 4.2.10 is used to compute the radius of influence:
•
A well 0.5 min diameter penetrates 33 m below the static water table. After a long period of pumping at
a rate of 80 m3/hr, the drawdowns in wells 18 and 45 m from the pumped well were found to be 1.8 and
1.1 m respectively. (a) What is the transmissivity of the aquifer? (b) What is the approximate drawdown
in the pumped well? (c) Determine the radius of influence of the pumping welL
SOLUTION (a) Use Equation 4.2.10 for steady-state radial flow to a well in an unconfined aquifer to compute the
hydraulic conductivity, where Q= 80 m3/hr = 1920 m3/day; h1 :::: 33- 1.8 = 31.2 m.; h2 = 33 - 1.1 =
31.9m;r2 =45mandr 1 = 18m:
3
1920 45
K= Q ln['2 J= m /day ln( )=12.7m/da
rt(h]- hn lj
2 2
n(31.9 - 31.2 ) 18 y
The transmissivity is computed as T= Kb = !2.7 m/day x 33m:;: 418m2/day.

(b) Ne)(t compute the head and drawdown at the well. First rearrange Equation 4.2.10 to solve for the
head at the well:
~-_g_ln('2 )= ln(~) =27.7m

3
1920
II,= 31.2 2 m /day
nK r 1 n(l2.68 mlday) 0.25 m
The drawdown is computed as sw =33 m- 27.7 m = 5.3 m.
l
(c) The radius of influence of the pumping well is computed by rearranging Equation 4.2.5:
R=(r1)exp
hJ- hnj = (45m)exp[n(l2.68 m/day)(332 -31.92 )
rtK(
=198m
3
[ Q I920m /day
•
4.2.3 Unconfined Aquifer with Uniform Recharge
Figure 4.2.9 shows a well penet:rating an unconfined aquifer that is recharged uniformly at rate
·now W from rainfall, excess irrigation water, or other surface-water sources. The flow Q toward the
g well well increases as the well is approached, reaching a maximum of Qw at the well. The increase
trating
n U.S .
.. After a
)mfrom
e aquifer
=25- 3
curve
1fluence: Unconfined
aquifer
Figure 4.2.9. Steady flow to a
well penetrating a uniformly
• Impermeable recharged unconfined aquifer.

in flow dQ through a cylinder of thickness dr and radius r comes from the recharged water
entering the cylinder from above; hence,
dQ =- 27trdr W (4.2.13)
Integrating, we obtain
(4.2.14)
but at the well r ~ 0 and Q =Qw, so that
Q=-7t?W+ Qw (4.2.15) 4.3
Substituting this flow in the equation for flow to the well (Equation 4.2.7) gives
-21trKh dh = -7tr2 W+ Q
dr w (4.2.16)
Integrating, and noting that h =h0 at r =r 0, yield the equation for the drawdown curve:
Qw ro
hQ2 -h 2 =W- (r 2 - r02) +-ln-
2K 1tK r (4.2.17)
By comparing Equation 4.2.17 with Equation 4.2.8, the effect of the vertical recharge becomes
apparent.
It follows that when r = r 0, Q =0, so that from Equation 4.2.15
(4.2.18)
Thus, the total flow of the well equals the recharge within the circle defined by the radius of
influence; conversely, the radius of influence is a function of the well pumpage and the
recharge rate only. This results in a steady-state drawdown; however, the analysis assumes an
idealized circular outer boundary with a constant head and no flow-conditions that rarely
occur in the field.
A pumping well is to be used to maintain a lowered water table at a construction site. The site is square,
50 m on a side, and the 25-cm diameter well is located at the center of the square, as shown in the figure.
The hydraulic conductivity of the unconfined aquifer is estimated to be about 1 x w-s m!s or 0.864
m/day. The bottom of the aquifer is approximately horizontal at a depth of 20 m below the ground
surface. Under natural conditions, the water table is nearly horizontal at a depth of 1 m below the ground
surface and the unconfined aquifer is uniformly recharged at a rate W::: 0.06 rnJday. During the con-
struction period, the water table must be lowered a minimum of 3 mover the site. Assuming steady-state
conditions, compute the minimum pumping rate required.
SOLUTION The given condition is satisfied if the drawdown at any of the comer points is 3m.
+---50m---.
The well discharge is expressed in terms of the radius of influence using Equation 4.2.17 as Qw =1tri W =
(0.06) nrl. Substitute this relationship along with ho =19m, h =16m, W=0.06 m/day, K =0.864 m/day,
and r = ~25 2 +25 2 = 35.35 minto Equation 4.2.17 to obtain
.
_::
4.3 Well in a Unifonn Flow 161
water
192-162= 0.06 (35.352-r.2)+(0.067tro2)ln(_!Q_)
2 X 0.864 O 1t X 0.864 35.35
2.13)
Solving the above equation using an iterative procedure yields r0 == 70 m. The minimum pumping rate is
Q" = n(702) (0.06) ==924m3/day or 0.01069 m3/s. •
2.14)
.2.15) 4.3 WELL IN A UNIFORM FLOW

Drawdown curves for well flow presented heretofore have assumed an initially horizontal
groundwater surface. A practical situation is that of a well pumping from an aquifer having a
unifonn flow field, as indicated by a uniformly sloping piezometric surface or water table.
.2.16)
Figure 4.3.1 shows sectional and plan views of a well penetrating a confined aquifer with a
sloping piezometric surface. It is apparent that the circular area of influence associated with a
radial flow pattern becomes distorted; however, for most relatively flat natural slopes the
Dupuit radial flow equation can be applied without appreciable error.
.2.17) For wells pumping from an area with a sloping hydraulic gradient, the hydraulic con-
;omes ductivity can be determined from Equation4.2.7 by inserting average heads and hydraulic gra-
dients. The resulting expression has the form
(4.3.1)
·.2.18)
lius of
for an unconfmed aquifer where Q is the pumping rate, hu and hd are the saturated thicknesses,
nd the
and i" and id are the water table slopes at distance r upstream and downstream, respectively,
nes an
from the well. For a confined aquifer, piezometric slopes replace water table slopes, and (hu +
rarely
hd) is replaced by 2b where b is the aquifer thickness.
In Figure 4.3.1, the groundwater divide marking the boundary of the region producing
inflow to the well is shown. For a well pumping for an infinite time, the boundary would
square, extend up to the limit of the aquifer. The expression for the boundary of the region producing
. figure. inflow can be derived by superposition of radial and one-dimensional flow fields to yield
r 0.864
_ ~ =tan[ 21t~bi
ground
ground
y) (4.3.2)
he con-
iy-state where the rectangular coordinates are as shown in Figure 4.3.1 with the origin at the well, b is the
aquifer thickness, Q is the discharge rate, i is the natural piezometric slope, and K is hydraulic con-
ductivity. From Equation 4.3.2, the boundary asymptotically approaches the finite limits
(4.3.3)
as x -t oo_ The boundary of the contributing area extends downstream to a stagnation point
where
Q
X - --- (4.3.4)
[,- 21tKbi
It follows that the upstream inflow zone equals 21tXL.

Equations 4.3.1 to 4.3.3 also apply to unconfined aquifers by replacing b by the uniform
rtrl W== saturated aquifer thickness h0 , providing the drawdown is small in relation to the aquifer thick-
~m/day, ness. An important practical application of these equations concerns determining whether an
upstream pollution source will affect a nearby pumping well (see Chapter 8).
4.4
4.4.1
Drawdown curve
-- I
- Confined aquifer
Impermeable
b
!
(a)
.~
Figure 4.3.1. Flow to a

well penetrating a confined
Groundwater divide aquifer having a sloping
plane piezometric surlace.
(a) Vertical section. (b)
(b) Plan view
A fully penetrating production well with a radius of 0.5 m pumps at the rate of 15 Us from a 35-m thick
~XAMPLE ./.3.1
confined aquifer with a hydraulic conductivity of 20 rnlday. If the distance and the observed piezometric
head drop between two observation wells were 1000 m and 3 m, respectively, before the production well
was installed, determine the longitudinal and transverse limits of groundwater entering the welL
SOLUTION First determine the slope of the piezometric surlace under natural conditions (i.e., before the production
well was installed):
i=M=~=0.003
Ill lOOOm
It is assumed that the observation wells were aligned with the groundwater flow direction. Then, using
Equations 4.3.3 and 4.3.4, compute the limits of groundwater entering the well on a horizontal plane (i.e.,
plan view) for Q = 15 Us= 1296 m3/day:
3
= +_Q_ = + 1296 m /day 8
h - 2Kbi - 2(20 rnlday)(35 m) x0.003 ±30 m
3
Q__
x ___ 1296m /day =- _ m
98 2
L- 2rtKbi- 2rt(20m/day)(35m)x0.003
A practical result is that contaminant sources farther than 98.2 m downstream of the well or ± 308 m in
the transverse direction do not impact the well. •
4.4 UNSTEADY RADIAL FLOW IN A CONFINED AQUIFER

4.4.1 Nonequilibrium Well Pumping Equation
When a well penetrating an extensive confined aquifer is pumped at a constant rate, the influ-
ence of the discharge extends outward with time. The rate of decline of head times the storage
coefficienrsummed over the area of influence equals the discharge. Because the water must
come from a reduction of storage within the aquifer, the head will continue to decline as long
as the aquifer is effectively infinite; therefore, unsteady, or transient, flow exists. The rate of
decline, however, decreases continuously as the area of influence expands.
The applicable differential equation (see Equation 3.9.12) in plane polar coordinates is
(J h
2
-+--=--
1 dh s dh
or2 r or T "Ot (4.4.1)
where h is head, r is radial distance from the pumped well, S is the storage coefficient, Tis
transmissivity, and tis the time since beginning" of pumping. Theis59 obtained a solution for
Equation 4.4.1 based on the analogy between groundwater flow and heat conduction. By
assuming that the well is replaced by a mathematical sink of constant strength and imposing
the boundary conditions h =ho fort =0, and h ~ h0 as r ~ oo fort~ 0, the solution
s=Q- J"" -e-"du

-= - Q W(u)
41tT
Q [
u u
= - -0.5772-lnu+u--+----+···
41tT
4rtT
2·2! 3·3! 4 ·4!

u2 u3 u4 l (4.4.2)
:oa is obtained, where s is drawdown, Q is the constant well discharge, and

tnfined 2
ling
r S
u=-
rface. 4Tt (4.4.3)
:h)
Equation 4.4.2 is known as the nonequilibrium, or Theis, equation. The integral is a function
of the lower limit u and is known as an exponential integral. It can be expanded as a conver-
m thick
gent series as shown in Equation 4.4.2 and is termed the well function, W(u).
ometric Alternatively, using U.S. customary units (gallon-day-foot system) where sis in ft, Qis in
on well gpm, Tis in gpdlft, u is in ft, and t is in days, we have
s =114.6Q W(u)
T (4.4.4a)
duction
l.87r2S
u= (tin days)
Tt (4.4.4b)
2693r2S
1,using u= (t in minutes)
Tt (4.4.4<:)
me (i.e.,
The nonequilibrium equation permits determination of the formation constants Sand Tby
means of pumping tests of wells. The equation is widely applied in practice and is preferred
over the equilibrium equation because (1) a value for Scan be detennined, (2) only one obser-
vation well is required, (3) a shorter period of pumping is generally necessary, and (4) no
assumption of steady-state flow conditions is required.
The assumptions inherent in Equation 4.4:2 should be emphasized because they are often
08 min overlooked in applying the nonequilibrium equation and thereby can lead to erroneous results.
• The assumptions include:
Table4.·
1. The aquifer is homogeneous, isotropic, of unifonn thickness, and of infinite areal extent.
2. Before pumping, the piezometric surface is horizontal. - u
3. The well is pumped at a constant discharge rate.
··:
--
X[
4. The pumped well penetrates the entire aquifer, and flow is everywhere horizontal within x w-'
the aquifer to the well. x w-z
5. The well diameter is infinitesimal so that storage within the well can be neglected. x w-J
xl0-4
6. Water removed from storage is discharged instantaneously with decline of head.
x w-
5
Seldom, if ever, are these assumptions strictly satisfied, but recognition of them can cre- x w-6
ate an awareness of the approximations involved for employing the nonequilibrium equation x w-7
under field conditions. Average values of SandT can be obtained in the vicinity of a pumped x w-s
well by measuring in one or more observation wells the change in drawdown with time under x w-9
X 10-IO
the influence of a constant pumping rate. Because of the mathematical difficulties encountered
x w-11
in applying Equation 4.4.2, or its equivalent, Equation 4.4.4, several investigators have devel-
x w-12
oped simpler approximate solutions that can be readily applied for field purposes. Three meth-
x w-u
ods, by Theis,59 Cooper and Jacob,9 and Chow,7 are described in the following sections with the
necessary tables and/or graphs. An illustrative example accompanies each method. x w-' 4
x w-15
4.4.2 Theis Method of Solution EXAM)

Equation 4.4.2 may be simplified to
s= (l)w(u)
41tT (4.4.5)
where W(u), termed the well function, is a convenient symbolic form of the exponential
integral. Rewriting Equation 4.4.3 as
(4.4.6)
we can see that the relation between W(u) and u must be similar to that between s and
?It because the terms in parentheses in the two equations are constants. Given this similarity,
Theis59 suggested an approximate solution for S and T based on a graphic method of
superposition.
A plot on logarithmic paper of W(u) versus u, known as a type curve, is prepared. Table
4.4.1 gives values of W(u) for a wide range of u. Values of drawdowns are plotted against val-
ues of ?It on logarithmic paper of the same size and scale as for the type curve. The observed
time-drawdown data are superimposed on the type curve, keeping the coordinate axes of the
two curves parallel, and adjusted until a position is found by trial whereby most of the plotted
points of the observed data fall on a segment of the type curve. Any convenient point is then
selected, and the coordinates of this match point are recorded. With values of W(u), u, s, and SOLUT/(
?It thus determined, S and T can be obtained from Equations 4.4.5 and 4.4.6. *
In areas where several wells exist near a well being test-pumped, simultaneous readings
of s in the wells enable distance-drawdown data to be fitted to a type curve in a manner iden-
tical to that for time-drawdown data.
'The computation of r'-!t values can be avoided by plotting s versus t rather than r'-!t. In this case, the type curve
must be turned over to obtain coincidence and a match point, but results are identical.
<tent. Table 4.4.1 Values of W(u) for Values of u
u 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
X 1 0.219 0.049 0.013 0.0038 0.0011 0.00036 0.00012 0.000038 0.000012

vithin X WI 1.82 1.22 0.91 0.70 0.56 0.45 0.37 0.31 0.26
x 2 w- 4.04 3.35 2.96 2.68 2.47 2.30 2.15 2.03 1.92
X w-3 6.33 5.64 5.23 4.95 4.73 4.54 4.39 4.26 4.14
x10-4 8.63 7.94 7.53 7.25 7.02 6.84 6.69 6.55 6.44
x w- 5
10.94 10.24 9.84 9.55 9.33 9.14 8.99 8.86 8.74
x w-6 13.24 12.55 12.14 11.85 11.63 11.45 11.29 11.16 11.04
r1cre-
•ation x w-7 15.54 14.85 14.44 14.15 13.93 13.75 t3.60 13.46 13.34
x w-s 17.84 17.15 16.74 16.46 16.23 16.05 15.90 15.76 15.65
mped
under x w-9 20.15 19.45 19.05 18.76 18.54 18.35 18.20 18.07 17.95
x w-lo 22.45 21.76 21.35 21.06 20.84 20.66 20.50 20.37 20.25
ttered
x w-11 24.75 24.06 23.65 23.36 23.14 22.96 22.81 22.67 22.55
level-
x w-12 27.05 26.36 25.96 25.67 25.44 25.26 25.11 24.97 24.86
rneth-
x w-13 29.36 28.66 28.26 27.97 27.75 27.56 27.41 27.28 27.16
th the
x w-14 31.66 30.97 30.56 30.27 30.05 29.87 29.71 29.58 29.46
x w-15 33.96 33.27 32.86 32.58 32.35 32.17 32.02 31.88 31.76
Drawdown was measured during a pumping test at frequent intervals in an observation well 200ft from a
EXAMPLE 4.4.1
well that was pumped at a constant rate of 500 gpm. The data for this pump test is listed in the table. These
measurements show that the water level is still dropping after 4,000 minutes of pumping; therefore, anal-
ysis of the test data requires use of the Theis nonequilibrium procedure. Determine T and S for this aquifer.
4.4.5) Pump test data

ential Time (min) Drawdown (ft)
0.05
2 0.22
3 0.40
4.4.6)
4 0.56
sand 5 0.70
larity, 7 0.94
od of 10 1.2
20 1.8
Table 40 2.5
;t val- 100 3.4
erved 300 4.5
of the 1,000 5.6
lotted 4,000 7.0
s then
s, and SOLUTION Step 1. Plot the time-drawdown data on log-Jog graph paper. The drawdown is plotted on the vertical axis
and the time since pumping started on the horizontal axis (not shown).
tdings Step 2. Superimpose this plot on the type curve sheet of the same size and scale as the tirne-drawdown plot,
iden- so that the ploned points match the type curve. The axes of both graphs must be kept parallel.
Step 3. Select a match point, which can be any point in the overlap area of the curve sheets. It is usually most
convenient to select a match point where the coordinates on the type curve are known in advance
(e.g., W(u) = I and 1/u =1 or W(u) = 1 and 1/u =10, etc.). Then determine the value of sand t for
this match point
ve
W(u) =1 s = 1ft 1/u =1 t= 2 min
Step 4. Determine T
T= li4.6QW(u)
s
114 6
= · X SOO X 1 =57300 gpd I ft
I
Step 5. Determine S 10
s-- I
n
5
-x2693r2
u
57300x2
2
- 1x2693x200 2
=l.06 x 10-3
• '5"
~ 1.0
0.5
A well penetrating a confined aquifer is pumped at a uniform rate of 2,500 m31day. Drawdowns during
the pumping period are measured in an observation well 60 m away; observations oft and s are listed in
Table 4.4.2. Using the Theis method determine T and S for this confined aquifer.
0.2
..,.... 0.1
SOLUTION Values of ?It in m21min are computed and appear in the right column ofTable 4.4.2. Values of s and ?It 10'
are plotted on Jogaritlunic paper. Values of W(u) and u from Table 4.4.1 are plotted on another sheet of
logaritlunic paper of the same size and scale, and a curve is drawn through the points. The two sheets are
superposed and shifted with coordinate axes parallel until the observational points coincide with the
4.4.3
curve, as shown in Figure 4.4.1. A convenient match point is selected with W(u) = 1.00 and u = 1 x w-2,
so that s =0.18 m and ?It= 150 m21min = 216,000 m2/day. Thus, from Equation 4.4.5,
Q 2500(1.00)
T=-W(u) = == 1110 m2 /day
41t.S 41t(O.l8)
and from Equation 4.4.6,
2
4Tu 4{l!IO)(tx10- )
S=-= 0.000206
r 2/t 216,000
•
Table 4.4.2 Pumping Test Data
(r=60 m)
t, min s,m ?It, m2/rnin t, min s,m ?tt, m2/min
0 0 00 18 0.67 200
0.20 3,600 24 0.72 150
1.5 0.27 2,400 30 0.76 120
2 0.30 1,800 40 0.81 90
2.5 0.34 1,440 50 0.85 72
3 0.37 1,200 60 0.90 60
4 0.41 900 80 0.93 45
5 0.45 720 100 0.96 36
6 0.48 600 120 1.00 30
8 0.53 450 150 1.04 24
10 0.57 360 180 1.07 20
12 0.60 300 210 1.10 17
14 0.63 257 240 1.12 15
• Match point
during
.sted in
Figure 4.4.1. Theis method of

md ?It superposition for solution of
heet of u the nonequilibrium equation.
:ets are
•ith the 4.4.3 Cooper-Jacob Method of Solution
x w-2,
It was noted by Cooper and Jacob9 that for small values of rand large values oft, u is small,
so that the series terms in Equation 4.4.4 become negligible after the first two terms. As a
result, the drawdown can be expressed by the asymptote
s =Q- [ -D.5772-ln-
47tT
r
4Tt
Sl
2
(4.4.7)
• Rewriting and changing to decimal logarithms reduce this to

-?1
s = 2.30Q log 2.25Tt
2
(4.4.8)
41tT r S
Therefore, a plot of drawdown s versus the logarithm oft forms a straight line. Projecting this
line to s =0, where t = t0 (see Figure 4.4.2), we have
1
0 ~ 2.30Q lo 2.25Tt0 (4 4 9)
4nT g r 2S ..
and it follows that for the above to hold true, log (1) = 0 and
2.2;Tto =1 (4.4.10)
rS
resulting in
- 2.25Tt0
S- 2 (4.4.11)
r
A value for T can be obtained by noting that if tlt0 = 10, then log tlt0 = 1; therefore, replacing
s by D.s, where D.s is the drawdown difference per log cycle oft, Equation 4.4.8 becomes33
T = 2.3dQ 1 . 10
~.
ti_ (4.4.1 2)
47t!ls 41T f>f.'o
*
Ul
....
E
0.8
vi Pumping test data
c 0.6
~
.g
~
0
a~~--~--~~----~~--~--~~~~--~~
0.1 2 2 5 10 2 5 10 3 Figure 4.4.2. Cooper-Jacob method for solu-
Time since pumping began, t, min tion of the nonequilibrium equation.
Thus, the procedure is first to solve for Twitb Equation 4.4.12 and then to solve for Swith
Equation 4.4.11. The straight-line approximation for this method should be restricted to small
values of u (u < 0.01) to avoid large errors.
llXAMPLE 4.'4.3 Rework Example 4.4.2 using the Cooper-Jacob method.
4.4.4
SOLUTION From the pumping test data in Table 4.4.2, sand tare plotted on seroilogarithmic paper, as shown in Fig-
ure 4.4.2. A straight line is fitted through the points, and til= 0.40 m and t0 = 0.39 min= 2.70 x w-4 day
are read. Thus,
T= ZJ0(2SOO) 1144 m2fday

41t(0.40)
and
_ 2.25Tto _ 2.25(1144)(2.70x!0-4) _
S - - 2-- - -0.000193
r (60)2
•
Using the Cooper-Jacob approximation, compute the rate of piezometric drawdown around a pumping
well with respect to time. If the well is pumping at a constant rate of 55 gpm from a sandy confined
aquifer with T =3,600 fe/day and S = 10-4, what is the time to reach near-steady-state conditions 200ft EXAN
from the pumping well? Assume that near-steady-state conditions are achieved when the drawdown rate
falls below 0.5 inlbr (based on accuracy of groundwater level measurements with the available equip-
ment). How does the answer change if the transmissivity of the aquifer is 1,200 f!2/day? SOLU1
SOLUTION Pirst, we must compute the critical time after which the Cooper-Jacob method becomes valid (i.e.,
u < 0.01) at 200ft:
The drawdown is approximated by
s=Q- ( -0.5772-ln-
. 41tT
r
2
4Tt
S)
which can be rearranged to

2
rs
s =Q- ( -0.5772-ln-+lnt )
4nT 4T
Taking the derivative of d.rawdown with respect to time yields

ds Q 1
-=--
dt 41tT t
This relationship implies that according to the Cooper-Jacob approximation, the rate of drawdown is
independent of radial distance and is inversely proportional with time. The change in d.rawdown with
respect to time and the time are, respectively,
ds::: _g__!
= 0.5 inlhr
dt 41tT I
3
10587 ft /day ! =Q2 ftlhr
olu- 41t(3600 ft 2 /day) t 12
fo=5.6hr
Note that the Cooper-Jacob approximation is satisfied so that the near-steady-state conditions at 200ft
Swith
will be reached after 5.6 hrs of pumping at this location. If the transmissivity were 1200 ft2/day, the
'small approximation would be valid when t ;::: 120 min and the drawdown rate at 200 ft would be negligible
after 16.8 hours of pumping. Thus, it would take longer to reach steady conditions with a lower
transmissivity. •
4.4.4 Chow Method of Solution

in Fig-
0-4 day Chow7 developed a method of solution with the advantages of avoiding curve fitting and being
unrestricted in its application. Again, measurements of drawdown in an observation well near
a pumped well are made. The observational data are plotted on semilogarithmic paper in the
same manner as for the Cooper-Jacob method. On the plotted curve, choose an arbitrary point
and note the coordinates, t and s. Next, draw a tangent to the curve at the chosen point and
determine the drawdown difference ru, in feet, per Jog cycle of time. Compute F(u) from
F(u) =_!__ (4.4.13)

• ru
and find the corresponding values of W(u) and u from Figure 4.4.3.* Finally, compute the for-
umping mation constant T by Equation 4.4.5 and S by Equation 4.4.6.
onfined
s 200ft EXAMPLE 4A.5 Repeat Example 4.4.2 using the Chow method.
•wn rate
:equip-
SOLUTION
In Figure 4.4.4 data are plotted from Table 4.4.2 and point A is selected on the curve where t =6 min= 4.2
X 1o-J day and s = 0.47 m. A tangent is constructed as shown; the drawdown difference per log cycle of time
is 6s = 0.38 m. Then F(u) = 0.47/0.38 = 1.24, and from Figure 4.4.3, W(u) = 2.75 and u = O.D38. Hence,
.id (i.e.,
2500
T =JL W(u) = 2.75 =1160 rn 1/day
4ru 41t(0.47)
and
4Ttu 4{1 160)(4.2 x 10-3 )(0.038)
S=--= =0.000206
(60)2
r2 ·
•
'For F(u) > 2.0. W(u)::; 2.30F(u), and u is obtained from Table 4.4.1.
10
5
<il
c
·o,_
....
· - Q)
0(1)
>
~-
2 oOl
-c:
(!) · -
.oa.
'5" -E:J
(I)
it >a.
~c:
.... 0
Ql c:
o\ 1.1 0 <?v iii
~ue$
3:~
5 o <? ·.s-
'-!
.>
0
~-"'
·o
Figure 4.4.3. Relation among F(u),

W(u) W(u), and u (after Chow\
Tangent line
OL-~--~--~--~--~~--~--~--~~----~~
0.1 2 s s 1o 2 s 2 s 1o3
Figure 4.4.4. Chow method for solution of the
Time since pumping began, t, min nonequilibrium equation.
4.4.5 Recovery Test

At the end of a pumping test, when the pump is stopped, the water levels in pumping and
observation wells will begin to rise. This is referred to as the recovery of groundwater levels,
while measurements of drawdown below the original static water level (prior to pumping) dur-
ing the recovery period are known as residual drawdowns. A schematic diagram of change in
water level with time during and after pumping is shown in Figure 4.4.5.
It is good practice to measure residual drawdowns because analysis of the data enables
transmissivity to be calculated, thereby providing an independent check on pumping test results.
Also, costs are nominal in relation to the conduct of a pumping test* Furthermore, the rate of SOLU1
*In addition, it should be noted that measurement of the recovery within a pumped well will provide an estimate of
transmissivity even without an observation well.
(iJ
c
'0>-
·r::: ID
0 >
!:~ Drawdown
oOJ
- c
Q) ·-
.oO.
-ID::J
E Recovery
> 0.
~c
'- 0
CD C
(ij
~~
------ Figure 4.4.5. Drawdown and recovery curves in an
1+---- Pumping period -----+11+---- Recovery period -----+1 observation well near a pumping well.
recb.arge Q to the well during recovery is assumed constant and equal to the mean pumping rate,
whereas pumping rates often vary and are difficult to control accurately in tbe field.
If a well is pumped for a known period of time and then shut down, the drawdown there-
after will be identically the same as if the discharge had been continued and a hypothetical
recharge well with the same flow were superposed on the discharging well at the instant the
discharge is shut down. From this principle, Theis59 showed that the residual drawdown s' can
be given as
s' = _Q_[ W(u)- W(u')] (4.4.14)

41tT
where
(4.4.15)
and t and (are defined in Figure 4.4.5. For r small and (large, the well functions can be approx-
imated by the first two terms of Equation 4.4.2 so that Equation 4.4.14 can be written as
S' =2.30Q log!_ (4.4.1 6)

41tT t'
of the
Thus, a plot of residual drawdown s' versus the logarithm of tit' fonns a straight line. The slope
of the line equals 2.30Q/41tT so that forM, the residual drawdown per log cycle of tit', the
transmissivity becomes
T= 2.30Q (4.4.17)
1g and 47t&'
levels,
No comparable value of Scan be determined by this recovery test method.
g) dur-
mgein
A well pumping at a uniform rate of 2,500 m3/day was shut down after 240 min; thereafter, measurements
:nables of s' and t' tabulated in Table 4.4.3 were made in an observation well Determine the transmissivity.
results.
rate of SOLUTION Values of tit' are computed, as shown in Table 4.4.3, and then plotted versus s' on semilogarithmic paper
(see Figure 4.4.6). A straight line is fitted through the points and M =0.40 m is determined; then,
nate of T =2.30Q =2.30(2500) 1140 m 2/day

4M.s' 47t(0.40)
Table4.4.3 Recovery Test Data (pump shut down at t =240 min)

10
t', min t, min tit' s',m
241 241 0.89

2 242 121 0.81
3 243 81 0.76
5 245 49 0.68
7 247 35 0.64
10 250 25 0.56
15 255 17 0.49
20 260 13 0.55
30 270 9 0.38
40 280 7 0.34
60 300 5 0.28
80 320 4 0.24
100 340 3.4 0.21
140 380 2.7 0.17
180 420 2.3 0.14
~
<II
'ai 0.8
E
--
r0
<ll
r: 0.6 ~s· =0.40 m

~
~
'0 0.4
!
~
'0
'iii
£ 0.2
Figure 4.4.6. Recovery test method

2 5
for solution of the nonequilibrium
Time ratio, tit' equation.
4.5 UNSTEADY RADIAL FLOW IN AN UNCONFINED AQUIFER

The previous solution methods for the nonequilibrium equation applied to pumping tests in
confined aquifers can also be applied to unconfined aquifers provided that the basic assump-
tions are satisfied. In general, if the drawdown is small in relation to the saturated thickness,
good approximations are possible.53
Where drawdowns are significant, the assumption that water released from storage is dis-
charged instantaneously with decline of head is frequently violated in unconfined aquifers.
Pumping test data reveal that as a water table is lowered, gravity drainage of water from the
unsaturated zone proceeds at a variable rate, known as delayed yield.4• 14• 42 In a series of con-
tributions, Boulton4-6 developed special type curves for analyzing pumping test data of uncon-
fined aquifers and for taking account of delayed yield. 65 These time~rawdown curves of
delayed yield are shown in Figure 4.5.l. The interpretation of any one curve can be considered
in three time segments. In the first segment, measured in seconds to a few minutes, water is
released essentially instantaneously from storage by compaction of the aquifer and by expan-
4.5 Unsteady Radial Flow in an Unconftned Aquifer 173
Theis type curve

for later unconfined
condition
Figure 4.5.1. Type curves of draw-
down versus time illustrating the effect
of delayed yield for pumping tests in
unconfined aquifers (after U.S. Bureau
Time, days of Reclamation61 ).
sion of entrapped air. This portion of the curve can be fitted by a type curve with a storage coef-
ficient equivalent to that of a confined aquifer. The second segment displays a flattening in slope
caused by gravity drainage replenishment from the pore space above the cone of depression.
Finally, in the third segment an equilibrium is approached between gravity drainage and the rate
of decline of the water table. This condition occurs after several minutes to several days and can
be fitted by a type curve with a storage coefficient for an unconfined aquifer.
From a water production standpoint, the storage coefficient obtained from the third seg-
ment of the curve in Figure 4.5. L, which is the specific yield, is the most reliable and hence
most important. For simplicity a pumping test should be continued sufficiently long enough to
define the third segment of the curve; then, by applying one of the solution methods previously
described for the nonequilibrium equation, a value for Scan be obtained.
The minimum length of pumping test to achieve an accurate estimate of S in an uncon-
fined aquifer depends on the transmissivity of the aquifer. One approach, based on an empiri-
cal study of various alluvial aquifer materials, is given by the graphs in Figure 4.5.2. The delay
index td is estimated in Figure 4.5.2a from the composition of aquifer material. Then knowing
the distance r between pumping and observation wells, and estimating SandT, an approxi-
:thod mation to the minimum pumping time tmin can be calculated from Figure 4.5.2b.
liD
Another approach is simply to ensure that the pumping test duration exceeds the follow-
ing suggested guidelines61 .
Predominant aquifer material Minimum pumping time, hours

~sts in Silt and clay 170
sump- Fine sand 30
kness, Medium sand and coarser materials 4
is dis-
Prickett47 developed a type curve solution for water table conditions based upon Boulton5.
uifers.
The following equation for drawdown in an unconfined aquifer with fully penetrating wells
•m the
and a constant discharge condition was presented later by Neuman:43
f con-
mcon- :1
ves of (4.5.1)
idered
2
ater is where u0 = r S (applicable for early drawdown data) (4.5.2)
:xpan- 4Tt ·
1
100
10 3 U)
U)
~ 10
c
c 0
8
.E 'iii
c
(])
~
~ E
~ :Js.,
x
10 2
(])
"0
.~
d'.~
6 iil
2-
>- .._E S:
ro
a5 0
0 U)
(]) 4
::l
«l
0.1
10 >
0.01
Fine 1(
Silt
O gravel
Materials through which 0.5 1.0 1.5 2.0 2.5 3.0 3.5
gravity drainage takes place Values of rlvTtd/S, dimensionless
(a) (b)
Figure 4.5.2. Empirical method for estimating the minimum length of a pumping test in an unconfined aquifer (after Prickett47 ). (a)
Empirical relation of delay index to character of materials through which gravity drainage occurs (b) Curve for estimating the mini-
mum time tmin at which effects of delayed gravity drainage cease to influence drawdown of a pumping well in an unconfined aquifer.
r 2S
uy =__Y (applicable for later drawdown data) (4.5.3)
4Tt
11-
r2K z
.,- b2Kh (4.5.4)
W(ua,uy,'ll) is the unconfined well function (Figure 4.5.3); Kh and Kv are the horizontal and ver-
tical hydraulic conductivities for an isotropic aquifer Kv =K2 and 11 =?Jb 2; and b is the initial
saturated thickness of the unconfined aquifer.
A well pumping at 144.4 ft 3/min fully penetrates an unconfined aquifer with a saturated thickness of 25
ft. Determine the transmissivity, storativity, specific yield, and horizontal and vertical hydraulic conduc-
(adapted from U.S. tivities using the tabulated time-drawdown data in Table 4.5.1 for an observation well located 73 ft away.
Department of the
Interior) 62 •
SOLUTION Time-drawdown data (Table 4.5.1) are plotted in Figure 4.5.4, which shows the typical three phases of
drawdown for unconfined aquifers. The early drawdown versus time data fit best on the type-a curves for
11 = 0.06. The selected match point in Figure 4.5.4 has the following coordinates: (t = 0.17 min, s = 0.57
ft) and (1/ua = 1.0, W(u 0 ,uy,Tl) = 1.0). Using Equation 4.5.1 with a discharge of Q = 144.4 ft 3/min, we find
the transmissivity to be
3
Q ( ) (144.4ft /min) 2 . 2
T=-W u0 ,uy,Tl = ( ) (1.0)=20.16ft /rmn:29,900ft /day
41tS 47t 0.57 ft
Next, the storativity value is computed using Equation 4.5.2:

4.5 Unsteady Radial Flow in an Unconfined Aquifer 175
10-1 10
1oo------~-----,------~-----,,-----~------.------.-----,
10
~
::;,;.,
.,
--
::) 0.4
0.6
S: 1008
0.1
Figure 4.5.3.
Theoretical curves of
W(ua,uy,Tl) versus llu0
0,01L--'ll1lLL__I__ _ _ __ j_ _ _ _ _ __ j __ _ _ __ j_ __ J __ _....L,__ _ _ __ j_ _ _ _ _ _....L,__ _ _ __ j
and lfuy for an uncon-
10-5 10-4 10-3 10-2 10-1 10 102 10 3
fined aquifer (after
11uy Neuman43).
3.5
Table 4.5.1 Time-Drawdown Data for Example 4.5.1.

t (min) s, feet t (min) s, feet t (min) s, feet t (min) s, feet
\(a)
0.165 0.12 1.68 0.82 10 1.02 200 1.52
mini-
quifer. 0.25 0.195 1.85 0.84 12 1.03 250 1.59
0.34 0.255 2 0.86 15 1.04 300 1.65
0.42 0.33 2.15 0.87 18 1.05 350 1.7
0.5 0.39 2.35 0.9 20 1.06 400 1.75
(4.5.3) 0.58 0.43 2.5 0.91 25 1.08 500 1.85
0.66 0.49 2.65 0.92 30 1.13 600 1.95
(4.5.4) 0.75 0.53 2.8 0.93 35 1.15 700 2.01
0.83 0.57 3 0.94 40 1.17 800 2.09
0.92 0.61 3.5 0.95 50 1.19 900 2.15
ad ver- 1 0.64 4 0.97 60 1.22 1,000 2.2
: initial 70 1,200 2.27
1.08 0.67 4.5 0.975 1.25
1.16 0.7 5 0.98 80 1.28 1,500 2.35
1.24 0.72 6 0.99 90 1.29 2,000 2.49
;s of25 1.33 0.74 7 1 100 1.31 2,500 2.59
:onduc- 1.42 0.76 8 1.01 120 1.36 3,000 2.66
ft away. 1.5 0.78 9 1.015 150 1.45
2
tases of 4Tu0 t 4(20.16ft /min)(1.0)(0.17min)
S =- 2- = 2 =0.00257
rves for r (73ft)
; = 0.57
we find Moving the data curve to the right on the type curve to the best late-time match (for TJ =0.06) where s =
0.57 ft (see the match point on Figure 4.5.5) yields (t = 13 min, s =0.57 ft) and (lluy = 0.1, W(uy, TJ) =
1). Inserting the appropriate values in Equation 4.5.1 does not change the transmissivity estimate, but
using Equation 4.5.3 yields
2
4Tu t 4(20.16 ft /min )(0.1)(13 min)
sY =--1-
r
= (73ft)
2 =0.02
1o.o ~
10-5 1 10-4 10-3 10 100
10~----~~--------.---------.-------~--------.---------.--------,
4.6
10,000
0.1 t(minutes)
O.D1
0.1 10 100 1000 10,000 100,000 1,000,000
1/ua
Figure 4.5.4. Type-a curve matching for Example 4.5.1.
11uy
1 o-s 10-2 10-1 10 100
10
100 10000 10,000

0.1 t(mlnutes)
10 100 1000 10,000 100,000 1,000,000
Figure 4.5.5. Type-y curve matching for Example 4.5. L

The horizontal hydraulic conductivity, K, or Kh, is computed using
20.16ft 21Jnin
T
Kh =- = =0.806 ft/rrun. or 1160 ft/day
b 25ft 2
and the vertical hydraulic conductivity, Kz or K., is computed using Equation 4.5.4:
2
flb 2Kh (0.06)(25 ft) (1160 ft/day)
K. = - - 2
- = =8.2 ftfday
(73ft) 2
r
•
4.6 UNSTEADY RADIAL FLOW IN A LEAKY AQUIFER

When a leaky aquifer, as shown in Figure 4.6.1, is pumped, water is withdrawn both from the
aquifer and from the saturated portion of the overlying aquitard, or semi pervious layer. Low-
ering the piezometric head in the aquifer by pumping creates a hydraulic gradient within the
aquitard; consequently, groundwater migrates vertically downward into the aquifer. The quan-
tity of water moving downward is proportional to the difference between the water table and
the piezometric nead8• 29• 58
Steady-state flow is possible to a well in a leaky aquifer because of the recharge through
the semipervious layer. The equilibrium will be established when the discharge rate of the
pump equals the recharge rate of vertical flow into the aquifer, provided that the water table
remains constant. Solutions for this special steady-state situation are available,25 · 33 but a more
general analysis for unsteady flow follows.
When pumping starts from a well in a leaky aquifer, drawdown of the piezometric surface
can be given by 19• 21• 25
s = _Q_ W(u,r!B) (4.6.1)

41tT
where s, Q, and r are defined in Figure 4.6.1, and again
r2S
u=- (4.6.2)
4Tt
The quantity riB is given by
r r
=
B ~TI( K'lb') (4.6.3)
where Tis the transmissivity of the leaky aquifer, K' is the vertical hydraulic conductivity of
the aquitard, and b' is the thickness of the saturated semipervious layer (see Figure 4.6.1). Val-
ues of the function W(u, riB) were tabulated by Hantusb. 19 It can be noted that Equation 4.6.1
has the form of the Theis equation (see Equation 4.4.5); in fact, for a confined aquifer, K' ____,
0, so that B ____, oo and riB ____, 0, thereby reducing Equation 4.6.1 to the Theis equation.
Employing this analogy and the Theis method of solution, Wa1ton64 prepared a family of
type curves for W(u, riB) as presented in Figure 4.6.2. Here values of W(u, riB) are plotted
against llu for various values of riB. On another sheet of logarithmic paper of the same scale,
s versus tis plotted. Superimposing the two sheets while keeping the coordinate axes parallel,
a position is found where most of the data points fall on one of the type curves. Selecting any
convenient match point, values of W(u, riB), llu, s, and t are noted. T is then found from
0 E'XAA
Groun d su rtace
II
t I ,----!.
(adapte
Depart
r Water table Interior
t I -------
-- - -- s -----------
' '
" /
/t-- \p·1ezometnc. su rtace
' \ I
Unconfined aquifer
l
;: - -
_yf} ~I! .~
r~·, I•
~··
i- """ .• ~
-
~uitard
---
---
-
---
- --
--
---
--
- Leaky
aquifer
-- Figure 4.6.1. Well pump-
--
ing from a leaky aquifer.
Equation 4.6.1 and S from Equation 4.6.2. Finally, from the value of riB belonging to the type
curve of best fit, it is possible to calculate K'!b' from Equation 4.6.3; and if b' is known from SOLUT1
field conditions, K' can be evaluated.
10
10 Non equilibrium 0.005 0.001
type curve
0.03 \ ""
~ 0.2 0.15
0.1 0.075 0.05 0.15 0.01
----------1.5
...----------2.0
. - - - - - - - - - - - r i B = 2.5
0.1
0.1
0.01~~~~~-L--~~--L-~--~~--~~~---L~~~~--~~--~~~---L~
10-1 1.0 10 102 1o3 104 105 106 107
O.D1
1/u 0.1
Figure 4.6.2. Type curves for analysis of pumping test data to evaluate storage coefficient and transmissivity of leaky aquifers
(after Walton64 ). Figure 4.6..:
·'
EXAMPLE 4.6.1 A well pumping at 600 ft3/min fully penetrates a confined aquifer overlain by a leaky confining layer of
14-ft thickness. Using the tabulated time-drawdown data for an observation well40 ft away from the
(adapted from U.S. pumping well, estimate the transmissivity and storage coefficient of the confined aquifer, and the per-
Department of the meability of the aquitard. Assume that the confining layer does not release water from storage.
Interior)62 .
Time (min) Drawdown (ft) Time (min) Drawdown (ft)

0 0.00 80 12.02
2 5.65 90 12.26
4 6.96 100 12.33
6 7.72 110 12.37
8 8.00 120 12.41
10 8.71 150 12.69
15 9.47 180 12.85
20 9.99 2l0 13.09
25 10.35 240 13.13
30 10.70 270 13.25
40 11.14 300 13.33
50 11.46 360 13.37
Jmp-
60 11.62 420 13.41
fer.
70 11.86
1e type
1 from
SOLUTION The time-<lrawdown field data were superimposed on the family type curves for leaky aquifers (Figure
4.6.3). Comparison shows that the best fit occurs for riB;;; 0.03. The coordinates of the match point
selected are
100.00.-------.....------,----------,
1.00'---------'--------.....__ _ _ _t---~
------1-------10-----100----~ 1000
nme (minutes)
0.1
0·0 6.L1_ _
__JfJJ_.L.L.__ _ j __ _ _ _ _ _.,o______,....oo------,o-'-o-o-----,:--'
o.ooo
1/u
!fS
Figure 4.6.3. Leaky type curve matching for Eltample 4.6.1.
~ =1000,
u
w(u.jJ= 1.0 Disch
t =59 min, s = 1.93 ft

_;~
Next we must perfonn the following unit conversions in order to obtain the lransmissivity in units of ~
ft2/day and hydraulic conductivity of the aquitard in units of ftlday for Q = 600 ft3/min = 864,000 ft3/day .· . J
:-.-·.. .
and t = 59 min= 0.041 days. The transmissivity and storage coefficient of the confined aquifer are com- .:··."
puted using Equations 4.6.1 and 4.6.2 rearranged respectively as
3
T=_Q_W(u,r!BJ= 864· 000 [1 /day (1.0)=35,624 ft 2/day
4ru 41t(l.93 ft)
2
4Ttu 4(35, 624ft /day)(0.041 days)( 0.001)
S =2- = 2 =0.00365
r (40ft)
The hydraulic conductivity of the aquitard is computed by rearranging Equation 4.6.3

2
, Tb'(r!B) 2 (35,624ft /day)(l4ft){0.03f
K = 2 = 2 0.28 ftlday Di~
r (40 ft)
• ro
4.7 WELL FLOW NEAR AQUIFER BOUNDARIES AqL

Where a well is pumped near an aquifer boundary, the assumption that the aquifer is of infinite
areal extent no longer holds. Analysis of this situation involves the principle of superposition by
which the drawdown of two or more wells is the sum of the drawdowns of each individual well.
By introducing imaginary (or image) wells, an aquifer of fmite extent can be transformed into
an infinite aquifer so that the solution methods previously described can be applied.
4.7.1 Well Flow Near a Stream

An example of the usefulness of the method of images is the situation of a well near a peren-
nial stream. 16· 22• 30 It is desired to obtain the head at any point under the influence of pumping
at a constant rate Q and to determine what fraction of the pumpage is derived from the stream.
Sectional views are shown in Figure 4.7.1 of the real system and an equivalent imaginary sys- I
tem. Note in Figure 4.7.lb that an imaginary recharge well* has been placed directly opposite ' \
and at the same distance from the stream as the real well. This image well operates simulta- \
\
neously and at the same rate as the real well so that the buildup (increase of head around a \
\
recharge well) and drawdown of head along the line of the stream exactly cancel. This fur-
nishes a constant head along the stream, which is equivalent to the constant elevation of the
stream forming the aquifer boundary. Thus, in the plan view of the resulting flow net, illus-
trated by Figure 4.7 .2, a single equipotential line is coincident with the axis of the stream. The I
resultant asymmetrical drawdown of the real well is given at any point by the algebraic sum of --i~
Eq~
the drawdown of the real well and the buildup of the recharge well, as if these wells were
located in an infinite aquifer.
Examples of hydraulically equivalent aquifer systems bounded by streams with various
configurations are shown in Figure 4.7.3. Note that combinations of both image recharge and
pumping wells are required. 33 For the single stream in Figure 4.7.3a, the steady-state draw- I
I
down at any point (x,y) is given by I
I
I
J
•A recharge well is a well through which water is added to an aquifer; hence, it is the reverse of a pumping well. I
4.7 Well Aow Near Aquifer Boundaries 181
Discharging well Perennial

I Ground surface
~~--y~
·· ---:--'-~..:O.:.:·~:....f_
.. level ·.. .. . ·:.; : ·
units of ... . .·
J ft3/day
:. · · ~~uifer .. ·..·: ·.. .. . ho·=.
re com-
Impermeable
(a)
Zero drawdown 0
boundary ~ 'L".--- Recharging
1\ 1 image well
Buildup component ess'0~ -1-}'
of image well o~"'({l~ _., _., I '-...,
~c~~-- 1
•
Discharging
real well t
0
----
__ - :I1
II
I
' ......... .....
- -
-- I
-- . -·~·---7. -.,...1 r-·--...,.-.~.~-:r-
~:~:J~v~l;:~.~.;~.~ 1.·-:;::-7~·~.~- : :.··,.::
. : . Drawdown ··..:. I 1 . : Aquifer ·:· hQ:
·.-.·: component: :. ::ll ·.-.·=·· · ·.·.. .- ·
:::. of real well: :. 1
.
t: ·. · .:·.·.. · .
r ~ ~
·infinite
1ition by
Impermeable
Ial well. Figure 4.7.1. Sectional views. (a) Discharging
Aquifer thickness h 0 should be
ned into well near a perennial stream (b) Equivalent
very large campared to resultant
drawdown near real well. hydraulic system in an aquifer of infinite areal
(b) extent (after Penis et al. 15)
a peren-
mmping
:stream.
tary sys- \
opposite '
simulta-
uound a
Ibis fur-
m of the
.et, illus-
:am. The
c sum of
!lls were
1 various
.arge and
I
tte draw- I
I
I
I Figure 4.7.2. Flow net for discharging real
I
J. well and recharging image well (after Ferris,
,g well. I et al. 15) .
fogs
(4.7.1)
where (x,., Yw) are the coordinates of the pumped well. Similarly, for the right-angle bound-
aries of Figure 4.7.3b,
2 2
Q [(x-xw) +(y+ Yw(j[(x+xw) +(y- Ywf]
(4.7.2)
s = 41tT In [(x- xw)2 +(y- Yw{j[(x+x.,f + (y+ Ywn
And for the strip aquifer bounded by two straight parallel streams (see Figure 4.7.3c),
1t(Y-Yw) 1t(x+xw)
Q cosh +cosh
s= --ln 2a 2a (4.7.3)
41tT cosh
1t(Y-Yw) - n(x+xw)
cosh---'-----'-.:...
2a 2a
and the angles are expressed in radians. Actually, in Figure 4.7.3c, the image wells extend to
infinity; however, in practice it is only necessary to include pairs of image wells closest to the
real well because others have a negligible influence on the drawdown.
The water level in the wells will draw down initially only under the influence of the
pumped well. After a time the effects of the recharge boundary will cause the time rate of
drawdown to decrease and eventually reach equilibrium conditions. This occurs when recharge
equals the pumping rate, as illustrated in Figure 4.7.4. The total drawdown for equilibrium
conditions can be expressed as
(4.7.4)
in which sr is the drawdown in an observation well near a recharge boundary, sP is the draw-
down due to the pumped well, and s; is the buildup due to the image well (recharge boundary).
The drawdown equation can be written as
•
•
0
•
(a) (b)
~ Real discharging well

o Image discharging well
• Image recharging well Figure 4.7.3. Image
well systems for aquifers
bounded by streams of
0 • -9-•
2a
0 • 0
•
•
0
u. • 0
0
• 0
•
various geometries. (a)
Unidirectional stream
(b) Rectangular stream
(c) Two parallel streams
(d) U-shaped stream
Theoretically, image
wells in (c) and (d)
extend left and right to
(c) (d) infinity.
4.7 Well Flow Near Aquifer Boundaries 183
logs
Drawdown due to
(4.7.1) pumped well
Buildup due to
Effects of recharge image well
:bound- boundary starts ,
5 . _../
Drawdown in obs.
f1_/ boood•~
(4.7.2) s well near a recharge
Figure 4.7.4. Recharge boundary effects on

log t time-ilrawdown curve.
(4.7.3)
(4.7.5)
xtend to
:St to the where Q is the constant pumping rate [L3ff], Tis the transmissivity [L2ff], W(up) and W(u;)
are dimensionless, uP and u; are
e of the
: rate of u =--
r;s and
,2s
U-=-'-
-echarge p 4Ttp 1 4Tt-
ilibriwn
'
in which r; and rP are in [L] and t is time in [T].
The drawdown in U.S. customary units (the gal]on-day-foot system) can be expressed as
(4.7.4)
1e draw- (4.7.6)
mdary).
1.87r;s 1.87r/S
where uP = and u. = -~-
TtP l Tt;
For large values of time, t, the well functions can be expressed as
W(up) =-{).5772 - ln uP (4.7.7)
and W(u;) = -().5772 - ln u; (4.7.8)

This allows Equation 4.7.5 to be simplified to
s, =_iL[-lnuP + lnu;] (4.7.9)

4nT
and Equation 4.7.6 is simplified to
age
l14.6Q [ ]
aquifers s,=-r- -lnuP+!nu; (4.7.10)
rns of
:s. (a)
·earn
Now using the ga1lon-day-foot system with time in minutes
tream
2693r:s
treams u
P
= Tt
(4.7.11)
am
1ge and
i) 2693r/S
~tto U· = ---'-- (4.7.12)
' Tt
Recharge line
Pumped well Image well

--
1----
Pumped
well
~';
er·
Figure 4.'i
Observation well Figure 4.7.5. Definition of
terms for Equation 4.7.15.
The drawdown from Equation 4.7.10 is expressed as
s ::; 114.6Q [ -In

' T
(2693r:sl
Tt
+In(2693r?s)]
Tt
' (4.7 .13)
which simplifies to
s -
r
- 528 [ 'i
- Qlog-
T rp
l (4.7.14)
Rorabaugh50 expressed this equation in tenns of the distances between the pumped well
and the line of recharges as
528Qlog~(4a2 + r; - 4arp cos Br h,

s ::; - - - - - = - - - - - - - - - - - - - (4.7.15)
' T
where a is the distance from the pumped well to the recharge boundary in ft, and Br is the angle
between a line connecting the pumped and image wells and a line connecting the pumped and
observation wells. Refer to Figure 4.7.5 for an explanation of terms.
A 0.5-m diameter well (200m from a river) is pumping at an unknown rate from a confined aquifer (see
Figure 4.7.6). The aquifer properties are T= 432m2/day and S = 4.0 x 10-4. After eight hours of pump-
ing, the drawdown in the observation well (60 m from the river) is 0.8 m. Compute the rate of pumping
and the drawdown in the pumped well. What is the effect of the river on drawdown in the observation
well and in the pumped well?
SOLUTION The following information is given in the above statement: r.. =0.25 m, T= 432m2/day= 5.0 X 10-3 m2/s,
S = 4 x 10-5, t = 8 hr =28,800 s, and s =0.8 m. A recharging image well is placed at the same distance
from the river as the pumped well as shown in Figure 4.7 .6b.
Equation 4.7.5 is used to compute the discharge from the pumped well knowing the above information:
River
- - - 140m - - - - + - 1 - 60 m
G-140 m---.e.-so mr+--..,...- - 200m --.!..
:--~
Pumped
well
Observation
well
--
Image
well
(a) (b)
Figure 4.7.6. Example 4.7.1 system. (a) Well locations (b) Image well location
ion of
7.15.
2 4
u; = r/S = (260) (4xl0 ) = _ x _2
4 69 10
4Tt 4(5xi0-3)(28800)
t7.13)
w(up) =3.79 for up= l.36x w-2 and W(u;) =2.54 for U; =4.69 X 10-2
Thus the discharge is computed using
0.14)
0 8- Q (3 79) Q (2 54)
· - 4~t(S x lo-3 ) · 4~t(5 x lo-3 ) ·
~dwell
so that Q =0.04 m3/s.
The d.rawdown in the pumped well is computed using equation 4.7.5:
2
:4.7.15) r 2S (0.25) (4x !0-4) -8
u =~- =4.34xl0
3
"' 4Tt 4(5 X 10- )(28800)
1eangle
(400)2(4 X 10-4)
ped and U· = -0. 11 )
.!·'' ' 4(5xlo-3 )(28800)
w(uw) =16.38 for uw = 4.39 X w-8 and W(u; )= 1.75 for U; = 0.111
lifer (see
>fpump- Thus the drawdown is
pumpipg
~ervatioo
The effect of the river on the wells is to decrease lhe d.rawdown, so lhe reduced d.rawdown in the obser-
.o-3 m2/s.. vation well is
: distance
Jrmation:
Similarly, in the pumped well, the reduced drawdown is
Sriver = (0.04 )(1.75)=-l.llm

41t Sx w-3.
•
4.7.2 Well FJow Near an Impermeable Boundary

In the case of an impermeable or barrier boundary, water cannot flow across the boundary; no
water is being contributed to the pumped well from the impervious formation. The cone of
depression that would exist for a pumped well in an aquifer of infinite areal extent is shown in
Figure 4.7.7b. Because of a barrier boundary, the cone of depression shown is no longer valid
since there can be no flow across the boundary. Placing an image well, discharging in nature
across the barrier boundary, creates the effect of no flow across the boundary. The image well
must be placed perpendicular to the barrier boundary and the same distance from the bound-
ary as the real well. The resulting real cone of depression is the summation of the components
of both the real and image well depression cones as shown in the Figure 4.7 .7b. Water levels
in wells will decline at an initial rate due only to the· influence of the pumpe~ well. As pump·
ing continues the barrier boundary effects will begin as simulated by the image well affecting
the real welL Wben the effects of the barrier boundary are realized, the time rate of drawdown
will increase (Figure 4.7 .8). When this occurs, the total rate of withdrawal from the aquifer is
equal to that of the pumped well plus that of the discharging image well causing the cone of
depression of the real well to be deflected downward.
The total drawdown in the real well can be expressed as
(4.7.16)
Ground surface
Average or effective position

of line of zero flow
(a)
Drawdown component
:~~:~5:,' :,:fJi[~~~ ;~¢~-~~',,H<>· e.:;j;<ho

:.: ·, .. . . . . . .. .. . • ..
~---a +----a----~
Figure 4.7.7. Sectional views. (a) Discharging
Aquifer thickness h0 should be very large compared well near an impermeable boundary (b)
to resultant drewdown near real well
Equivalent hydraulic system in an aquifer of infi-
(b) nite areal extent (after Ferris et al. 15)
logs
.ary; no Total drawdown s8
:one of
IOWnin
!r·valid
1 nature
Drawdown in obs.
ge well well due to pumping
bound- production well
JOnents
r levels
:pump· Figure 4.7.8. Barrier
ffecting boundary effects on
log t time-drawdown curve.
.wdown
tuifer is
cone of in which sb is the total drawdown, sP is the drawdown in an observation well due to pumping
of the production well, and S; is the drawdown due to the discharging image well (barrier
boundary). The total drawdown can be expressed as
(4.7.16) (4.7.1 7)
where Q is the constant pumping rate [L3rr], Tis the transmissivity [L2fr], W(up) and W(u;)
are dimensionless, and uP and u; are
2
u
r
= -P-
S
and U __
I_r..Zs
P 4Ttp ; - 4Tt.
I
in which r; and rP are in [L] and tis time in [T].

Drawdown Equation 4.7.16 can also be expressed in U.S. customary units (gal-day-foot)
system where S is in ft, Q is in gpm, Tis in gpd/ft, r is in ft, and t is in days:
_ 114.6Q w( u )+
sb - - -
114.6Q
-- w( u- )
T p T I
(4.7.18)
11 6
= ~ Q [w(uP)+ w(u;)]
where u = 1.87 r; S and u. = l.S 7'/S

p Ttp I Tt;
Now suppose that we choose drawdowns at times tP and t; such that sP = s;. then W(uP) =
W(u;) and uP=u;. Then
which reduces to
(4.7.19)
targing Equation 4. 7.19 defines the law of times which states that for a given aquifer, the times of
occurrence of equal drawdown vary directly as the squares of distances from an observation
:r of infi· well to a production well and an image well of equal discharge. The law of times can be used
to determine the distance from an image well to an observation well, using
(4.7.20)
in which r; is the distance from the image well to the observation well in ft, rP is the distance
from the pumped well to the observation well in ft, tP is the time after pumping started and
before the barrier boundary is effective, and ti is the time (after pumping started and after the
barrier boundary becomes effective) where sP =s;. 4.7.3
A well is pumping near a barrier boundary (see Figure 4.7.9) at a rate of 0.03 m3/s from a confined
EXAMPLE 4.7.2
w-
aquifer 20m thick. The hydraulic conductivity of the aquifer is 27.65 rnlday and its storativity is 3 x 5•
Determine the drawdown in the observation well after 10 hours of continuous pumping. What is the frac-
tion of the drawdown attributable to the barrier boundary?
SOLUTION The following information is given in the above problem statement: Q = 0.03 m3/s, b =20m, K = 27.65
mlday =3.2 x 10-4 rn!s, S =3 x w-5, t = 10 hrs =36,000 s. An image well is placed across the boundary
at the same distance from the boundary as the pumped well (as shown in Figure 4.7.9b). The drawdown
in the observation well is due to the real well and the imaginary well (which accounts for the barrier
boundary). Hence, using Equation 4.7. 17
Next compute the distance from the observation weJI to the image well: r/ = 6002 + 2402 -
2(600)(300) cos 30° = 16,y85 m2 so r; =410 m. Using r;, compute
168185(3xlo-5 ) _
U·= ( )
' 4(20) 3.2x10-4 {36,000)
=5.47 X 10 3
The well functions are now computed or obtained from Table 4.4.1 as W(up) = 5.72 for uP= 1.88 x w-3
and W(u1) = 4.64 for u; =5.47 x 10-3.
The drawdown at the observation well is computed as
03
s= ('
41t(20) 3.2xl0
-4) =(5.72 +4.64) =3.86 m.
The drawdown attributable to the barrier boundary is computed as
Q ( ) 0.03
s;=-Wu1 = ( -4)(4.64)=1.73m
41tT 41t(20) 3.2xl0
Barrier
Barrier
boundary
--- --boundary
--- . . .'!...
~
-- -- ---
1 - - - - - 300m - - - - + f . < - - - 300m--~
(a) (b)
Figure 4.7.9. Example 4.7.2 system. (a) Well locations (b) Image well location
and the fraction of drawdown attributable to the impermeable boundary is
!.!._= 1.7) =0.45(45%).

·.7.20)
stance
s 3.86
•
!d and
ter the
4.7.3 WeJJ Flow Near Other Boundaries
In addition to the previous two examples, the method of images can be applied to a large num-
mfined ber of groundwater boundary problems. As before, actual boundaries are replaced by an equiv-
X 10-5. alent hydraulic system, which includes imaginary wells and permits solutions to be obtained
1e frac- from equations applicable only to extensive aquifers. Two boundary conditions to suggest the
adaptability of the method are shown in Figure 4.7.10. Figure 4.7.l0a shows a discharging well
in an aquifer bounded on two sides by impermeable barriers. The image discharge wells 11 and
= 27.65 ! 2 provide the required flow but, in addition, a third image welll3 is necessary to balance draw-
mndary
downs along the extensions of the boundaries. The resulting system of four discharging wells
.wdown
irian extensive aquifer represents hydraulically the flow system for the physical boundary con-
barrier
ditions. Finally, Figure 4.7.lOb presents the situation of a well near an impermeable boundary
and a perennial stream. The image wells required follow from the previous illustrations.
For a wedge-shaped aquifer, such as a valley bounded by two converging impermeable
barriers, the drawdown at any location within the aquifer can be calculated by the same method
of images. 54 Consider the aquifer fonned by two baniers intersecting at an angle of 45 degrees
shown in Figure 4. 7.11. Seven image pumping wells plus the single real pumping well form a
circle with its center at the wedge apex; the radius equals the distance from the apex to the real
pumping weU. 15 The drawdown at any point between the two barriers can then be calculated
by summing the individual drawdowns. In general, it can be shown that the number of image
wells n required for a wedge angle e is given by
360°
n=---1 (4.7.21)
e
8 X 10-3
where eis an aliquot part of 360 degrees.
If the arrangement of two boundaries is such that they are parallel to each other, analysis
by the image-well theory requires the use of an image-well system extending to infi.nity. 15 Each
successively added secondary image well produces a residual effect at the opposite boundary.
Figure 4.7.10. Image well

systems for a discharging well
near aquifer boundaries. (a)
Aquifer bounded by two
impermeable barriers intersect-
ing at right angles (b) Aquifer
~ Discharging bounded by an impermeable
real well barrier intersected at right
Aquifer angles by a perennial stream.
Open circles are discharging
image wells; filled circles are
recharging image wells (after
(a) (b) Perris et ai. 15).
t
A Impermeable
barrier
B
.
+--- ~-
~
Q)
c.
~
E
CD
~ 0
c.
(I)
~ 0
Figure 4.7.11. Image E
well system for a dis-
charging well in an
aquifer bounded by two •
impermeable barriers
Open circles signify

intersecting at an angle •
of 45 degrees (after
discharging wells
Ferris et al. 1\
It is only necessary to add pairs of image wells until the next pair has negligible influence on
the sum of all image-well effects out to the point. The use of spreadsheets makes the compu-
tation much easier. Image-well systems for several parallel-boundary situations are shown in
Figures 4.7.12 and 4.7.13.
Discharging
real well 4.7.4
·~i
~
)
Ill
'0
«1::l c
- Image pattern repeats to infinity coo
.Q
:::J
0 Image pattern repeats to infinity -
.0
0
• • 0 Q)
e> • • 0 0
Is 16 14 12 a! /1 13 Is 17
.r:::.
Aquifer u
CD
a:
Image wells,/, are numbered in the sequence

In which they were considered and located.
o Signifies discharging well
• Signifies recharging well Discharging
real well
0 0 0 0 0 0 0 0 0 0 0 0
-Image pattern repeats to infinity Image pattern repeats to infinity-
Barrier boundary Figure 4.7.12. Plans

I I of image-well systems
I I
I I for selected parallel-
0 0 0 0 0 0 o I o I o o 0 0 0 0 boundary situations
I I
I I (after Ferris et al. 15).
I
0 0
• • 0 0
• I • 0 0 • • 0 0
I
I
0 0
• • 0 0
• I • 0 0 • • 0 0
t Barrier
I
I
I
I
boundaries
• • 0 0
• •
i • .
Ql
~
E 1 Recharge boundaries
Ql
~ 0 0 •• 00 ,1,
I
00 el 00
0.
Ql I
g 0 0 • • 0 0 ,I.
I
0 0 • • 0 0
mage E I I
l dis- -+-- Image pattern repeats to infinity -
t I
an I I
by two • • o o 1 • ol o
J
I
I
• • o o • •
riers I I
1 angle • • o o 1 • ol o
I
I
I
• • o o • •
fter Figure 4.7.13. Plans of image-well systems
o Signifies discharging well • Signifies recharging well for a rectangular aquifer (after Ferris et al. 15 ).
ence on The above equations involve T for confined aquifers. To adapt them to unconfined
compu- aquifers, s should be replaced by s'' =s- Sl!2h0 where ho is the initial saturated aquifer thick-
hown in ness. Storage coefficients cannot be calculated from steady-state boundary equations.
Procedures for analyzing unsteady well flows near aquifer boundaries, involving graphic
solutions, are available.15· 21
4.7.4 Location of Aquifer Boundary

Permeable aquifer boundaries such as streams would normally be visible near a pumping
well; however, impermeable subsurface boundaries such as faults or dikes may not be appar-
ent. Where this situation is encountered, the location and orientation of such a barrier can be
defined by careful analysis of pumping test data. 15• 61 In the Cooper-Jacob method (see Equa-
tion 4.4.8) the slope of the straight line on semilogarithmic paper depends only on the pump-
ing rate and the transmissivity. If an impermeable boundary is present, the rate of drawdown
in an observation well will double under the influence of an image pumping well (see Figure
4.7.14a).* To determine the location of the image well, straight lines are fitted through the two
legs of the data- An arbitrary drawdown sA is selected and a timet, for this to occur under the
influence of the real well is measured (see Figure 4.7.14a). Similarly, a timet; for the same
drawdown to be produced by the image well is defined. Then, knowing the distance rr between
the real well and the observation well, the distance r; to the image well (see Figure 4.7 .14b) can
be found from the law oftimes (Equation 4.7.20). The distance ri defines only the radius of a
circle on which the image well lies. It requires measurements in two more observation wells
in order to define uniquely by intersection of three arcs the location of the image well (see Fig-
ure 4.7.14b). The boundary then lies at the midpoint of and perpendicular to a line connecting
~lans
the real and image wells.
tstems
tllel-
·It should be noted that if the boundary is a stream recharging the aquifer, an image recharge well is introduced.
ons
1.15).
This produces a slope of equal but opposite sign on the drawdown curve, resulting in a horizontal asymptote.
1.0
~ 1*-------1;--------~
~ -----
-----
E
josL
3:
ctl
t,---J
o I
I
Composil
curve f1
wells
Time, t, min
(a)
Image
pumping ---------• Obs. well2
well I
I
I
I
-- r, ~ 1
"'.__ '-, -
--+1Obs. well 1
I
'a
' . . . . ,,...rl
Unknown impermeable _@
aquifer boundary , --- Pumping Figure 4.7.14. Diagrams illustrating procedure to
'...----- well locate an unknown impermeable aquifer boundary
Obs. well3 near a pumping well. (a) Cooper-Jacob drawdown
curve showing effect of an impenneable boundary
(b) Field situation required to locate an unknown
(b) impermeable aquifer boundary
4.8 MULTIPLE WELL SYSTEMS

Where the cones of depression of two nearby pumping wells overlap, one well is said to SOLUT[(
interfere with another because of the increased drawdown and pumping lift created. For a
group of wells forming a well field, the drawdown can be detennined at any point if the well
discharges are known, or vice versa. From the principle of superposition, the drawdown at any
point in the area of influence caused by the discharge of several wells is equal to the sum of
the drawdowns caused by each well individually. Thus,
(4.8.1)
where sr is the total drawdown at a given point and s1, s?J s3, ... , snare the drawdowns at the
point caused by the discharge of wells 1, 2, 3, ... , n, respectively. The summation of draw-
downs may be illustrated in a simple way by the well line of Figure 4.8.1; the individual and
composite drawdown curves are given for Q1 = Q2 = Q3. Clearly, the number of wells and the
geometry of the well field are important in detennining drawdowns. Solutions can be based on
the equilibrium or nonequilibrium equation. Equations of well discharge for particular well
patterns have been developed. 41 · 49
In general, wells in a well field designed for water supply should be spaced as far apart as
possible so their areas of influence will produce a minimum of interference with each other. On
4.8 Multiple Well Systems 193
Q2 Ground surface
..---..,.---=-...., ,__ _~t..--.......,......_, ..,....r;,._....,....-...,...........~
£Original piezometric
- - - - - ---===:::
-------
- ---
---- --- ---
,--
'
......
><
/
/
' '\
I
Composite drawdown
\
\ Drawdown curve
for Q3 only
curve for all three
wells pumping
Impermeable
Confined aquifer
Figure 4.8.1. Individual and composite
--- - .-. . .:. drawdown curves for three wells in a line.
the other hand, economic factors such as cost of land or pipelines may lead to a least-cost well
layout that includes some interference. 21 For drainage wells designed to control water table ele-
vations, it may be desirable to space wells so that interference increases the drainage effect.
Three pumping wells located along a straight line are spaced at 200m apart. What should be the steady-
EXAMP/.E 4.8.1
state pumping rate from each well so that the near steady-state drawdown in each well will not exceed
2 m? The transmissivity of the confined aquifer, which all the wells fully penetrate, is 2400 m2/day and
all the wells are 40 em in diameter. The thickness of the aquifer is 40 m and the radius of influence of
each well is 800 rn.
:dure to
Jundary
awdown
)Undary
10 20 30
known
I· 200m
)I· 200m
~I
SOLUTION The following information is given in the above problem statement: s~ ~ 2m, s2 ~2m, and s3 ~ 2m, T=
said to
2,400 m2/day = 27.8 x 10-3 m2/s, rw = 0.2 m, b "' 40 m, r0 = 800 m, and r = 200 m. Let Q be the pump-
I. For a ing rate from each well and h0 be the head before pumping started. For well I, s~ = s 11 + s 12 + s13 where
the well su is the drawdown in well i due to pumping in we!Ij. Thus, for the other wells, s2 = s21 + s22 + s23, and
nat any s3 = s3 ~ + s32 + s33' By symmetry, s~ = s3. The drawdowns in well! due to pumping in wells l, 2, and 3
sum of are respectively
(4.8.1) Qln(~J Qm(~~) 47.48Q

1S at the s, 1
= 27tT = 21t(27.8xl0-3 )
>f draw-
lual and
and the
Qln(~J- QlnG:J 1.94Q
>ased on 21tT 27t( 27.8 x w-3 )
1lar well
Qln(;,oJ Qln(800)
apart as s13 = ll3 = 400 ;: 3.97Q
27tT 21t(27.8x10-3 )
•ther. On
The drawdowns in wells I and 3 are identical so total drawdown in the wells is s1 = s3 = 47.48Q + 7.94Q
+ 3.97Q = 59.39Q. The drawdowns in we112 due to pumping in wells I, 2, and 3 are respectively
Qlnl( ;o J Qtn(800J
s, = 12 = 200 = 7.94Q
·' 2rr.T 2rr.(27.8x w-3)
s22 = s11 = 47.48Q
s23 =s21 = 7.94Q
The total drawdown in well 2 is s2 = 7.94Q + 47.48Q + 7.94Q::: 63.36Q. The relationships for s 1 =
59.39Q and s 2 ::: 63.36Q show that for the same discharge from all the wells, more drawdown results at
the middle well; therefore, the drawdown in this well governs. So using s2 :S: 2 or 63.36Q :S: 2, then the
Steady-State pumping rate from each well should be Q :S: 3.16 X 10-2 m3/s::: 113 m3/hr. •
. .
EXAMPU.' 4.8.2 It is required to dewater a construction site 80 m by 80 m. The bottom of the construction will be 1.5 m below
the initial water surface elevation of90 m. Four pumps are to be used in 0.5-m diameter wells at the four cor-
ners of the site. The transmissivity and the storage coefficient of the aquifer are 1,600 m2/day and 0.16,
respectively. The site needs to be ready after one month of pumping. Detennine the required pumping rate. 4.9 l
SOLUTION To solve this problem, the least drawdown at the site should be greater than 1.5 m. It can be shown that the
potential points of interest that may have the least drawdown are the center of the square (point a) and the
midpoint on each side of the square (points b). Approximation is made using the Cooper-Jacob method.
b
r:-----< }----:"l 2
.
•.. a .•·
·.o:..... b
4 3
b
At point a (the center of the square),
r S
u=-=
2
4Tt
r::: ~402 +40 2 = 56.6 m, and
(56.6 m?(O.l6)
4(1600 m2/day)(30 days)
=0.00267
Since u < 0.01, we can use the approximate solution by Cooper-Jacob expressed by Equation 4.4.7:
-
-
sa =L(-0.5772-ln(u))= ( Q ){-0.5772-ln(0.00267))=0.0002661Q
41tT 41t 1600 m2/day
Using the principle of superposition and by symmetry, the drawdown caused by the four wells is sr = 4
X sa= 4 X 0.0002661Q = 0.0010643Q and Sr = 0.00to643Q = 1.5 m ~ Q =1409 m /day.
3
At any of the four points represented by b, r 1 = 40 m for two of the wells and r2 = ~80 2 + 402
-
= 89.44 m for the remaining two wells. Then
2 2
rS (40 m) (0.16)
u, =-:;:; 0.0013333
4Tt 4(1600 m2 /day)(30 days)
u2 = r2 S = (89.44 m)2 (0.16) = 0.006666
Figure'
4Tt 4(1600m 2/day)(30days) urations
4.9 Partially Penetrating Wells 195
7.94Q Since u1 < 0.01 and u2 < O.Ql, the Cooper-Jacob method of solution can be used again:
sb = 2[
4;T (-0.5772 -ln( u 1))] + 2[
4;T (-0.5772 -ln( u2 )) ]
=2[ ( Q )(-0.5772-ln(0.0013333))]+
41t 1600 m2/day
J( Q 2 ) ( -Q5772 -In(0.006666))]
141t 1600 m /day
·or s 1 =
:sults at = 2 X 0.0003Q + 2 X 0.0002205Q
hen the :~~ =1.041 x 10-3 Q=1.5 m -1 Q= 1441 m3/day
• Thus the points represented by bare critical and a discharge of 1,441 m3/day from each well is required .
n below
our cor-
•
ld 0.16,
.ng rate. 4.9 PARTIALLY PENETRATING WELLS
A well whose length of water entry is less than the aquifer it penetrates is known as a partially
that the penetrating well. Figure 4.9.1 illustrates the situation of a partially penetrating well in a con-
and the fined aquifer. The flow pattern to such wells differs from the radial horizontal flow assumed
tethod. to exist around fully penetrating wells. The average length of a flow line into a partially pene"
trating well exceeds that into a fully penetrating well so a greater resistance to flow is thus
encountered. For practical purposes this results in the following relationships between two
similar wells, one partially and one fully penetrating the same aquifer: if QP =Q, then sP > s;
and if sP =s, then QP < Q. Here Q is well discharge, sis drawdown at the well, and the sub-
script p refers to the partially penetrating well. The effect of partial penetration is negligible on
the flow pattern and the drawdown beyond a radial distance larger than 0.5 to 2 times the sat-
urated thickness b, depending on the amount of penetration.
4.7:
b
is sr= 4
Confined aquifer
(1) (2J
(a) (b)
Figure 4.9.1. Partially penetrating wells in a confined aquifer. (a) Effect of partially penetrating well on drawdown (b) Two config-
urations of partially penetrating wells
The drawdown sPat the well face of a partially penetrating well in a confined aquifer (see
Figure 4.9.la) can be expressed as
sP=s+tis (4.9.1)
where !Js refers to the additional drawdown resulting from the effect of partial penetration. It
can be shown for steady-state conditions and the typical situation· in Figure 4.9.lb(1),28
(4.9.2)
where Tis transmissivity, pis the penetration fraction (p =hjb), and hs and rware shown in (a) Well
Figure 4.9.lb( 1). Equation 4.9.2 applies where p> 0.20.
For the case of a well screen centered in the thickness of the aquifer [see Figure 4.9.1 b(2)],
the value of~ is given by
Figure •
!Js = QP 1- p In (1- p )hs
2nT p 2rw (4.9.3)
Equation 4.9.2 can be modified for a well in an unconfined aquifer by defining
tis2h = QP I- p ln (l- p)hs

w
nK p rw (4.9.4)
where hw is the saturated thickness at the well with full penetration and the hydraulic conduc-
tivity K =Tfhw Then
(4.9.5)
and similarly for Equation 4.9.3.
Detailed methods for analyzing effects of partial penetration on well flow for steady and
unsteady conditions in confined, unconfined, leaky, and anisotropic aquifers have been out-
lined by Hantush23• 27 and others.32• 34· 57 Although evaluating the effects is complicated except
for the simplest cases, common field situations often reduce the practical importance of par-
tial penetration. t One occurs where an observation well is located more than 1.5 to 2 times the
saturated aquifer thickness from a pumping well; in this situation the effect of partial penetra-
tion can be neglected for homogeneous and isotropic aquifers. Another applies to many allu-
vial aquifers with pronounced anisotropy. Here the vertical flow components become small,
thereby enabling a pumping well to be approximated as a fully penetrating well in a confined
or leaky aquifer with a saturated thickness equal to the length of the well screen.
Compare the four cases that are depicted in Figure 4.9.2 where a 1-m diameter well fully/partially pen-
etrates vertically through a confined aquifer whose thickness is 28 m. Comment on their relative effi-
ciencies by evaluating the specific capacity (Q/s) for each case. Take the radius of influence for all cases
as 150m. The pumping rates and the aquifer transmissivity are the same for all cases.
SOLUTION Case (a): The drawdown in this case where the pumping well fully penetrates is computed by rearrang-
ing the Thiem equation (4.2.4):
s= _Q_[ln[!!_Jl 150
= _g_ (In( j'J= 0.9078( Q)
2ttT rw 2rtT 0.5 T
'The drawdown increment is the same whether partial penetration starts from the top or from the bottom of the
aquifer.
t It should be noted that any well with 85 percent or more open or screened hole in the saturated thickness may be
considered as fully penetrating.
4.9 Partially Penetrating Wells 197
er (see
(4.9.1)
tion. It
28 28m ~~
~ 14m
28m
l_l
(4.9.2)
ownm (a) Well fully penetrating (b) Well penetrating the top (c) Well penetrating 14 m {d) Well penetrating 14m
14 m of the confined of the confined aquifer of the confined aquifer
aquifer with the well screen with two 7 m screened
lb(2)], centered intervals
Figure 4.9.2. Example 4.9.1 cases
(4.9.3) Case (b): The additional drawdown is computed by rearranging Equation 4.9.2:
l:ls =[.&_Il- plln(1- p)hs ]~[_g_ll-0.5l1 n (1-0.5)(14)]

2nT p r.., 2nT 0.5 -l (0.5)
(4.9.4) =0.4201~)
onduc-
So the total drawdown becomes sb (0.9078 + 0.4200)(Q/1) =1.3278 (Q/TJ.
Case (c): The additional drawdown in the case of a well screen centered in the thickness of the aquifer
(4.9.5)
is given by Equation 4.9.3:
tdy and l:ls =[.&_I.!..:.E.Iln(1- p)hsJ=

2nT p 2rw
[_g_I
2rtT
1- 0.5lln (1 - 0.5)(14)]
0.5 2(0.5)
en out-
except
of par- == 0.3097( ~)
nes the
The total drawdown for this case becomes sc (0.9078 + 0.3097) (Q/1) =1.2175 (QIT)
•enetra-
ty allu- Case (d): This case is equivalent to a screen length of 7 m centered in a 14-m thick aquifer. Again using
: small, Equation 4.9.3,
Jnfined
l:ls == [.&_I.!..:.E_IIn{1- p)hs]== [_g_I 1- 0.Siln (1- O.SX7)]
27tT p 2rw 21tT 0.5 2(0.5)
illy pen- = 0.1994( ~)

jve effi-
all cases The total drawdown for this case becomes sd (0.9078 + 0.1994) (Q/1) =1.1072 (Q/1).
Calculate the specific capacity (Qis) for each case:
Q/sa =1.1016(1)
~ang-
Qlsb = 0.7531(1)
Qfsc = 0.8214(1)
=
Q/sd 0.9032(1)
Or taking case (a), fully penetrating well, as the maximum potential specific capacity, cases (b), (c), and
the (d) yield 68 percent, 75 percent, and 82 percent of the maximum capacity, respectively. So among the
three cases, case (d) is the most effective one. Also, centering the well screen yields higher capacity
nay be compared to the case where the same partial penetration starts from the top or from the bottom of the
'
·: aquifer. •
4.10 WELL FLOW FOR SPECIAL CONDITIONS

A variety of solutions to well flow problems have been derived for special aquifer, pumping,
and well conditions. 23 ·38 Inasmuch as these are of less general application than those outlined
heretofore and involve more extensive mathematical treatment, they will be omitted here. It is
worth noting, however, that solutions have been obtained for the following special conditions:
1. Constant well drawdown 2•21
2. Varying, cyclic, and intennittent well discharges20• 36 · 39• 52- 55 - 56
3. Sloping aquifers21
4. Aquifers of variable thickness 21
5. Two-layered aquifers24•44
rial of'
6. Anisotropic aquifers 10· 23 •27•43• 66 surface
7. Aquifer conditions varying with depth40 •51 m, resp<
voir to t
8. Large-diameter wells46• 67
4.1.9 0
9. Collector wells (see Chapter 5) 28 · 45 into the
10. Wells with multiple-sectioned well screens57 the char
13 and 1
tionA i~
PROBLEMS
4.1.10/J
4.1.1 A confined aquifer is 18.5 m thick. The potentiometric sur- 4.1.3 Two observation wells are to be constructed to determine the fully pel
face elevations at two observation wells 822 m apart are 25.96 and natural slope of the piezometric surface in a confined aquifer. shown i
24.62 m. If the horizontal hydraulic conductivity of the aquifer is Assuming a steady unidirectional flow, if the accuracies of piezo- ductivi~
25 m/day, determine the flow rate per unit width of the aquifer, metric head measurements and horizontal distance measurements tion of 1
specific discharge, and average linear velocity of the flow as sum- are ±3 em and ±5 em, respective! y, determine the percent error in m andir
ing steady unidirectional flow. estimating the slope of the piezometric surface as a function of the the river
4.1.2 Two confined aquifers are separated by an aquitard as shown distance between the two wells. What would be the accuracy of a taminati
below. The piezometric head difference between the upper and hydraulic gradient estimate if a piezometric head drop of 3 m is . the cana
lower aquifer measured along a vertical line is 6.5 m. If the verti- observed between two wells 100m apart? travel ti1
cal hydraulic conductivity of confining unit is 0.046 m/day, deter- 4.1.4 Rework Problem 4.1.1 if a ±10° error is also possible in (ne = 0.:
mine the direction and magnitude of leakage per km2 between the estimating the groundwater flow direction. advectio
upper and lower confined aquifers through the confming unit. 4.1.5 Near steady-state conditions, explain how the hydraulic gra- prevent 1
Also, estimate the travel time of a water particle through the con- dient changes in the flow direction in 4.2.1A ·
fining layer between the two aquifers. Estimated thickness of the trates a c
(a) a confined aquifer;
separation is 4.15 m. pumpini
(b) an unconfined aquifer.
at two ol
4.1.6 Consider two strata of the same soil material that lie between 0.9 and(
two channels. The first stratum is confined and the second one is
ity and 1
unconfined, and the water surface elevations in the channels are
influenCt
24 and 16 m above the bottom of the unconfined aquifer. What drawdov;
should be the thickness of the confined aquifer for which
l) the discharge through both strata are equal,
2) the discharge through the confined aquifer is half of that
through the unconfined aquifer?
4.1.7 An unconfined aquifer in a stratum of clean sand and gravel
has a hydraulic conductivity of 10-2 em/sec. From two observa-
[
tion wells 200m apart, the observed water table elevations are 11
.
and 7 m measured from the bottom of the stratum. Determine the ·.
discharge per unit width of the aquifer. .
4.1.8 An earthen darn is 200 m across (i.e., the distance from the
upstream face to the downstream face) and underlain by imper-
. ·. .. · meable bedrock. The average hydraulic conductivity of the mate- - - _,
Problems 199
1ping,
tlined
~. It is
.tions:
Figure to accompany Problem

4.1.9
rial of which the dam is composed is 0.065 m/day. If the water 4.2.2 A confined aquifer of lO·m thickness and 16.43 m/day
surface elevations in the reservoir and the tailwaters are 25 and 4.5 hydraulic conductivity is fully penetrated by a pumping well of
m, respectively, estimate the magnitude of leakage from the reser- 0.5 m radius operating at Q = 425 m3/day. Determine the draw-
voir to the tailwaters per 100-m width of the dam. down under steady-state conditions in the pumping well and 50 m
4.1.9 Compute the volume of water that seeps from the channel away from the well. Take the radius of influence of the pumping
into the river in the figure above. The water surface elevations in as 300m.
the channel and river with respect to the underlying bedrock are 4.2.3 What percent increase/decrease would occur in the draw-
13 and 10.5 m, respectively. The hydraulic conductivity of forma- down of the pumping well if the radius of the well is doubled and
tion A is 5.6 m/day and that of formation B is 12.3 m/day. the pumping rate is kept the same in Problem 4.2.2? Assume the
same radius of influence.
4.1.10 A canal is constructed parallel to a river 460 m away both
tine the fully penetrating an unconfined aquifer of clean sand and gravel as 4.2.4 What percent increase/decrease would occur in the well flow
tquifer. shown in the illustration below. The aquifer has a hydraulic con- if the well diameter is doubled and the drawdown in the well is
· piezO:- ductivity of K = 18.5 m/day and is subject to an average infiltra- kept constant in Problem 4.2.2? Assume the same radius of influ-
em~nts tion of 1.6 m/year. The water surface elevation in the canal is 8.5 ence.
~rror·.in m and in the river it is I 0 m. If the mound between the canal and 4.2.5 The initial piezometric surface in a confined aquifer of 20-m
10f the the river gets contaminated and the river is to remain free of con- thickness is 34 m above the bottom. After a long period of pump-
tcy o(a tamination, (a) det~rmine the daily discharge of groundwater into ing, the piezometric surface stabilizes at 29.3 m above the bottom.
3m is the canal and into the river per kilometer of both; (b) estimate the The hydraulic conductivity of the aquifer is 12.2 m/day. If the
travel times from the water divide to the canal and to the river radius of the well is 0.5 m and the radius of influence of the pump-
;ible in (ne = 0.35); (c) assu'fuing that the contaminant travels mainly by ing is 500 m, what is the steady-state well discharge?
advection, propose any operationiu changes to the given layout to 4.2.6 The initial piezometric head in a confined aquifer that has a
lie gra- prevent the river from being contaminated. thickness of 11.6 m is 85.7 m above sea level. A well with a radius
4.2.1 A well that.pumps at a constant rate of 0.5 m3/s fully pene- of 0.5 m pumps at a constant rate of 1,240 m3/day. After the cone
trates a confined aquifer of 34-m thickness. After a long period of of depression has achieved equilibrium, the piezometric heads at
pumping, near steady-state conditions, the measured drawdowns two observation wells 40 and 95 m from the pumping well are
at two observation wells 50 and 100m from the pumping well are measured as 78.9 and 83.4 m above sea level, respectively. Also,
etween 0.9 and 0.4 m, respectively. (a) Calculate the hydraulic conductiv- the piezometric head in the pumping well during equilibrium
I one is ity and transmissivity of the aquifer, (b) estimate the radius of remains at 46.6 m. Determine (a) the aquifer transmissivity, (b)
tels are influence of the pumping well, and (c) calculate the expected the radius of influence of the pumping, and (c) the total well
'· What drawdown in the pumping well if the radius of the well is 0.4 m. losses in and around the pumping well.
W = 1.6 m/year
of that
~ ~ ~ ~ ~ ~ ~
I gravel
bserva- -~. ~----~-
· ...
; are l1
1ine the
·om the
imper-
~ '"'_!_~~~~~ ---=--.- 1 - ---=--.
~ ·-=- -____L._~~~~~~
Figure to accompany Problem
e mate-
.l
~-----1--
-- _,_
'
~:--1---
1~ - - · - ~ -
1 Impermeable bedrock
==:_...~.._-
·-=- ~ - _ ,,_,.,
_I
-==-·_.1.._------~--
_.L...._ -. ·-
-- -r=::::===-
-==..----J--
~ ----- . -
T
4.1.10
4.2.7 A pumping well of0.75-m radius fully penetrates an uncon- m/day. A capture well is proposed to clean up the contamination. 4.4.5
fined aquifer of 24-m thickness and produces at a rate of 10 Us. The coordinates of the limits of contamination with respect to the fined
After a long period of pumping, the drawdown in an observation capture well are tabulated below. Determine the minimum required from
well 30 m from the pumping well is 1.6 m. The drawdown in pumping rate for the cleanup and delineate the capture zone. data f
another observation well 60 m from the pumping well is 1.1 m. Assume that the plume remains relatively still before or during the aquif1
Calculate (a) the hydraulic conductivity of the aquifer, (b) the operation of the capture well. Note that the capture well is located
expected drawdown in the pumping well, and (c) the radius of at (0 m, 0 m).
influence of the pumping well.
x (m) Y (m)
4.2.8 After a long period of pumping from an unconfined aquifer
at a constant rate of 850 m3/day, the cone of depression reaches -15 35
equilibrium. The aquifer has an initial saturated thickness of 20 m -29 9.5
and a hydraulic conductivity of 8.65 m/day. During the equilib- -20 -28.8
rium, the water levels in an observation well 50 m away and in the 5 -26
pumping well are measured as 18.4 and 9.9 m. Determine (a) the 38 -29
radius of influence of the pumping, (b) the radial distance where 62 -1
the steady state drawdown is 5 em, (c) the expected drawdown in 14.2 33.4
the pumping well (rw = 0.4 m), and (d) the total well head losses.
-15 35
4.2.9 Water is pumped at a constant rate of 500m3/day from an
4.4.1 A fully penetrating production well pumps from a confined
unconfined aquifer whose thickness is 15 m and hydraulic con-
aquifer at a constant rate of 64 Us. If the coefficients of trans-
ductivity is 5.5 m/day. Given that the steady-state drawdown
missivity and storage of the aquifer are 1,240 m2/day and 4 x
50 m from the pumping well is 2.5 m, plot the water table profile
1~, respectively, estimate the drawdowns at a distance of200 m
under steady-state conditions for r > 1.5H, where His the initial
from the pumping well for pumping periods of 8 hours, 30 days,
saturated thickness of the aquifer.
and 6 months using the Theis equation. Also, estimate the radius
4.2.10 Rework Problem 4.2.5 in the absence of the recharge. (Hint: of influence of the pumping well after 1 hour of pumping. 4.4.6
You may have to guess the radius of influence for this problem.)
4.4.2 Rework Problem 4.4.1 using the Cooper-Jacob method of aquife
4.2.11 Rework Problem 4.2.5 with the recharge rate doubled (i.e., solution. comp1
W = 0.12 m/day). Approximately what percentage of the appro;
4.4.3 Given that T = 125 m 2/day, S = 1~. t = 2,693 min, Q =
extracted water comes from the aquifer itself?
5,500 m3/day, and r =305 m for a confined aquifer, compute the comp1
4.2.12 A well with a radius of 0.5 m pumps at the rate of 15 Us drawdown. appro)
from an unconfined aquifer that is uniformly recharged at a rate radius
4.4.4 Using the Cooper-Jacob method of solution, plot draw-
of 0.6 m/day. Without pumping from the well, the water table is down versus time for four observation wells 5, 50, 100, and 500
nearly horizontal and the aquifer thickness is 30m. The hydraulic m from a well pumping at a constant rate of 300 m3/day in a Tim
conductivity of the aquifer is 20 m/day. Determine the radius of sandy confined aquifer with b = 12m, K = 25 m/day, and S =
influence of the well and the approximate drawdown at the well 10-4. Given that depths to water in wells and piezometers can be
location near steady-state conditions. measured with an accuracy of about 0.5 em, how long does it
4.3.1 A confined aquifer of 35-m thickness is contaminated as take to reach near-steady conditions? Also, calculate the volume
shown in the plan view below. The regional groundwater gradient is of water accumulated above ground until the near-steady condi-
i =0.003 and the confined aquifer has a hydraulic gradient of 20 tions are achieved at those distances.
Contaminant
plume
-
-
......--capture well
-20 20
x(m)
Problems 201
ali.on. 4.4.5 Drawdown was measured during a pumping test in a con- 4.4.7 For the time-drawdown data recorded during the recovery
to the fined aquifer at frequent intervals in an observation well 200 ft phase of a pump test, determine the transmissivity and storativity
iuired from a well that was pumped at a constant rate of 500 gpm. The of the confined aquifer. The pumping rate is 162.9 ft3/rnin and the
wne. data for this pump test is listed below. Determine T and S for this pump is shut off at 800 min.
1g the aquifer.
!Cated Recovery of test well Recovery of observation well
Pump test data Time (min) s' (ft) Time (min) s' (ft)
Time (min) Drawdown (ft)
800 12.5 800 1.86
0.05 803 20 805 1.78
2 0.22 808 5 810 1.64
3 0.40 813 0.5 815 1.53
4 0.56 820 1.5 820 1.45
5 0.70 880 1 825 1.37
7 0.94 940 0.80 830 l.32
10 1.2 995 0.69 840 1.22
20 1.8 1,055 0.59 850 1.15
lfmed 4() 2.5 1,115 0.51 860 1.09
trans- 100 3.4 1,175 0.49 870 1.03
d4 X 300 4.5 1,235 0.46 880 0.97
~OOm 1,000 5.6 1,295 qJ8 890 0.94
days, 4,000 7 1,360 0.34 900 0.90
-adius 1,416 0.33 910 0.87
4.4.6 For the time-<l.rawdown data listed below for a confined 1,418 0.33 920 0.85
od of aquifer, calculate T and S using Jacob's approximation. After 1,527 0.22 980 0.70
computing T and S, check to see that the basic assumption of this 1,600 0.22 1,040 0.61
,Q:::; approximation is satisfied. For the values of T and S that you 1,100 0.54
te the computed, after how many minutes of pumping would Jacob's 1,160 0.49
approximation be valid? The discharge is Q:::; 1,500 gpm and the 1,220 0.46
radius r = 300 ft. 1,280 0.40
iraw-
d 500 1,340 0.36
r in a Time after pumping started (min) Drawdown (ft) 1,400 0.36
dS:::: 1 0.45 1,460 0.34
an be 1,520 0.31
2 0.74
oes it 1,600 0.29
3 0.91
•lwne
4 1.04
ondi- 4.4.8 (The data given in this problem also refer to Problems 4.4.9
6 1.21
and 4.7 .1 below.) A fully penetrating pumping well in a confined
8 1.32
aquifer is located between a suspected barrier boundary and a
10 1.45 known recharge boundary as shown in the figure on the nex.t
30 2.02 page. Drawdown data from the observation well in the figure
40 2.17 during a typical pumping test of the well in question are given in
50 2.30 the following table. If the pumping is constant at a rate of 100
60 2.34 gpm and the observation well is 200 ft away from the pumping
80 2.50 well, detennine the transmissivity and the storage coefficient of
100 2.67 the aquifer using the given aquifer test data.
200 2.96 Data from observation well
400 3.25
600 3.41 Time (min) Drawdown (ft)
800 3.50 2 0.025
1,000 3.60 3 0.04
1,440 3.81 (continues)
well, is given in the following table. The average thickness of the

Data from observation well
aquifer near this observation well is 84 ft. The test also reported
Time (min) Drawdown (ft) that the Galena-Platteville Aquifer has a heterogeneous and
anisotropic nature and consists of two media, fractures and
4 0.06 porous rock. The estimated hydraulic conductivity and specific
5 0.07 yield of the aquifer are respectively 8 ft/day and 0.049 from the
6 0.085 PZ-1 observation well data. Using the given time-drawdown data
7 0.095 and the Neuman Type A and Type Y curves for unconfined
8 0.105 aquifers, make your own estimate of the aquifer hydraulic con-
9 0.115 ductivity and specific yield.
10 0.12
Time (min) Drawdown (ft) Time (min) Drawdown (ft)
15 0.16
20 0.18 0.9 0.012 100 0.805
30 0.22 1 0.014 230 0.842
40 0.25 1.5 0.032 320 0.86
50 0.27 2 0.052 410 0.877
60 0.285 2.5 0.072 450 0.883
70 0.3 3 0.093 530 0.891
80 0.33 3.5 0.111 630 0.903
90 0.35 4 0.13 810 0.946
100 0.37 4.5 0.148 890 0.969
150 0.45 5 0.165 980 0.98
200 0.54 6 0.198 1070 0.976
4.7.1 "
300 0.58 7 0.254 1190 0.991 .,
1em4.
400 0.64 8 0.277 1350 1.006 acont
500 0.69 9 0.299 1525 1.031 of l ft
600 0.74 10 0.319 1720 1.065 ft awa
20 0.509 2130 1.085 4.7.2(
30 0.687 2250 1.114 Draw<
40 0.754 2400 1.136 thepu
50 0.774 2500 1.149 typeo
60 0.783 2600 1.159 the ol:
70 0.788 2810 1.175 need t•
80 0.791 3140 1.204 the em
•
Pumped
90 0.8
Known recharge
well boundary 4.6.1 (This problem is adapted from the Ground Water Man-
ual.62) The drawdown v~rsus time data for an observation well
160 feet from the pumping well of the same aquifer as in Exam-
ple 4.6.1 is given below. Estimate the transmissivity and storage
coefficient of the confined aquifer and the permeability of the
aquitard. Compare the results with the answers to Example 4.6.1.
4.4.9 Determine the aquifer properties in Problem 4.4.8 using
the Cooper-Jacob method of solution. Tune (min) Drawdown (ft)
4.5.1 (This problem is adapted from U.S. Geological Survey 4 2.15
Water-Resources Investigations Report 89-4081 (1989).) The
6 2.86
Galena-Platteville Aquifer Test was performed on June 8-15,
1987, with a 6-inch-diameter fully penetrating well pumping 8 3.46
under a constant rate (Q) of 19.4 gal/min for 54 hr. Eight partially 10 3.78
penetrating observation wells in the Galena-Platteville Aquifer 15 4.58
were monitored during the pumping test. The time-drawdown 20 5.09
data for observation well PZ-1, which is 108 ft from the pumping (continues)
Problems 203
~ss of the 120 1.25
reported Time {min) Drawdown (ft)
180 1.51
:ous and 25 5.49 240 1.70
Jres and 30 5.85 300 1.87
specific 40 6.37 360 1.99
from the
50 6.64 420 2.10
Jwndata
:onfined
60 6.8 480 2.20
!lie con- 70 6.96 540 2.28
80 7.16 600 2.36
90 7.36 660 2.46
own (ft) 100 7.44 720 2.50
).805
llO 7.52 840 2.63
).842
120 7.56 960 2.77
).86 150 7.64
).877 180 7.88 4.7.3 A production well fully penetrating a nonleaky isotropic
210 7.92 artesian aquifer delimited by two barrier boundaries (perpendi-
).883
240 7.96 cular to each other) was continuously pumped at a constant rate
).891
270 7.96 of 1,485 gpm for a period of four hours. The drawdowns in the
).903 following table were observed at a distance of 300 ft in a fully
300 7.96
1.946 penetrating observation well. Compute the coefficients of trans-
360 7.95
1.969 missivity and storage of the aquifer and the distances to each
420 7.96
1.98 image well from the observation well.
1.976
4.7.1 What would the drawdown in the pumping well be in Prob-
1.991 t (min) s (ft)
Iem 4.4.8 at the end of pumping at a constant rate of 100 gpm for
.006 a continuous period of 180 days? The pumping well with a radius 2 0.80
.031 of I ft is located 500ft away from the barrier boundary and 1,000 3 0.92
.065 ft away from the recharge boundary. 4 1.06
.085 4.7.2 (This problem is adapted from the Ground Water Manua/.62 ) 5 1.17
.114 Drawdown versus time data for an observation well 100 ft from 6 1.23
.136 the pumping well in a pump test are tabulated below. Identify the 7 1.32
.149 type of boundary and determine the radius of the image well from 8 1.37
.159 the observation well. What additional information would you
9 1.43
.175 need to locate the boundary? (Hint: Use the method described at
10 1.48
.204 the end of Section 4.7)
20 1.88
Time (min) Drawdown (ft) 30 2.11
40 2.34
r Man- 5 0.08
m well 50 2.52
10 0.22
Exam- 60 2.70
15 0.32
storage 70 2.83
20 0.41
of the 80 3.00
25 0.49
~ 4.6.1. 90 3.17
30 0.56
100 3.30
40 0.67
200 4.21
50 0.77
300 4.43
60 0.85
70 0.95 4.8.1 It is required to dewater a construction site 80 m by 80 m.
80 1.01 The bottom of the construction will be 1.5 m below the initial
90 1.08 water surface elevation of 90 m. Four pumps are to be used in
100 1.14 0.5-m diameter wells at the four comers of the site. Detennine
110 1.20 the required pumping rate. The aquifer hasT= 1600 m2/day and
tinues) (continues) the wells each have a radius of influence of 600 m.
4.8.2 Reposition the wells in Example 4.8.1 such that they form 4.8.4 Two pumping wells 1,000 m away fully penetrate the same 34. Lai
an equilateral triangle (same spacings). For the same restrictions confined aquifer. One of the wells pumps at a rate of 1,240 tirne·d•
on the drawdown, will the discharge decrease or increase? If so, m3/day. The second well pumps at 850m3/day. If the aquifer has anisotn
by what percent? If not, what difference do you perceive between a transmissivity of 2,000 m2/day and a storage coefficient of 4 x 35. Le•
the two problems? 10~. when would the wells start interfering with each other? York. I
4.8.3 Rework Problem 4.8.1 as if the site is to be ready after one 36. Le1
month of pumping. Assume the storage coefficient of the aquifer ing, J01
35-51,
isS== 0.16.
37. Lol
Profess.
REFERENCES 38. Mil
Hydrau
1. Barlow, P. M. and A. F. Moench, WTAQ-A Computer Program for 16. Glover, R. E. and G. G. Balmer, River depletion resulting from 1963.
Calculating Drawdowns and Estimating Hydraulic Properties for Con· pumping a well near a river, Trans. Amer. Geophysical Union, v. 35, 39, Me
fined and Water· Table Aquifers, U.S. Geological SulYey Water-Resources pp.468-470, 1954. pu mpa~
Investigarions Report 99-4225, I999. 17. Halford, K. J. and E. L. Kuniansky, Spreadsheets for the Analysis of 40. Mo
2. Bennett, G. D. and E. P. Patten, Jr., Constant-Head Pumping Test of Aquifer-Test and Slug-Test Data, U.S. Geological Survey Open·File undergc
a Multiaquifer Well to Determine Characteristics of Individual Aquifers, Report 02-197, httpJ/water.usgs.gov/pubs/oflofr02197, 2003. Resourc
U.S. Geological SulYey Water-Supply Paper 1536-G, pp. 181-203, 1962. 18. Hall, P. and J. Chen, Water Well and Aquifer Test Analysis, Water 41. Mu
3. Bentall, R., Methods for Determining Permeability, Transmissibility, Resources Publications, Littleton, CO, 1996. Media,)
and Drawdown, U.S. Geological SulYey Water-Supply Paper 1536-I, pp.
19. Hantush, M.S., Analysis of data from pumping tests in leaky aquifers, 42. Ner
243-341, 1963.
Trans. Amer. Geophysical Union, v. 37, pp. 702-714, 1956. delayed
4. Boulton, N. S., The drawdown of the water-table under non-steady
20. Hantush, M. S., Drawdown around wells of variable discharge, Jour. 1031-H
conditions near a pumped well in an unconfined formation, Proc. lnst.
Geophysical Research, v. 69, pp. 4221-4235, 1964. 43 New
Civil Engrs., v. 3, pt. ill, pp. 564-579, 1954.
21. Hamush, M.S., Hydraulics of wells, in Advances in Hydroscience, V. fi ned aq
5. Boulton, N. S., Analysis of data from non-equilibrium pumping tests
T. Chow, ed., v. l, Academic Press, New York, pp. 281-432, 1964. Researc.
allowing for delayed yield from storage, Proc. Inst. Civil Engrs., v. 26, pp.
469-482, 1963. 22. Hantush, M. S., Wells near streams with semipervious beds, Jour. 44. Neu
Geophysical Research, v. 70, pp. 2829-2838, 1965. hydrauli
6. Boulton, N. S. and T. D. Streltsova, New equations for determining the
Researci
formation constants of an aquifer from pumping test data, Water 23. Hantush, M. S., Wells in homogeneous anisotropic aquifers, Water
Resources Research, v. 11, pp. 14B-153, 1975. Resources Research, v. 2, pp. 273-279, 1966. 45. Ne\1
South C.
7. Chow, V. T., On the determination of transmissibility and storage coef- 24. Hantush, M. S., Flow to wells in aquifers separated by a semipervious
Repon I
ficients from pumping test data, Trans. Amer. Geophysical Union, v. 33, layer, Jour. Geophysical Research, v. 72, pp. 1709-1720, 1967.
pp.397-404, 1952. 46. Pap;
25. Hantush, M. S. and C. E. Jacob, Non-steady radial flow in an infinite large dia
8. Cooley. R. L. and C. M. Case, Effect of a water table aquitard on draw- leaky aquifer, Trans. Amer. Geophysical Union, v. 36, pp. 95-l 12, 1955.
down in an underlying pumped aquifer, Water Resources Research, v. 9, 47. Pric
26. Hantush, M. S. and I. S. Papadopulos, Flow of ground water to conditio1
pp. 434-447, 1973. collector wells, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 8-8. no.
9. Cooper, H. H., Jr., and C. E. Jacob, A generalized graphical method for 48. Pruc
HY5, pp. 221-244, 1962.
evaluating formation constants and summarizing well-field history, Trans. Data alU
27. Hantush, M. S. and R. G. Thomas, A method for analyzing a draw- South·Ct
Amer. Geophysical Union, v. 27, pp. 526-534, 1946.
down test in anisotropic aquifers, Water Resources Research, v. 2, pp. fnvestiga
10. Dagan, G., A method of detennining the permeability and effective 2Sl-285, 1966.
porosity of unconfined anisotropic aquifers, Water Resources Research, 49. Rao,
28. Huisman, L., Groundwater Recovery, Winchester, NY, 336 pp., 1972. Ground I
v. 3, pp. 1059-1071, 1967.
11. Dawson, K. and J. Istok, Aquifer Testing: Design and Analysis of 29. Jacob, C. E., Radial flow in a leaky artesian aquifer, Trans. Amer.
Pumping and Slug Tests, Lewis Publishers, Boca Raton, FL., 1991. Geophysical Union, v. 27, pp. 19&-208, 1946.
12. Driscoll, F. G, Groundwater and Wells, 2nd ed., Johnson Division, St. 30. Jenkins, C. T., Techniques for computing rate and volume of stream
Paul, MN, 1986. depletion by wells, Ground Water, v. 6, no. 2, pp. 37-46, 1968.
13. Dupuit, J., Etudes Thioriques et Pratiques sur La Mouvement des 3l. Kasenow, M.,lntroduction to Aquifer Analysis, Water Resources Pub-
Eaux dans l..es Canaux Dlcouverts eta Travers les Terrains Permiables, lications, Littleton, CO, 1997.
2nd ed., Dunod, Paris, 304 pp., 1863. 32. Kipp, K. L., Jr., Unsteady flow to a partially penetrating, finite radius
14. Ehlig, C. and J. C. Halepaska, A numerical study of confined-uncon- well in an unconfined aquifer, Water Resources Research, v. 9, pp.
fined aquifers including effects of delayed yield and leakage, Water 448-462, 1973.
Resources Besearch, v. 12, pp. 1175-1183, 1976. 33. Kruseman, G. P. and N. A. de Ridder, Analysis and evaluation of
15. Ferris, I. G. et al., Theory of aquifer tests, U.S. Geological SulYey pumping test data, International Institute for Land Reclamation and
Water-Supply Paper 1536-B, pp. 69-174, 1962. Improvement, The Netherlands, 1994.
References 205
~same 34. Lakshminarayana, V. and S. P. Rajagopalan, Type-curve analysis of 50. Rorabaugh, M. 1., Graphical and theoretical analysis of step-
1,240 time-drawdown data for partially penetrating wells in unconfined drawdown test of artesian well, Proc. Amer. Soc. Civil Engrs., v. 79, sep.
fer has anisotropic aquifers, Ground Water, v. !6, pp. 328-333, 1978. 362, 23 pp .. 1953.
of 4 x 35. Lee, J., Well Testing, Society of Petroleum Engineers of AIME, New 51. Rushton, K. R. andY. K. Chan, Pumping test analysis when param-
~r? York, 1982. eters vary with depth, Ground Wmer, v. 14, pp. 82-87. 1976.
36. Lennox, D. H. and A. Vanden Berg, Drawdowns due to cyclic pump- 52. Sheahan, N. T, Determining transmissibility from cyclic discharge,
ing, Jour. Hydraulics Oiv., Amer. Soc. Civi l Engrs., v. 93, no. HY6, pp. Ground Water, v. 4, no. 3, pp. 33-34, 1966.
35-51, 1967. 53. Stallman, R. W., Effects of water table conditions on water level
37. Lohman, S. W., Ground-Water Hydraulics, U.S. Geological Survey changes near pumping wells, Water ResourCI!S Research, v. I, pp.
Professional Paper 108, 70 pp., 1972. 295-312, 1965.
38. Milojevic, M., Radial collector wells adjacent to the river bank. Jour. 54. Stallman, R. W. and I. S. Papadopulos, Measurement of Hydraulic
Hydraulics Oiv., Amer. Soc. Civil Engrs., v. 89, no: HY6, pp. 133-15 1, Diffusiviry of Wedge-Shaped Aquifers Drained by Streams, U.S. Geolog·
~from 1963. ical Survey Professional Paper 514, 50 pp., 1966.
v. 35, 39. Moench, A. Ground-water fluctuations in response to arbit.rary 55. Sternberg, Y. M., Transmissibility determination from variable dis-
pumpage, Ground Water, v. 9, no. 2, pp. 4-8, 1971. charge pumping tests, Grotmd Water, v. 5. no. 4, pp. 27-29, 1967.
lysis of 40. Moench, A. F. and T. A. Prickeu, Radial flow in an infinite aquifer 56. Sternberg, Y. M., Simplified solution for variable rate pumping test,
!n-File undergoing conversion from artesian to water table conditions, Water Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 94, no. HY I, pp.
Resources Research, v. 8, pp. 494-499, 1972. 177-180, 1968.
Water 41. Muskat, M., The Flow of Homogeneous Fluids through Porotu 57. Sternberg, Y. M., Efficiency of partially penetrating wells, Ground
Media. McGraw-Hill, New York, 763 pp., 1937. Water, v. II, no. 3, pp. 5-8, 1973.
Juifers, 42. Neuman, S. P., Theory of flow in unconfined aquifers considering 58. Streltsova, T. 0 ., On the leakage assumption applied to equations of
delayed response of the water table, Water Resources Research, v. 8, pp. groundwater flow, Jour. Hydrology, v. 20, pp. 237-253. 1973.
:, Jour. 1031- 1045, 1972. 59. Theis, C. V., The relation between the lowering of the piezometric sur-
43 Neuman, S. P., Analysis of pumping test data from anisotropic uncon- face and the rate and duration of discharge of a well using ground-water
nee, V. fined aquifers considering delayed gravity response, Water Resources storage, Trans. Amer. Geophysical Union, v. 16, pp. 519-524, 1935.
Research, v. 11. pp. 329-342, 1975. 60. Thiem, G. Hydrologische Methoden, Gebhardt, Leipzig, 56 pp.,
44. Neuman, S. P. and P. A. Witherspoon, Field determination of the 1906.
, Jour.
hydraulic properties of leaky multiple aquifer systems, Water Resources 61. U.S. Bureau of Reclamation, Ground Water Manual, U.S. Dept. Inte-
Research, v. 8, pp. 1284-1298, 1972. rior, 480 pp., 1977.
Water
45. Newcome, R., Jr., Pumping Tests of the Coastal Plain Aquifers in 62. U.S. Bureau of Reclamation, Ground Water Manual, U.S. Govern-
South Carolina, State of South Carolina Water Resources Commission ment Printing Office, Denver, CO, 1981.
Nious
Report 174, 1993. 63. Walton, W., Aquifer Test Analysis with Windows™, CRC Press, Boca
46. Papadopulos, I. S. and H. H. Cooper, Jr., Drawdown in a well of Raton, FL. 1996.
nfinite
large diameter, Water Resources Research, v. 3, pp. 241-244, 1%7. 64. Walton, W. C., Leaky Artesian Aquifer Conditions in Illinois, lllinois
1955.
47. Prickett, T. A., Type-curve solution to aquifer tests under water-table State Water Survey Rept. Invest. 39, Urbana, 27 pp., 1960.
tier to conditions, Ground Water, v. 3, no. 3, pp. 5-14, 1%5. 65. Walton, W. C., Comprehensive analysis of water-table aquifer test
38. 00 . data, Ground Wmer, v. 16. pp. 311-317. 1978.
48. Prudic, D. E., EstinwlesofHydrau/ic Conductivity from Aquifer-Test
Data and Specific-Capacity OaJa, Gulf Coast Regional Aquifer Systems, 66. Weeks, E. P., Determining the ratio of horizontal to vertical perme-
draw- South-Central United States, U. S. Geological Survey W01er-Resources ability by aquifer-test analysis, Water Resourcts Research, v. 5, pp.
2, pp. Investigations Report 90-4121, 1991. 1%-214, 1969.
49. Rao, D. B. et al.. Orawdown in a well group along a straight line, 67. Wigley, T. M. L., Flow into a finite well with arbitrary discharge,
1972. Ground Water, v. 9, no. 4, pp. 12- 18, 1971. Jour. Hydrology, v. 6, pp. 209-213, 1968.
Amer.
;tream
s Pub-
radius
9, pp.
ion of
nand
5~
Chapter 5
Water Wells
A water well is a hole or shaft, usually vertical, excavated in the earth for bringing ground-
water to the surface. Occasionally wells serve other purposes, such as for subsurface
exploration and observation, artificial recharge, and disposal of wastewaters. Many methods
exist for constructing wells; selection of a particular method depends on the purpose of the
well, the quantity of water required, depth to groundwater, geologic conditions, and economic
factors. Shallow wells are dug, bored, driven, or jetted; deep wells are drilled by cable tool or
rotary methods. Attention to proper design can ensure efficient and long-lived wells. After a
well has been drilled, it should be completed, developed for' optimum yield, and tested. Wells
should be sealed against entrance of surface pollution and given periodic maintenance. Wells
of horizontal extent are constructed where warranted by special groundwater situations.
5.1 TEST HOLES AND WELL LOGS

Before drilling a well in a new area, it is common practice to drill a test hole. The purpose of
a test hole is to determine depths to groundwater, quality of water, and physical character and
thickness of aquifers without the expense of a regular well, which might prove to be unsuc-
cessful. Diameters seldom exceed 20 em. Test holes may be put down by any method for well
construction; however, cable tool, rotary, and jetting methods are commonly employed. If the
test hole appears suitable as a site for a finished well, it can be reamed with hydraulic rotary
equipment to convert it to a larger permanent well.
During drilling of a test hole or well, a careful record, or log, is kept of the various geo-
logic formations and the depths at which they are encountered (see Chapter 11). A helpful
method is to collect samples of cuttings in containers, labeling each with the depth where
obtained. Later these can be studied and analyzed for grain size distribution. Most states
require licensed well drillers to submit logs-recording depth, color, character, size of mate-
rial, and structure of the strata penetrated-for wells they drill. Proper identification of strata
in the hydraulic rotary method requires careful analysis because drilling mud is mixed with
each sample. A drilling-time log (again see Chapter 11) is sometimes helpful in this respect
5.2 METHODS FOR CONSTRUCTING SHALLOW WELLS

Shallow wells, generally less than 15m in depth, are constructed by digging, boring, driving,
or jetting.38• 83· 96 The methods are briefly described in the following paragraphs; Table 5.2. t
lists their applications.
206
••
Vo<
::l.
'2(li~l=iSSc->B""<r<~
0.. ~· "' ~ (ii "' ac:;;;:;~~~ ~~~85~g.~~ I
N
•
:;;-
,.....
CIQ
~
:- - · I» ~ ~ ~ S" 8
9-S,<n<~>-•
Sg.~~~o
Qn>::::<?CLa I = :::::
oo~-::l=O~»::l
~ ~ ~
0
~
~-
n ;;;;T
~
0.. (")
~ o 0..
1
• · ··· WI
Table 5.2.1 Water Well Construction Methods and Applications (after U.S . Soil Conservation Service97)
Method Materials Water table Usual Usual Usual

for which _d epth for which maximum diameter casing Customary Yield
best suited best suited (m) depth (m) range (em) material use (m3/day 3 ) Remarks
Augering Clay, silt, sand, 2-9 10 5-20 Sheet metal Domestic, 15-250 Most effective for penetrating
Hand auger gravel less drainage and.removing clay. Limited
than 2 em by gravel over 2 em. Casing
required if material is loose.
Power auger Clay, silt, sand, 2-15 25 15-90 Concrete, steel Domestic, 15-500 Limited by gravel over 5 em,
gravel less or wrought irrigation, otherwise same as for hand
than 5 em iron pipe drainage auger.
Driven wells Silt., sand, 2-5 15 3-10 Standard Domestic, 15-200 Limited to shallow water
Hand, air gravel less weight pipe drainage table, no large gravel.
hanuner than 5 em
Jetted wells Silt, sand, 2-5 15 4-8 Standard Domestic, 15-150 Limited to shallow water
Light, gravel less weight pipe drainage table, no large gravel.
portable rig than 2 em
Drilled wells Unconsolidated Any depth 450b 8-60 Steel or All uses 15-15,000 Effective for water exploration.
Cable tool and consolidated wrought- Requires casing in loose mat-
mediwnhard iron pipe erials. Mudscow and hollow rod
and hard rock bitsdeveloped for drilling uncon-
solidated fine to medium sedi-
ments.
Rotary Silt, sand, Any depth 450b 8-45 Steel or All uses 15-15,000 Fastest method for all except
gravel less than wrought- hardest rock. Casing usually not Vl
2 em; soft to iron pipe N

required during drilling. Effec-
hard consoli- tive for gravel envelope wells. ~
~
dated rock ::r
Reverse- Silt, sand, 2-30 60 40-120 Steel or Irrigation, 2,500-20,000 Effective for large-diameter holes 8.
circulation gravel, cobble wrought- industrial, in unconsolidated and partially "'
rotary iron pipe municipal
...
0'
consolidated deposits. Requires (')
0
large volume of water for drilling. ::l
Effective for gravel envelope wells. "'

2
(")
Rotary- Silt, sand, Any depth 600b 30-50 Steel or Irrigation, 2,500-15,000 Now used in oil exploration. a.
:::0
percussion gravel less than wrought- industrial, Very fast drilling. Combines ()Q
(/)
5 em; soft to iron pipe municipal rotary and percussion methods :::>"
a.
hard consoli- (air drilling) cuttings removed 0
dated rock ~
by ak Would be economical for
deep water wells. ~
V>
• Yield influenced primarily by geology and availability of groundwater.
N
b Greater depths reached with heavier equipment. 0
'-I
208 Chapter 5 Water Wells
5.2.1 Dug Wells

Dating from Biblical times, dug wells have furnished countless water supplies throughout the
world. • Depths range up to 20 m or more, depending on the position of the water table, while
diameters are usually 1 to 10m. Dug wells can yield relatively large quantities of water from
shallow sources and are most extensively employed for individual water supplies in areas con-
taining unconsolidated glacial and alluvial deposits. 106 Their large diameters pennit storage of
considerable quantities of water if the wells extend some distance below the water table.
In the past, all dug wells were excavated by hand, and even today the same method is
widely employed. A typical dug well in under-developed portions of the world is often no more
than an irregular hole in the ground that intersects the water table (see Fig. 5.2.1). A pick and
shovel are the basic implements. Loose material is hauled to the surface in a container by means
of suitable pulleys and lines. Large dug wells can be constructed rapidly with portable excavat-
ing equipment such as clamshell and orange-peel buckets. For safety and to prevent caving,
lining (or cribbing) of wood or sheet piling should be placed in the hole to brace the walls.
A modem dug well is permanently lined with a casing (often referred to as a curb) of
wood staves, brick, rock (Fig. 5.2.2), concrete, or metal. Curbs should be perforated or contain
openings for entry of water and must be firmly seated at the bottom. Dug wells should be deep
enough to extend a few meters below the water table. Gravel should be backfilled around the
curb and at the bottom of the well to control sand entry and possible caving. A properly
5.2.2
Figure 5.2.1. Women gathering

water from a crude dug well in the
Shinyanga Region of Tanzania,
East Africa (courtesy DHV
Consulting Engineers, Amersfoort,
The Netherlands).
'See also the description of qanats in Chapter I.

ut the
while
·from
scon-
lgeof
1od is
,more
:kand
neans
cavat-
aving,
Is.
rb) of
ontain
!deep
nd the
:Jperly
Figure 5.2.2. A domestic

dug well with a rock curb,
concrete seal, and hand
pump.
constructed dug well penetrating a permeable aquifer can yield 2,500 to 7,500 m3/day,
although most domestic dug wells yield less than 500m3/day.
A serious limitation of large open dug wells involves the ease of their pollution by surface
water, airborne material, and objects falling or finding entrance into the wells.
5.2.2 Bored Wells

Where a water table exists at a shallow depth in an unconsolidated aquifer, bored wells can fur-
nish small quantities of water at minimum cost. Bored wells are constructed with hand-
operated or power-driven earth augers. Hand augers are available in several shapes and sizes,
all operating with cutting blades at the bottom that bore into the ground with a rotary motion.
When the blades are full of loose earth, the auger is removed from the hole and emptied; the
operation is repeated until the desired hole depth is reached. Hand-bored wells seldom exceed
20 em in diameter and 15 m in depth. Power-driven augers will bore holes up to 1 m in
ering diameter and, under favorable conditions, to depths exceeding 30 m. The auger consists of a
ll in the cylindrical steel bucket (Figure 5.2.3) with a cutting edge projecting from a slot in the bottom.
tnia, The bucket is filled by rotating it in the hole by a drive shaft of adjustable length. When full,
the auger is hoisted to the surface and the excavated material is removed through hinged open-
:rsfoort,
ings on the side or bottom of the bucket. Reamers, attached to the top of the bucket, can
enlarge holes to diameters exceeding the auger size.
A continuous-flight power auger (Figure 5.2.4) has a spiral extending from the bottom of
the hole to the surface. Cuttings are carried to the surface as on a screw conveyor, while sec-
Reamer
5.2.
Cutting
blade
Figure 5.2.3. Augers for boring
Auger with Auger with wells. Spiral flight augers are
bottom open adjustable also widely employed for small-
for dumping reamer diameter holes.
Figure 5.2.4. Hollow-

stem auger drilling. The
hollow-stem, continuous-
flight auger bores into
soft soils, carrying the
cuttings upward along
the flights. When the
desired depth is reached,
the plug is removed from
the bit and withdrawn
from inside the hollow
stem. A well point on a
casing ( 11/4 in or 2 in)
can then be inserted to
the bottom of the hollow
stem and the auger
pulled out, leaving the
small-diameter monitor-
ing well in place (M. L.
Sc~\
tions of the auger may be added as depth increases. The simple equipment, usually truck-
mounted, can be operated rapidly by one person and functions to depths exceeding 50 m in
unconsolidated formations that do uot contain large boulders.
Augers work best in formations that do not cave. Where loose sand and gravel are encoun-
tered in a large-diameter hole, or the boring reaches the water table, it may be necessary to
lower a concrete or metal casing to the bottom of the hole and continue boring inside. After the
desired depth is reached, the permanent well casing and screen are centered in the hole, the
outer casing is removed, and the peripheral space is backfilled with gravel. Augers sometimes
supplement other well-drilling methods where sticky clay formations are encountered; here
augers are more effective than any other penetrating device.
5.2.3 Driven Wells

A driven well consists of a series of connected lengths of pipe driven by repeated impacts into
the ground to below the water table. Water enters the well through a drive (or sand) point at
the lower end of the well (Fig. 5.2.5). This consists of a screened cylindrical section protected
during driving by a steel cone at the bottom. Diameters of driven wells are small, most falling
in the range of 3 to 10 em. Standard-weight water pipe having threaded couplings serves for
ring casing. Most depths are less than 15m, although a few exceed 20m. As suction-type pumps
e
extract water from driven wells, the water table must be near the ground surlace if a continuous
lall-
Weight
Drive cap
Coupling
Drive cap
low- Pipe
:. The Coupling
.nuous-
into
: the
.ong
!he
ached,
:d from
lWfl
.Uow Water table
~ on a
~in)
~to
hollow Screen
r
; the
:mitor-
:M. L. Drive point
Figure 5.2.5. A driven weU with driving mechanism.
Water tabla
Topump.
Static water
table
Figure 5.2.6. A well-point system to dewa-
Well point ter an excavation site.
water supply is to be obtained. For best results the water table should be within 3 to 5 m of
ground surface in order to provide adequate drawdown without exceeding the suction limit.
Yields from driven wells are small, with discharges of about 100-250 m3/day.
Driven wells are best suited for domestic supplies, for temporary water supplies, and for
exploration and observation. Batteries of driven wells connected by a suction header to a sin-
gle pump are effective for localized lowering of the water table. Such installations, known as
well-point systems, are particularly advantageous for dewatering excavations for foundations
and other subsurface construction operations.74 Figure 5.2.6 illustrates how a well-point
installation reduces the groundwater level to furnish a dry excavation.
Driven wells are limited to unconsolidated formations containing no large gravel or rocks
that might damage the drive point. To drive a well, the pipe casing and threads should be pro-
tected at the top with a drive cap (see Figure 5.2.5). Driving can be done with a maul, sledge,
drop hammer, or air hammer. It is usually good practice to place (by boring or driving) an outer
protective casing to at least 3 m below ground surface. Screens are available in a variety of
opening sizes, the choice depending on the size of particles in the water-bearing stratum.
Important advantages of driven wells are that they can be constructed in a short time, at
minimum cost, and even by one person.
5.2.4 Jetted Wells

Jetted wells are constructed by the cutting action of a downward-directed stream of water. The
high-velocity stream washes the earth away, while the casing, which is lowered into the deep-
ening hole, conducts the water and cuttings up and out of the well. Small-diameter holes of 3
to 10 em are formed in this manner (although the method is capable of producing diameters up
to 30 em or more) to depths greater than 15m. Jetted wells have only small yields and are best
adapted to unconsolidated formations. Because of the speed of jetting a well and the portability
of the equipment, jetted wells are useful for exploratory test holes, observation wells, and well-
point systems. II. 66
Figure 5.2.7. Various designs ofjetting

drill bits.
Various types of jetting drill bits are shown in Figure 5.2.7. In penetrating clays and hard-
pans, the drill pipe is raised and lowered sharply, causing the bit to shatter the formation. Dur-
ing the jetting operation, the drill pipe is turned slowly to ensure a straight hole. To complete
a shallow jetted well after the casing extends to below the water table, the well pipe with screen
attached is lowered to the bottom of the hole inside the casing. The outer casing is then pulled,
gravel is inserted in the outer space, and the well is ready for pumping.
A simplification of the above procedure can be obtained by using a self-jetting well point.
This consists of a tube of brass screen ending in a jetting nozzle, which is screwed to the well
ewa-
Suction
header
m of
limit.
nd for
a sin-
·wn as
ations
-point
Well pipe
·rocks
e pro-
ledge,
1 outer
.ety of
n.
me, at Gravel
!r. The
:deep-
!S of3 Screen
ters up
re best
:ability Figure 5.2.8. Jetted well
:1 well- Self-jetting with self-jetting well
well point
point.
pipe (Figure 5.2.8). As soon as the well point has been jetted to the required depth, the well is
completed and ready for pumping. Gravel should be added around the drill pipe for permanent
installations.
5.3 METHODS FOR DRILLING DEEP WELLS

Most large, deep, high-capacity wells are constructed by drilling. Construction can be accom-
plished by the cable tool method or by one of several rotary methodsY·47· 53 Each method has
particular advantages, so experienced drillers endeavor to have equipment available for a diver-
sity of drilling approaches. 1· 2· 47 Applications of drilling methods are listed in Table 5.3.1,
while Table 5.3.2 indicates the performance of the methods in various geologic formations.
Examples of the construction of deep wells in unconsolidated and consolidated formations
are shown in Figures 5.3.1(a)-5.3.l(l), respectively, taken from standard specifications for
wells prepared by the American Water Works Association. The construction procedure of a
successful well is dependent on local conditions encountered in drilling; hence, each well
should be treated as an individual project. Figure 5.3.2 illustrates modern truck-mounted drill
rigs. Construction methods differ regionally within the United States and also from one driller
to another. General construction methods are described in the following sections.
Table 5.3.2 Perfonnance of Drilling Methods in Various Types of Geologic Fonnations Grave~p
place an
(after Speedstar Div.88)
Drilling Method
Type of fonnation Cable tool Rotary
Rotary percussion•
Dune sand Difficult Rapid NR c
Loose sand and gravel Difficult Rapid NR
Quicksand Difficult, except in thin streaks Rapid NR
Requires a string of drive pipe
Loose boulders in alluvial Difficult; slow but generally Difficult, NR
fans or glacial drift can be handled by driving pipe frequently
impossible
Clay and silt Slow Rapid NR
Firm shale Rapid Rapid NR
Sticky Shale Slow Rapid NR
Brittle shale Rapid Rapid NR
Sandstone, poorly cemented Slow Slow NR
Chert nodules Rapid Slow NR
Limestone Rapid Rapid Very rapid
Limestone with Rapid Slow Very rapid
chert nodules
Limestone with small Very rapid
cracks or fractures
Limestone, cavernous Rapid Slow to Difficult Gra
casi
impossible !ern
Dolmite Rapid Rapid Very rapid
Basalts, thin layers in Rapid Slow Very rapid Figure!
sedimentary rocks Associa1
7.
Basalts, thick layers Slow Slow Rapid Denver,
Metamorphic rocks Slow Slow Rapid
Granite Slow Slow Rapid
~
·~
"NR: not recommended.
.veil is
1anent
.ceom- ·;··~~-- Shoe

od has f>ump-
IF.f.<,~~~
,diver-
Cemented
5.3.1, In Place
ions. . . Gravel·
Feed
lations Line
Jns for
re of a
:h well
ed drill
:driller
(a) Type 1 (b) Type 2

Gra118l·packed weU with conductor casing grouted In Gravel-packed well with well casing cemented In
place and gravel enY8Iope exten<ing to SUiface. place and grawtl 80118lope l&rmlnated above the top
of the screen with gnl11811eed line.
~
~'tn
as Grout
Is Plaoed
Shoe
Pu~
f
If
Housing
Casing
Cemented
In P13Q8
Shoe
Gravel
Envelope
Screen
'I rapid
y rapid
y rapid
(c) Type 3 -(d) Type 4
ficult Gravel-packed weN wllh telescoped screen, we11 Naluraly dewloped well with lel.seoped saeeo,
casing cemented In place, and gravel envelope weU p!M71p--1\ouslng casing drtwn or jacked Into
terminated abcwe the top Of screen. place, and the conductor aeaJed as locally required.
y rapid
y rapid Figure 5.3.1. Examples of well construction in unconsolidated and consolidated formations (reprinted from American Water Works
Association AWWA Standard for Deep Wells, 3 by permission of the American Water Works Association, 6666 West Quincy Avenue,
Denver, CO 80235).
Pump-
Housing
Casing
Temporary
Casing
May Be
~.:.....,..:.,....,.:... Withdrawn
as Grout
Is Placed .. ~..• ·.~ ....~..
.~·
1.5 in. ~<."".-·="';:_-'-·;"'"-·.,_· Shoe

~.;..;...,r., at Casing
Minimum .•
...:.- ....:. ~· ~:~.
. . .. ~ :. : .....
Couplings .-" ... ~ ,- : ·'-.-. ~
Pump-
~..;;;;;;.~~Shoe
::.'~·-;>·--~:;.
.... ... ~.:: ....
~----~ Housing
Casing With
Perforation
Packer Made
in Place
(e) Type 5
(f) Type 6 wen
Naturally developed well with telescoped screen, and1
temporary casing driven or jacked into place, and Naturally developed well with well casing advanced
the pump-housing casing sealed In to prevent by driving or Jacking and perforated in place.
contamination.
Over
Temporary
Castng
:::7~"-':·::f,lfU___..R&~ ~!:wn
as Grout
Is Placed
c IIIU~~~Shoe
0
PumJ>-
~
I
Pump- 111~~;..;:;..:.,<.:.- Housing ~
18;:..:.;,~:...;.- Housing Casing
& Casing Cemented !'
Cemented in Place ~
In Place
.. ~
Gravel '
Envelope
~~.:,;..:,-,...... In Under-
reamed
1
~
Hole -.-~
Cl
Blank ~
i!Rf"~~-Screen Casing ::l
Screen ~
-~-
"'
E
·~:
., tf
(g) Type 7 '·
{h) Type 8
Gravel-packed well with undemtamed borehole for Open I
screen and pump-housing casing cemented In place. Gravel-packed well with underreamed borehole lor aquifer
screens In multiple unconsofidated aquifers. ground
Figure 5.3.1 (continued) Figun

i!SSt
!,~
:'; ::.~:-: .. ::_·
'·,·..:-;..~....
8.s Grout
Js Ptaced
;.: -··~i=i¥·
:.:·. ' ,.., ~-· ~ Pump.
;.~ •• ••••. ~ng
;, •• •• Cemented Pump-
~ •-t •.::,.. -. Jn Place
~ I t- 1- J
~
/,
•'
I I• ,,.
,_, ,_ ""'' .
I•
~ted
· I ~t ,_•-• ,.., ,_ In Place
:, ,_, , _, n ..
• I• II• "'"' •
(i) Type 9 (j) Type 10 ·

Well with open hole completlon in consolidated rock Gravel-pad<ed well oompl&ted In consojldated rock
and well caslng cemented In place.
with well casing cemented In plaoe.
~----Pump
Er.lll---- Column or
AowPipe
Packer
or Seal
ir
as Grout
Is Placed
lfiww.;............ ~~
or Casing
Cemented
In Place
::aii-.-E~?:- Shoe
(k) Type 11 (I) Type 12

Open hole or screened wei COillllellon In an art&sJan Naturally del/eloped well with saeen and casing
a~lter where lhe piezometric level Is above the
ground eleva!lon. lnslaled ln place In an open hole. Blantt casing 1o
non-wat«-beartng formation Is Ol)lional.
Figure 5.3.1 (continued)

Figure 5.3.2. Rotary drill rig designed for air/mud

applications. This rig is truck-engine powered and the
single-tubed derrick accommodates a casing harruner,
dual wall systems, or drill and drive systems (cour-
tesy of Inger.;olJ Rand).
5.3.1 Cable Tool Method

Wells drilled by the cable tool (also percussion or standanf) method are constructed with a
standard well-drilling rig, percussion tools, and a bailer.38• 40• 95 The method is capable of
drilling holes of 8 to 60 em in diameter through consolidated rock materials to depths of 600
m. In unconsolidated sand and gravel, especially quicksand, it is least effective because the
loose material slumps and caves around the bit. Drilling is accomplished by regular lifting and
dropping of a string of tools. On the lower end, a bit with a relatively sharp chisel edge breaks
the rock by impact.
From top to bottom, a string of tools consists of a swivel socket, a set ofjars, a drill stem,
and a drilling bit (Figure 5.3.3). The total weight may amount to several thousand kilograms.
Drill Drilling bit Bailer Figure 5.3.3. Basic well-drilling tools for
Swivel Jars
socket stem the cable tool method.
Tools are made of steel and are joined with tapered box-and-pin screw joints. The most impor-
tant part of the string of tools is the bit, which does the actual drilling. Bits are manufactured
in lengths of lto 3m and weigh up to 1,500 kg. Variously shaped bits are made for drilling in
different rock formations. The drill stem is a long steel bar that adds weight and length to the
drill so that it will cut rapidly and vertically. A set of jars consists of a pair of narrow connect-
ing links. They have no direct effect on the drilling; their purpose is only to loosen the tools
should they stick in the hole. Under normal tension on the drilling line the jars remain fully
extended. When tools become stuck, the line is slackened to allow the links to open to their full
length, whereupon an upstroke of the line will cause the upper section of the jars to impart an
upward blow to the tools. The swivel socket attaches the drilling cable to the string of tools.
Drill cuttings are removed from the well by a bailer or sand bucket (Figure 5.3.3).
Although several models are manufactured, a bailer consists essentially of a section of pipe
with a valve at the bottom and a ring at the top for attachment to the bailer line. When lowered
into the well, the valve permits cuttings to enter the bailer but prevents them from escaping.
After filling, the bailer is hoisted to the surface and emptied. Bailers are available in a range
\Ud of diameters, lengths of 3 to 8 m, and capacities up to 0.25 m3.
1d the The drilling rig for the cable tool method consists of a mast, a multiline hoist, a walking
nmer, beam, and an engine. In most present-day designs the entire assembly is truck-mounted (again
•ur- see Figure 5.3.2) for ready portability.
During drilling, the tools make 20-40 strokes per minute, ranging from 40 to 100 em in
length. The drilling line is rotated so that the bit forms a round hole, and additional line is let
out as needed so that the bit will always strike the bottom of the hole. Water should be added
to the hole if none is encountered to form a paste with the cuttings, thereby reducing friction
.vith a on the falling bit. After the bit has cut 1 or 2 m through a formation, the string of tools is lifted
ble of to the surface and the hole is bailed. In unconsolidated formations, casing should be main-
>f 600 tained to near the bottom of the hole to avoid caving. Casing is driven down by means of drive
se the clamps fastened to the drill stem; the up and down motion of the tools striking the top of the
1g and casing, protected by a drive head, sinks the casing. On the bottom of the first section of cas-
Jreaks ing, a drive shoe (see Figure 5.3.1) with a beveled cutting edge is fastened to protect the cas-
ing as it is being driven.
~stem,
As casing is driven deeper in unconsolidated formations, the vibration causes the sides of
~rams.
the hole to collapse against it. Frictional forces increase until further driving becomes impos-
sible. When this occurs, a smaller diameter casing is telescoped inside of the casing already in
the hole; thereafter, drilling is continued using a smaller diameter bit. In deep holes, several
such reductions in casing may be required.
In drilling any deep well it is important that proper alignment be maintained so as not to
interfere with pump installation and operation. The greatest problem occurs in drilling through
rock formations. Some drillers have found that boles tending to bend can be corrected by
detonating explosives at the bottom. This shatters the surrounding rock and permits drilling to
progress vertically.
The cable tool rig is highly versatile in its ability to drill satisfactorily over a wide range
of geologic conditions ... Its major drawbacks are its slower drilling rate, its depth limitation,
the necessity of driving casing coincidentally with drilling in unconsolidated materials, and the
difficulty of pulling casing from deep holes. The simplicity of design, ruggedness, and ease of
maintenance and repair of the rigs and tools are important advantages in isolated areas. 94 Also,
less water is required for drilling than with other methods, a matter of concern in arid and
semiarid regions. Furthermore, sampling and fonnation logging are simpler and more accurate
with a cable tool rig.
)Ols for '1n particular, cable tool rigs can drill through boulders and fractured, fissured, broken, or cavernous rocks, which
often are beyond the capabilities of other types of equipment (see Table 5.3.1).
5.3.2 Rotary Method

A rapid method for drilling in unconsolidated strata is the rotary method. 73 Deep wells up to
45 em in diameter, and even larger with a reamer, can be constructed. The method operates
continuously with a hollow rotating bit through which a mixture of clay and water, or drilling
mud, is forced. Material loosened by the bit is carried upward in the hole by- the rising mud.
No casing is ordinarily required during drilling because the mud forms a clay lining, or mud
cake, on the wall of the well by filtration. This seals the walls, thereby preventing caving, entry
of groundwater, and loss of drilling mud.
Drill bits are available in various forms; a group of conical roller gears with teeth that
scrape, grind, and fracture the rock is a common design (see Figure 5.3.4). The typical string
of tools consists of a bit, a drill collar (which adds weight to the bit and aids in maintaining
hole alignment), and a drill pipe that extends to the ground surface. The upper end of the drill
pipe is attached to the kelly-a square section of drill rod. The drill is turned by a rotating table
that fits closely around the kelly and allows the drill rod to slide downward as the hole deep-
ens. The drilling rig for a rotary outfit consists of a derrick, or mast, a rotating table, a pump
for the drilling mud, a hoist, and the engine.
Drilling mud consists of a suspension of water, bentonite, clay, and various organic addi-
tives. Maintenance of the correct mud in terms of weight, viscosity, jelling strength, and low
percentage of suspended solids is important for trouble-free drilling. 94 Organic additives that
degrade with time and thereby cause the mud cake to break down within a few days are a recent
innovation. The drilling mud leaves the drill pipe through the bit, where it cools and lubricates
the cutting surface, entrains drill cuttings, and carries the drill cuttings upward within the
annular space between drill pipe and hole wall as the fluid returns to ground surface (see Fig-
ure 5.3.5). The drilling mud then overflows into a ditch and passes into a settling pit. Here the
cuttings settle out; thereafter, the mud is picked up by the pump for recirculation in the hole.
Rotary drilling is employed for oil wells and its application to water-well drilling is
steadily increasing. Advantages are the rapid drilling rate, the avoidance of placement of a
casing during drilling, and the convenience for electric logging (see Chapter 12). Disadvan-
tages include high equipment cost, more complex operation, the need to remove the mud cake
during well development, and the problem of lost circulation in highly permeable or cav-
ernous geologic formations.
5.3.3 Air Rotary Method

Rotary drilling can also be accomplished with compressed air in place of drilling mud. The
technique is rapid and convenient for small-diameter holes in consolidated formations where
a clay lining is unnecessary to support the walls against caving. Larger diameter holes can be
drilled by employing foams and other air additives. 21 Drilling depths can exceed 150m under
5.3.4
Figure 5.3.4. Examples
of rotary drill bits. (a)
Fishtail bit (b) Cone-type
rock bit (c) Carbide but-
ton bit (after Speedstar
(a) (b) (c) Div. 88 )
sup to
perates
frilling
gmud.
or mud
~. entry
:th that
1 string
.taining
he drill
1g table 00
e deep-
a pump
ic addi-
md low
ves that
arecent
bricates
hin the
n
-
;ee Fig-
Jere the ..........
hole. \
...___ _\-~!!11Ill \ Screen
lling is ' t t
Settling pit
!nt of a
sadvan-
ud cake
or cav-
f f
ud. The
s where Figure 5.3.5. Drilling mud circulation
; can be system for the rotary method (after
n under Speedstar Div. 88).
favorable circumstances. An important advantage of the air rotary method is its ability to drill
through fissured rock formations with little or no water required.
5.3.4 Rotary-Percussion Method

.amples
). (a)
A recently developed rotary-percussion procedure using air as the drilling fluid provides the
)ne-type fastest method for drilling in hard-rock fonnations. A rotating bit, with the action of a pneumatic
tde but- hammer, delivers 10 to 15 impacts per second to the bottom of the hole. Penetration rates of as
edstar much as 0.3 rnlmin have been achieved. Where caving formations or large quantities of water
are encountered, a change to conventional rotary drilling with mud usually becomes necessary.
5.3.5 Reverse-Circulation Rotary Method

The reverse-circulation rotary method has become increasingly popular as a means for drilling
large-diameter holes in unconsolidated formations. Water is pumped up through the drill pipe
employing a large-capacity centrifugal or jet pump similar to those for gravel dredges. Dis-
charge from the hole flows into a large pit where cuttings settle out. Thereafter the water
runs through a ditch and back into the hole so that the water level in the hole is maintained at
ground surface.
To avoid erosion of the sides of the hole, downward velocities must be restricted; there-
fore, the minimum hole diameter is about 40 em. Drilling bits range in diameter from 0.4 to
1.8 m. The velocity of water up the drill pipe usually exceeds 2 m/s.
The water table should be 3 to 4 m below ground surface in order to obtain an effective
head differential between well and aquifer. With this difference, fine-grained particles sus-
pended in the water aid in stabilizing the walls. Where the water table is closer to the surface,
casing can be extended above ground surface to increase the head. On the other hand, where
the water table is deep, it may be necessary to place surface casing to minimize water loss.
For unconsolidated formations the reverse-circulation rig is probably the most rapid
drilling equipment available. It requires a large volume of readily available water. Such rigs
norma!Jy can drill to depths of 125m; modifications with airlift pumping can extend this depth
range. • Because reverse-circulation holes have large diameters, completed wells are usually
gravel packed.
5.4 WELL COMPLETION

After a well has been drilled, it must be completed. This can involve placement of casing,
cementing of casing, placement of well screens, and gravel packing; however, wells in hard-
rock formations can be left as open holes so that these components may not be required. 5.4.2 c
5.4.1 Well Casings

Well casing serves as a lining to maintain an open hole from ground surface to the aquifer. It
seals out surface water and any undesirable groundwater and also provides structural support
against caving materials outside the well. Materials commonly employed for well casings are
wrought iron, alloyed or unalloyed steel, and ingot iron. 5 Joints normally consist of threaded
couplings or are welded, the object being to secure watertightness. In cable-tool drilling, the
casing is driven into place; in rotary methods, the casing is smaller than the drilled hole and 5.4.3 s
hence can be lowered into place. Polyvinyl chloride pipe is widely employed as casing for
shallow, small-diameter observation wells.
Surface casing is installed from ground surface through upper strata of unstable or frac-
tured materials into a stable and, if possible, relatively impermeable material. 94 Such surface
casing serves several purposes, including: (1) supporting unstable materials during drilling, (2)
reducing loss of drilling fluids, (3) facilitating installation or removal of other casing, (4) aid-
ing in placing a sanitary seal, and (5) serving as a reservoir for a gravel pack. This casing may
be temporary during drilling or it may be permanent. Recommended minimum diameters of
surface casing are given in Table 5.4.1.
'A versatile modification of reverse rotary drilling is the dual-tube method. This involves a rotating outer casing
connected to an inner concentric pipe with air serving as the circulating fluid. Rapid drilling rates are possible with
diameters up to 40 em and depths to 400 m. A particular advantage of the method is that continuous and accurate
geologic and groundwater quality samples can be obtained as a function of depth.
Table 5.4.1 Recommended Minimum Diameters for Well Casings and Screens
(after U.S. Bureau of Reclamation94)
drilling
ill pipe
,,,: Surface Casing Diameter (em)
;s. Dis- Nominal pump Naturally Gravel- Nominal
e water Well yield chamber casing developed packed screen
lined at (m3/day) diameter (em) wells wells diameter (em)
< 270 15 25 45 5
1; there- 270-680 20 30 50 10
n 0.4 to 680-1,900 35
25 55 15
1,900-4,400 30 40 60 20
ffective 4,400-7,600 35 45 65 25
les sus- 7,600-14,000 40 50 70 30
surface, 14,000-19,000 50 60 80 35
., where 19,000-27,000 60 70 90 40
loss.
st rapid
1ch rigs Pump chamber casing comprises all casing above the screen in wells of uniform diame-
is depth ter. For telescoping wells it is the casing within which the pump bowls are set. Recommended
usually minimum diameters are listed in Table 5.4.1. Normally, the pump chamber casing should have
a nominal diame!er at least 5 em larger than the nominal diameter of the pump bowls. Non-
metallic pipes are sometimes employed where corrosion or incrustation is a problem. Possi-
bilities include ceramic-clay, concrete, asbestos-cement, plastic, and fiberglass-reinforced
plastic pipe; strengths of these materials are not comparable to steel pipe, however.
casing,
in hard-
ed. 5.4.2 Cementing
Wells are cemented in the annular space surrounding the casing to prevent entrance of water
of unsatisfactory quality, to protect the casing against exterior corrosion, and/or to stabilize
caving rock formations. Cement grout, consisting of a mixture of cement and water and some-
tuifer. It times various additives, can be placed by a dump bailer, by a tremie pipe, or by pumping.5· 21
support It is important that the grout be introduced at the bottom of the space to be grouted to ensure
ings are that the zone is properly sealed.
hreaded
ling, the
tole and
5.4.3 Screens
sing for In consolidated formations, where the material surrounding the well is stable, groundwater can
enter directly into an uncased well. In unconsolidated fonnations, however, wells are equipped
or frac- with screens. These stabilize the sides of the hole, prevent sand movement into the well, and
smface allow a maximum amount of water to enter the well with a minimum of hydraulic resistance.
ling, (2) In the cable-tool method of drilling, screens are normally placed by the pullback method.
(4) aid- After casing is in place, the screen is lowered inside, and the casing is pulled up to near the top
ing may of the screen. A lead packer ring on the top of the screen is flared outward to form a seal between
teters of the inside of the casing and the screen. For the rotary method of drilling without casing, screens
are lowered into place as drilling mud is diluted and again are sealed by a lead packer to an upper
permanent casing. Screens are also sometimes placed by the bail-down method, involving bail-
ing out material below the screen until the screen section is lowered to the desired aquifer depth.
The placement of the well screen or open interval of a discharging well can greatly influence
:asing
iblewith
areas contributing recharge to the well and vulnerability of the well to contamination. The
:curate "degree of separation" refers to a combination of two factors. The first is the distance between
the water table and the top of the well screen or open interval of the well (greater distance
Figure 5.4.1. Degrees of separation between the water table and the well screen of a cased well or
the open hole of an uncased well. 34
implies greater separation). The second is the thickness and vertical and horizontal conductivi-
ties of confining units and aquifers between the water table and open interval of the well (in par-
ticular, greater thickness and lower hydraulic conductivity of low-conductivity confining units
imply greater separation). The wells in Figure 5.4.1 part a illustrate various degrees of separa-
tion in a layered aquifer. In general, greater degrees of separation lead to a more complex three-
dimensional configuration of groundwater flowpaths, longer travel times from the water table
or surface-water body to a discharging well, extension of the contributing area farther from the
discharging well, and possibly a more discontinuous and complex shape of the contributing
area. A separate sketch for bedrock wells is provided in part b (well6) to illustrate a practice in
some areas to install well casing only to the top (surface) of consolidated rock. As a result, the
bottom of the well casing can be above or only a few feet below the water table. In these situa-
tions, the well can be particularly vulnerable to possible contamination from groundwater near
the water table; that is, no separation exists between the water table and the top of the operation
interval of the bedrock well. Homeowner wells and virtually all public-supply wells tapping
unconsolidated deposits (wells 1-5) usually are screened some distance below the water table.34
In the past, well casings were often perforated in place by a special cutting knife. This
practice is now generally discontinued because of the large irregular openings created, the
small percentage of open area obtained, and the difficulty of controlling entry of sand with
water during pumping. More common is the use of preperforated casing, constructed by saw-
ing, machining, or torch-cutting slots in the casing. Slot openings range from less than 1 to 6
mm; with larger slots the maximum percentage of open area is about 12 percent.94 Openings
by sawing or machining can be properly sized, whereas torch-cut slots tend to be large, irreg-
ular, and conducive to sand entry.
Figure 5.4.2. Continuous

slot wire-wound well screen
in an unconsolidated forma-
tion. The grain size distribu-
tion around the screen illus-
trates a properly developed
well (courtesy Johnson Div.,
UOP Inc.5\
A major factor in controlling head loss through a perforated well section is the percentage
of open area. For practical purposes a minimum open area of 15 percent is desirable; this value
is readily obtained with many commercial screens but not with preperforated casing. 92· 94
Manufactured screens are preferred to preperforated casing because of the ability to tai-
lor opening sizes to aquifer conditions and because of larger percentages of open area that can
be achieved. Several types are available: punched, stamped, louvred, wire-wound perforated
pipe, and continuous-slot wire-wound screens. The latter type, consisting of a continuous
winding of round or specially shaped wire on a cage of vertical rods, is the most efficient, pos-
sesses the largest open area, and can be closely matched to aquifer gradations (see Figure
5.4.2). Although such screens are more expensive, they may prove to be more economical,
especially for thin but highly productive aquifers.
Screens are available in a range of diameters; selection of screen diameter should be made
on the basis of the desired well yield and aquifer thickness. Recommended minimum screen
diameters are included in Table 5.4.1. To minimize well losses and screen clogging, entrance
velocities should be kept within specified limits. Because aquifers composed of finer grained
materials tend to clog more easily than do those of coarser materials, field experience has indi-
:llor cated that there is a relationship between hydraulic conductivity of an aquifer and screen
entrance velocity, as shown in Table 5.4.2.
To express the velocities in Table 5.4.2 in terms of screen size, the following equation can
.ductivi- be applied:
(in par-
ng units
v = Q (5 .4.1)
s C1tdsLsP
separa-
!X three- where vs is the optimum screen entrance velocity, Q is well discharge, c is a clogging coef-
ter table ficient (estimated at 0.5 on the basis that approximately 50 percent of the open area of a screen
From the
tributing
actice in Table 5.4.2 Optimum Entrance Velocity of Water Through a Well Screen (after Walton 101 )
!SU1t, the Hydraulic conductivity Optimum screen entrance
:se situa- of aquifer (m/day) velocity (rnlmin)
aternear
>250 3.7
•peration
; tapping 250 3.4
200 3.0
:r table.34
160 2.7
tife. This
120 2.4
ated, the
100 2.1
and with
80 1.8
I by saw-
60 1.5
,an 1 to 6
40 1.2
)penings
20 0.9
ge, irreg-
<20 0.6
will be blocked by aquifer material), ds is the screen diameter, Ls is the screen length, and P is 100
the percentage of open area in the screen (available from manufacturer's specifications). Thus,
for a given aquifer material, aquifer thickness, well yield, and type of screen, the appropriate
80
diameter and length of well screen can be selected.
Screens are made of a variety of metals and metal alloys, plastics, concrete, asbestos- Ol
c::
cement, fiberglass-reinforced epoxy, coated base metals, and wood. 94 Because a well screen is 'Ci)
Cl) 60
ro
particularly susceptible to corrosion and incrustation, nonferrous metals, alloys, and plastics 0.
are often selected to prolong well life and efficient operation. Table 5.4.3 lists the more com- c
Q)
f:! 40
mon metallic screen materials in the order of increasing cost. Q)
0..
A significant characteristic of a well screen is its slot size, which should be determined
from mechanical analyses of formation samples obtained during drilling of the well or a pilot 20
hole.* If the uniformity coefficient of an aquifer sample for a naturally developed well (with-
out a gravel pack) is 5 or less, the selected slot size should retain 40 to 50 percent of the aquifer 0
materiaL For a uniformity coefficient greater than 5, the slot size should retain 30 to 50 per- 0
cent of the aquifer material. 94 In essence, the screen permits finer material to enter the well and
to be removed by bailing during development of the well. 36• 48 · 87 • 98 But coarser material is
retained outside, fanning a penneable envelope around the well (Figure 5.4.2). An illustration
of how a well screen can fonn a natural gravel pack around a well is shown in Figure 5.4.3.
Here, with 40 percent of the aquifer material retained, the resulting gravel envelope has a per-
meability more than 30 times greater than that of the aquifer.
Where a well screen is to be surrounded by an artificial gravel pack, the size of the screen
openings is governed by the size of the gravel (see next section).
5.4.4 Gravel Packs

A gravel-packed well is one containing an artificially placed gravel screen or envelope sur-
rounding the well screen (see Figures 5.3.l(a-c)). A gravel pack (l) stabilizes the aquifer,
(2) minimizes sand pumping, (3) pennits use of a large screen slot with a maximum open area,
Table 5.4.3 Metallic Well Screen Materials and Their Resistance to Acid and
Corrosion (after U.S. Bureau of Reclamation94)
Corrosion resistance
Material a Acid resistance in normal groundwater
Low-carbon steel Poor Poorb.c
Uniform
Toncan and Armco iron Poor Fairb,c
(U)
Admiralty red brass Good Goode
Silicon red brass Good Goode
304 stainless steel Good Very good
Everdure bronze Very good Very goodd
Monel metal Very good Very goodct
Super nickel Very good Very goodd 2
•Materials are listed in order of increasing cost.

'Not recommended for permanent installations where incrustation is a serious problem.
"Not recommended for permanent installations where sulfate-reducing or similar bac-
teria are present or where water contains more than 60 mg/l S04 .
~ecommended only in areas where corrosion is very aggressive.
•Screen manufacturers will often recommend the most satisfactory slot size based on the grain size analysis of a
given aquifer.
mdPis 100
). Thus,
·. ,.
ropriate .....:.
80 20
>bestos- Ol -o
c (I)
creen is 'iii
1/) 60 40 -~
~
!';l
plastics c.
.....
recom-
c
Q) Gravel envelope
c
!: 40 -60 ~....
(I) adjacent to 1.55 mm Q)
0.. 0..
screen slot size
~mrined Uc = 1.4
r a pilot 20 -80
K = 1,900 rn/day
ll (with-
:aquifer 0 Figure 5.4.3. Development of a natural gr:avel
0 2 3 pack around a well screen from a well-graded
50 per-
Slot size (mm) unconsolidated aquifer {after Mogg71 ) .
Nell and
tterial is
tstration and (4) provides an annular zone of high permeability, which increases the effective radius and
re 5.4.3. yield of the well. Maximum grain size of a pack should be near 1 em, while the thickness
iS a per-
should be in the range of 8 to 15 em.
Various formulas for relating gravel pack grain-size gradations to aquifer grain-size gra-
e screen dations have been developed. 52• 60• 71 · 84 Criteria confonning to U.S. Bureau of Reclamation
field experience are summarized in Table 5.4.4. The selected gravel should be washed and
screened siliceous material that is rounded, abrasive-resistant, and dense. Gravel should be
placed in such a manner as to ensure complete filling of the annular space and to minimize seg-
regation. A common procedure is to extend two trernie pipes to the bottom of the well on oppo-
ope sur- site sides of the screen. Gravel is poured, washed, or pumped into the tremie pipes; these are
aquifer, then withdrawn in stages as the pack is placed. In cable-tool holes, the inner casing and screen
>en area, are set inside the blank outer casing, the annular space is filled with gravel, and thereafter the
outer casing is pulled out of the well. In sandy aquifers, where a gravel pack is most essential,
deep wells should be constructed by the rotary or reverse-circulation rotary method. The
drilling fluid should be circulated and diluted with water before the gravel is introduced.
Table 5.4.4 Criteria for Selection of Gravel Pack Material (after U.S. Bureau of Reclamation94)
Uniformity coefficient
(Uc) of aquifer Gravel pack criteria Screen slot size
< 2.5 (a) Uc between 1 and 2.5 with the 50% size not greater than ::; 10% passing size of the gravel pack
six times the 50% size of the aquifer
(b) If (a) is not available, Uc between 2.5 and 5 with 50% size
not greater than nine times the 50% size of the aquifer
---- ------- ..-.... --.. -------------------------------- ------------------------------------------------------------------------------- --------- ------- ----------- ------- -------------------- --------
2.5-5 (a) Uc between 1 and 2.5 with the 50% size not greater than s 10% passing size of the gravel pack
nine times the 50% size of the formation
(b) If (a) is not available, Uc between 2.5 and 5 with 50% size
not greater than 12 times the 50% size of the aquifer
--- .... ______ -------------- ----·--........ -----.......................................... -- ----. ----- -........ --- ... --------------- ..-----.. --- ...-.------- --------- ------ ----------- ---------------------------- --·--- -----
~
>5 (a) Multiply the 30% passing size of the aquifer by 6 and 9 and :s; 10% passing size of the gravel pack
locate the points on the grain-size distribution graph on the
same horizontal line.
(b) Through these points draw two parallel lines representing
materials with Uc::; 2.5 .
>is of a
(c) Select gravel pack material that falls between the two lines.
5.5 WELL DEVELOPMENT

Following completion, a new well is developed to increase its specific capacity, prevent sanding,
and obtain maximum economic well life. These results are accomplished by removing the finer
material from the natural formations surrounding the perlorated sections of the casing. Where a
well has been gravel packed, much of the same purpose has been accomplished, although devel-
opment is still beneficial. The importance of developing wells cannot be underestimated; all too
often development is not carried out adequately to produce full potential yields.
Development procedures are varied and include pumping, surging, use of compressed air,
hydraulic jetting, addition of chemicals, hydraulic fracturing, and use of explosives. 53 · 94 These
are briefly described in subsequent paragraphs.
5.5.1 Pumping
This procedure involves pumping a well in a series of steps from a low discharge to one exceed-
ing the design capacity.* To be most effective, the intake area of the pump should extend to near
the center of the screened section. At each step the well is pumped until the water clears, after
which the power is shut off and water in the pump column surges back into the well. The step
is repeated until only clear water appears. The discharge rate is then increased and the procedure
repeated until the final rate is the maximum capacity of the pump or well. This irregular and
noncontinuous pumping agitates the fine material surrounding the well so that it can be carried
into the well and pumped out. The coarser fraction entering the well is removed by a bailer or
sand pump from the bottom. This development method by pumping is recommended as a fin-
ishing procedure after any of the development techniques described subsequently.
5.5.2 Surging
Another method for developing a well is by the up-and-down motion of a surge block attached
to the bottom of a drill stem. Such blocks are particularly applicable with a cable tool rig. 94 Solid,
vented, and spring-loaded surge blocks, often constructed by well-drilling contractors, are
employed.t The cylindrical block is 2 to 5 em smaller than the well screen and fitted with belt-
ing, rubber, or leather that will not damage the screen. As the block is moved up and down in the
screen, a surging action is imparted to the water. The downstroke causes backwash to break up
any bridging that may occur, while the upstroke pulls dislodged sand grains into the well.
Initially, surging should begin with a slow stroke at the bottom of the screen and progress
to the top of the screen. This should then be repeated with increasingly faster strokes. The pro-
cedure is completed when material accumulating in the bottom of the well becomes negligible.
For wells in rock aquifers, surging can be accomplished in the casing above open holes.
5.5.3 Surging with Air

To develop wells with compressed air, an air compressor is connected to an air pipe into the
well. Around the air pipe a discharge pipe is fitted, as shown in Figure 5.5.1. (Both pipes
should be capable of being shifted vertically by clamps. Initially, the pipes extend to near the
bottom of the screened section; for efficient operation, the water depth in the discharge pipe 5.5.4 B
*In the field this technique is sometimes referred to as "rawhiding" a well. If a high discharge occurs initially,
"bridging" (wedging sand grains around individual perforations formed by the sudden pull on the sand toward the
well) can prevent fine material from being removed and reduce the effectiveness of the development process.
1Surging can also be accomplished with a flap-valve bailer, if a close fit exists within the well screen.
Clamp
mding,
1e finer
Vhere a Discharge-
1 devel-
; all too
sed air,
4
These
/"Well casing
exceed-
!to near
rs, after
~he step
ocedure Well screen
1lar and
:carried
'ailer or
Discharge pipe
as a fin-
Air pipe
attached Fine sand brought

94 Solid, into well
tors, are Ball
•ith belt- Figut"e 5.5.1. Installation for well devel-
vn in the opment with compressed air.
Jreak up
:11.
progress should exceed two-thirds the length of the pipe. To begin the development, the air pipe is
Thepro- closed and the air pressure is allowed to build up to 0.7 to 1.0 x 106 Pa, whereupon it is
gligible. released suddenly into the well by means of a quick-opening valve. The inrush of air creates a
es. powerful surge within the well, first increasing then decreasing the pressure as water is forced
up the discharge pipe. The process loosens the fine material surrounding the perforations; the
material may then be brought into the well by continuous air injection, creating an airlift
pump. The operation is repeated at intervals along the screened section until sand accretion
into the becomes negligible.
tth pipes
near the
rrge pipe 5.5.4 Back washing with Air
In the backwashing method, the top of the well is fitted with an airtight cover. Discharge and
11ly, air pipes are irlstalled similar to the previous method, together with a separate short air pipe and
ward the a three-way valve, as shown in Figure 5.5.2. Compressed air is released through the long air
~s. pipe, forcing air and water out of the well through the discharge pipe. After the water clears,
the air supply is shut off and the water is allowed to return to its static level. The three-way
Solid flange Discharge
Connecting flange
Discharge pipe
Air pipe
5.5.6
Drive shoe
Figure 5.5.2. Installation for

well development by back-
washing with air.
valve is then turned to admit air into the top of the well through the short air pipe. This
backwashes the water from the well through the discharge pipe and at the same time agitates 5.5.7
the sand grains surrounding the well. Air is forced into the well until it begins escaping from
the discharge pipe, after which the three-way valve is turned and the air supply is again
directed down the long air pipe to pump the well. Backwashing is repeated until the well is
fully developed.
5.5.5 Hydraulic Jetting

Jetting with a high-velocity stream of water is an effective development technique in open rock 5.5.8
holes and in wells containing screens with large percentage openings (see Figure 5.5.3). The
jet nozzle, mounted horizontally, is attached to a string of pipe, which is connected through a
swivel and hose to a high-pressure, high-capacity ptimp. The jet head is slowly rotated to suc-
cessively higher levels. Fine-grained material from unconsolidated aquifers is carried into the
well by the turbulent flow; in addition, the method is particularly effective in developing
gravel-packed wells.
Figure 5.5.3. High-velocity hydraulic jetting

through a continuous slot wire-wound well
screen for well development (after Johnson Div.,
UOP Inc.5\
5.5.6 Chemicals
Open-hole wells in limestone or dolomite fonnations can be developed by adding hydro-
chloric acid to water in the well. The solvent action removes fine particles and tends to widen
fractures leading into the well bore. Normally this procedure would be followed by one of the
previously described development methods. Hydrofluoric acid can be similarly employed for
rocks containing silicates.
For most development methods, adding one of the polypbosphates* to water in the well
will aid the development process.94 These compounds act as deflocculants and dispersants of
clays and other fine-grained materials, thereby enabling the mud cake on the wall of a hole and
the clay fractions in an aquifer to be more readily removed by the development.
Blocks of solid carbon dioxide (dry ice) are sometimes added to a well after acidizing and
tion for
Jack- surging with compressed air to complete well development. The accumulation of gaseous car-
bon dioxide released by sublimation builds up a pressure within the well; upon release this
causes a burst of muddy water from the well.
i)e. This
agitates 5.5.7 Hydraulic Fracturing
ng from Hydraulic fracturing, a technique borrowed from the petroleum industry, is occasionally
is again employed to enhance the yield of open-hole rock wells. Inflatable packers on a pipe ex.tending
! well is
to ground surface isolate a section of aquifer. After filling the pipe and isolated section with
water, pump pressure is applied to fracture the rock. Sand is sometimes pumped into the sec-
tion to force the grains into the rock fractures, to maintain the openings.
pen rock 5.5.8 Explosives

i.3). The
nrough a Detonation of explosives in rock wells often increases yields by enlarging the hole, increasing
d to suc- rock fractures, and removing fine-grained deposits on the face of the well bore.
1into the
veloping 'Common household phosphate-based detergents can serve as a substitute; excessive foaming when the well is
pumped may result.
5.6 TESTING WELLS FOR YIELD

Following development of a new well, it should be tested to determine its yield and drawdown.
This information provides a basis for determining the water supply available from the well, for
selecting the type of pump, and for estimating the cost of pumping. A test is accomplished by
measuring the static water level, after which the well is pumped at a maximum rate until the
water level in the well stabilizes. The depth to water is then noted (techniques for measuring
water well depths are described in Chapter 12). The difference in depths is the drawdown, and
the discharge-draw down ratio is an estimate of the specific capacity of the well. The discharge
can be determined by any of several measuring devices connected to the discharge pipe.
,.......
.0
"'<.l0
5.7 PUMPING EQUIPMENT .E
0
r/)
Well pumps produce flow by transforming mechanical energy to hydraulic energy. A wide -s-.;;
variety of pumps are produced annually by many manufacturers. The selection of a particular :r:
0
size and type of pump depends on several factors, including: (1) pumping capacity, (2) -~
:0
well diameter and depth, (3) depth and variability of pumping level, (4) straightness of the ~
well, (5) sand pumping, (6) total pumping head, (7) duration of pumping, (8) type of power r/)
available, and (9) costs. 94 Table 5.7.1 summarizes characteristics of pumps most frequently :::)
....
employed in wells.
~
'-"
V>
5.7.1 Total Pumping Head ~

.8
"0
The total pumping head, or total dynamic head, of a pump represents the total vertical lift of Ill
;;.,
the water from the well. As shown in Figure 5.7.1, this total head consists of three components: ,Q
(1) the drawdown inside the well (including aquifer and well losses); (2) the static head, being ~
ll.l
the difference between the static groundwater level and the static discharge elevation; and (3)
friction losses due to flow through the intake and discharge pipes. As indicated in the figure, -...
;;.,
Q
0
;:I
the total pumping head increases with discharge. 0"'
J:
s"'
0.
(il
5.7.2 Pumps for Shallow Wells ~
;:I
....0 ;::J"'
For shallow wells where only small discharges are needed, hand-operated pitcher pumps, tur-
"'
u
bine pumps, and gear pumps may be installed. Discharges range up to 500m3/day. Suction lifts ·a
<ll
·~::
should not exceed about 7 m for efficient and continuous service. ~
Where a larger discharge is required from a shallow well, a centrifugal pump is commonly G (il::l
.l:l
employed. The assembly may be mounted with a horizontal or a vertical shaft. The horizontal u ;::J"'
design is efficient, easy to install and maintain, and usually connected directly to an electric 1""(
motor. Because of the low suction head, the pump is often placed a short distance above the
r-;
Ill
water level in large diameter wells. .!
~
~ Oi
·.ou .
u
5.7.3 Pumps for Deep Wells £
In deep wells requiring high lifts, large-capacity pumps serving irrigation, municipal, or
industrial water requirements are installed. Several types of pumps are suitable for deep-well
operation: plunger, displacement, airlift, jet, and, most importantly, deep-well turbine and
submersible.
The deep-well turbine pump has been widely adopted for large, deep, high-capacity wells
(see Figure 5.7.2a). This type of pump has its impeller suspended vertically on a long drive
shaft within a discharge pipe. The bowl of the pump contains the impeller and guide vanes;
0
<0<1'<
:::J :-
s· ~ -· 0
<
n> ::J. <:!
o-N a0 o-~ ::::n§r-o-- co "' 0 Q :=. ..- l" n
:r.
:5 tn ;::;. ~
s:: ::::.:8_.:-o
~ c..
~o.?f~::::
0-o"Cl :::1 Y'
0. :lE • Q) o - . - <">:lE
0 (") 0 ~ ....., ~
o._. 0<1s· a:': ~ ~~s--=-o.: ::J. -
Q)
g ::J. n> Q) ~ .:-
s-Eiag. -·E "' -
5_!!.0
.... 0 (") t:-'< a'";' • 8., '<~o~e:n> 0 0.0<1 0'< ....
Table 5.7.1 Characteristics of Pumps Frequently Employed in Wells (after U.S. Public Health Service96)
Practical Usual Usual

suction well-pumping pressure
Type of pump lifta depth heads Advantages Disadvantages
Reciprocating:
Shallow well 6-7 m 6-7m 30-60m Positive action; discharge against variable Pulsating discharge; subject to vibration
Deep well 6-?m Up to 180m Up to 180m heads; pumps water containing sand and silt; and noise; maintenance cost may be
above especially adapted to low capacity and high high; may cause destructive pressure
cylinder lifrs. if operated against closed valve
Centrifugal:
Shallow well
Straight 6mmax. 3-6m 30-45m Smooth, even flow; pumps water containing Loses prime easily; efficiency depends
centrifugal sand and silt; pressure on system is even and on operating under design heads and
(single stage) free from shock; low starting torque; usually speed
reliable and good service life
Regenerative 8mmax. 8m 30-60 m Same as straight centrifugal except not Same as straight centrifugal except
vane turbine type suitable for pumping water containing sand maintains priming easily
(single impeller) or silt; self-priming
Deep well
Vertical line Impellers 15-90m 30-250 m Same as shallow well turbine Efflciency depends on operating Wlder
shaft turbine submerged design head and speed; requires straight
(multistage) well large enough for turbine bowls and
housing; lubrication and alignment of
shaft critical; abrasion from sand
Submersible Pump and 15- 120 m 15-120 m Same as shallow well turbine; easy to frost- Repair to motor or pump requires
turbine motor proof installation; short pump shaft to motor pulling from well; sealing of electrical
(multistage) submerged equipment from water vapor critical;
abrasion from sand
Jet:
Shallow well 4-6 m below Upto4-6 m 25-45 m High capacity at low heads; simple in Capacity reduces as lift increases; air in
ejector below ejector operation; does not have to be installed over suction or return line will stop pumping Ul
well; no moving parts in well ~
Deep well 4-6 m below 7-35 m 25-45 m Same as shallow well jet Same as shallow well jet 2'
ejector 60 mrnax. 3
'0
Rotary: :;·
<ro
Shallow well 7m 7m 15-75 m Positive action; discharge constant under Subject to rapid wear if water contains sand CT1
J:>
(gear type) variable heads; efficient operation or silt; wear of gears reduces efficiency ~
.a·
Deep well (helical Usually 15-150 m 30-150 m Same as shallow well rotary; only one Same as shallow well rotary except no s0
rotary type) submerged moving pump device in well gear wear ~
"Practical suction lift at sea level. Reduce lift 0.3 m for each 300 m above sea level. N
~
~
Friction
loss
Storage
reservoir
t
c Pressure
b--- iw
Static
head head
To use Total I
pumping
Pump
head
Pump axis
Figure 5.7.1. Diagram illustrating total

pumping head for a well supplying astor-
age reservoir. Note the increase in head as
Discharge ---+ a function of well discharge.
several bowls connected in series for higher heads form a multiple-stage pump. The pump is
usually driven by an electric motor at the ground surface and connected by a long vertical shaft
positioned by bearings within the discharge pipe. Deep-well pumps, being submerged, require
no priming and are capable of operating under a wide range of water levels without having to
be reset.
The submersible pump is simply a deep-well turbine pump close-coupled to a small-
diameter submersible electric motor, as shown in Figure 5.7 .2b. The efficiency of the pump is
increased by direct coupling, while effective cooling results from complete immersion. Pump
sizes range from small units that fit inside an 8-cm casing to large-capacity units involving
numerous stages. An advantage of submersible pumps is that they can lift water from deep
wells where long shafts in crooked casings might prohibit installation of a deep-well turbine
pump. Other advantages, which account for increasing installations of this type of pump,
include ease of maintenance, freedom from noise, protection from weathering and flooding,
and avoidance of a large above ground installation.
5.8 PROTECTION OF WELLS

5.8.1 Sanitary Protection
Wherever groundwater pumped from a well is intended for human consumption, proper
sanitary precautions must be taken to protect the water quality. Pollution sources may exist
either above or below ground surface (see Chapter 8). Precautions apply equally to springs;
Figure 5.8.1 shows, for example, a typical method for protecting a spring water supply.
Surface pollution can enter wells either through the annular space outside of the casing or
through the top of the well itself. To close avenues of access for undesirable water outside of
the casing, the annular space should be filled with cement grout as shown for deep wells in Fig-
5.8 Protection of Wells 235
Pump
l total
g a star-
head as
Motor
JUmp is
al shaft
require
.ving to
small-
lUmp is
Figure 5.7.2. Pumps for
. Pump
deep wells. (a) Turbine
tolving
(b) Submersible (after
n deep (a) (b) Anderson6)
turbine
pump,
lOding,
ure 5.3 .1. Entry through the top of the well can be avoided by providing a watertight cover to
seal the top of the casing. 12 Some pumps are available with closed metal bases that provide the
necessary closure. For pumps with an open-type base, or where the pump is not placed directly
over the well, a seal is required for the annular opening between the discharge pipe and cas-
ing. • Seals may be made of metal or lead packing; asphaltic and mastic compounds are also
satisfactory. Covers around the well should be made of concrete, should be elevated above the
proper adjacent land level, and should slope away from the well (see Figure 5.8.2).
y exist Whenever a new well is completed or an old well repaired, contamination from equipment,
Jrings; well materials, or surface water may be introduced into the well. Addition and agitation of a
chlorine compound will disinfect the well. Following disinfection, the well should be pumped
)ing or
5ide of
in Fig- It is desirable to provide a small opening in or below the pump base to allow for periodic water level measurements.
Surface water
, diversion
\ ditch
\
\
Cobb I·
\
I
I
I
I
I
I
I
I
I Fence
I
I
~
I I
I I
I I
_ _ _ _ _j
I
I
Cleanout Screened
drain
~
II
drain Perforated
pipe
Plan
Surface water
diversion ditch
~}......._ Perforated
pipe
Water-bearing gravel
5.8.3 j
Figure 5.8.1. Plan and elevation views of a

developed spring showing a typical method
Cleanout drain for providing sanitary protection (after U.S.
Elevation
Public Health Service 96).
to waste until all traces of chlorine are removed. As a final check on the potability of the water, 5.9 W1
a sample should be collected and sent to a laboratory for bacteriological examination.
5.8.2 Frost Protection

In regions where winter frost occurs, it is important to protect pumps and water lines from
freezing. A typical method for frostproofing a domestic well is shown in Figure 5.8.3. The pit-
less adapter, attached to the well casing, provides access to the well, while the discharge pipe
runs about 2 m underground to the basement of the house.
5.9 Well Rehabilitation 237
Reinforced concrete
cover slab sloped
away from pump
· ...... Figure 5.8.2. A drilled well showing grout

.. •.:. -:· . seal, concrete slab, and well seal for sanitary
protection (after U.S. Public Health
Service96).
5.8.3 Abandonment of Wells

Whenever a well is abandoned, for whatever reason, it should be sealed by filling it with clay,
iews of a
l method
concrete, or earth. Not only is surface contamination then unable to enter the well, but sealing
liter U.S. serves other useful purposes: prevents accidents, avoids possible movement of inferior water
from one aquifer to another, and conserves water in flowing wells?0
5.9 WELL REHABILITATION

1e water,
. A new well, properly drilled, cased, and developed, will give years of satisfactory service with
little attention. Many wells fail, however; that is, they yield decreasing quantities of water with
time." Well rehabilitation refers to the treatment of a production well by mechanical, chemi-
cal, or other means to recover as much as possible of the lost production capacity.58• 59 Table
nes from 5.9.llists well rehabilitation methods and their applications to various types of aquifers.
. The pit-
arge pipe 'Frequently the pLIUlp rather than the well is at fault; hence, it should be checked before beginning any extensive
well repair.
1
Ground
Electrical surface MetI
control box _ _j__
Muriati
Conduit for wiring I folio
chlo1
Polyph•
foUo
chlo1
Pressure Dynam
switch
Compn
casing
Dry ice
Surging
Chiorin
Caustic
Sand and gravel

aquifer
Figure 5.8.3. Diagram of a domestic well
installation with a pitless adapter to protect the
well from frost (after Gibb37).
One cause of failure is depletion of the groundwater supply. Not a fault of the well, this
trouble can sometimes be remedied by decreasing pumping drafts, resetting the pump, or
deepening the well. A second cause of well trouble results from faulty well construction. Such 5.10
items as poor casing connections, improper perforations or screens, incomplete placement of
gravel packs, and poorly seated wells are typical of difficulties encountered. Depending on the
particular situation as determined from a television or photographic survey of the well (see
Chapter 12), it may be possible to repair the well, but sudden failures involving entrance of
sand or collapse of a casing often require replacement of the entire well.
The third and most prevalent cause of well failure results from corrosion or incrustation
of well screens. 8· 62• 81 Corrosion may result from direct chemical action of the groundwater or
from electrolytic action caused by the presence of two different metals in the well. The effects 5.10.1
of corrosion can be minimized by selecting nonmetallic well screens or ones of corrosion-
resistant metal (such as nickel, copper, or stainless steel), and by providing cathodic protec-
tion. • If the damage is localized, it may be possible to insert a liner inside the screen to prevent
excessive sand pumping.
Incrustation is caused by precipitation on or near well screens of materials carried in solu-
tion by groundwater.72 The sudden pressure drop associated with water entering a well under
heavy pumping can release carbon dioxide and cause precipitation of calcium carbonate.
Another cause of incrustation stems from the presence of oxygen in a well; this can change
soluble ferrous iron to insoluble ferric hydroxide. Screens can be cleaned by shooting a string
'One method of providing cathodic protection for a well is to introduce a metal low on the electrochemical scale
that will be corroded instead of the well casing. Rods of magnesium suspended in the well water serve this purpose.
5.I 0 Horizontal Wells 239
Table 5.9.1 Rehabilitation Methods and Their Applications to Various Types of Aquifers (after Erickson30)
Method Unconsolidated aquifers Consolidated sandstone Consolidated limestone

Muriatic acid" Removes iron, sulfur, and carbonate Not usually effective Sometimes beneficial; best results
follmyed by deposits obtained by pressure acidizing
chlorine
Polyphosphate Removes fine silt, clay, colloids, Not usually effective Not usually effective
followed by disseminated shale, and soft
chlorine iron deposits
Dynamiting Not recommended Effective for aU types of well- Effective when very large charges
screen deposits are used
Compressed air Removes plugging deposits of silt and Not used Not used
fine sand in areas adjacent to screens
Dry ice Same as compressed air Used only rarely, to remove Not usually effective
cuttings from the face of a
new production well
Surging Same as compressed air Rarely used Rarely used
Chlorineb Removes iron and slime-forming Removes iron and slime- Removes iron and slime-forming
bacteria forming bacteria bacteria
Caustic soda Removes oil scum left by oil- Re111oves oil scum left by oil- Removes oil scum left by oil-
lubricating pumps lubricating pumps lubricating pumps
'Not to be used with concrete screens.
bUsually used in a concentration of 500 mg/1.
ell
of vibratory explosives in the well or by adding hydrochloric acid (HCI) or sulfamic acid
:eel the (H2NS0 3H) to the well, followed by agitation and surging. Where slime-forming organisms
block screens, particularly in recharge wells, treatment with chlorine gas or hypochlorite
solutions94 can remedy the problem. For improving yields of rock wells, acidizing or shooting
:ll, this with explosives is generally effective. 102
mp, or
1. Such 5.10 HORIZONTAL WELLS
1ent of
Subsurface conditions often preclude groundwater development by normal vertical wells. Such
on the
conditions may involve aquifers that are thin, poorly permeable, or underlain by permafrost or
:ll (see
saline water. In other circumstances, where groundwater is to be derived primarily from infil-
mce of
tration of streamflow, a horizontal well system may be advantageous. Also, in developing
areas of the world, a horizontal well may be far less costly than a vertical well.
station
'ater or
effects 5.10.1 Infiltration Galleries
rosion· An infiltration gallery is a horizontal conduit for intercepting and collecting groundwater by
protee· gravity flow. 9 Qanats, described in Chapter 1, illustrate one type of gallery. Galleries, nonnally
Jrevent constructed at the water table elevation, discharge into a sump where a pump lifts the water to
ground surface for use. In Europe and the United States, many infiltration galleries are laid par-
n solu· allel to riverbeds, where with induced infiltration an adequate perennial water supply can be
l under obtained (see Chapter 13). Depending on the type of aquifer penetrated, galleries may be
Jon ate. unlined or lined with vitrified clay, brick concrete, or cast iron. •
change
1 string
'Deep infiltration galleries exist in the chalk aquifers of southeastern England. Here the permeability of solid chalk
;cale is quite low, and unlined horizontal tunnels (or adits), roughly 2 min diameter and extending for distances up to 2
Jurpose. km, intersect a maximum number of fissures from which roost of the water is obtained.
In Alaska, galleries have been widely employed to obtain water supplies where underly-
ing permafrost (see Chapter 2) would not contribute groundwaterY On oceanic islands, gal-
leries have the particular advantage of enabling fresh water to be collected with little
disturbance of underlying saline water (see Chapter 14). Such installations are well known in
Hawaii (see Figure 5.10.1), for example, where they are unlined in basalt,78· 103 and in Barba-
dos, where they are unlined in coral limestone.
5.10.2 Horizontal Pipes

On sloping ground surfaces, small-diameter horizontal holes can be drilled by the rotary
method. 93· 107 Perforated pipes placed in these holes tap groundwater that would otherwise be
discharged by seepage or from small springs. Two examples are shown in Figure 5.10.2. Such
horizontal pipes provide sanitary and lowcost water, and, in addition, enable flow to be con-
trolled by valves at the discharge ends. The same technique has also been widely employed to
drain side slopes, such as in highway cuts, in order to reduce the possibility of landslides.
5.10.3 Collector Wells

For cities and industries located near rivers, the problem of obtaining high-quality, low-
temperature water at reasonable cost has become increasingly difficult. In Europe and the
United States, groundwater pumped from collector wells tapping permeable alluvial aquifers
has often proved to be a successful solution.39· 89 If located adjacent to a surface water source,
a collector well lowers the water table and thereby induces infiltration of surface water through
the bed of the water body to the well (see Chapter 13 ). In this manner greater supplies of water
... -:< . .
Ventilating pipe
Discharge ·.·· · -·
: : .4 .• • : •
pipe .. .
··· ..·
Pumps
Figure 5.10.1. Cross section of an· infiltra-
Water table """\ tion gallery in Hawaii. These are locally
Basalt
aquifer known as Maui-type wells because they
were first constructed on the island of Maui
to provide water for irrigation of sugar cane
(after Watson 103).
Figure 5.10.2. Examples of horizontal

wells consisting of small-diameter perf-
orated pipes drilled into hillsides. These
wells should have a downward slope into
the aquifer to prevent fonnation of a vacu-
um inside the pipes (after Welchert and
Freeman 107). (a) Contact spring fonnation
(a) (b) (b) Dike spring formation
5.10 Horizontal Wells 241
erly- can be obtained than would be available from groundwater alone. Analytic solutions of the
. gal- flow to a coUector well have been developed (see Chapter 4).
little Plan and elevation views of a collector well are shown in Figure 5.10.3. The central cylin-
vnin der, consisting of a monolithic concrete caisson about 5 m in diameter, is sunk into the aquifer
arba- by excavating the inside earth material. After the requisite depth is reached, a thick concrete
plug is poured to seal the bottom. Perforated pipes, 15 to 20 em in diameter, are jacked hydrauli-
cally into the water-bearing formation through precast portholes in the caisson to form a radial
pattern of horizontal pipes. During construction, fine-grained material is washed into the cais-
otary
se be
Such
con-
·ed to
/"' High water level
s.
Jow-
td the
uifers
mrce,
rough
water
tfi.ltra-
llly
ey
fMaui
ucane . ·.· ···
.. ··. ·. ·.: .;·.
tal <IIIII II I II II I II !Ill II II II II p I ·!

!rf-
ese
into
,,.acu-
·•d Figure 5.10.3. A collector well located
·ion near a surface water body (courtesy
Ranney Method Water Supplies, Inc.).
son so that natural gravel packs form around the perforations (again see Figure 5.10.3). The
number, length, and radial pattern of the collector pipes can be varied to obtain the maximum
capacity; usually more pipes are extended toward than away from the surface water source.
The large area of exposed perforations in a collector well causes low inflow velocities,
which minimize incrustation, clogging, and sand transport. Polluted river water is filtered by
its passage through the unconsolidated aquifer to the well. The initial cost of a collector well
exceeds that of a vertical well; however, advantages of large yields, reduced pumping heads,
and low maintenance costs are factors to be considered. Yields vary with local conditions; the
average for a large number of such wells approximated 27,000 m3/day. Installations bordering
streams with regulated stages may encounter decreases in yield with time because of sediment
deposition on streambeds. 39
Collector wells can also function in permeable aquifers removed from surface water. Sev-
eral such installations gave an average yield of about 15,000 m3/day.
5.11 CHARACTERISTIC WELL LOSSES

5.11.1 Well Losses
The drawdown at a well includes not only that of the logarithmic drawdown curve at the well
face, but also a well loss caused by flow through the well screen and flow inside of the well to
the pump intake. Because the well loss is associated with turbulent flow, it may be indicated
as being proportional to an nth power of the discharge, as Q", where n is a constant greater than
one. Jacob51 suggested that a value n = 2 might be reasonaoly assumed, but Rorabaugh82
pointed out that n can deviate significantJy from 2. An exact value for n cannot be stated
because of differences of individual wells; detailed investigations of flows inside and outside
of wells show that considerable variations occur from assumed flow distributions.
Taking account of the well loss, the total drawdown swat the well may be written for the
steady-state confined case:
Q ro n
s =-ln-+CQ
w 2rtT r... (5.11.1)
where Cis a constant governed by the radius, construction, and condition of the well. For sim-
plicity let
(5.11.2)
so that
(5.11.3)
Therefore, as shown in Figure 5.11.1, the total drawdown sw consists of the formation loss BQ
and the well loss CQ".
Consideration of Equation 5.11.3 provides a useful insight to the relation between well
5.11.2
discharge and well radius. From Equations 4.2.3 and 4.2.7, it can be seen that Q varies
inversely with ln(r0/rw), if all other variables are held constant. This shows that discharge
varies only a small amount with well radius. For example, doubling a well radius increases the
discharge only 10 percent When the comparison is extended to include well loss, however, the
effect is significant. Doubling the well radius doubles the intake area, reduces entrance veloc-
ities to almost half, and (if n =2) cuts the frictional loss to less than a third. For axial flow
within the well, the area increases four times, reducing this loss an even greater extent.
It is apparent that the well loss can be a substantial fraction of total drawdown when
pumping rates are large, as illustrated by Figure 5.11.2. With proper design and development
5.11 Characteristic Well Losses 243
'he
urn
tes,
by
1el!
1dS,
the ,.
ing
.ent
ev-
Confined aquifer
Figure 5.11.1. Relation of well loss C(l' to draw-
down for a well penetrating a confined aquifer.
60r--------,---------.--------~
vell
li to
tted
han
~82
1ted
iide
the
l.l)
;im- Figure 5.11.2. Variation of total draw-
down s..,, aquifer loss BQ, and well loss
5 10 15 C(l' with well discharge (after
Well discharge, Q, 100m3/day Rorabaugh82 ).
1.2)
of new wells, well losses can be minimized. Clogging or deterioration of well screens can
increase well losses in old wells. 54 Based on field experience, Walton 101 suggested criteria for
1.3) the well loss coefficient C in Equation 5.11.3. These are presented in Table 5.11.1 to aid in
evaluating the condition of a well.
:BQ
well
5.11.2 Evaluation of Well Loss
tries To evaluate well loss, a step-drawdown pumping test is required. This consists of pumping a
arge well initially at a low rate until the drawdown within the well essentially stabilizes. 51 • 64· 86 The
s the discharge is then increased through a successive series of steps as shown by the time-drawdown
·,the data in Figure 5.11.3a. Incremental drawdowns 6.s for each step are determined from approxi-
~Joe mately equal time intervals. The individual drawdown curves should be extrapolated with a
flow slope proportional to the discharge in order to measure the incremental drawdowns.
From Equation 5.11.3 and letting n =2, we have
vhen s
nent .J£. = B+ CQ (5.11.4)
Q
Table S.ll.l RelationofWell Loss Coefficient to Well Condition (afterWalton 101 )

Well loss coefficient C (min2/m5) Well condition
< 0.5 Properly designed and developed

0.5 to 1.0 Mild deterioration or clogging
1.0 to 4.0 Severe deterioration or clogging
> 4.0 Difficult to restore well to original capacity
EXJ
0
.601 =500 m3/day 1 t.s1 = 1.0 m I I
t.0 2 = 500 m3/day

----
.652 =1.6 m
4r-
' --r-- -
t.Q3 = 1,000 m3/day

r63m
-r---.
8i- -
12 r-
16 r-
.604 =500m3/day .654 =5.1 m
"-r--
I
I
I
--
l
-
sou
.605 =250 m3/day 1 t.s 5 = 4.6 m
I I I
tj_ I --~--·
20
20 50 100 200 500 1000 2000
Time (min)
(a)
N
E 8.E
~
~ 7.E
"0
"'0I 4
,....
~
I C=
I
0.001 = 2 0 10-6
500 . X
E
......
S.E
(/)
~ S.E
Sw= 8.0 X 10-4 Q + 2.0 X 10-G Q 2 ""0
<')
I
4.E·
0
!:.. 3.E·
0
~ 2.E·
Figure 5.11.3. Step-drawdown pumping test analysis !/)
00 1000 2000 3000 to evaluate well loss (after Bierschenk1\ (a) Time- 1.E-
Well discharge, Q (m3/day) drawdown data from step-drawdown pumping test O.E ~
(b) Determination of Band C from graph of S)Q
(b) versus Q
S.ll Characteristic Well Losses 245
Therefore, by plotting s.JQ versus Q (see FigureS .11.3) and fitting a straight line through the
points, the well loss coefficient Cis given by the slope of the line and the fonnation loss coef-
ficient B by the intercept Q = 0.
Rorabaugh82 presented a modification of this graphic analysis to detennine n in cases
where it deviates significantly from 2.
·. The time-drawdown data from a step-drawdown pumping test in an existing well is given below. Deter-
mine the well loss coefficient C and the fonnation loss coefficient B assuming n =2, and find the corre-
sponding contributions of the formation loss and well loss for the given drawdown data.
Q (m3/day) s,. (m)
500 1.14
1,000 2.66
1,500 5.57
2,000 8.82
2,500 13.54
3,000 18.79
3,500 23.67
SOLUTION Using the given time-drawdown data, we plot values of s,jQ versus Q in Figure 5.11.4. Then, a straight
line is fitted through the data points.
The equation for the best-fit straight line is given by
y = 1.597 x l~x + 0.001307

or
s,.JQ = 1.597 x l0-6Q + 0.001307
Thus, from Equation 5.11.4, the fonnation loss coefficient is B = 0.001307 days/m2 and the well loss
coefficient is C = 1.597 x t0-6 day 2/m5 or C = 3.31 min2/m5. Note that this coefficient indicates severe
deterioration or clogging in the well according to Walton criteria (see Table 5.11 .1). 101
Using the coefficients computed above, we can calculate the drawdown due to fonnation loss and the
additional drawdown due to well loss for any discharge. The theoretical drawdown data based on the esti-
mated coefficients are reproduced for the given discharges in Table 5.11.2. As seen in the table, the
theoretical drawdowns do not match the measured data exactly because of then = 2 assumption. On the
other hand, they are close enough for practical purposes.
8.E- 03 ....--..---.---..,...-- -.----.-- -,-- -.-- - ,

7.E- 03 y = 1.597E - 0.6x + 1.307E - 03
16.E- 03
<n
~ 5. E- 03
"0
7 4.E- 03
0
~ 3.E-03
0
'1 2.E- 03
talysis <I)
me- 1.E - 03
est O.E + 00
IQ 0 500 1000 1500 2000 2500 3000 3500 4000
3
Well discharge, 0 (m /day) Figure 5.11.4. Drawdown data for Example 5.11.1.
Thble 5.11.2 Calculated Well Loss Parameters for Example 5. l 1.1 SOLl
Q (m3/day) BQ(m) CQ2 {m) s,.,(m)
500 0.65 0.40 1.05

1,000 1.31 1.60 2.90
1,500 1.96 3.59 5.55
2,000 2.61 6.39 9.00
2,500 3.27 9.98 13.25
3,000 3.92 14.37 18.29
3,500 4.57 19.56 24.14
Finally, the theoretical drawdown curves relating the fonnation and well losses to the well discharge are
shown in Figure 5.11.5. As can be seen, the well losses become more significant as the well discharge
increases and even larger than the fonnation loss beyond approximately Q =800m3/day. •
10
• • • • • • Formation loss
---Well loss
8
- - Total loss
~
...
<ll
<ll
~
6
Cl)
c 5
!:
0
u
~ 3
Cl
__ .......
2 ......
Figure 5.11.5 Example

500 1000 1500 2000 5.11.1: Comparison of for-
Well discharge, Q (m3/day) mation and well losses.
The purpose of this example is to illustrate the Rorabaugh graphical procedure82 to detennine the value
of n, the formation loss coefficient B, and the well loss coefficient C in Equation 5.11.3 using the time--
drawdown data from a step-drawdown pumping test (listed below)
Q (ro3/day) sw(m)
500 2.40
1,000 5.38
1,500 9.28
2,000 14.36
2,500 20.82
3,000 28.87
3,500 38.70
5.II Characteristic Well Losses 247
SOLUTION Step 1 Rearrange Equation 5.11.3 as follows:

s._ =BQ+CQ''
~-B= CQn-1
Q
log(~ -B}=log(C)+(n-l)log(Q)
This represents a straighHine plot of (s,./Q- B) versus Q on log-log scale with slope {n- I) and inter-
cept C. However, such a plot cannot be made since the formation loss coefficient B is not known.
Step 2 Rorabaugh's procedure requires a family of plots for incremental values of the fonnation loss coef-
ficient 8 until a straight line plot is obtained. This is done for the given time-drawdown data and the
:e are resulting plots are shown in Figure 5.11.6.
1arge Figure 5.11.6.
• o---
Example
5.11.2: Plots
for Rorabaugh
o.oo2 procedure.
Q) 0.003
I
o 1.E -03
......
~ 0.004
(/)
0.0045
0.006
1.E- 0 4 ' - - - - - - - - - - - - - - - ' - - - - - - - - - - - - '
100 1000 10,000
Q(m3/day)
Step 3 Referring to Figure 5.11.6, we observe a straight line for 8 ::: 0.0045 days/m2. The slope of this
line is 1.6; n - I = 1.6 and n = 2.6 is obtained.
·or- Step 4 The value of C is obtained by taking an arbitrary combination of Q and sw and applying Equa-
tion 5.11.3: Q= 2000 m3/day and sw = 14.36 m,
s,.,=BQ + CQ"
14.36 m =(0.0045 days/m2){2000 m3/day) + C(2000 m3/day) 2·6
c =1.4 x w-s
:value
time-- Step 5 Using the values of B, C, and n, the formation and well losses for the given discharges are calcu-
lated and tabulated in Table 5.11.3.
Table 5.11.3 Calculated Drawdown Parameters for Example 5.11.2
Q (m3/day) BQ(m) c(f (m) Sw (m)
500 2.25 0.15 2.40

1,000 4.50 0.88 5.38
1,500 6.75 2.53 9.28
2,000 9.00 5.36 14.36
2,500 11.25 9.57 20.82
3,000 13.50 15.37 28.87
3,500 15.75 22.95 38.70
I
5.12 SPECIFIC CAPACITY AND WELL EFFICIENCY

If discharge is divided by drawdown in a pumping well, the specific capacity of the well is
obtained. This is a measure of the productivity of a well; clearly, the larger the specific capac-
ity, the better the well. Starting from the Cooper-Jacob approximation of the nonequilibrium
equation (Equation 4.4.8) and including the well loss,
s =2.30Q l 2.25Tt CQn

- - og--+ (5.12.1)
w 4nT r;s
so that the specific capacity is
Q I
::::-----~--~~--- (5.12.2)
(2.30/41tT) log(2.25Ttlr;s) + CQ"- 1
This indicates that the specific capacity decreases with Q and t; the well data plotted in Figure
5.12.1 demonstrate this effect. For a given discharge a well is often assumed to have a constant
specific capacity. When using this approximation for u < 0.01, it can be seen that the change
with time is minor.
Any significant decline in the specific capacity of a well can be attributed either to a
reduction in transmissivity due to a lowering of the groundwater level in an unconfined aquifer
or to an increase in well loss associated with clogging or deterioration of the well screen.
If a pumping well is assumed to be 100 percent efficient (C(t = 0), then the specific
capacity from Equation 5.12.2 can be presented in the graphic fonn of Figure 5.12.2. Here
speciftc capacity at the end of one day of pumping is plotted as a function of S, T, and a well
diameter of 30 em.65 This graph provides a convenient means for estimating T from existing
pumping wells; any error inS has a small effect on T 35• 49 Refer to Prudic79 for application
-lan!
PotE
of estimating hydraulic conductivity from aquifer-test data and specific-capacity data. The
factors that affect estimates of transmissivity based on specific capacity are shown in Fig-
ure 5.12.3.
Figure 5.12.2 yields a theoretical specific capacity (QIBQ) for known values of Sand Tin
an aquifer. This computed specific capacity, when compared with one measured in the field
(Q/sw)• defines the approximate efficiency of a well. 13 Thus, for a specified duration of pump-
ing, the well efficiency Ew is given as a percentage by
Coni
-LL
a.
500 I I
~ 400
J?.
=- Q- 1000 m3/day
-
E
~ 300 - -
~g-
0-2000
Pre
o 200
.g - - z:or
uCD Q:= 5000
0.
en 100 - Q; 10,000 -
Figure 5.12.1. Variation in specific capaci-

10 100 1000 ty of a pumping well with discharge and
Tlme (days) time (after Jacob51 ).
5.12 Specific Capacity and Well Efftciency 249
5000 - I I I I
tell is
apac- T 2000 m2/day
mum
>;
1000 F T 1500
T 1000 -
ro 500 1- T 500
.12.1) J?-
E
.~
T- 200
~
a.
r5 100 T 100 --
--
(.)
;o:::
·o
-----
.12.2) ~ 50 f- T 60
(/)
T- 20 Figure 5.12.2. Graph relat-

~igure
ing specific capacity to trans-
nstant T 10 missivity and storage coeffi-
10 1- t 1 day -
hange rw- 15 em cient from the Cooper-Jacob
5 I I I I nonequilibrium equation for
.f to a 10-s 10-4 10-3 10-2 a confined aquifer (after
quifer Storage coefficient (S) Benta11 1\).
n.
lecific
. Here
a well
;isting
cation
t The
land surface
Potentiometric surface
----------------
- -Cone- o; _ ---------------
(nonpumplng)
Drawdown In _ - - -
- -
n Fig-
td Tin
depreas-:-
'on .... ,
'
' 1the aql!if6_!.... ....
/
,.
,.. b. Magnitude of the
wellloaa compared
to the drawdown In
the aquifer
1
! field \
?Ump- Well
loss
a. Th iclmen of t11e producing lone c. The difference b~n the

compared to the length of the •nominal• radius and the
screen or open hole effective radiue
-- -- -----
Confined
aquifer
Producing length of
zone screen
:apaci- ---
.... - - --- Figure 5.12.3. Factors
affecting estimates of trans-
md '27lr77777777777777777//////77777777777777777 missivity based on specific
Con imng bed capacity (Heath45).
E =tOO Q/sw = 100 BQ orE = 100( BQ

w Q!BQ sw w BQ+ CQ2
l (5.12.4)
Another method for recognizing an inefficient well is to note its initial recovery rate when
pumping is stopped. Where the well loss is large, this drawdown component recovers rapidly
by drainage into the well from the surrounding aquifer. A rough rule of thumb for this purpose
is if a pump is shut off after one hour of pumping and 90 percent or more of the drawdown is
recovered after five minutes, it can be concluded that the well is unacceptably inefficient.
5.13 SLUGTESTS
5.13.1 Definition
Pump tests are typically expensive to conduct because of the installation costs of wells. Slug
testing has been used for several years as a cost-effective and quick method of estimating the
hydraulic properties of aquifers. More recently (since the L980s) it has gained even more pop-
ularity (a) in obtaining estimates of hydraulic properties of contaminated aquifers where treat-
ing the pumped water is not desirable or feasible and (b) in field investigations of
low-permeability materials, particularly for studies of potential waste storage or disposal sites. 5.13.
The materials at these sites may have a hydraulic conductivity that is too low for pump tests.
Slug tests consist of measuring the recovery of head in a well after near instantaneous
change in the head at that well. A solid object (slug) is rapidly introduced into or removed from
the well, causing a sudden change (increase or decrease) in the water level in the well. Tests
can also be performed by introducing an equivalent volume of water into the well; or, an
equivalent volume of water can be removed from the well, causing a sudden decrease in the
water level. Following the sudden change in head, the water level returns to the static water
level. While the water level is returning to the static level, the head is measured as a function
of time (referred to as the response data). This response data is used to determine the hydraulic
properties of the aquifer using one of several methods of analyses. Butler17 presents a com-
prehensive summary of slug-test methodology and analysis.
5.13.2 Design Guidelines

Butler17 presented the following guidelines for slug tests (also see Butler et al./ 9 Dawson and
Istok,zs Driscoll/9 and Weight and Sonderegger105 for further discussions):
1. Well-drilling methods, such as the driving based cable-tool. pneumatic/hydraulic hammering,
or rotasonic methods that minimize the generation of drilling debris should be employed.
2. Well development efforts should focus on developing discrete intervals along the well
screen. Special care should be given to prevent the vertical flow within the filter pack from
weakening the effectiveness of development activities. Among the measures for doing so
are developing before the filter pack is installed, using the filter packs that minimize the
vertical flow, or using natural filter packs in unstable formations.
3. Further development activities may be needed in some cases. The identification of a low
permeability skin is the clearest indication of such a need. A preliminary analysis of the
response data using a theoretical model and a resulting implausible specific storage esti-
mate is the strongest evidence of a low-K skin.
4. The nominal screen length should be used as the effective screen length in the analyses.
5.13 Slug Tests 251
•' ~-
5. The effective screen radius should be taken as tbe radius of the filter pack in wells with arti-
2.4) ficial filter packs. On tbe other hand, the nominal screen radius may be more appropriate to
use in tbe analyses for wells with natural filter packs where development has been limited.
'hen 6. The effective radius of the well casing should be taken as the nominal radius of the well
·idly casing. However, a comparison of the theoretical and actual rise/drop of water level in the
X>se well following tbe injection of the slug will indicate the reasonableness of this replacement.
m is
7. A minimum number of three slug tests with initial displacement values varying by at least
a factor of two should be performed at each well. Both the rising head and falling head
tests should be employed to reveal a possible skin-related directional dependence. The
possible existence of a dynamic skin can be identified if the frrst and last tests of the series
at the same well employ the same initial displacement. All these variations and combina-
tions should rate the effectiveness of well-development efforts and the validity of conven-
Slug tional slug test theories.
pbe 8. A series of tests consisting of a combination of rising and falling head tests should be
pop- planned and employed.
reat-
s of
;ites. 5.13.3 Performance of Slug Tests
;ts. The equipment (Figure 5.13. L) used in slug tests include devices for initiating the test, devices
eous for measuring changes in head during the test, and devices for storing the head measurements.
from The most common method for initiating the test is to use a solid object (or slug) and introduce
rests (or remove) it into (from) the well. A conventional slug would be a piece of stainless steel or
r,· an PVC pipe filled with sand or similar material and capped at both ends. Both conventional and
a tbe streamlined solid slugs have been used. The majority of slug tests are perfonned with sensors
vater (pressure transducers) that measure the pressure exerted by the overlying column of water. The
::tion pressure transducer most commonly used is the semiconductor strain-gauge transducer.
aulic Electric tapes are also commonly used to measure water levels in slug tests of moderate to low
:;om- hydraulic conductivity.
Slug tests may also be performed pneumatically. This method involves pressurizing the air
column in a sealed well by injecting compressed air or nitrogen gas to depress the water level
Jand
Rope
!ring,
:d.
well
from
ng so
;e the
a low
>f the
: esti-
rses.
Figure 5.13.1. Slug testing schematic
(Weight and Wittman; 106 used with permis-
sion of Groundwater Publishing Co.).
Air pressure
transducer
Release
valve Air
compressor
Static water
level)_
Figure 5.13.2. Hypothetical cross sec-

tion displaying a well at which the 5.13
Pressure.:----,. pneumatic method is being used for
transducer . · "' test initiation. Note that there must be
an airtight seal at the point at which
t
Well screen
the cable for the submerged transducer
passes through the top of the well head
(figure not to scale; after McLane et
al. 69 as presented in Butler 1\
by driving water out of the well into the aquifer (see Figure 5.13.2). Water levels continue to
drop until the decrease in pressure head of the water is equal to the magnitude of the increase
of pressure head of the air column. Then the test is initiated by rapidly depressurizing the air
column. A variation of this procedure would be to apply a vacuum to the air column followed
by a sudden release of the vacuum to initiate the test.
Advantages of the pneumatic approach include the following: no water is handled during
the test, test initiation is very rapid, and only the transducer and cable have to be cleaned prior
to moving to a new well. Another advantage is that if both the pressurization and depressur-
ization are done rapidly with respect to the formation response, then a pair of falling-head and
rising-head tests, respectively, can be performed. The pneumatic approach requires a special
airtight well-head apparatus, which can be a major disadvantage for the method. Greene and
Shapiro41 • 42 describe methods for conducting pneumatic slug tests and computation of type
curves for estimating transmissivity and storativity. Butler17 provides a more detailed discus-
sion of the various devices used for different slug tests.
Butler17 provides the following four performance guidelines for slug tests:
1. The measurement of the response of a formation to a slug test must be done with appro-
priate equipment, usually selected based on the expected hydraulic conductivity of the for-
mation. A pressure transducer connected to a data-acquisition device is most appropriate
for obtaining and storing head measurements in a slug test where the formation has a mod-
erate to high hydraulic conductivity. For formations that are less permeable, use of an elec-
tric tape is also acceptable. The equipment used for head measurements should always be
checked before and after the slug test at a particular well.
2. The initial displacement of water in the well and the resulting rise or drop of the water level
must be instantaneous relative to the response of the formation. Pneumatic- or packer-based
systems should be used for test initiation in formations of very high permeability. In less
5.13 Slug Tests 253
permeable formations, use of a solid slug is also acceptable. Techniques such as the addition
or removal of water that may threaten the instantaneous test initiation should be avoided.
3. The initial head following the instantaneous displacement of water in the well must be
measured accurately. A comparison of the expected initial head based on the known vol-
ume of displacement (i.e., slug volume) and the measured initial head should reveal the
appropriateness of using the nominal casing diameter as the effective casing diameter. If
the nominal casing diameter together with the known volume of displacement does not
provide a reasonable estimate of the initial head, the effective radius can be adjusted based
on a mass balance.
4. There should be enough time before a slug test is repeated in the same well that the resid-
ual deviation from the static level is less than 5% of the initial head change. If it is sus-
pected that the fonnation response is affected by an incomplete recovery of the water level
from the previous test, the test data can be plotted as in the Hvorslev method (mentioned
. in the next section). A near-linear plot of the log of the deviation of head from the static
water level versus time would deny an incomplete recovery from the previous test.
. sec-
5.13.4 Methods for Analyzing Slug Test Data
X
Slug-test analysis methods were first developed during the 1950s by Hvorslev46 and Ferris and
tbe
:h
Knowles. 31 Hvorslev's simple method led to its use for both confined and unconfined aquifers
lucer as an estimate of the hydraulic conductivity within the screened interval. Cooper et al.26 and
head Papadopulos et al.76 made improvements of the methods by Ferris and Knowles. Later, Bouwer
et and Rice 15 and Bouwer14 developed a method for analysis in confined and unconfined aquifers
that takes into account aquifer geometry and partial penetration effects. Campbell et al.22 sum-
marized various methods with their relative merits and an extensive list of references.
Methods for analyzing slug test data have been reported in various reports and books,
1ue to including references 17, 28, 42, 43, 85, and 100. Other references on aquifer tests include Hall
Tease and Chen44, Kasenow55, Kruseman and de Ridder61 , and Lee63.
he air Many government computer software packages and spreadsheets have been developed,
owed including those by Greene and Shapiro42, Halford and Kuniansky43 , and Sepulveda. 85 The
private sector has also developed many computer software packages and spreadsheets with
luring graphic user interfaces. Figure 5.13.3 presents eight conditions for slug tests. Various methods
. prior for the analysis of slug tests in these eight cases (discussed in the following sections of this
~ssur chapter) may be outlined as follows.
•d and
Case 1: Unconfined aquifer, partially/fully penetrating well, water level variations above the
pecial
well screen (rising head test).
teand
f type Available methods:
iscus-
(i) Bouwer and Rice method 14· 15
(ii) Dagan method 17
(iii) KGS (Kansas Geological Survey) model 17
lppro-
1e for- Case 2: Unconfined aquifer, partially/fully penetrating well, water level drops below the top of
>priate well screen (rising head test). (Note: A double straight line effect or a concave-upward curva-
tmod- ture may be observed when the response data is plotted as the logarithm of head vs. time. This
n elec- is more likely if there is an artificial filter pack in a well screened across a low hydraulic con-
aysbe ductivity unit.)
Available methods:
~r level
·-based (i) Bouwer and Rice method (an effective casing radius may be needed)
In less (ii) Dagan method
5.14 Slug Tests for Confined Formations 255
urtace Case 5: Confined aquifer, fully penetrating well, rising head test.
.·,
Applicabk methods:
able•:· (i) Cooper-Bredehoeft-Papadopulos method26
(ii) Hvorslev method46
...·· .. :·
Case 6: Confined aquifer, fully penetrating well, falling head test The methods for Case 5 also
apply to this case.
Case 7 and Case 8: Confined aquifer, partially penetrating well, falling head test.
Applicabk rMtlwds:
(i) Hvorslev method (different shape factors are used than that for fully penetrating wells)
(ii) Confined extensions of the Dagan method 17
;urface (iii) KGS model (developed by extending the Cooper et al. method to the case of a partially
penetrating well 17)
ial · (iv) Bouwer and Rice method (while, theoretically, the accuracy of the hydraulic conductivity
netric -_
e~--~-,- estimates increase as the distance between the top of well screen and the upper confining
layer increases)
5.14 SLUG TESTS FOR CONFINED FORMATIONS

ned.:...c
fer Butler17 points out that the vast majority of slug tests performed on fully penetrating wells are
analyzed using the Cooper, Bredehoeft, and Papadopulos26 or Hvorslev46 methods, which can
be used for fully or partially penetrating wells. Both these methods are discussed in the sec-
tions below. A newer method by Peres et al.,77 based upon a deconvolution approach, also has
potential, as discussed by Butler, 17 for both partially and fully penetrating wells. Also, meth-
ods by Dagan27 and the KGS model (Hyder et al.50), described later, can be used for partially
penetrating wells in confined aquifers.
5.14.1 Cooper, Bredehoeft, and Papadopulos Method

This method (Cooper et al.26) is based upon a fully penetrating well or open borehole, as illus-
trated in Figure 5.14.1. A slug or a known volume of water is introduced into the well over a
short time period of a few seconds, causing the water to rise to a maximum height H0 above
the initial water level. As the water level declines, the head above the original level, H(t), is
measured as a function of time. A plot of the ratio H(t)/80 as function of time can be developed
with HIH0 on an arithmetic scale and time with a log scale.
This methodology is based upon assumptions that include a homogeneous formation and
Darcian flow, instantaneous introduction of the slug, hydrologic boundaries in the plane of the
fonnation a large distance from the test well, and negligible well losses. Cooper et al. 26 devel-
oped the following analytical solution to the mathematical model:
~t) =/(~,a) (5.14.1)

ove the 0
pis a dimensionless time parameter defmed as

e water R- K,Bt- Tt
uctivity 1-'- 2 - 2 (5.14.2)
rc rc
the sat-
se zone where K, is the radial component of hydraulic conductivity, B is the formation thickness, and
tturated t is the time since the initial displacement of water in the well. a is the dimensionless storage
parameter defined as
v 1.
I Water level immediately
t( after injection
-.-------1-----+---'----1 Water level at time t

0.:
Head in aquifer
at timet
O.E
H
Ho
O.L
Figure 5.14.1. Well into which a

volume, V, of water is suddenly 0.~
injected for a slug test of a con-

fined aquifer (Cooper et al. 26).
0.0
1.0E
(5.14.3)
where rs is the effective radius of well screen, rc is the effective radius of well casing, and Ss
is the specific storage. If time is measured in seconds and all dimensions are in centimeters,
then the value of T is cm2/s. If time is in seconds and all dimensions are in feet, then trans-
missivity Tis in ft 2/s.
Tables 5.14.la and b provide values of H(t)IH0 for Equation 5.14.1. Figure 5.14.2 presents
the type curves developed by Papadopulos et al. 76 for slug tests in wells of finite diameter.
Steps in the Cooper, Bredehoeft, Papadopulos method are as follows:
1. Plot the normalized response (field) data (H(t)/H0 ) versus the logarithm of the thne since
the test began on the same scale and size as the type curves.
2. Overlay the data plot on the type-curves moving parallel along the log-time axis of the
response data until one of the a curves approximately matches the plot of the response
(field) data. Record the value of a.
3. Select match points from each plot. As an example, set ~ = 1 and read the corresponding
real time t t.o·
4. Use Equation 5.14.2 to solve for the hydraulic conductivity and then solve for the
transmissivity.
5. Use Equation 5.14.3 to solve for the specific storage (storage coefficient).
EXAMPLE 5.14.1 A slug test is performed in a confined aquifer of 5 m thickness with a casing radius of 8.8 em and a screen
radius of 5.6 em. After the slug of water is injected in the well, the water level rises by 0.52 m following
which the water level drops with time as shown in Table 5.14.2. Estimate the transmissivity, hydraulic
conductivity, and storage coefficient using the Cooper-Bredehoeft-Papadopulos method.
SOLVTION Using a plot of H!H0 versus time on a semilogarithmic paper, this field curve is superimposed on the type
curves (Figure 5.14.2) with the arithmetic axis coincident and shifted horizontally until a best fit is
obtained on one of the type curves. In this case the field data fits best to the type curve for a= 10-4. The
5.14 Slug Tests for Confined Fonnations 257
1.0 I
r-.::::::: r- 1"--
r-
r1
~ r--- ~ ~ ~ ~t'
~
a= 10-
~ ~ 1\1\
1'-t'-.. v~---' 1- =10-2
0.8
f'-,... ~ v =10-3
vv
~~
'\
1\ 1\[\
v I. . .Y 1- =10-
4
.I
~ vr\ ~ vi'\
i\ v~---'
1-
~
= 10-5
=10-6
0.6 =10-7
~ ~~~~~
\ 1-
v ~ =10-8
H f\
Ho I 1\ 1-
1-
=10-9
= 10-10
0.4
" 1\
~ ~
~ 1\l\
I ~~
l\\
~\
ich a
nly 0.2 1\ [\
) 0-
~).
0.0
1.0E- 02
I 1.0E- 01 1.0E + 00
""
1\ [\f\
~ 1\f'l
r--~
i[
1.0E + 01 1.0E + 02
Figure 5.14.2. Selected
type curves for the Cooper,
d4.3) Bredehoeft, and
Papadopulos method
(Papadopulos et al. 76).
and Ss
neters, Table 5.14.1a Values of H(t)/H0
trans-
HIH0
·esents Ttlr02 a= w-t a= w-2 a= w-3 (X = 10-4 a= w-s
:ter.
1.00 X 10-J 0.9771 0.9920 0.9969 0.9985 0.9992
2.15 X 10-J 0.9658 0.9876 0.9949 0.9974 0.9985
~since 4.64 X 10-3 0.9490 0.9807 0.9914 0.9954 0.9970
1.00 x w-2 0.9238 0.9693 0.9853 0.9915 0.9942
of the 2.15 x w-2 0.8860 0.9505 0.9744 0.9841 0.9888
sponse 4.64 X 10-l 0.8293 0.9187 0.9545 0.9701 0.9781
1.00 X 10-t 0.7460 0.8655 0.9183 0.9434 0.9572
2. 15 x w-r 0.6289 0.7782 0.8538 0.8935 0.9167
onding
4.64 x w- 1 0.4782 0.6436 0.7436 0.8031 0.8410
1.00 X 10° 0.3117 0.4598 0.5729 0.6520 0.7080
for the 2.15 X 10° 0.1665 0.2597 0.3543 0.4364 0.5038
4.64x 10° 0.07415 0.1086 0.1554 0.2082 0.2620
7.00 X 10° 0.04625 0.06204 0.08519 0.1161 0.1521
1.00 X 101 0.03065 0.03780 0.04821 0.06355 0.08378
1.40 X 10 1 0.02092 0.02414 0.02844 0.03492 0.04426
a screen 2.15 X 10 1 0.01297 0.01414 0.01545 0.01723 0.01999
·llowing
3.00x 101 0.009070 0.009615 0.01016 0.01083 0.01169
ydraulic
4.64 x 101 0.005711 0.005919 0.006111 0.0063 19 0.006554
7.00x 101 0.003722 0.003809 0.003884 0.003962 0.004046
1.00 X 102 0.002577 0.002618 0.002653 0.002688 0.002725
the type 2.15 X 102 0.001179 . 0.001187 0.001194 0.001201 0.001208
:st fit is Source: From Cooper et at. 26
0-4. The
Table 5.14.1b Values of H(t)IH0

Tt/rc2 a= w-6 a= w-7 a= w-s a= w-9 a= 10- 10
0.001 0.9994 0.9996 0.9996 0.9997 0.9997

0.002 0.9989 0.9992 0.9993 0.9994 0.9995
0.004 0.9980 0.9985 0.9987 0.9989 0.9991
0.006 0.9972 0.9978 0.9982 0.9984 0.9986
0.008 0.9964 0.9971 0.9976 0.9980 0.9982
0.01 0.9956 0.9965 0.9971 0.9975 0.9978
0.02 0.9919 0.9934 0.9944 0.9952 0.9958
0.04 0.9848 0.9875 0.9894 0.9908 0.9919
0.06 0.9782 0.9819 0.9846 0.9866 0.9881
0.08 0.9718 0.9765 0.9799 0.9824 0.9844
0.1 0.9655 0.9712 0.9753 0.9784 0.9807
0.2 0.9361 0.9459 0.9532 0.9587 0.9631
0.4 0.8828 0.8995 0.9122 0.9220 0.9298
0.6 0.8345 0.8569 0.8741 0.8875 0.8984
0.8 0.7901 0.8173 0.8383 0.8550 0.8686
1 0.7489 0.7801 0.8045 0.8240 0.8401
2 0.5800 0.6235 0.6591 0.6889 0.7139
3 0.4554 0.5033 0.5442 0.5792 0.6096
4 0.3613 0.4093 0.4517 0.4891 0.5222
5 0.2893 0.3351 0.3768 0.4146 0.4487
6 0.2337 0.2759 0.3157 0.3525 0.3865 5.14.2
7 0.1903 0.2285 0.2655 0.3007 0.3337
8 0.1562 0.1903 0.2243 0.2573 0.2888
9 0.1292 0.1594 0.1902 0.2208 0.2505
10 0.1078 0.1343 0.1620 0.1900 0.2178
20 0.02720 0.03343 0.04129 0.05071 0.06149
30 0.01286 0.01448 0.01667 0.01956 0.02320
40 0.008337 0.008898 0.009637 0.01062 O.OJL90
50 0.006209 0.006470 0.006789 0.007192 0.007709
60 0.004961 0.005111 0.005283 0.005487 0.005735
80 0.003547 0.003617 0.003691 0.003773 0.003863
100 0.002763 0.002803 0.002845 0.002890 0.002938
200 0.001313 0.001322 0.001 330 0.001339 0.001348
Source: From Papadopulos et al 76
time that corresponds to the vertical axis is for P= 1.0. At the axis for P= 1.0, the t = 133 seconds. Sub-
stituting t = l33s and rc = 8.8 em into Equation 5.14.2 yields the hydraulic conductivity
~'c
1 2
K, = = (l.0)(8·8 cm) =1.1645 x l0-3 crnls = l.006 m/day
Bt1.o (500 cm)(l33 s)
and the transmissivity is given by

T = K,B = (1.006 m/day)(S m) = 5.03 m2/day
Substituting rc = 8.8 em and rs =5.6 em, and o: =10-4 into Equation 5.14.3 and solving for S yields
rc1a (8.8 cm) (

2
10-4) -4
S=- = = 2.47x!O
's2 (5.6 em)2 ·
•
5.14 Slug Tests for Confined Formations 259
Table 5.14.2 Water Level after Slug Test for Example 5.14.1
·10 Time (s) H(m) H!Ho
2 0.51 0.99
4 0.51 0.98
6 0.50 0.97
8 0.50 0.96
10 0.49 0.95
20 0.48 0.92
30 0.46 0.89
50 0.43 0.83
1 80 0.40 0.76
4 100 0.37 0.72
7 120 0.35 0.68
1 150 0.32 0.62
8 200 0.29 0.55
4 250 0.25 0.48
6 300 0.22 0.42
1 400 0.17 0.33
9 500 0.14 0.27
6 800 0.07 0.13
2 1,000 0.06 0.11
7
;5
5.14.2 Hvorslev Method
·7
:8 This method differs from the Cooper, Bredehoeft, and Papadopulos method in the following
15 three aspects:
'8 l. Although the hydraulic head in the formation still varies with time, the specific storage is
49.' assumed to be so small that its effects can be neglected, so that any change in head in the
i20 well is instantly propagated throughout the flow system (a quasi-steady-state representa-
.90 tion of the slug-induced flow).
1709
2. The slug test need not be introduced in an instantaneous fashion.
i735
1863 3. Lateral constant-head boundaries are at a finite distance from the test well.
~938
The analytical solution for this methodology is expressed as23
1348
ln[H(t)]-_ 2K,Let (5.14.4)
Ho - r} In( Relrs)
ds. Sub-
where Re is the effective radius of the slug test and Leis the effective length of the well screen.
A semilog plot of the normalized heads H(t)IH0 , versus time, is a straight line. This
method calculates the slope of that line, which in turn is used to estimate the radial component
of the hydraulic conductivity. The steps in the Hvorslev method are as follows:
1. Plot the logarithm of the nonnalized response data versus the time since the test began. A
straight line is fit to the plot.
2. Calculate the slope of the straight line. One method is to estimate the time (referred to as
e1ds T0 ) at which the normalized head equals 0.368 (natural log of 0.368 is -1), which is
defined as the basic time lag according to Hvorslev.46 At the start of the test, the logarith-
mic head tenn is one and the time tennis zero, so the slope is simply the log 10 of 0.368
• over T0 , which is -1/T0 in terms of the natural logarithm.

Cement grout
/Casing
Figure 5.14.3.
Piezometer geometry for
Hvorslev method. Note
that for a piezometer
installed in a low-per-
meability unit the value
R is the radius of the
highest permeable zone
that includes the gravel
pack zone and Le is the
length of the gravel
pack zone. 32
3. Calculate the radial component of the hydraulic conductivity using Equation 5.14.4 writ-
ten in terms of the slope calculated using a normalized head of 0.368,
(5.14.5)
1.
in which T0 is the time at which a nom1alized head of 0.368 is obtained. If a gravel pack
zone exists, which is usually installed while testing a low permeability unit, the radius of
the well screen should include the gravel pack zone. Similarly, Le is the length of the
gravel pack zone in such a case (see Figure 5.14.3). H!H0
o.·
Equation 5.14.5 requires an estimate of the effective radius of the slug test, Re, an empir-
ical parameter that is a function of a. Typically values of either the well screen length or 200
times the effective radius of the well screen are used for Re (Butler 17 ). The assumption of a
straight-line fit of the semilog plot of the test data is appropriate as long as the effect of the
elastic storage mechanisms can be neglected. When the storage effects have an effect on the
response data, the data will display a distinct concave-upward curvature, which is primarily a
o.c
function of the dimensionless storage parameter, a. Chirlin23 provides a theoretical explana-
tion for the behavior. The Hvorslev method does not require that the slug be introduced nearly
instantaneously as in the Cooper, Bredehoeft, and Papadopulos method. As previously
mentioned this is a quasi-steady-state representation of the slug-induced flow, eliminating the
need for a nearly instantaneous initiation, with the only assumption that the slug introduction 5.15 ~
is completed prior to the collection of the response data. The Hvorslev method can therefore be
used to analyze response data that have been affected by noninstantaneous slug introduction.
A slug test is performed by adding a slug of 3,000 cm3 of water into a piezometer that is screened 12m
vertically in medium sand. The injection of the slug causes the water level in the well to rise 0.56 m. The
radius of the well screen is 5.8 em while the radius of the well casing is 4.1 em. The water level recov-
ery in the well is recorded using a pressure transducer as tabulated on the next page. Detennine the
hydraulic conductivity in the vicinity of the screened portion of the formation using the Hvorslev method.
5.15 Slug Tests for Unconfined Formations 261
Elapsed Time (s) Change in water level (m) -H!Ho
0 0.560 1.000
0.387 0.691
2 0.198 0.354
3 0.141 0.252
4 0.087 0.155
5 0.050 0.089
6 O.Q35 0.062
7 0.017 0.030
etry for
8 0.010 0.018
. Note
9 0.006 0.011
eter
10 0.000 0.000
-per-
: value
the
From the plot (Figure 5.14.4) of the data (i.e., normalized deviation of head from static water level ver-
e zone
sus time), the time for the head to drop to 37 percent of the initial head change is T0 = 2 seconds. Also
gravel
given in the problem statement is the well construction information:
is the
1el rc=4.lcm, r5 =5.8cm, Le=12m
Finally, taking the effective radius of the slug test, R,, as the well screen length, L,, and substituting into
Equation 5.14.5 yields
.4 writ- 2
r, 1n( R.Jr,) (4.1 cm)2 1n(l200 cm/5.8 em)
K ~ ~ 0.01867 cm/s = 16.13 m/day
' 2LJ0 (2)(1200 cm)(2.0 s)
5.14.5)
1.00
~
........
el pack ......
tdius of
'
1 of the
H!H0 ""' ~
. empir-
1 or 200
0.10
"" ........
'-,.,.
.on of a
t of the
ton the
narily a
0.01
0 2 4
I
6
""" ~
8 10
xplana- Time(s) Figure 5.14.4. Dataplot for Example 5.14.2.
i nearly
viously
•
ting the
-duction 5.15 SLUG TESTS FOR UNCONFINED FORMATIONS
~fore be
Methods for unconfined formations are primarily classified on the basis of whether or not the
luction.
well screen intersects the water table (see Figure 5.15.la and b). Butler17 classifies these as:
a) well screen below the water table, b) well screen intersects the water table, and c) well screen
below the water table, in which case the methods can be subdivided according to their handling
ted 12m
6m. The
of the storage mechanism. When the well screen is below the water table, the change in satu-
el recov- rated thickness during a test is typically quite small, so a linear model can be used. When a well
mine the screen intersects the water table, assumptions of a constant saturated thickness or a constant
method. effective screen length may not be correct, resulting in the need for a nonlinear model.
A B
..
~ .- . ..
'· -· Gravel pack or

. . -- ~d~veloped zon.e~ I i
..
• • ~ I
·-: : :·. ·
,l ·. :; • ,
· . -.- .. -.
Impermeable
(a) (b)
Figure 5.15.1. (a) Geometry and symbols for a slug test on a partially penetrating screened well in an unconfined aquifer with a
gravel pack around the screen 14 (b) Slug test for borehole with groundwater level below A and above B top of screen or perforated
section 14
According to Butler, 17 the vast majority of slug tests performed in unconfined formations
in wells screened below the water table are analyzed using the Bouwer and Rice method 15 , the
Dagan method27 , or the KGS model with the unconfined variant (Hyder et al. 50) .
5.15.1 Bouwer and Rice Method

The two key assumptions of this method are that the effects of elastic storage mechanisms can
be neglected and the water table level does not change; consequently the saturated thickness of
the formation does not change during the test. The original model of Bouwer and Rice was
defined for isotropic conditions. Zlotnik 110 extended the method to the general anisotropic case.
As pointed out by Bouwer and Rice 15, the slug test applies theoretically to aquifers where
the upper boundary is a plane source (rising water level test) or sink (falling water-level test),
as in an unconfined aquifer. Because most of the head difference, H, between the static water
table and the water level in the well is dissipated in the vicinity around the screen of the per-
forated section, the method should also be applicable to situations where the upper boundary
of the aquifer is an impermeable or semi-impenneable upper confining layer. Thus the method
should give reasonable results for the hydraulic conductivity in confined, semiconfined, or
stratified aquifers. Theoretically, the greater the distance between the top of the screened or
open section of the well and the upper confining layer, the more accurate the estimate of the
hydraulic conductivity. However, as Bouwer14 points out, source boundaries of groundwater
flowing into the well in response to lowering the water level are difficult to define because of
the elastic deformation of the aquifer material and confining and interbedded fine-textured
layers, and because of leakage through semiconfining layers.
The analytical solution of the method is (Bouwer and Rice 15 ; Bouwer14; and Zlotnik 110):
(5.15.1)
5.15 Slug Tests for Unconfmed Formations 263
where the symbols are defined as follows:

K, hydraulic conductivity of the fonnation [Lff]
rc radius of the well casing [L]
rw radius of the gravel envelope [L]
Re the effective radial distance over which head is dissipated [L)
Le the length of the screen or open section of the well through which water can enter
H0 the drawdown at time t = 0 [L)
H, the drawdown at timet= t (L]
the time since H =H0 [T]
This method is based upon the plot of the nonnalized head (H/H0) versus time as a straight
line and involves calculating the slope of the straight-line fit to the response data. Then the slope
is used to estimate the hydraulic conductivity. The steps in the method are as follows (Butler 1\
1. Plot the logarithm of the nonnalized response data {H,IH0) versus the time since the test
began and fit a straight line through the data.
2. Calculate the slope of the straight-line fit. Using the time lag T0 , write the slope in tenns
ith a of the natural logarithm as -l!T0.
orated 3. Assume the anisotropy ratio to be l. Estimate the effective radius parameter (Butler1\
4. Compute the radial component of the hydraulic conductivity using Equation 5.15.1.
As with the Hvorslev method, the effective radius, Re• is considered as a parameter.
Bouwer and Rice 15 and Bouwer14 presented a method for estimating the dimensionless ratio,
ln(R)rw). If Lw is less than h, the saturated thickness of the aquifer, then the well is partially
penetrating and
(5.15.2)
:ms can
:ness of
ice was If Lw is equal to h, then the well is fully penetrating and
·iccase.
~where (5.15.3)
el test),
c water
~he per- where A, B, and Care dimensionless numbers found from Figure 5.15.2.
mndary Theoretically, this slug test method applies to any diameter of the well, and from a prac-
method tical viewpoint the well diameter should be selected so that the geometric parameters apply to
.ned, or Figure 5.15.2. The larger r 111 and Le, the larger the portion of the aquifer on which the hydraulic
ened or conductivity is detennined.
e of the Bouwer14 discussed the fact that users of the method have observed that when plotting the
1dwater Jog H(t) versus t as in Figure 5.15.3, sometimes a double straight-line effect results, as
:ause of illustrated in Figure 5. 15.4. This discussion assumes that a bail-down test (water level is low-
extured ered by bailing or pumping so that water flows from the aquifer into the well) was performed.
The steep straight-line segmentA-B is most likely due to a highly permeable zone around the
well (gravel pack or developed zone), which sends water into the well immediately after the
water level has been lowered (A in Figure 5.15.1b). When the water level in the permeable
:5.15.1) zone around the well has been drained to the water level in the well, the flow into the well
slows down and fonns the less steep segment B-C shown in Figure 5.15.4. This less steep
c ~,"""
12 /
I
I
I
I
I
I
I
I
A 8 I
and I
I
c 6 I 8 ./
/
/.
/.
/
2
--.... --·-·--
-- ...-..-. ..,.
./·
Figure 5.15.2. Dimensionless parameters A, B,

5 10 500 1000 5000 and C plotted as a function of Lelr,.. These param-
eters are used in the detennination of ln(R/r,..). 14
segment is indicative of the flow from the undisturbed aquifer into the well; therefore segment
B-C should be used in calculating the hydraulic conductivity.
Next consider the situation when the groundwater level is above the top of the screen or
perforated section (B in Figure 5.15.lb) and during the test the water level in the well is not
lowered enough to drop below the top of the open section. In this situation the gravel envelope
or developed zone around the open section cannot drain and the inflow into the hole is then
immediately controlled by the aquifer, such that the double straight-line effect does not occur.
In the situation that the double straight line still occurs, this may indicate leakage around the
casing or grouting above the gravel pack.
0.5
SOLUTJ
0.1
0.01
Figure 5.15.3. Head in a borehole as a function of

time for a well in Salt River Bed, 27th Avenue,
Phoenix, AZ. Note that the data form a straight line
10 30 40 during the early part of the test but eventually devi-
t(seconds) ate from the straight line. 14
5.15 Slug Tests for Unconfined Formations 265
Figure 5.15.4. Head in a borehole as a function

of time fonns two straight-line segments during
B. the early part of the test but eventually deviates
tram- 0 from the straight line (Bouwer 14).
.).14
Bouwer14 points out that when the double straight line is caused by a gravel pack around
~ent the well, the effective well radius should be taken as the radial distance from the center of the
well to the outer surface of the gravel pack. When the double-straight line is caused by a nat-
!en or urally developed zone around the well, r..., is more difficult to evaluate, resulting in the need for
is not experience and good judgment. It may be possible to estimate r..., from the value of Hat point
1elope Bin Figure 5.15.4.
s then
occur.
nd the I!XAMPLE 5.15.1 A slug test is perfonned by bailing 1,230 cm3 of water from the test well. The bailing causes a near instan-
taneous drop of 0.61 m of the water level in the well. Head measurements during the recovery phase of the
water level using an electric tape are presented in Table 5.15.1. Using the given well construction infonna-
tion, determine the hydraulic conductivity of the formation using the Bouwer and Rice method.
SOLUTION First, detennine if the measured drop of water level agrees with removed volume of water.
A conceptual picture of the problem shows that the water level is still above the screened interval fol-
lowing the bailing (Figure 5.15.5). Thus, the expected drop of Vv'ater level corresponding to a removal of
1,230 cm3 water can be calculated utilizing the casing radius:
Ho = ~= 1230 cm3 2-60.2 em

7tr; 1t(2.55 em)
This result is consistent with the measured drop of water level, 61 em, given in the problem statement.
The logarithm of the normalized response data versus time is plotted in Figure 5.15.6.
Next, we must rearrange Equation 5.15.1:
Talcing H 1 = 0.368H0 yields
:tion of
K= r} In(Relr..,) 1 1n[ Ho ] = r/ In( R,lr..,) 1
2L, T0 0.36880 2L, T0
ue,
ght line From the given geometry,
ly devi-
rc=2.55cm, r,=l0.15cm, L,=3m, Lw=4.4m
Table 5.15.1 Recovery from Bailing Test (Example 5.15.1) 1.(
Elapsed Time (s) Change in water level (rn) H!H0
0 0.610 1.000
2 0.583 0.956
6 0.551 0.903 H!H0
10 0.526 0.862
IS 0.508 0.833
26 0.482 0.790
42 0.450 0.738
61 0.418 0.685 0.1
77 0.396 0.649
92 0.377 0.618
115 0.350 0.574
135 0.329 0.540
147 0.318 0.521
168 0.300 0.492
189 0.284 0.466
207 0.271 0.445
227 0.258 0.423
H!H0
257 0.240 0.394
286 0.224 0.368
315 0.209 0.343
346 0.196 0.321
386 0.180 0.295
0.164 0.269 0.1(
430
574 0.121 0.198
668 0.099 0.162
731 0.092 0.150
938 0.077 0.126
1,115 0.066 0.109
Boring diameter = 20.3 em
Casing diameter = 5.1 em
Depth to bottom of well = 8.2 m BGS (below ground surface)
Depth to top of casing = 5.2 m BGS
Depth to water = 3.8 m BGS
Saturated thickness = 4.4 m
8.2 m
Figure 5.15.5. Conditions for slug test in

Example 5.15.1.
5.15 Slug Tests for Unconfined Fonnarions 267
HIH0
0
0
0
0
200 400 600 BOO 1000 1200 Figure 5.15.6. Recovery from bailing test for
Time(s) Example 5.15.1.
H!H0
0.100
200 400 600 800 1000 1200 Figure 5.15.7. Response to bailing test for
Time (s) Example 5.15.1.
Since Lw is equal to h, the saturated thickness of the aquifer, Equation S.1 5.3 must be used:
[
-1
1n Re _
'w -
1.1 +-C-
ln(L,.Ir,.) LJrw l
L/rw =300 cm/10.15 ern:: 30 ~ Figure 5.15.2 ~ C =2.2
-1
ln~= 1.1 + 22
· =2.73
r,. I
ln(440 cm/10.15 em) 300 cm/10.15 em 1
Note that the points of the response data do not follow a straight line exactly in Figure 5.15.7. Instead,
an early quick response is followed by a relatively straight line segment. Finally, the data points start
deviating from this straight line for large t values. Overall, the response data indicate a double straight-
line effect. The relatively quick early time response may be attributed to the effect of the gravel pack
zone, if there is any. However, because the water level changes occur above the screened portion of the
well in this case, it is more likely that there is leakage around the casing or grouting above the gravel
pack. A sensitivity analysis could be made using different parts of the data to find the hydraulic conduc-
tivity. However, the response data will be separated into three segments for time intervals t =0-26 s,
42-430 s, and 574-1,115 sin the current solution.
Note that a new h0 value emerges when the straight line is extrapolated backward because of the dou-
=
ble straight-line effect, such that (H0)new 0.778H0.
T0= 388.5 s for the first straight line segment based on the new h0 value. Thus,
r} ln(R.fr.. ) 1 (2.55 cmf(2.73) l _5

K= = ( )( ) =7.6 x lO cm/s = 0.0658 m/day
2Le T0 2 300 em 388.5 s •
5.15.2 Dagan Method

Dagan27 presented a method based on a mathematical model similar to that for the Bouwer and
Rice method with the only difference that a constant-head boundary is not assumed at the finite
radial distance from the test well, so that the hydrologic boundaries in the lateral plane are
assumed to be at an infinite distance from the well.
The steps in applying this method are as follows (B utler 17):
1. Plot the log of the normalized response data versus the time since the test began and fit a
straight line to the plot. Compute the slope of the straight line.
2. Estimate the parameter \jl for the particular well formation configuration using
where K/Kr is the anisotropy ratio. In most cases, 1 is used for the anisotropy ratio.
3. Using \If, the normalized distance from the water table (Lw + Le)/Le, and the normalized
length of the well screen L/h, select a value of P, the dimensionless flow parameter, from
Table 5.15.2. Table 5.15.3 is used to determine values of P for confined aquifer conditions.
4. Estimate the hydraulic conductivity (radial component) using
rc2 (11P)
K=-- (5.15.4)
r 2LeTo
EXAMI
Table 15.15.2a Tabulated Values of the Dimensionless Table 15.15.2b Tabulated Values of the Dimensionless
Flow Parameter, P, Used in Dagan Method for Wells Flow Parameter, P, Used in Dagan Method for Wells
Screened Below the Water Table Screened Below the Water Table SOLUT/1
(Lw+ Le)/Le L/h
o/ 8.0 4.0 2.0 1.5 1.05 o/ 1.0 0.83 0.67 0.50 0.20 0.10
0.20 0.646 0.663 0.705 0.756 1.045 0.20 1.289 0.723 0.631 0.576 0.510 0.492
0.10 0.477 0.487 0.505 0.531 0.687 0.10 0.800 0.510 0.460 0.428 0.390 0.380
0.067 0.409 0.416 0.429 0.446 0.562 0.050 0.536 0.384 0.354 0.335 0.312 0.306
0.050 0.367 0.373 0.385 0.397 0.491 0.025 0.388 0.305 0.286 0.273 0.258 0.254
0.033 0.322 0.325 0.335 0.352 0.414 0.010 0.279 0.238 0.227 0.219 0.209 0.206
O.D25 0.294 0.297 0.305 0.322 0.370 Note: Values generated with the semianalytical solution of Cole and
0.020 0.276 0.278 0.287 0.301 0.342 Zlotnik25, courtesy of K. D. Cole (as presented in Butler 1\
0.013 0.247 0.249 0.255 0.269 0.300 Values for (L, + L,) =h.
0.010 0.230 0.231 0.238 0.250 0.276
0.0067 0.211 0.210 0.213 0.2?:7 0.248
0.0050 0.198 0.199 0.201 0.213 0.230
Note: Values generated with the semianalytical solution of Cole and
Zlotnik25 , courtesy ofK. D. Cole (as presented in Butler 1\ Values for
L/h ~ 0.05.
Table 5.15.3 Tabulated Values of the Dimensionless Flow

Parameter, P, Used in Confined Formation for the Dagan Method
(L"'+L,)IL,
• "'
8.0 4.0 2.0 1.5 1.05
0.20 0.741 0.727 0.681 0.640 0.561
0.10 0.539 0.533 0.505 0.483 0.432
0.067 0.458 0.455 0.432 0.416 0.377
er and 0.050 0.412 0.408 0.390 0.378 0.345
; finite 0.033 0.359 0.357 0.343 0.331 0.307
ne are 0.025 0.328 0.325 0.314 0.305 0.285
0.020 0.307 0.305 0.295 0.288 0.270
0.013 0.275 0.273 0.263 0.259 0.245
0.010 0.254 0.254 0.246 0.240 0.230
1d fit a
0.0067 0.232 0.230 0.224 0.218 0.211
0.0050 0.218 0.2 16 0.210 0.205 0.199
Note: Values generated with the semianalytical solution of Cole and
Zlotnik,25 courtesy of K D. Cole (as presented in Butler1\ Values for
L/h ~ 0.05.
~K/Kr
:~
I. 'I'=
L/ rw
talized
~.from
Practical issues of concern discussed for the Bouwer and Rice method are also applicable
litions.
to this method. These include the uncertainty about the anisotropy, which interval to fit with a
straight line, and the impact of the noninstantaneous slug.
5.15.4)
Using the slug test data in Table 5.15.4, determine the hydraulic conductivity of the fonnation by the
Dagan method.
EXAMPLE 5.15.2
)S
SOLUTION A conceptual picture of the slug test is shown in Figure 5.15.8.

From this picture,
0.10 'c= 2.55 em, rw = 10 em, L, =3m, Lw=10.3 m, h =15m

The response data on semilogarithmic scale is plotted in Figure 5.15.9. From this plot, for H, =0.368H0,
0.492
T0 = 7.9 seconds. Next, determine the ljf parameter for the given well geometry. Assuming the anisotropy
0.380•
ratio is one,
0.306
0.254 ..JKlK:
'I' =
L/rw
=
300 cm/10 em
0.0333
0.206
Lw + L, =10.3 m+3 m =4.43
L, 3m
Using the above parameters, the dimensionless flow parameter from Table 5.15.2 can be determined:
p =0.325.
Finally, the hydraulic conductivity estimate can be made using Equation 5.15.4:
2
r} (liP)
{2.55 cm) (1/0.325)
K, = 2L,To = (2X300 cm)(7.9 s) 0.00422l cm/s = 3.65 m/day
•
Table 5.15.4 Slug Test Data for Example 5.15.2 5.15.3
Elapsed Time (s) Change in water level (m)

0 0.640
I 0.617
2 0.491
3 0.433
4 0.385
5 0.340
6 0.305
7 0.268
8 0.241
9 0.211
10 0.189
11 0.169
12 0.147
13 0.130 1.00
14 0.115
15 0.101
16 0.089 0.80
17 0.078 "0
«!
18 0.068 Q)
.I::
19 0.058 "0
Q)
.!:!
20 0.051 (ij
21 0.044 E
0
22 0.038 z
23 0.032
Boring diameter = 20 em
Casing diameter = 5.1 em 0.00
0.00
Depth to bottom of well = 17 m BGS (below ground surface)
Depth to top of casing = 14m BGS
Depth to water= 6.7 m BGS
Saturated thickness= 15 m
1.00 -
""""' ~
~
~
HIH0
0.10
~
17m -
15m
I
Figure 5.15.8. Slug test for Example 5.15.2.
5 10
Time (s)
15
Figure 5.15.9. Normalized response data for Example 5.15.2.

20 25
Figure 5.15
5.15.3 KGS Model

Unlike the Bouwer and Rice method or the Dagan method, the KGS (Kansas Geological
Survey) method incorporates the storage properties of the media into the analysis. Hyder et
aJ. 50 presented the analytical solution to this model as
H(t)
- =/(~. u, 'If, L/4-. L/h) (5.15.5)
Ho
Plotting the normalized head versus the logarithm of~ for a particular well-formation con-
figuration defined as ('If, L/Lw, L/h) develops a series of type curves (Figure 5.15.10), each
referring to a different Ct.. The analysis procedure consists of fitting one of the Ct. curves to the
field data (normalized response data versus logarithm of the time since the test began) and
KGS Model Type Curves KGS Model Type Curves

1/t= 0.0635 ift = 0.00635
KGS Model Type Curves

"'"' 0.0217
--a::: 10- 1
- - - a:::1o-2
0.80
----a::: 10-3
"0
!IS ------ a= 10-4
Q)
of: 0.60 --a~ 10-5
"0
Q) .... · .. · Cooper et al.
.~
'i
E
0
z
10 100
25
{3
Figure 5.15.10. Normalized head (H(t)!Ho) vs. logarithm of Ptype curves generated using the KGS model for different values of ljf. 17
select match points (e.g., l 0) and read the value of t1.o· Using the definition of the radial com-
ponent of the hydraulic conductivity is
2
_l_ (5.15.6)
Kr :::
Letl.o
And the specific storage is
r2
ss = - 2' - (5. 15.7)
rwLe
5.16 SLUG TESTS FOR HIGH CONDUCTIVITY FORMATIONS

Slug tests performed in high transmissivity (high conductivity) aquifers may produce oscilla-
tory data. In fact, the data may exhibit an underdamped oscillatory behavior, where the oscil-
lations occur as a damped sine wave, indicating that inertial forces are significant. Figure
5. 16.1 illustrates the slug test results with underdamped oscillatory data. Methods have been
developed for the analysis of underdamped oscillatory slug-test response data.57• 67· 70· 90· 99
These methods are based upon solutions of the conservation of momentum written in
terms of the dimensionless deviation of the water level (wd =w!H0) from the static level in the
test well, where w is the actual deviation, and dimensionless time, td == (g/Le)0·5t. The conser- 5.17
vation of momentum is expressed as
2
d wd dwd
-2-+Cd--+wd =0 (5.16.1)
dtd dtd
where cd is the dimensionless dampening parameter.

For the van der Kamp model99, the dimensionless dampening parameter, Cd << 2 (for
response data that are well within the underdamped region), is expressed as
(5.16.2) PROB
5.11.1 T
91-6 91-6 are tabu]
0 calculate
0.8 . Silty sand with gravel
L19 =74ft 11 the valid
0.6 S= 0.001 and (d) c
Sandy silt with gravel
a= 3226 for a pur
0.4
T =14,200 ft 2/day 34
0.2 Sand with gravel and cobbles
~ 0~--------------~--~~~~r---~ 53
l: -0.2
Silty sand with gravel
-0.4
Water level
-0.6 86
-0.8
Silty gravel with sand
-1~~~~~UL~~~~~L-~~~~~
0.1 10 100
120 Screen length: 10 feet
Dimensionless time
0 25 50 feet
5.11.2 u
Figure 5.16.1. Slug test results and drillhole schematic for well 91-6. 106 loss coef
5.17 Well Skin Effect 273
com- For the Kipp model57
.15.6) (5.16.3)
For the Springer and Gelhar model 90
.15.7)
(5.16.4)
For the McElwee et al. model67
scilla-
oscil- (5.16.5)
=<igure
~been
'· 99
The Kipp method accounts for both well-bore storage and inertial effects, producing a series
of type curves that allow for oscillatory responses.
ten in
in the
onser- 5.17 WELL-SKIN EFFECT
The process of drilling, installing, and developing a well often causes the material in the
immediate vicinity of the well to have altered characteristics from that of the original form-
i.16.1)
ation. These altered characteristics cause the well-skin effect, which may have a large impact
on the estimate of the hydraulic conductivity from a slug test. The method by Ramey et at.80
has been the most common approach for incorporating the well-skin effect in fully penetrat-
2 (for ing wells. In partially penetrating wells, the method by Ramey et al. 80 and the KGS model can
be used. See Butler17 for more detail on the well-skin effect.
).16.2) PROBLEMS
5.11.1 The drawdown data from a step-drawdown pumping test in Equation 4.67 for the drawdown data from a step-drawdown
are tabulated below. Using the given data and assuming n = 2, (a) pumping test below.
calculate the fonnation and well loss coefficients, (b) comment on
Q (m3/day) S.., (m)
the validity of then = 2 assumption, (c) assess the well condition,
and (d) calculate the specific capacity and efficiency of the well 250 0.32
for a pumping rate of 1,250 m3/day. 500 0.87
750 1.78
Q (m3/day) Sw(m) 1,000 3.12
500 1.14 1,250 4.95
1,000 2.5 1,500 7.34
1,500 4.22 1,750 10.33
2,000 6.43
5.11.3 A new pumping well with a radius of 0.5 m is constructed
2,500 9.12
and the drawdown data during the first six months are given on
3,000 11.97
page 274. The well pumps at a constant rate of 1,500 m3/day from
3,500 14.87 a confined aquifer with a transmissivity of 640m2/day and astor-
age coefficient of 4.5 x 10-4. If the aquifer is isotropic, homoge-
5.11.2 Using the Rorabaugh procedure, determine the formation neous, and infinite in areal extent, assess the well condition for the
loss coefficient, well loss coefficient, and the power n of discharge first six months following its construction.
t (days) sw (m) water level recovery in the well is tabulated below. Estimate the 5.14.4
transmissivity, hydraulic conductivity, and storage coefficient in the'
2 3.34 using the Cooper-Bredehoeft-Papadopulos method. level n
4 3.47 well cc
Time (s) H (m) H!Ho tivity o
6 3.55
8 3.60 3 0.46 0.910
10 3.64 9 0.39 0.765
15 3.72 15 0.34 0.660
20 3.79 30 0.23 0.450
30 3.88 45 0.16 0.310
60 4.13 60 0.1 1 0.220
90 4.36 90 0.06 0.125
120 4.59
150 4.87 5.14.3 The results of a slug test are given in Table P5.14.3. Well
180 5.34 construction infonnation is as follows:
Boring diameter= 20 em
5.11.4 The drawdown in an existing pumping well with a radius
Casing diameter = 5.1 em
of 0.5 m is measured as 3.88 m. If the well has been pumping at
Depth to bottom of well =17 m BGS
a constant rate of 750 m3/day for a long period of time so that
near steady-state conditions are established (the water level in the Depth to top of casing= 14m BGS
pumping well is essentially stabilized), estimate the transmissiv- Depth to water= 6.7 m BGS
ity of the confined aquifer. Assume the well is 100 percent effi- Saturated thickness =15 m
cient. Would the actual transmissivity be larger or smaller than If the initial water level rise following the injection of the slug is
your estimate considering the well losses? 0.64 m, determine the hydraulic conductivity of the formation Bori
5.11.5 An existing well has been pumping from a confined using the Hvorslev method. Casi
aquifer at a constant rate of I ,500m3/day for three months. The Table P5.14.3 Slug Test Results for Problem 5. 14.3
Dep
drawdown in the pumping well with a radius of 0.5 m is meas- Dep
ured as 4.42 m at the end of three months. Estimate the aquifer Elapsed time (s) Change in water level (m) Dep
transmissivity assuming the well losses are negligible. Perf01m a 0 0.640 Satu
sensitivity analysis considering the well losses. 1 0.617
5.15.1
5.11.6 Resolve Problem 4.7.1 including well losses. Use C = 0.8 2 0.491
x 10-4 ft!gpm2 and n = 2. method
3 0.433 Are the.
5.14.1 The water level data from a slug test are tabulated below. 4 0.385
5.15.2 ~
The initial water level following the injection of the slug is 0.46 5 0.340 lated in
m. The casing radius of the well is 7.8 em and the radius of the 6 0.305 Figure]
screen is 5.4 em. If the test is performed in a confined aquifer
7 0.268 m. Asse
with a thickness of 12m. determine the transmissivity, hydraulic
8 0.241 well ge<
conductivity, and storage coefficient using the Cooper-
9 0.211 1. C<X
Bredehoeft-Papadopulos method.
10 0.189
Time (s) H (m) HIH0 2. Hvc
11 0.169
3. Bor
2 0.41 0.89 12 0.147
13 0.130 4. Da~
6 0.36 0.78
10 0.31 0.68 14 0.115 Determi
20 15 0.101 feasible
0.25 0.54
45 0.15 0.32 16 0.089
Table P
80 0.09 0.20 17 0.078
100 0.06 0.14 18 0.068
120 0.05 0.11 19 0.058
20 0.051
5.14.2 A casing with a radius of 6.8 em is installed through a con- 21 0.044
fined aquifer. A screen with a radius of 5.6 em penetrates the full 22 O.Q38
thickness of the formation, 8 m. A slug of water is injected 23 0.032
instantaneously, raising the water level in the well 0.51 m. The
Problems 275
tate the
5.14.4 A slug test is performed by lowering the static water level Table P5.15.2 (continued) Slug Test Data for Problem 5.15.2
in the well by 0.60 m at time t = 0 sec. Following this, the water
fficient Time (s) HJH0
level recovery in the well occurs as tabulated below. Given the
well construction information, detenn.ine the hydraulic conduc- 15 0.829
tivity of the formation using the Hvorslev method. 20 0.792
Elapsed time (s) Change in water level (m) 29 0.732
39 0.675
0 0.600 48 0.629
22 0.550 57 0.588
43 0.510 67 0.548
65 0.475 78 0.509
92 0.436 87 0.480
134 0.387 97 0.451
167 0.344 115 0.407
.3. Well
225 0.298 133 0.369
263 0.258 156 0.332
298 0.228 179 0.302
350 0.192 219 0.263
415 0.156 255 0.239
472 0.134 292 0.220
542 0.104 355 0.197
628 0.082 449 0.176
e slug is
mnation Boring diameter = 20 em 563 0.160
Casing diameter= 5.2 em 626 0.154
Depth to bottom of well = 8.2 m BGS 738 0.145
Depth to top of casing = 5.2 m BGS 795 0.140
n) Depth to water= 3.8 m BGS 879 0.135
Saturated thickness= 4.4 m
5,15.1 Rework Problem 5.14.3 using the Bouwer and Rice

method. Compare your results to the results of Problem 5.14.3.
Are they consistent? Boring diameter"' 23 em
5.15.2 The response data from a slug (bail-down) test are tabu- Casing diameter= 10.2 em
lated in Table P5.1S.2. The well construction details are shown in Depth to bottom of well "' 36.6 m BGS
Figure P5.15.2. The initial head drop following the bailing is 0.99 Depth to top of casing"' 24.4 m BGS
m. Assess the feasibility of the following methods for the given Depth to water = 24.4 m BGS
well geometry and response data. Saturated thickness"' 15.2 m
1. Cooper-Bredehoeft-Papadopulos method
2. Hvorslev method
3. Bouwer and Rice method
4. Dagan method
m
Determine the hydraulic conductivity of the formation using the
feasible methods and compare the results.
36.6m
Table P5.15.2 Slug Test Data for Problem 5.15.2
Time (s) HIH.o
0 1.000
2 0.960
4 0.930
5 0.918
10 0.867 Figure PS.15.2. Well construction details for Problem P5. L5.2
5.16.1 Perform a literature search in order to describe the van der 5.16.5 Describe the mathematical formulation of the Hvorslev 37. Gib
Kamp method99 for performing slug tests in high-conductivity method and describe the differences in the mathematical formu- Illinois ;
formations. Start with the reference Wylie and Magnuson 108 lations between this method and the Cooper eta!. method. 38. Gib
5.16.2 Perfonn a literature search in order to describe the Kipp 5.16.6 Describe the mathematical formulation of the Bouwer and Berkele:
methodologr 7 for slug tests in high conductivity formations. Rice method. Describe how the two key assumptions of this 39. Gid
Start with the reference by Weight and Wittman. 106 method are reflected in the mathematical formulation. lectorw
pp. 120(
5.16.3 Develop a write-up on the well-skin effect. 5.16.7 Describe the mathematical formulation of the Dagan 40. Gor
5.16.4 Describe the mathematical formulation of the Cooper et method. Describe the differences in assumptions with the Co., Sou
al. method and describe the assumptions. Bouwer and Rice method. 41. Greo
surized ~
missivit:
REFERENCES 95-424,
42. Gre(
I. Ahmad, N., Tubewells, Construction and Maintenance, Scientific 19. Butler, J. J., Jr., C. D. McElwee, and W. Liu, Improving the quality of the com1
Research Stores, Lahore, Pakistan, 250 pp.• 1969. parameter estimates obtained from slug tests, Ground Water, v. 34, no. 3, ofprema
2. Ahrens, T. P., Basic considerations of well design, Water Well Jour., pp.480-490, 1996. no. 2,pp
v. 24, no. 4, pp. 45-50; no. S, pp. 49-51; no. 6, pp. 47-51; no. 8, pp. 20. California Dept. of Water Resources, Water well standards: State of 43. Half
35-37, 1970. California, Bull. 74, Sacramento, 205 pp., 1968. Aquifer-
3. Amer. Water Works Assoc., AWWA Standard for Water Wells, 21. Campbell, M. D., and I. H. Lehr, Water Well Technology, McGraw- Report D.
ANSJJAWWA AI00--97, Denver, CO, Feb. 1998. Hill, New York, 681 pp., 1973. 44. Hall
4. Amer. Water Works Assoc., Ge11i•1g the Most from Your Well Supply, 22. Campbell, M. D., M. S. Starrett, J. D. Fowler, and J. J. Klein, Slug Resourct
Proceedings AWwA Seminar, New York. 62 pp., 1972. tests and hydraulic conductivity, Ground ~ater Managemelll, v. 2, 1990. 45. Heat
S. Amer. Water Works Assoc., Ground water, AWWA Manual M21, 130 23. Chirlin, G. R., A critique of the Hvorslev method for slug test analysis: Survey "ij
pp.• 1973 The fully penetrating well, Ground Water Monitor Rev., v. 9, no. 2, 1989. 46. Hro'
6. Anderson, K. E., Water Well Handbook, 2nd ed, Missouri Water Well 24. Chirlin, G. R., The slug test: The first four decades, Ground Water observati
& Pump Contractors Assoc., Rolla, 281 pp., 1967. Management, v. I, 1990. Station, I
7. Barlow, P. M. and A. F. Moench, WTAQ-Computer Program for 25. Cole, K. D., and V. A. Zlotnik, Modification of Dagan's numerical 47. Huis
Calculating Drawdowns and Eslimllting Hydraulic Properties for Con- method for slug and packer test interpretation, in Computational Methods 336 pp.,.
fined and Water-Table Aquifers, U.S. Geological Survey Water-Resources in Water Resources X. Peters et al., ed., K.luwer Academic, Dordrecht, 48. Hunt
Investigations Report99-4425, 1999. 1994. no. 1, pp.
8. Barnes, I. and F. E. Clarke, Chemical Propenies of Ground Water and 26. Cooper, H. H., Jr., J.D. Bredehoeft, and I. S. Papadopulos, Response 49. Hurr,
Their Corrosion and Encrustation Effects on Wells, U.S. Geological Sur- of a finite-diameter well to an instantaneous charge of water, Water cific capa
vey Professional Paper498-D, 58 pp., 1969. Resources Research, v. 3, no.1, pp. 11-21, I %7. 50. Hyde
9. Bennett, T. W., On the design and construction of infiltration galleries, 27. Dagan, G., A note on packer, slug, and recovery tests in unconfined in partial!
Ground Water, v. 8, no. 3, pp. 16--24, 1970. aquifers, Water Resources Research, v. 14, no. 5, 1978. 1994.
10. Bentall, R., Methods for Determing Permeability, Transmissibility, 28. Dawson, K., and J. Istok, Aquifer Testing: Design and Analysis of 5!. Jacot
and Drawdown. U.S. Geological Survey Water-Supply Paper 1536-1, pp. Pumping and Slug Tests, Lewis Publishers, Boca Raton, FL. 1991. well, Trar
243-341, I 963. 29. Driscoll, F. G., Groundwater and Wells, 2nd ed., Johnson Division, St. 52. Johm
11. Bentall, R., Methods of Collecting and Interpreting Ground-Water Paul, MN, 1986. Well Des;
Data. U.S. Geological Survey Water-Supply Paper 1544-H, 97 pp., 1963. 30. Erickson, C. R., Cleaning methods for deep wells and pumps, Jour. Water-S~q
12. Bernhard, A. P., Protection of water-supply wells from contamination Amer. Water Works Assoc., v. 53, pp. 155-162, 1961. 53. Johns
by wastewater, Ground Water, v. 11, no. 3, pp. 9-15, 1973. 31. Ferris, 1. G., and D. B. Knowles, Slug Test for Estimating Transmissi- E. Johnsa
13. Bierschenk, W. H., Determining well efficiency by multiple step-- bility, U.S. Geological Survey Note 26, 1954. 54. Karru
drawdowo tests, Inti. Assoc. Sci. Hydrology Publ. 64, pp. 493-507, 1964. 32. Fetter, C. W., Applied Hydrogeology, Prentice Hall, Upper Saddle Ground V,
14. Bouwer, H., The Bouwer and Rice Slug Test-An Update, Ground River, NJ, 200 I. 55. Kaser
Water; v. 27, no. 3, pp. 304-309, 1989. 33. Feulner, A. J., Galleries and Their Use for Development of Shallow lications,l
15. Bouwer, H., and R. C. Rice, A slug test for determining hydraulic con- Groundwater Supplies, with Special Reference to Alaska, U.S. Geological 56. Kemt
ductivity of unconfined aquifers with completely or partially penetrating Survey Water-Supply Paper 1809-E, 16 pp.. 1964. uation of~
wells, Water Resources Research, v. 12, pp. 423-428, 1976. 34. Franke, 0. L., Estimating Areas Contributing Recharge to Wells: 57. Kipp.
16. Butler, J. J., Jr.. The role of pumping tests in site characterization: Some Lessons from Previous Studies, U.S. Geological Survey Circular 1174, of a well
theoretical considerations, Ground Water, v. 28, no. 3, pp. 394-402, 1990. 1998. 1397-140:
17. Butler, J. J. Jr., The Design, Peifonnance, and Analysis ofSlug Tests, 35. Gabrysch, R. D., The relation between specific capacity and aquifer 58. Koeni
Lewis Publishers, CRC Press LLC, Boca Raton, FL, 1998. transmissibility in the Houston area, Texas, Ground Water. v. 5, no. 4., pp. Jour.AmeJ
18. Butler, I. 1., Jr., and I. M. Healey, Relationship between pumping-test 9-14, 1968. 59. Koeni
and slug-test parameters: Scale effect or artifact, Ground Water, v. 36, no. 36. Garg, S. P., and J. Lal, Rational design of well screens, Jour. lrrig. achieved b
2, pp. 305-313, 1998. Drainage Div., Amer. Soc. Civil Engrs., v. 97, no.IR I, pp. l 3 l -147, 1971. 652--{;70,:
References 277
>rslev 37. Gibb, J. P, Wells (lJl([ pumping systems for domestic watu supplies, 60. Kruse, G., Selection of gravtl packs for wells in unconsolidated
>nnu- Illinois State Water Survey Circular I 17, Urbana, 17 pp.. 1973. aquifers, Colorado State Univ. Exp. Sta. Tech. Bull. 66, Fort Collins. 22
38. Gibson, U. P., and R. D. Singer. Water Well Manual, Premier Press, pp.. 1960.
Berkeley, CA. 156 pp., 1971. 61. Kruseman, G. P., and N. A. deRider, Analysis and Evaluation of
~rand
39. Gidley, H. K., and J. H. Millar, Performance records of radial col- Pumping Test Data, International Insttitue for Land Reclamation and
f this Improvement, The Nethedands, 1994.
lector wells in Ohio River Valley, Jour. Amer. Water Works Assoc., v. 52,
pp. 1206-1210, 1960. 62. Larson, T. E.. Corrosion by dcmesric waters, Bull. 59, Illinois State
>agan 40. Gordon, R. W., Water well drilling with cable tools, Bucyrus-Erie Water Survey, Urbana, 48 pp., 1975.
h the Co., South Milwaukee, WI, 230 pp., 1958. 63. Lee, J., Well Testing, Society of Petroleum Engineers of AIME. New
41. Greene, E. A., and A. M. Shapiro, Methods of Conducting Air-Pres- York, 1982.
surized Slug Tests and Computarion ofType Curves for Estimating Trans· 64. Lennox, D. H., Analysis and application of step-drawdown test, Jour.
missivity and Storativiry, U.S. Geological Survey Open-File Report Hydraulics Div., Amer. Soc. Civil Engrs., v. 92, no. HY6, pp. 25-48, 1996.
95-424, 1995. 65. Logan, J., Estimating transmissibility from routine production tests
42. Greene, E. A., and A. M. Shapiro, AlRSLUG: A Fortran program for of water welLs, Ground Water, v. 2, no. I, pp. 36-37, 1964.
uity of the computation of type curves to estimate transmissivity and storativity 66. Matlock, W. G.• Small diameter wells drilled by jet-percussion
. no. 3, of prematurely terminated air-pressurized slug tests, Ground Water, v. 36, method, Ground Water. v. 8, no. I, pp. 6-9, 1970.
no.2,pp.373-376, 1998. 67. McElwee, C. D., J. J. Butler. Jr., G. C. Bohling, Nonlinear Anal)'sis
'tate of 43. Halford, K. J.• and E. L. Kuniansky, Spreadsheets for the Analysis of of Slug Tests in Highly Permeable Aquifers Using a Hvorslev-rype
Aquifer-Test and Slug-Test Data, U.S. Geological Survey Open-File approach, Kansas Geological Survey, Open-File Report 92-39,
:Graw- Report 02-197, http://pubs. usgs.gov/of/ofr02197, 2002. Lawrence, KS, 1992.
44. Hall, P., and J. Chen, Water Well (lJl([ Aquifer Test Analysis, Water 68. McElwee, C. D., J. I . Butler, Jr., and G. C. Bohling, Sensitivity
1, Slug Resources Publications, Littleton, CO, 1996. Analysis of slug tests: Part I, The slugged well, Journal of Hydrology, v.
1990. 45. Heath, R. C., Basic Ground-Water Hydrology, U.S. Geological Water 164, pp.53-67, 1995.
1alysis: Survey Water-Supply Paper 2220, 1983. 69. McLane, G. A., D. A. Harrity, and K. 0. Thomsen, A pneumatic
1989. 46. Hrovslev, M. I ., Time lag and soil permeability in groundwater method for conducting rising and falling head tests in highly permeable
1
Water observations, U.S. Army Corps of Engineers Waterway Experimentation aquifers, Proc. 1990 NWWA Outdoor Action. Conf., National Water Well
Station, Bulletin 36, Vicksburg, MS 1951. Association, 1990.
nerical 47. Huisman, L., Groundwarer Recovery, Winchester Press, New York. 70. McElwee, C. D., and M. Zenner, Unified analysis of slug tests
rethods 336 pp.• 1972. including nonlinearities, inertial effects, a11d turbulence, Kansas Geolog-
'drecht, 48. Hunter Blair, A., Well screens and gravel packs, Ground Water, v. 8, ical Survey Open-File Report 93-45, 1993. ·
no. I, pp. 10-21, 1970. 71. Mogg, J. L., The technical aspects of gravel well construction, Jour.
:sponse 49. Hurr, R. T., A new approach for estimating transmissibility from spe- New England Water Works Assoc., v. 77, pp. 155-164, 1963.
Water cific capacity, Water Resources Research. v. 2, pp. 657~. l96fi. 72. Mogg, J. L.• Practical construction and incrustation guide lines for
50. Hyder, z.. J. J. Butler, Jr., C. D. McElwee, and W. Z. Liu, Slug tests water wells, Ground Water. v. 10, no. 2, pp. ~11. 1972.
Jnfined in partially penetrating wells, Water Resourcts Research, v. 30, no. I I , 73. Moore, P. L., Drilling Practices Manual, Petroleum Publishing,
1994. Thlsa, OK, 448 pp.• 1974.
lysis of 51. Jacob, C. E., Drawdown test to determine effective radius of artesian 74. Noble, D. G.• Well points for dewatering, Ground Water, v. 1. no. 3,
well, Trans. Amer. Soc. Civil Engrs., v. 112, pp. 1047-1070, 1947. pp. 21-26, 1963.
;ion, St. 52. Johnson, A. I., et al., Laboratory Study of Aquifer Propenies and 75. Pandit, N. S., and R. F. Miner, Interpretation of slug test data, Ground
Well Design for an Artificial-Recharge Site, U.S. Geological Survey Water, v. 24, pp. 743-749, 1986.
s, Jour. Water-Supply Paper 1615-H, 42 pp., 1966. 76. Papadopulos, S. S., J. D. Bredehoeft. and H. H. Cooper, Jr., On the
53. Johnson Div., UOP Inc., Ground Water and Wells, 2nd ed., Edward analysis of "slug test" data, Water Resources Research, v. 9, no. 4, pp.
csmissi: E. Johnson, St Paul, MN, 440 pp., 1972. 19-21, 1973.
54. Karanjac, J., Well losses due to reduced formation permeability, 77. Peres, A. M.. M. Onur, and A. C. Reynolds, A new analysis proce-
Saddle Ground Water, v. 10, no. 4, pp. 42-49, 1972. dure for determining aquifer properties from slug test data, Water
55. Kasenow, M.lnJroducrion to Aquifer Analysis, Water Resources Pub- Resources Research. v. 25. no. 7, 1989.
)hallow lications, Littleton, CO, 1997. 78. Peterson, F. L., Water development on tropic volcanic islands-Type
1/ogical 56. Kemblowski, M. W., and C. L. Klein, An automated numerical eval- example: Hawaii, Ground Water, v. 10, no. 5, pp. 18-23, 1972.
uation of slug test data, Ground Water, v. 26, pp. 435-438, 1988. 79. Prudic, D. E., Estimates ofHydraulic Conductivity from Aquifer-Test
rWells: 57. Kipp, K. L., Jr., Type curve analysis of inertial effects in the response Data and Specific-Capacity Data, GulfCoast Regional Aquifer Systems,
r 1174, of a well to a slug test, Water Resources Research. v. 21, no. 9, pp. South-Central United Sates, U.S. Geological Survey Water-Resources
1397-1408, 1985. Investigations Report 90-4121, 1991.
aquifer 58. Koenig, L., Survey and analysis of well stimulation performance, 80. Ramey, H. J., Jr., R.G. Agarwal, and I. Martin, Analysis of "slug
•.4., pp. Jour. Amer. Water Works Assoc., v. 52, pp. 333- 350, 1960. test" or DST now period data, Journal Canadian Petroleum Technology,
59. Koenig, L., Relation between aquifer permeability and improvement vol. 24, 1975.
tr. lrrig. achieved by well stimulation, Jour. Amer. Water Works Assoc., v. 53, pp. 81. Ritchie, E. A., Cathodic protection wells and ground-water pollution,
7, 1971. f, 652-670, 1961. GroundWarer, v. 14, pp. 146-149, 1976.
82. Rorabaugh, M. I., Graphical and theoretical analysis of ste[Hiraw- 97. U.S. Soil Conservation Service, Engineering Field Manual for Cl
down test of artesian well, Proc. Amer. Soc. Civil Engrg., v. 79. Sept 352., Conservation Practices, 995 pp., 1969.
23 pp., 1953. 98. Vaadia, Y., and V. H. Scott, Hydraulic properties of perforated well
83. Scalf, M. L. et al., Manual of Ground Water Sampling Procedures, casings, Jour. lrrig. Drainage Div., Amer. Soc. Civil Engrs .. v. 84, no.
National Water Well Association, 1981. IRJ, Paper !505, 26 pp., 1958.
84. Schwartz, D. H. Successful sand control design for high rate oil and 99. van der Kamp, G., Determining aquifer transmissivity by means of
water wells, lour. Perr. Tech., v. 21, pp. 1193-1198, 1969. well response tests: The underdamped case. Water Resources Research,
85. Sepulveda, N., Computer Algorirhm for the Analysis of Under- v. 12, no. I, pp. 71-77, 1976.
damped and Overdamped Water-Level Responses in Slug Tests, U.S. Ceo- 100. Walton, W., Aquifer Test Analysis with Windows™, CRC Press.
logical Survey Water-Resources lnvesrigations Report 91-4162, 1992. Boca Raton, FL, 1996.
86. Sheahan, N. T., Type-curve solution of step-drawdown test, Ground 10!. Walton, W. C., Selected analytical methods for well and aquifer
Warer. v. 9, no. I. pp. 25-29, 1971. evaluation, Bull. 49, Illinois State Water Survey, Urbana, 81 pp., 1962.
87. Soliman, M. M., Boundary flow considerations in the design of 102. Walton. W. C., and S. Csatlany, Yields of deep sandstone wells in
wells, Jour. lrrig. Drain. Div., Amer. Soc. Civi\Engrs., v. 91, no. IR1, pp. Northern Illinois, Rept. of Inv. 43, Illinois State Water Survey, Urbana,
159-177, 1965 47 pp., 1962.
88. Speedstar Div., Well Dril/ing Manual, Koehring Co., Enid, OK, 72 103. Watson, L. J., Development of ground water in Hawaii. lour.
pp. (no date). Hydraulics Div., Amer. Soc. Civil Engrs., v. 90, no. HY6, pp. 185-202,
89. Spiridonoff, S. V., Design and use of radial collector wells, Jour. 1964
Amer. Water Works Assoc., v. 56, pp. 689-698, 1964. 104. Watt, S. B., and W. E. Wood, Hand Dug Wells and Their Construc-
90. Springer, R. K., and L. W. Gelhar, Characterization of Large-Scale tion, Intermediate Technology, London, 234 pp., 1976.
Aquifer Heterogeneity in Glacial Outwash by Analysis of Slug Tests with 105. Weight, W. D., and J. L. Sonderegger, Manual of Applied Field
Oscillatory Responses, Cape Cod, Massachusetts, U.S. Geological Sur- Hydrogeology, McGraw-Hill, New York, 2001 .
vey Water Resources Investigations Report 91-4034, 1991. 106. Weight, W. D., and G. P. Wittman, Oscillatory slug-test data sets: A
91. Stow, G.R.S., Modem water-well drilling techniques in use in the comparison of two methods, Ground Water; v. 37, no. 6, pp. 827-835,
United Kingdom, Ground Water, v. I, no. 3, pp. 3-12, 1963. 1999.
92. Stramel, G. J., Maintenance of well efficiency, Jour. Amer. Water I 07. Weichert, W. T., and B. N. Freeman, Horizontal wells, Jour. Range
Works Assoc., v. 57, pp. 996-1010, 1965. Management, v. 26, pp. 253-256, 1973.
93. U.S. Bureau of Mines, Horizomal Boring Technology: A Stare of the 108. Wylie, A. and S. Magnuson, Spreadsheet modeling of slug tests
An Study, Info. Circular 8392, 86 pp., 1968. using the van der Kamp method, Ground Water; v. 33, no. 2, pp. 326-329,
94. U.S. Bureau of Reclamation, Groului Water Manual, U.S. Dept. of 1995. 6.1 T
the Interior, 480 pp., 1977. 109. Yang, Y. 1., and T. M. Gates, Wellbore skin effect in slug-test data
95. U.S. Environmental Protection Agency, Manual of Water We// analysis for low-permeability geologic materials, Ground Water. v. 35, 6.1.1
Construction Practices, Rep/. Environmental Protection Agency-570/9- no.6,pp. 931-937,1997.
75-001, Washington, DC, !56 pp., 1976. 110. Z1otnik, V., Interpretation of slug and packer tests in anisotropic
96. U.S. Public Health Service, Manual of Individual Water Supply aquifers, Ground Water, v. 32, no. 5, 1994.
Systems, Pub!. no. 24, 121 pp., 1962.
E~
:lf!
EQ)
·x a;
IDE
E~
-'iii
0 >
cQl
·--o-
( ]....
)
·-ro-
ro
~ ~
'0'0
'0 §
! e-
~Ol
:l-
ECI!
:l :::J -
(J c
(,) c
<(ro
I for
well
Chapter 6
, no.
ns of
1rth,
'ress,
Groundwater Levels and
·uifer
•62.
lis in
Environmeiital Influeiices
ban a,
]OI4r.
-202,
A groundwater level, whether it be the water table of an unconfined aquifer or the piezometric
surface of a confined aquifer, indicates the elevation of atmospheric pressure of the aquifer.
Any phenomenon that produces a change in pressure on groundwater will cause the ground-
Field water level to vary. Differences between supply and withdrawal of groundwater cause levels
to fluctuate. Streamflow variations are closely related to groundwater levels. Other diverse
!ts:A
-835, influences on groundwater levels include meteorological and tidal phenomena, urbanization,
earthquakes, and external loads. Finally, subsidence of the land surface can occur due to
lange changes in underlying groundwater conditions.
: tests
-329,
6.1 TIME VARIATIONS OF LEVELS
t data
V, 35, 6.1.1 Secular Variations
Secular variations of groundwater levels are those extending over periods of several years or
tropic
more. Alternating series of wet and dry years, in which the rainfall is above or below the mean,
will produce long-period fluctuations of levels. The long records of rainfall and groundwater lev-
els shown in Figure 6.1.1 illustrate this point. Rainfall is not an accurate indicator of groundwa-
ter level changes. Recharge is the governing factor (assuming annual withdrawals are constant);
it depends on rainfall intensity and distribution and amount of surface runoff. Figure 6.1.2 illus-
trates the recharge occurring during and immediately following periods of precipitation.
Cii
+50 ::l
§'E
til.£.
'Oc:
0 c: al
2~
.!!!QJ
>.r:::.
-50 a>-
"E
'Co
(1) ....
«;:::
-100 :; ]i
E.£:
6~
(J
Figure 6.1.1. Secular variations of maxi-
-150 <(
-20L---~----L_--~----~----~----~
mum annual groundwater level and annual
1900 1910 1920 1930 1940 1950 rainfall in San Bernardino Valley,
Years California.
279
280 Chapter 6 Groundwater Levels and Environmental Influences
--::;;:-Rechorge
..."'' Well PI- 533
1\ 11nal
4J•
(i) TO
......
Q) 60 Prtclpltotio~ at Wa1hin9tan, N~-
.....
Q)
50 1-
E 40
- -
~ 30 -
........
-
I~
20
-a:
<a
c
'«i
10
0 I ill
JAN
I
I
ld I
fES MAR
I
[
Ll
APR
111
MAY
II Ill
JUNE
I .I II
JULY
o1 I I 1.
AUG I SEPT OCT I
II
NOV
I,
I
I,
DEC
1918
Figure 6.1.2. Fluctuation of the water table in the coastal plain of North Carolina (Heath34).
In other instances, pronounced trends may be noted. Thus, in overdeveloped basins where
draft exceeds recharge, a downward trend of groundwater levels may continue for many years.
Figure 6.1.3a dramatically illustrates the decline in piezometric surface of a deep sandstone
aquifer as a result of nearly a century of intensive pumping in the Chicago metropolitan area.
Maximum declines in heads occurred around 1980, after which public water suppliers shifted Baseft·
their source of water from groundwater to additional withdrawals from Lake Michigan. This Albers I
Slanda.
shift resulted in a significant decrease in aquifer withdrawals and a recovery of heads as shown
in Figure 6.1.3b.
Figures 6.1.4-6.1.6 show, respectively, the inferred potentiometric surface of the Mem-
phis (Tennessee) aquifer prior to groundwater development, the potentiometric surface in 1995
with cones of depression, and the declining water levels in two observation wells.
6.1.2 Seasonal Variations

Many groundwater levels show a seasonal pattern of fluctuation. This results from influences
such as rainfall and irrigation pumping that follow well-defined seasonal cycles. The variations
shown in Figure 6.1. 7 are typical for areas subject to frozen ground in winter. Highest levels
occur in late spring and are lowest in winter. In irrigated areas where frozen ground is not a fac-
tor, lowest levels normally occur during fall at the end of the irrigation season. The amplitude
depends on recharge, pumpage, and the type of aquifer; confined aquifers normally display a
greater range in levels than do unconfined aquifers.
6.1.3 Short-Term Variations

Groundwater levels often display characteristic short-term fluctuations governed by the pri-
mary use of groundwater in a locality. Clearly defined diurnal variations may be associated
with municipal water-supply wells. Similarly, weekly patterns occur with pumping for indus-
trial and municipal purposes.
42'
where
years.
jstone
1 area
;hifted Base Iran U.S. GedogQJ Sol\'0!' 1:2,000.000 l:';gllll Dola
Abel$ Eq./ai·Mea C<ric projedlon
t. This Slandard pardel$ 33' and w, eemral meridian -$•
shown
Explanation Figure 6.1.3a. Decline in heads (water levels) in the
-700-- Line of equal water-level decline, 1864-198G-Oashed Cambrian-Ordovician confined aquifer, Chicago and
Mem- where approximate. Interval, in feet, Is variable Milwaukee areas, 1864-1980 (modified from Avery3 as
11995 Major ground-water divide presented in Alley el al. 1) .
250
Jences
~
lations ~ 200.
~
levels «l
([)
tafac- "' 150
<D
Jlitude ;>
0
.0
;play a «l 100
~
Qi
;>
~
50
~
([)
(ii Figure 6.1.3b. Representative
~ 0
he pri- trend of water levels for a deep
>ciated well in Cook County, Chicago
-50
indus- 1940 1950 1960 1970 f980 1990 2000 area, since 1940 (from
Year Visocicy89).
282 Chapter 6 Groundwater Levels and Environmental [nfluences
llllf
ExPanatJon
-23()- ""llllllaolltllc-
Showso-alwNd>
-IIYtl would n...
-
slood In lillhUY .-d
wells. Ccnt.our lnlarvll
10feolO.CUmls-
IS1l' ~-
Figure 6.1.4. Inferred poten-

tiometric surface of the
Memphis aquifer prior to
I
groundwater development.
I The observation wells shown
I
I were selected for their early
I
I TI!NNI'SSI!£ I records away from initial
---r~--r-----
, I pumping centers (modified
I \ from Criner and Parks, 1976,
as presented in Taylor and
fo'o/o'o fMUS
0 SII.IMTBtS Alley~.
....
Q)
ro
s
Figt
in F:
.......
~
2
g2
(!)
:0
2:1
,_ 3
2
Figure 6.15. ·Potentiometric t1l
~
surface of the Memphis .9 4
aquifer in 1995 showing .t::
li
cones of depression and loca- <ll
0
tion of observation wells 5
Sh:P-76 and Sh:Q-1 (modi-
fied from Kingsbury, 1996,
as presented in Taylor and 6
1,., •'' '•I UJI41tJtS
o
' Jr.!S Allel0).
..--..
(l)
0
~
t:
::;)
(/)
-o
c
.sll
~
0
a;
.0
(l)
>
(l)
._
(l)
(ij
poten- ~
to 1930 1935 1940 1~5 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995
! 01. Years
;hown Q) 70 8
early
.al
tied
-...
0
co
::j
(/) 80
lJ
1976, c
~
and 90
~
.9
(])
.0
100
-a;
.t:
~ no:
...
-
(])
Ill 120
~ 19l5 1930 1935 1940 1945 1950 1955 1960 1965 1970 . 1975 1980 1985 1990 1995
Years
Figure 6.1.6. Declining water-level trends in two long-tenn observation wells in the Memphis area. Locations of wells are shown
in Figure 6.1.5 (Taylor and AUey80).
ometric
tis
iog
il!d loca-
ells 5
(modi-
1996, Figure 6.1.7. Seasonal fluctuations of the
rand 6
1951 water table in a glacial till aquifer in Ohio.
Years Well depth is 9 m (after KJein and Kase~ 1 ).
6.2 STREAMFLOW AND GROUNDWATER LEVELS

Where a stream channel is in direct contact with an unconfined aquifer, the stream may
recharge the groundwater or receive discharge from the groundwater, depending on the rela-
tive levels. A gaining stream is one receiving groundwater discharge; a losing stream is one
recharging groundwater (see Figure 6.2.1). Often a gaining stream may become a losing one,
and conversely, as the stream stage changes.5°· 59 Figure 6.2.2 illustrates water table contours
and groundwater flow direction in relation to stream stages.
The term rising water is applied to marked increases in streamflow in reaches where a
subsurface restriction forces groundwater to the surface. Figure 6.2.3 illustrates the phenome-
(a) Gaining Steam
-
-
losing Steam
(b)
Figure 6.2.1. Interaction of

streams and ground water
(modified from Winter et al. 96,
as presented in Alley et al. 1) .
t
Gaining streams (a) receive
water from the groundwater
system, whereas losing streams
(b) lose water to the groundwa-
ter system. For groundwater to
discharge to a stream channel,
Losing Steam That Is Disconnected the altitude of the water table in
(c) From The Water Table the vicinity of the stream must
Aow direction be higher than the altitude of
the stream-water surface.
Conversely, for surface water to
seep to ground water, the alti-
tude of the water table in the 6.2.1 B
vicinity of the stream must be
lower than the altitude of the
stream surface. Some losing
streams (c) are separated from
the saturated groundwater
system by an unsaturated zone.
Cross
may sections
rela-
. one
one,
:ours
::re a
)me- Plan views
Figure 6.2.2. Water table contours and ground-

water flow directions in relation to stream
(a) (b) stages. (a) Losing stream (b) Gaining stream
Streambed
/ Water table
-...... Groundwater
flow __.;!
Impermeable
{a)
of ~- .
t
re
er
·earns
ldwa-
L No flow _0---=""
~
Distance downstream -
/Noflow
Figure 6.2.3. Ulustration of rising water in a
stream channel from emerging groundwater
flow. (a) Cross section along stream charmel in
:erto an alluvial valley (b) Streamflow as a function
mel, {b) of distance along the stream
tble in
must
:of non for a situation where a dry stream channel exists above and below the convergent section.
Figure 6.2.4 illustrates the dynamic interface between groundwater and streams.
ater to
alti-
the 6.2.1 Bank Storage
;t be
the During a flood period of a stream, groundwater levels are temporarily raised near the channel
.ng by inflow from the stream. The volume of water stored and released after the flood is referred
from to as the bank storage. Field data are rarely adequate to evaluate bank storage and its rate of
inflow and outflow; therefore, analytic or model approaches are necessary to obtain quantita-
zone. tive estimates for specified boundary conditions.
t
Q)
E
:J
0
terface of groundwater- >
Q)
flow system, hyporheic zone, Cl)
~
and stream 0
(i)
j£
c
<0
mo
0
t
Q)
~
(1)
Figure 6.2.4. The dynamic interface between .r::.
0
groundwater and streams (modified from U)
0
Winter et al.,96 as presented in Alley et al. 1) . 0
Streambeds are unique envirorunents where
groundwater that drains much of the subsur-
face of landscapes interacts with sutface
water that drains much of the surface of land-
scapes. Mixing of surface water and ground-
water takes place in the hyporheic zone where
microbial activity and chemical transforma-
tions commonly are enhanced.
Figure 6.2.5 illustrates idealized groundwater conditions adjacent to a flooding stream. A

flood hydrograph of sinusoidal fonn (Figure 6.2.5a) was superimposed on an aquifer and
stream situation sketched in Figure 6.2.5b. As a result of the flood, the bank storage increased
and then decreased; the variation of the volume of water in storage is depicted in Figure 6.2.5c.
The derivative of the volume curve yields the groundwater flow curve (Figure 6.2.5d). From
this it can be seen that a stream fluctuation produces large variations in magnitude and direc-
tion of local groundwater flow.
Cooper and Rorabaugh13 derived solutions for changes in groundwater head near the
stream, groundwater flow to the stream, and bank storage. Their comprehensive analysis also
included a family of asymmetric flood-wave stage hydrographs, which facilitate study of the
effects of a wide variety of flood shapes on groundwater.
6.2.2 Base Flow

Streamflow originating from groundwater discharge is referred to as groundwater runoff or
base flow. During periods of precipitation, streamflow is derived primarily from surface runoff;
'~
Crest stage / Ground surlace
Initial v---Fiood amplitude

and
final
stage
\. \7 / Initial water table
Time in flood periods

(a)
Stream
channel l Aquifer
Impermeable
(b)
Time in flood periods

(c)
t
Water leaving bank storage
ween
,,t).
2 3
~re
Time In flood periods
;ur-
(d)
land-
md- Figure 6.2.5. Groundwater in relation to a flood-
Water entering bank storage
o,~here ing stream. (a) Flood hydrograph (b) Vertical
na- cross section of field conditions (c) Volume of
bank storage as a function of time (d)
Groundwater flow to and from bank storage.
Results are from a laboratory model (after
Todd81 ).
m.A
rand
~ased whereas during extended dry periods, all streamflow may be contributed by base flow. 32· 60
.2.5c. Typically, base flow is not subject to wide fluctuations and is indicative of aquifer character-
F'rom istics within a basin.49•57• 58
lirec- To estimate base flow, a rating curve of groundwater runoff can be prepared by plotting
mean groundwater stage (water table level) within the basin against streamflow during periods
lf the when all flow is contributed by groundwater.74 Figure 6.2.6a shows rating curves for a small
; also drainage basin in Dlinois. Data were fitted by two curves: one for the April-October period
1fthe when evapotranspiration from groundwater is significant, and the other for the November-
March period when evapotranspiration is minimal. With these rating curves and the mean
groundwater stages for one year (see Figure 6.2.6b), the separation of surface runoff and base
flow hydrographs shown in Figure 6.2.6c could be achieved. It can be noted that frozen ground
impeded groundwater recharge during February and March and that base flows were largest
off or during the spring and summer months. Groundwater contributed 33 percent of total streamflow
moff; for the year.
Streamflow at any instant contains groundwater contributed at previous times and differ-
ent locations within the drainage area. During and after a storm period in a small drainage
basin, the water table will rise, causing the base flow to increase also (Figure 6.2.6a). But
superimposed on this will be the bank storage fluctuation (Figure 6.2.5d). The effects of these
two variations are shown schematically in Figure 6.2.7.
An alternative approach to determining the separation of total streamflow into surface-
runoff and groundwater components during flood periods can be accomplished from meas-
urements of chemical concentrations. 53• 88 Total dissolved solids or any major ion will serve the
purpose with the equation
(6.2.1)
- • Data for periods when

evapotranspiration !s very small
- - o Data for periods when
evapotranspiration Is great
0 2 4 6 8 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Groundwater runoff (m3/sec) Months
(a) (b)
liD Estimated 6.2.:

groundwater
100
0.1
Figure 6.2.6. Estimate of base flow for Panther Creek near Peoria, illinois (drainage area: 246 km2). (a) Rating curves of mean
groundwater stage versus groundwater runoff (base flow) (b) Mean groundwater stage for 1951 (c) Streamflow hydrograph for 1951
showing surface runoff and base flow components (after Schicht and Walton73)
iffer-
inage
•. But
/Total hydrograph
these
rface-
.neas-
ve the
t
6.2.1)
Baseflow
hydrographs
Bank storage
Figure 6.2.7. Schematic diagram of
neglected
the variation of base flow during a
I I "' flood hydrograph with and without
/
effects of bank storage (after
Time- Singh7\
where Cis ionic concentration, Q is streamflow, TR is total runoff, GW is groundwater con-

tribution (base flow), and SR is surface runoff. Solving for the base flow,
(6.2.2)
where
(6.2.3)
Values of Ccw are measured during rainless periods, CsR is measured in small tributary streams
Dec during storm events, and C1R is measured during the peak flow period in the main stream.
Measurements by this method for three small basins (6 to 13 km2) showed that groundwater
contributed from 32 to 42 percent of the total flow at peak discharge. 64
6.2.3 Base Flow Recession Curve

A recession curve shows the variation of base flow with time during periods of little or no rain-
fall over a drainage basin (see Chow et al.9). In essence, it is a measure of the drainage rate of
groundwater storage from the basin?~· 94 If large, highly permeable aquifers are contained
within a drainage area, the base flow will be sustained even through prolonged droughts; if the
aquifers are small and of low permeability, the base flow will decrease relatively rapidly and
rnay even cease. 17• 83
Analyses of streamflow hydrographs show that the recession curve can often be fitted by
the equation
(6.2.4)
where Q is streamflow at timet in days after a given discharge Q0, and K is a recession con-
stant governed by the hydrogeologic characteristics of the basin.4• 9· 52 The value of K can be
empirically determined from the slope of a straight line fitted to a series of consecutive dis-
!an charges plotted on semilogarithmic paper, as shown in Figure 6.2.8. Typical values lie in the
or 1951 range of 0.89 and 0.95. Thus, prior knowledge of the shape of the recession curve enables
future estimates to be made of streamflow during rainless periods.
6.3 I
t
Ideal base flow
6.3.1
Stream becomes
losing Figure 6.2.8. Base flow recession curves of
streamflow for varying magnitudes of evapo-
Time--+ transpiration losses from groundwater.
v-- Water !able divide
-~,- ------
J Initial water table 1 "' Initial ------------ -- - 6.3.2.
! Final
~----~:~w~~;,;:;;;:::x-- -------
I
1 Unconfined isotropic aquifer I
(a) (b)
Figure 6.2.9. Water table and stream channel conditions affecting base flow. (a) Fully penetrating stream (b) Partially penetrating
stream (after Singh75)
An analytic study of base flow by Singh75 demonstrated that base flow recession curves
depend on the degree to which a stream channel is entrenched in an aquifer. For a fully pene-
trating stream (see Figure 6.2.9a), recession curves do not plot as straight lines on semiloga-
rithmic paper; instead, the recession rate continuously decreases with time, forming a concave
curve. But for deep aquifers and partially penetrating streams (see Figure 6.2.9b), the straight-
line approximation is generally applicable. The value of K in Equation 6.2.4 varies directly
with the degree of stream entrenchment.
These approaches to base flow assume that groundwater drains only toward the stream
channel. Groundwater also can flow downward to an underlying leaky aquifer and can be lost
by evapotranspiration to the atmosphere.75 Where these diversions are significant, the reces-
sion curve will be deflected downward. In semiarid regions where streamflow is intermittent,
evapotranspiration losses become significant; this causes the recession curve to steepen (see
Figure 6.2.8) until streamflow finally ceases.
6.3 Fluctuations Due to Evapotranspiration 291
6.3 FLUCTUATIONS DUE TO EVAPOTRANSPIRATION

Unconfined aquifers with water tables near ground surface frequently exhibit diurnal fluctua-
tions that can be ascribed to evaporation and/or transpiration. Both processes cause a discharge
of groundwater into the atmosphere and have nearly the same diurnal variation because of their
high correlation with temperature.
6.3.1 Evaporation Effects

Evaporation from groundwater increases as the water table approaches ground surface. The
rate also depends on the soil structure, which controls the capillary tension above the water
table and hence its hydraulic conductivity (see Chapter 3). Computation of actual evaporation
from bare soil is complicated by variations in external evaporative conditions at the soil sur-
face?6· 35 · 87 For isothennal conditions, upward movement is essentially all in the liquid phase,
but a soil may have a high surface temperature, causing it to dry out, establishing upward vapor
·es of movement in response to a vapor pressure gradient.69
:vapo- Field measurements of groundwater evaporation from tanks filled with soil (lysimeters)
have been made. Water tables were maintained at prescribed depths below ground surface.
Results, expressed as a percentage of pan evaporation at ground surface, are shown in Figure
6.3.1. For water tables within one meter of ground surface, evaporation is largely controlled by
atmospheric conditions, but below this soil prop.erties become limiting and the rate decreases
markedly with depth. ~
I
I 6.3.2 Transpiration Effects
( Where the root zone of vegetation reaches the saturated stratum, the uptake of water by roots
I equals (for practical purposes) the transpiration rate. Figure 6.3.2 shows water level variations
measured in a well in a thicket of willows. Rapid foliage growth during August (Figure 6.3.2a)
I caused daily fluctuations averaging about 10 em with the water table between 1.5 and 1.8 m
~ below ground surface. Heavy frosts occurred in early October and most leaves had fallen
1ting
:::urves
pene-
tiloga-
mcave
raight-
irectly
stream
be lost
reces-
.littent, Figure 6.3.1. Groundwater evaporation, expressed as a
:n (see 0 40 60 percentage of pan evaporation, as a function of depth
Percent of pan evaporation to water table (after White95).
by mid-October; thereafter, diurnal fluctuations were negligible (Figure 6.3.2b), with the veg-
etation dormant.
Magnitudes of transpiration fluctuations depend on the type of vegetation, season, and
weather. Hot, windy days produce maximum drawdowns, whereas cool, cloudy days show
only small variations. Fluctuations begin with the appearance of foliage and cease after killing
frosts. Cutting of plants eliminates or materially reduces amplitudes. Transpiration discharge
does not occur in nonvegetated areas, such as plowed fields, or in areas where the water table
is far below ground surface. After rain on high water table vegetated land, the water table
rises sharply as the increased soil moisture meets the transpiration demand and reduces the
groundwater discharge; but on cleared land or when vegetation is dormant, little or no rise is
evident.
The effect of transpiration on the interaction of groundwater with prairie pothole wetlands
is illustrated in Figure 6.3.3. The wetland, near Jamestown, North Dakota, receives ground-
water discharge most of the time; however, transpiration of groundwater by plants around the
perimeter of the wetland can cause water to seep from the wetland, creating cones of depres-
sion. Seepage from wetlands commonly is assumed to be groundwater recharge, but in this
case the water is actually lost to transpiration. This process results in depressions of the water
table. The transpiration-induced depressions in the water table commonly are filled by
recharge during the following spring and then form again by late summer almost every year. 96
Figu
6.3.3 Evapotranspiration Effects prairi
From a practical standpoint it is often difficult to segregate evaporation and transpiration Thei
losses from groundwater; therefore, the combined loss, referred to as evapotranspiration (Wetl
23 26
Days of August
(a)
Figure 6.3.2. Effect of

transpiration discharge
on groundwater levels
near Milford, Utah. (a)
In summer (b) After frost
(after White95)
6.3 fluctuations Due to Evapotranspiration 293
~ veg-
1, and
SEMIP£1tMANUn WETlAND
show
illing se.ASONALWftLANO
barge - 1~....,. W"TERTMI.Et:ON'TOUR· ~
·table b-. r"' IJ..IIrlhltl .1,640,r-t
~-IIYM O~ltMrvll
table Sfeec
es the ~~~~AT£R
rise is ~ DlVtbE
tlands
ound-
nd the
epres-
.n this
water
ed by
tear.96
Figure 6.3.3. Transpiration directly from groundwater causes cones of depression to form by late summer around the perimeter of
prairie pothole Wetland Pl in the Cottonwood Lake area in North Dakota (modified from T. C. Wmter and D. 0. Rosenberry, 1995).
ration The interaction of ground water with prairie pothole wetlands in the Cottonwood Lake area, east-central North Dakota, 1979-1990:
(Wetlands, v. 15, no. 3, pp. 193-211, as presented in Winter et al.96)
ration
(or consumptive use) is typically the quantity normally measured or calculated. The variation
of evapotranspiration with water table depth is sketched in Figure 6.3.4 for three groundcover
conditions. It is apparent that the deeper the roots, the greater the depth at which water losses
occur. Even with relatively deep water tables, evapotranspiration does not necessarily become
zero because upward transport can still occur, albeit minimally, in the vapor phase.
The pattern of diurnal fluctuation resulting from discharge of groundwater is nearly iden-
tical for evaporation and transpiration. The maximum water table level occurs in midmorning
(see Figure 6.3.5) and represents a temporary equilibrium between discharge and recharge
from surrounding groundwater. From midmorning until early evening, losses exceed recharge
and the level falls. The steep slope near midday indicates maximum discharge associated with
highest temperatures. The evening minimum again represents an equilibrium point, while the
rise during the night hours is recharge in excess of discharge.
White95suggested a method for computing the total quantity of groundwater withdrawn by
evapotranspiration during a day. If it is assumed that evapotranspiration is negligible from mid-
night to 4 A.M. and, further, that the water table level during this interval approximates the
mean for the day, then the hourly recharge from midnight to 4 A.M. may be taken as the aver-
age rate for the day. If we let h equal the hourly rate of rise of the water table from midnight
to 4 A.M., as shown by the upper curve in Figure 6.3.5, and s the net fall or rise of the water
:ct of table during the 24-hour period, then as a good approximation the diurnal volume of ground-
arge water discharge per unit area becomes
vels
(6.3.1)
t. (a)
er frost where SY is the specific yield near the water table. Actually, as pointed out by Troxell,82 the rate
of groundwater recharge to the vegetated area varies inversely with the water table level. Thus,
294 Chapter 6 Groundwater Levels and Envirorunental Influences
Evapotranspiration - 6.4 F
6.4.1
Deep-rooted
vegetation
Figure 6.3.4. Generalized variation of evapotranspi-

ration from groundwater with water table depth for
three groundcover conditions (after Bouwe~).
I I
I I
l
I I /
I I /
I 1/
I ).I
I /I
I / I
I/ I
v I 24 h
/1 I
/I I
// I I
Water table level
l
11 I
I
I
I
fl
I
Area equals daily
volume of water
transpired
il Transpi ration Slope of water
I rate ~
table level curve
Midnight
I
Noon
i I
Midnight
Figure 6.3.5. Interrelations of water table level,
recharge, and evapotranspiration fluctuations
6AM 6 PM (after Troxell82).
the difference between the recharge rate and the slope of the groundwater level curve gives the 1:
evapotranspiration rate. The lower portion of Figure 6.3.5 illustrates this; the area between the
two curves is a measure of the daily volume of water released to the atmosphere.
.~
6.4 FLUCTUATIONS DUE TO METEOROLOGICAL PHENOMENA

6.4.1 Atmospheric Pressure
Changes in atmospheric pressure (barometric tides) produce sizable fluctuations in wells
penetrating confined aquifers. 10· 28 The relationship is inverse; that is, increases in atmospheric
pressure produce decreases in water levels, and conversely. When atmospheric pressure
changes are expressed in terms of a column of water, the ratio of water level change to pres-
sure change expresses the barometric efficiency of an aquifer. Thus,
B = yt.h (6.4.1)
t'lpa
where B is barometric efficiency, y is the specific weight of water, l.lh is the change in piezo-
metric level, and f¥Ja is the change in atmospheric pressure. Most observations yield values in
the range of 20 to 70 percent.
The effect is apparent in data shown in Figure 6.4.1. The upper curve indicates observed
ranspi- water levels in a well in Iowa City, Iowa, penetrating a confined aquifer. The lower curve
th for shows atmospheric pressure inverted, expressed in meters of water, and multiplied by 0.75. A
close correspondence of major fluctuations exists in the two curves; the equality of amplitudes
indicates that the barometric efficiency of the aquifer is about 75 percent.
The phenomenon can be explained by recognizing that aquifers are elastic bodies.70· 71 If
!J.pa is the change in atmospheric pressure and !J.p,., is the resulting change in hydrostatic pres-
sure at the top of a confined aquifer, then
(6.4.2)
where !J.s, is the increased compressive stress on the aquifer (Figure 6.4.2). At a well pene-
trating the confined aquifer, the relation
(6.4.3)
~ 1.8
Q5
s
E
:l
co
"0
Q)
>
.8a:l
ai 1.6
>
.!!!!
..._
Q)
le level, ~ 1.5
ai
.tions 3:
Figure 6.4.1. Response of water
level in a wel.l penetrating a con-
fined aquifer to atmospheric pres-
gives the 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 2 3 4 5 6 7 8 9 10 sure changes, showing a baro-
tween the March 1939 April 1939 metric efficiency of 75 percent
Days of the month (after Robinson71 ).
Ground~
surface '..
-~--
Confining layer
\_ Upper boundary -~-- -----

of aquifer t Piezometric /
Sand grains
I surface
Aquifer
Figure 6.4.2. Idealized distribution of forces

_,--- Aquifer ~
at the upper boundary of a confined aquifer
resulting from a change in atmospheric
pressure.
(a) (b)
Figure 6.4.3. Effect of an increase in atmospheric pressure on the water

level of a well penetrating a confined aquifer. 6.4.:
exists as shown in Figure 6.4.3a, where y is the specific weight of water. Let the atmospheric
pressure increase by !1pa, then
Pw + flrJw =Pa + !1pa + yh' (6.4.4)
as shown in Figure 6.4.3b. Substituting for Pw from Equation 6.4.3 yields
flrJw =!1pa + y(h'- h) (6.4.5)
But from Equation 6.4.2 it is apparent that !1pw < !1pa, indicating that h' <h. Generally, there-
fore, the water level in a well falls with an increase in atmospheric pressure. It follows that the
converse is also true. •
Jacob48 developed expressions relating barometric efficiency of a confined aquifer to
aquifer and water properties, including the storage coefficient. Gilliland28 showed that changes
in soil moisture from infiltrating precipitation can affect the magnitude of barometric efficiency.
For an unconfined aquifer, atmospheric pressure changes are transmitted directly to the
water table, both in the aquifer and in a well; hence, no pressure difference occurs. Air
entrapped in pores below the water table is affected by pressure changes, however, causing
fluctuations similar to but smaller than that observed in confined aquifers.59• 63 Temperature
fluctuations in the capillary zone will also induce water table fluctuations where entrapped air
is present.78• 84
Atmospheric pressure fluctuations do affect water tables substantially on small, penneable
oceanic islands. The response of sea-level changes to atmospheric pressure is essentially
E
~
isostatic; that is, sea level adjusts to a constant mass of the ocean-atmosphere column. This § 10
causes the ocean to act as an inverted barometer with sea level rising about 1 em to compen- ~
·o.. B
"5
sate for a drop in atmospheric pressure of 1 mb. These fluctuations amount to about 20 em in ~
the open ocean and are transmitted as long-term tides to the water table (see the subsequent a. 6
Ql
section on ocean tides). Data from Bermuda by Vacher86 shown in Figure 6.4.4 illustrate the 01
ro
~ 4
fluctuations. Ill
~ 2
<ll
Ql
~ 0
•Atmospheric pressure waves created by nuclear explosions in the Soviet Union have caused fluctuations of the
piezometric surface in limestone aquifers in England.38 One fluctuation displayed an amplitude of 0.46 em in Figur
response to a pressure wave of 900-1,000 microbars. Rasm1
-~ ~~
~~ 10 em {
E._
_a.
<(
~ §1
«l.-
~~ 10 em{
Figure 6.4.4. Variations in atmospheric
al- pressure and water table elevation dur-
:c;:a>
ing October 1973 at Devonshire Post
Office, Bennuda. Note that the varia-
tion is inverse with a rise in pressure
10 20 31 associated with a decline in water level
Days of October (after Vache~6).
water
6.4.2 Rainfall
As described previously, rainfall is not an accurate indicator of groundwater recharge because
of sudace and subsurface losses as well as travel time for vertical percolation. The travel time
·spheric
may vary from a few minutes for shallow water tables in permeable formations to several
months or years for deep water tables underlying sediments with low vertical permeabilities.
(6.4.4) Furthermore, in arid and semiarid regions, recharge from rainfall may be essentially zero.
Shallow water tables show definite responses to rainfall, as Figure 6.4.5 indicates. Water lev-
els shown are the average for 25 observation wells; greatest fluctuations occurred in the upper
(6.4.5) portions of the basin and smallest near the basin outlet.
{,there-
that the
uifer to
15
changes
}ciency.
E"
;g
y to the CD
~
h • ',, 14.5
urs. Air ~ : \,•" I I .s:
tl I ~ 1:
causing ~ :~ : ; 0
perature \',
' 1\
/~
~ ~:1
I f
' I \
I \ ~Q>
pped air lo,;o,""'\• I I '- II I I 1I \ .-_ I \ 14 ai
~
~ '~ J.
·~·
: ~' "'\
I \ I \ 1 0 I 0 1 • ' I \ CD
.. '
', '
, II.'~ r, ~ ·~ : \ \\./r ; :0
\ ~'-. .~ \ .. ~I \ \ 1\ : \ !!!
....
rmeable 0 :~ \ \ •\<I, I
.PJ
\.1\ l1J
.:
\ /'., I \ I I
;entially ~ f ' \ \\ : ~
on. This \'.\' '",'.. ~

:
13.5
CD
Ol
~
:ompen- \ ~' . \~ ', Q)
>
\~~ ! ', ' <(
Wcmin
>sequent ' ••• ' . 13
trate the
:>fthe
tm Figure 6.4.5. Variation in average water table level and weekly average precipitation, Beaverdam Creek Basin, Maryland (after
Rasmussen and Andreasen68).
y- Ground surface
------\---------
Saturated zone resulting
from infiltration rainfall
H
::_. .c
~- Zoneof - a.
Q)
compressed air 0
Water table ~
Figure 6.4.6. Water table rise in an
observation well resulting from infil-
trating rainfall sealing the ground sur-
face and compressing air above the
water table.
Groundwater levels may show seasonal variations due to rainfall, but often these include
natural discharge and pumping effects as well. Droughts extending over a period of several
years contribute to declining water levels.
Where the unsaturated zone above a water table has a moisture content less than that of
specific retention (see Chapter 2), the water table will not respond to recharge from rainfall
until this deficiency has been satisfied. Thereafter, the rise !lh will amount to
(6.4.6)
where Pi is that portion of precipitation that percolates to the water table and Sy is specific yield.
An interesting phenomenon occasionally noted in observation wells is a nearly instanta-
neous response of shallow water tables to rainfall. This may be explained by the pressure
increase of air trapped in the zone of aeration when rainfall seals surface pores and infiltrating
water compresses the underlying air. If the zone containing interconnected air-filled pores (H 6.5 FI
in Figure 6.4.6) is compressed to a thickness H- m, then the pressure above the water table is 6.5.1 (
increased by mi(H- m) of an atmosphere, causing the water level in an observation well to rise
m
t:.h=--(lO)m (6.4.7)
H-m
For shaJlow water tables the rise (which occurs only in the well) can be an order of magnitude
larger than the depth of infiltrating rainfall; however, escaping air soon dissipates the effect
Similarly, when water is applied uniformly to the top of a dry column of sand in the laboratory,
the air is compressed until released by a spontaneous upward eruption.
6.4.3 Wmd
Minor fluctuations of water levels are caused by wind blowing over the tops of wells. The effect
is identical to the action of a vacuum pump. As a gust of wind blows across the top of a casing,
the air pressure within the well is suddenly lowered and, as a consequence, the water level
quickly rises. After the gust passes, the air pressure in the well rises and the water level falls.
6.4.4 Frost
In regions of heavy frost, it has been observed that shallow water tables decline gradually dur-
ing the winter and rise sharply in early spring before recharge from the ground surface could
Unsaturated
,nan
. infti-
nd sur- Zone of saturation
: the
Figure 6.4.7. Sketch illustrating the variation in
January February March April depth to water table in response to winter frost
Months conditions (after Schneider74).
nclude
several
occur (see Figure 6.4.7).74 This fluctuation can be attributed to the presence of a frost layer
that of above the water table. During winter, water moves upward from the water table by capillary
rainfall movement and by vapor transfer to the frost layer, where it freezes. Vapor migration occurs in
response to the thermal gradient and to the fact that vapor pressure over ice is less than that
(6.4.6) over liquid water at ooc. In early spring, approximately when the mean air temperature reaches
0°C, the frost layer begins thawing from the bottom; consequently, meltwater percolates down-
.c yield. ward to rejoin the water table.
1Stanta-
>ressure
iltrating
•ores (H
6.5 FLUCTUATIONS DUE TO TIDES
·table is 6.5.1 Ocean Tides
!1 to rise
In coastal aquifers in contact with the ocean, sinusoidal fluctuations of groundwater levels
(6.4.7) occur in response to tides. If the sea level varies with a simple harmonic motion, a train of sinu-
soidal waves is propagated inland from the submarine outcrop of the aquifer. With distance,
inland amplitudes of the waves decrease and the time lag of a given maximum increases. The
1gnitude problem has been solved by analogy to heat conduction in a semi-infinite solid subject to peri-
.e effect.
odic temperature variations normal to the infinite dimension. 31. 93
)Oratory, For simplicity, consider the one-directional flow in a confined aquifer as shown in Figure
6.5.la. From Equation 3.9.11, the applicable differential equation governing the flow is
d2h s dh (6.5.1)
dx 2 =Tat
'he effect
a casing, where h is the net rise or fall of the piezometric surface with reference to the mean level, x is
1ter level the distance inland from the outcrop, S is the storage coefficient of the aquifer, T is transmis-
el falls. sivity, and tis time. If we let the amplitude, or half-range, of the tide be h0 (see Figure 6.5. I a),
the applicable boundary conditions include h =h0 sin rot at x =0 and h =0 at x =oo. The angu-
lar velocity is ro; for a tidal period t0,
2n
1ally dur- (1)=- (6.5.2)
ace could fo
300 Chapter 6 Groundwater Levels and Envirorunental Influences
Ground surface
Tidal range 2h0
=-t _____ . .
Range fluctuation 2h
=========
Piezometric surface r------ _________ _
x - --
'
H
~ Mean sea level
Confined aquifer Ocean
(a)
Ground surface Tidal range 2h0
Unconfined aquifer Ocean
(b)
Figure 6.5.1. Groundwater level fluctua-

tions produced by ocean tides. (a)
(c) Confined aquifer (b) Unconfined aquifer
(c) Loading of a confined aquifer
The solution of Equation 6.5.1 with these boundary conditions is
h -- l._
''Oe -x~nSitoT sm
. [21tt
- - x-y1
-C' I t 0
II•J
r) (6.5.3)
to
From this it follows that amplitude hx of groundwater fluctuations at a distance x from the
shore equals
(6.5.4)
The time lag tL of a given maximum or minimum after it occurs in the ocean can be obtained
by solving the quantity within the parentheses of Equation 6.5.3 fort, so that
(6.5.5)
The waves travel with a velocity
(6.5.6)
and the wavelength is given by
(6.5.7)
Substituting the wavelength for x in ~uation 6.5.4 shows that the amplitude decreases by a fac-
tor e-211, or 11535, for each wavelength. Water flows into the aquifer during half of each cycle and
out during the other half. By Darcy's law, the quantity of flow V per half-cycle per foot of coast is
3r 0 /8 I3t /8(dh)
0
V=
J-r0 /8
q dt == T
-r0 /8
-
dx x=O
dt (6.5.8)
where q is the flow per foot of coast. Differentiating Equation 6.5.3 to obtain dhl'dx and inte-
grating yield
(6.5.9)
The above analysis is also applicable as a good approximation to water table fluctuations
of an unconfined aquifer if the range of fluctuation is small in comparison to the saturated
thickness (Figure 6.5.lb). Figure 6.5.2 shows fluctuations in wells penetrating an unconfined
aquifer at various distances from a surface water level varying approximately with a sinusoidal
pattern.
Just as atmospheric pressure changes produce variations of piezometric levels, so do tidal
fluctuations vary the load on confined aquifers extending under the ocean floor (Figure 6. 5.1 c).
Contrary to the atmospheric pressure effect, tidal fluctuations are direct; that is, as the sea level
increases, the groundwater level does also. Figure 6.5.3 illustrates the effect for a well only 30
m from shore. The ratio of piezometric level amplitude to tidal amplitude is known as the tidal
fluctua- Elbe River
aquifer
20.0
(6.5.3)
from the
(6.5.4)
obtained
(6.5.5)
Well at4600 m
17.o l=±=t=t:::r::::r:::::I:::::D:=r:::J Figure 6.5.2. Fluctuations of the Elbe River and
2 3 4 5 6 7 8 9 10 water table levels in wells at various distances from
(6.5.6) Days of November, 1910 the river (after Werner and Noren93).
'-E
!!!~
1.5
ro E
~ ~
o-
_ro
s::;"O 2.0
-~
g-.Q
O<ll
.0
2.5
.!:!
~ E' -21.o
E:::-
o (J)
N
(j) Q)
>
a::- -21.5 Figure 6.5.3. Tidal fluctuations and induced
piezometric surface fluctuations observed in a
6 7 8 9 10 11 well 30 m from shore at Mattawoman Creek,
Days ol July, 1938 Maryland (after Meinzer56).
efficiency of the aquifer. Jacob~ showed that tidal efficiency Cis related to barometric effi-
ciency B by
C=l-B (6.5.10)
Thus, tidal efficiency is a measure of the incompetence of overlying confining beds to resist
pressure changes. Aquifer response to loading rather than head change at the outcrop requires
that the amplitude given by Equation 6.5.3 be multiplied by C. 6.6 u
6.5.2 Earth Tides

Regular semidiurnal fluctuations of small magnitude have been observed in piezometric sur-
faces of confined aquifers located at great distances from the ocean. 70 After correcting well
levels for atmospheric pressure changes, these fluctuations appear quite distinctly in certain
wells where the phenomenon has been investigated. Figure 6.5.4 shows fluctuations over a
lunar cycle from a 250-m well tapping a confined aquifer in Iowa City.
Time of the moon's

transit at upper
culmination
March 1939
Days of the month
Figure 6.5.4. Water level fluctuations in a confined aquifer produced by earth tides (after Robinson 71 ).
These fluctuations result from earth tides, produced by the attraction exerted on the earth's
crust by the moon and, to a lesser extent, the sun. Robinson's observations, 71 based on analy-
ses of well records, make convincing evidence: (1) two daily cycles of fluctuations occur
about 50 min later each day, as does the moon; (2) the average daily retardation of cycles agree
closely with that of the moon's transit; (3) the daily troughs of the water level coincide with
the transits of the moon at upper and lower culmination; and (4) periods of large regular fluc-
tuations coincide with periods of new and full moon, whereas periods of small irregular fluc-
tuations coincide with periods of first and third quarters of the moon. All of these facts may be
ICed noted in the data of Figure 6.5.4. Bredehoeft6 has pointed out that wells serve as sensitive indi-
i in a cators of this dilatation of the earth's crust.
·eek, At times of new and full moon, the tide-producing forces of the moon and sun act in the
same direction; then ocean tides display a greater than average range. But when the moon is
in the first or third quarter, tide-producing forces of the sun and moon act perpendicular to each
tric effi- other, causing ocean tides of smaller than average range. The coincidence of the time of low
water with that of the moon's transit can be explained by reasoning that at this time tidal attrac-
(6.5.10) tion is maximum; therefore, the overburden load on the aquifer is reduced, allowing the aquifer
to expand slightly.
to resist
requires
6.6 URBANIZATION
The process of urbanization often causes changes in groundwater levels as a result of
decreased recharge and increased withdrawal. In rural areas, water supplies are usually
obtained from shallow wells, while most of the domestic wastewater is returned to the ground
~tric sur-
through cesspools or septic tanks. Thus, a quantitative balance in the hydrologic system
ling well
remains. As population increases, many individual wells are abandoned in favor of deeper pub-
n certain
lic wells. Later, with the introduction of sewer systems, storm water and wastewater typically
lS over a
discharge to a nearby surface water body (see Figure 6.6.1). Thus, three conditions disrupt the
subsurface hydrologic balance and produce declines in groundwater levels:
1. Reduced groundwater recharge due to paved surface areas and storm sewers
2. Increased groundwater discharge by pumping wells
3. Decreased groundwater recharge due to export of wastewater collected by sanitary sewers
Effects of the urbanization trend are well illustrated on Long Island, New York, in Figure
6.6.2. 12· 20 Here the above conditions have all been present, leading not only to a decline in
water tables but also to groundwater pollution (see Chapter 8), seawater intrusion (see Chap-
ter 14), and reduced streamflow. Artificial recharge efforts (see Chapter 13) are underway to
counteract these undesirable results of urbanization.
Precipitation
! ! ! ! !
Figure 6.6.1. Schematic
diagram illustrating how
urbanization can cause
....____.,_ Recharge
Original lowering of water table
Shallow well
water table elevation. (a) Rural
(a) situation
304 Chapter 6 Groundwater Levels and Environmental lnfluences
Precipitation
1 1 1
Water supply Figure 6.6.1 (continued).

Stormwater sewer
Sanitary sewer Schematic diagram illus-
trating how urbanization
can cause lowering of
Lowered water table
Well water table elevation. (b)
(b) Urban development
Principal
aquifers
tO 20 30MILES
I I I
I I I
t 0 'lO 30 K'll()ME'ffiRS
COUI-m' Sawered area
n
70
61
Figure 6.6.2. Water-level

li2 record for a well complet-
ed in the upper glacial
50 aquifer in west-central
Nassau County, Long
Island, New York {as pre-
1945 1950 1955 1880 1i65 1971) 1975 1980 1i65 IWO 1996
sented in Alley et al. 1). Figure 6.6.3.
Figure 6.6.3 illustrates the effects of urbanization on Las Vegas, Nevada. Figure 6.6.4
illustrates the differences in water budget components for predevelopment times and the pres-
ent. In the present water budget, even though the usage is greatly increased, only a small frac-
tion of the water in the Las Vegas Valley is consumed. Most of the water is either returned to the
aquifer system or evaporated or discharged into the Colorado River system (Galloway et al.23).
inued).
1 illus-
zation
:of
)0 . (b)
lt w
Faults (arrow indicates

direction of relative movement)
Las Vegas Wash

E
Principal
aquifers
Predevelopment
Near surface reservoir (from urban runoff)

Some near-surface water
recharges the principal aquifers
ater-level
complet-
0 SMiles Present
lacial
:entral 0 5 Kilometers
Long Vertical exaggeration ISx
'(as pre-
t al. 1). Figure 6.6.3. Effect of urbanization on Las Vegas, Nevada (Galloway et al.23).
Predevelopment
,- <;• '
Atmosphere
..'• . ~
·Natural
recharge. '
Eart
1000
Thousands of
a:cre.leet
Subsurface
~------------£-~~--------~· o~ow
~--------Pr_in_c_ip_a_la_q_u_ire_r_s______~r~
Present
·1 .,,
Evaporation
90
.c 80
..···
§ 70
2 60
(J)
~50
~
((I
40
5- 30
..c
@ 20
Figure 6.6.4. Comparison of water budgets for prede-
w 10
velopment period and the present for Las Vegas, 0
Principal aquifers
Nevada (Galloway et al?\
35
6.7 EARTHQUAKES 30
Observations reveal that earthquakes have a variety of effects on groundwater. 90• 91
Most
"0
(J)
25
(J)~
spectacular are sudden rises or falls of water levels in wells, changes in discharge of springs, ·c-s
·;:: § 20
appearance of new springs, and eruptions of water and mud out of the ground. More commonly, 2E
however, earthquake shocks produce small fluctuations (hydroseisms) in wells penetrating con- ~~ 15
mE
en~
fined aquifers. A good example is furnished by the water level record on an expanded time scale ~ 10
shown in Figure 6.7.1. This earthquake was centered on the Argentina-Chile border, nearly 5
8,000 km from the recording well in Milwaukee. Although little is known of the quantitative
0
effects of earthquakes on groundwater, these fluctuations result from compression and expan-
sion (dilatation) of elastic confined aquifers by the passage of earthquake (Rayleigh) waves. 14
6.7 Earthquakes 307
Rise
Figure 6.7.1. Water level fluctua-

tions in a well at Milwaukee,
Wisconsin, resulting from an
Earthquake on December 9, 1950;
epicenter at
Time earthquake centered on the
24°S lat. 67°W long. 11 1 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 Argentina-Chile border (after
Fall 1 - - - - - 20 min. Vorhis91)
These waves travel at speeds of approximately 200 krn/min so that fluctuations appear after lit-
tle more than one hour even from the most distant earthquake centers.
Looking at the converse situation, field studies have revealed that injection of wastewater
into a deep well can trigger earthquakes.33 Evidence stems from injection of chemical-
manufacturing waste fluids near Denver, Colorado, into a well 3,671 m deep, penetrating sed-
imentary rocks into Precambrian crystalline rocks. Figure 6.7.2 shows the time variation of
fluid injection and earthquake frequency. Most earthquake magnitudes were small-within the
range of 1.5 to 4.4 on the Richter scale. It is believed that the mechanism by which fluid injec-
tion triggered the earthquakes stems from a reduction of frictional resistance to faulting, a
reduction that occurs with increase in pore pressure. Knowledge ofthis phenomenon has stim-
ulated research into the possibility of injecting water into potentially dangerous fault zones.
90
.c. 80
§ 70
~ 60
(I)
a. 50
(J)
~ 40
ro
5- 30
€ro 20
L1J 10
JL
I
or prede- ..,.,i
0
.as, 1962 1963 1964
(a)
35
). 91Most
,r springs,
ommonly,
Figure 6.7.2. (a) Earthquake frequency
-ating con- (b) Volume of injected wastewater. Near
time scale Denver, Colorado, during the period
ier, nearly 1962-1965 (after Healy et al}3
uaotitative copyright© 1968 by American
0
mdexpao- 1962 1963 1965 Association for the Advancement of
14 (b)
1) waves. ... Science).
This might trigger minor earthquakes, ease stresses along a fault, and hence prevent the sudden
release of accumulated energy that results in disastrous earthquakes.
The Hegben Lake, Montana, earthquake of August I959 caused fluctuations in wells
throughout the United States and even in Hawaii and Puerto Rico. 15 The main shock was
recorded by water level fluctuations that ranged to more than 3.0 m; values exceeded 0.3 m in
nine states. Even more dramatic were the hydroseismic responses to the Anchorage, Alaska,
earthquake of March 27, 1964. This large·magnitude earthquake (8.4 to 8.8 on the Richter
II
scale) affected groundwater levels throughout the United States, with the largest fluctuation
exceeding 7.0 m at a well in South Dakota. 92 Hydroseisms were also recorded at such distant
locations as Denmark, Egypt, South Africa, the Philippines, and Australia.
6.8 EXTERNAL LOADS

The elastic property of confined aquifers results in changes in hydrostatic pressure when
changes in loading occur. Some of the best examples are exhibited by wells located near rail·
jj
roads where passing trains produce measurable fluctuations of the piezometric surface. Figure
6.8.1 illustrates changes in water level produced by a train stopping and starting near a well at Load
·~
Smithtown, New York.
The application of a load compresses the aquifer and increases the hydrostatic pressure.
Thereafter the pressure decreases and approaches its original value asymptotically as water
flows radially away from the point where the load is applied. Thus, initially the load is shared
by the confined water and the solid material of the aquifer; however, as the water flows radi·
ally outward, an increasing proportion of the load is borne by the structure of the aquifer. The
schematic diagram after Jacob47 in Figure 6.8.2 shows this effect. Here a point load is instan-
taneously applied. The lower surface of the aquifer is assumed fixed; lengths of arrows indi·
cate the relative magnitudes of flow velocities at various distances from the load. During the
interval from A to B, the hydrostatic pressure decreases and the deflection of the upper surface
of the aquifer increases. Subsequently, when the load is removed, the pressure drops to a min·
imum and then recovers toward its initial value as shown by times C and D.
6.9 LAND SUBSIDENCE AND GROUNDWATER

Land subsidence is a gradual settling or sudden sinking of the earth's surface caused by the
subsurface movement of earth materials. Land subsidence is a global problem; in the United
States more than 17,000 square miles in 45 states have been directly affected by subsidence.
Principal causes are aquifeHystem compaction, mining of groundwater, drainage of organic
soils, underground mining, hydrocompaction, natural compaction, sinkholes, and thawing per-
-------
udden ~ Load applied
wells
:k was
/ Load removed
3 min
~laska,
_......__1
tichter
:uation
distant A 8 0
I
~when Time-
:ar rail-
Figure
well at
L d !A {_ Ground
oa surface +Ia
-~-'--.._...,
c D
:essure.
L<:essu~
s water
(
:shared
VS radi-
Figure 6.8.2. Hydrostatic pressure variations
fer. The and aquifer deflections resulting from a point
instan- load applied and later removed from the
NS indi- [" ground surface above a confined aquifer
ting the (after Jacob47).
surface
) a min-
mafrost. More than 80 percent of the identified subsidence in the United States is a conse-
quence of our exploitation of groundwater. Figure 6.9.1 shows the locations in the United
States where subsidence has been attributed to groundwater pumpage. The continuing and
increasing development of land and water resources threatens to worsen existing land-
d by the subsidence problems and initiate new problems. Galloway et al.23 reported on the land subsi-
! United dence due to mining groundwater in several areas of the United States, including Santa Clara
sidence. Valley, California; San Joaquin Valley, California; Houston-Galveston, Texas; Las Vegas,
organic Nevada; and south-central Arizona.
ting per- The relation between changes in groundwater levels and compression of the aquifer sys-
tem is based upon the principle of effective stress. When groundwater levels are lowered, the
support provided by the pore-fluid pressure is transferred to the skeleton of the aquifer system,
which compresses as shown in Figure 6.9.2. When the pore-fluid pressure is increased, such
as when groundwater recharges the aquifer system, support previously provided by the skele-
ton is transferred to the fluid and the skeleton expands. The skeleton alternately undergoes
compression and expansion as the pore-fluid pressure fluctuates with aquifer-system discharge
and recharge. This fully recoverable deformation occurs in all aquifer systems, commonly
resulting in seasonal, reversible displacements in land surface of up to one inch or more in
response to seasonal changes in pumpage23 .
Preconsolidation stress is the maximum level of past stressing of a skeletal element. When
ns in a the load on an aquitard skeleton exceeds the preconsolidation stress, the aquitard skeleton can
tstopping undergo irreversible compaction, resulting in a permanent reduction of pore volume as the pore
1 (after fluid is forced out of the aquitards into the aquifers. This process is illustrated in Figure 6.9.3.
In confined aquifer systems subject to large-scale overdraft, the volume of water resulting from
New Jersey
Atlantic City-Oceanside area
Barnegat Bay-New York Bay
coastal area \
Nevada
Las Vegas Valley'\
California - -<L-
Antelope Valley
Coachella Valley
Elsinore Valley 6.9.1
La Verne area
Lucerne Valley
Mojave River Basin
Oxnard Plain
Pomona Basin
Sacramento Valley
Salinas Valley VIrginia
Sanbenito Valley Franklin-SuHolk area
San bernadino area Williamsburg-West Point
San Gabriel Valley
San Jacinto Basin area
San Joaquin Valley South Central
San Luis Obispo area Arizona
Santa Clara Valley Avra Valley
Temecula East Salt River Valley
Valley Eloy Basin
Wolf Valley Gila Bend area Louisiana
Harquahala Plain Baton Rouges area
San Simon Valley New Orleans area
Stanfield Basin
Tucson Basin
West Salt River Valley Houston-Galveston
Willcox Basin Hueco Bolson-EI Paso, Juarez
When I
pumpir
• Major alluvial aquifer systems
ground
in the coterminous United States
and rai
on the .
23
Figure 6.9.1. Areas in the United States where subsidence has been attributed to groundwater pumpage (Galloway et al. ). beyonc
precon:
threshc
aquitat•
When water levels drop. due mainly and the
to seasonal increases in groundwater When groundwater is recharged
surface
pumping, some support for the over- and water levels rise, some sup-
perman
lying material shifts from the piessur- port for the overlying material
ized fluid filling the pores to.the gran- shifts from the granular skeleton
ular skeleton of the aquifer system. to the pressurized pore fluid.
Confined
aqlifer
Decreased fluid pressure Increased fluid pressure

causes the skeleton to expands the skeleton,
contract. creating some creating some small uplift
small subsidence of land of land surface.
5\rlace.
23 Figure ·
Figure 6.9.2. Relation between changes in groundwater level and compression of aquifer system (Galloway et al ).
irreversible aquitard compaction is essentially equal to the volume of subsidence. This volume
typically can range from 10 to 30 percent of the total volume of water pumped, representing a
one-time mining of the stored groundwater and resulting in a small pennanent reduction in
storage capacitl3. The concepts of aquitard drainage and aquifer-system compaction are fur-
ther illustrated in Figure 6. 9.4.
Figure 6.9.5 illustrates the predeveloped conditions and postdeveloped conditions (with
land subsidence) of the San Joaquin Valley in California. At least four distinct phenomena have
been identified. 2
e
1rea
6.9.1 Lowering of Piezometric Surface
ea
Land subsidence has been observed to accompany extensive lowering of the piezometric sur-
face in regions of heavy pumping from confined aquifers. Figure 6.9.6 and 6.9.7, respectively,
illustrate the trend in subsidence in the Houston area due to groundwater pumpage and the
arrested subsidence by substituting imported water. Figure 6.9.8 shows land subsidence in
rea
;t Point south-central Arizona, illustrating another regional subsidence.
Recoverable land subsidence caused by

When long-term
Land surface :::::::::::::::::-:+:t:: J reversible elastic deformation
pumping lowers Permanent land subsidence caused by
groundwater levels
and raises stresses
on the aquitards
t-------..,.-...1...-rl J irreversible inelastic deformation
beyond the
preconsolidalion-stres:siJI!-···~jj~
thresholds, the
aquitards compact
and the land
surface subsides
permanently.
Compaction of the aquifer system
is concentrated in the aquitards.
IJifer-
on
Depth
to water
Time
Granular aquitard Rearranged, compac-
skeleton defining fluid- ted granular aquitard Long-tenn decline in water level
d pressure
filled pore spaces skeleton with reduced modulated by the seasonal cycles
skeleton, of groundwater pumpage
e small uplift storing groundwater porosity and ground-
:e. water storage capacity
Figure 6.9.3. Inelastic compaction irreversibly altering aquifer system (Galloway et al. 23).
This principle describes the relation between changes in water levels and deformation of the aquifer system.
Land surface Land surface
For any arbitrary plane below the
water table, the total stress repre-
Aquifer system sented by the weight ()fthe over-
TOTAL STRESS lying rock and water is balanced
(weight or load of the by the pore-fluid pressure and the Thick aquitard
overlying rocks intergranular or effective stress. (laterally Pred(
and water) extensive)
Grou1
· Thick aquitard them
- cente
·. (laterally
Confined· ·.. discontinuous) disch
aquifer throlJl
system
EFFECTJVE PORE-FLUID
STRESS STRESS
(portion of the total (porton of the total
stress th.rt is borne stress that is borne
by the 5okl gnios by tt-e water filling
in the skeleton of the pores of the
the rod< matrix) «- matrix )
Posldt
Figure 6.9.4a. Aquitacd drainage and aquifer-system compaction; the principle of effective stress (Galloway et al. 23). Groun
down¥
pumpl1
Prior to the extensive development of
groundwattr' resources, water levels are ·During development ofground· After ground-water pumping slows
r~latively stable-though s.ubject to sca- water resources, water levels decline- or decreases, water levels stabiliu but
·sonal and longer-term climatk variability. and land subsidence begins. land subsidmce may continue.
I •
Figure
Groundwater withdrawal from Under the principle of effective

~U5dl.'ft confined aquif~ reduces fluid stress. the compaction of a thick
pressures (p). As the total stress sequence of interbedded aqUfers
l.=.l=,.· (or) remans nearly constant a and aqUtards can proceed only
portion of the load is shifted from as rapidly as pore pressures
the confined fluid to the skeleton · throu~ the sequence can de·
of the aquifer system, increasng · i:ay toward equilbium with re-
the effective stress (CJe) and duced pressures in the po.rnped
.causing some convession. aquifers. Most of the land svbsi·
dence occurs as a reslit of the
pennanent compaction of the
water aquitardS. which may be delayed
dve to their slow drainage.
Figure 6.9.4b. Aquitacd drainage and aquifer-system compaction; prolonged changes in groundwater levels induce subsidence
(Galloway et a1. 23).
Predeve lopment
Groundwater flowed from
the mountains toward the
center of the valley where It
discharged into streams or
;)
Postdevelopment
Groundwater flow generally
downward and toward
pumping centers.
It
Figure 6.9.5. Comparison of pre- and postdevelopment concepts for the San Joaquin Valley in California (Galloway et al.23).
The explanation for this subsidence is based on fundamentals of soil mechanics. 16• 54 Con-
etrectNe . sider the pressure diagram for a confined aquifer overlain by an unconfined aquifer shown in
ofa thick J
ed aqUfers · Figure 6.9.9. Initially, the total (geostatic) pressure p1 at any depth (see Figure 6.9.9a) is
eed only
;ures Pr =Ph+ Pi (6.9.1)
Ke can de- ·
I with re- ' where Ph is the hydraulic pressure and Pi is the intergranular pressure. If pumping in the con-
pumped fined aquifer lowers the piezometric surface while the water table remains unchanged due to
ltld subsi·
1lt of the an impermeable clay layer separating the aquifers (Figure 6.9.9b) then Equation 6.9.1 becomes
r1 of the
>e delayed (6.9.2)
lag!!.
Note that ph' <ph andp/ > P; for both the confined aquifer and the clay layer. Adjustments
to these new pressure distributions will take place essentially instantaneously in the penneable,
~nee coarse-grained aquifer. But in the relatively impermeable, fine-grained clay, this adjustment
may take months to years. Because clayey materials are highly compressible, the increased
Subsidence trends reflect patterns of resource

development that shifted inland from coastal
oil and gas extraction to groundwater extrac-
tion for muicipal and industrial supplies.
~ ! \
Line ofequal~~~
subsidence ; .
Contin1
1974-1
Subside
(fee(
Figure 6.
Subsidem
water. A r
increasinB
6.9.2 HJ
Figure 6.9.6. Subsidence trends in
Houston, Texas area (Galloway et al. 23).
intergranular pressure (p/- p;) causes the clay layer to be compacted. This reduces its poros-
ity, while water contained in the clay pores is squeezed downward into the confined aquifer.
The volume of water displaced from the clay equals the reduction in clay volume and also
the volumetric land surface subsidence. Similarly, the reduction in thickness of the clay layer
equals the vertical land subsidence. The amount of compaction is a function of the thickness
and vertical permeability of the clay, of the time and magnitude of piezometric surface decline,
and of the microstructure of the clay.66 Because sand and gravel deposits are relatively incom-
pressible, the increased intergranular pressure has a negligible effect on the aquifer itself.
Change ln groundwater levels in wells in the

Evangeline aquifer, 19n-1997
Some arrested subsidence

1974-1997
Continuing subsidence
Subsidence 0.0~
1974-1997
_ C _
(feet)
0.8 --.--r--.----..,;---.--
1975 1985 1995
Subsidence
(feet) 1.2------>t-----
Extensometer sites
for measuring subsidence
® Addicks
® lake Houston
©Baytown
@Clear lake
1975 1985 1995 ®Texas City
Figure 6.9.7. Houston, Texas, area subsidence and arrested subsidence (Galloway et al. 23). The Harris-Galveston Coastal
Subsidence District has arrested subsidence along the western margins of Galveston Bay by substituting imported water for ground-
water. A new challenge is to manage groundwater use north and west of Houston where water levels are declining and subsidence is
increasing.
Land subsidence resulting from compaction of fine-grained sediments occurs at several

locations in the United States22 and throughout the world. 19· 25 The problem has been exten-
sively studied by Poland and others.46• 65•66•67* Compaction of clay is largely inelastic and per-
manent; consequently, the only effective control measure for this type of subsidence is
increasing piezometric levels by reducing pumping and by recharge of water through injection
wells (see Chapter 13). Withdrawal of oil and gas produces the same problem of land subsi-
dence; major areas affected exist in the United States, Italy, Japan, and Venezuela. 65
6.9.2 Hydrocompaction
Collapse of the ground surface has been observed to occur when water is applied to certain
types of soils. Particularly susceptible are (1) loose, moisture-deficient alluvial deposits,
including mud flows; and (2) moisture-deficient loess deposits. 55 · 77 An example of this type
of shallow subsidence is found on the arid west side of the San Joaquin Valley, California. Soils
s poros-
characteristically are desiccated with a high void content and low density (1.1 to 1.4 g/cm 3) .
tquifer.
Most of these soils have never been saturated since deposition, but when irrigation water, for
and also
lay layer
~ickness
'The periodic flooding of Piazza San Marco in Venice from the Grand Canal provides dramatic evidence that subsi-
decline, dence from groundwater pumpage threatens this beautiful city. A decline of 20 em over the last .SO years has been
vincom- reported. Recent restrictions 011 pumpage in the industrial suburb of Porto Marghera are expected to halt this subsi-
:self. dence rate 25
Luke
Air Focce
Base
o -·~~------~
r-o
~---
Figure 6.9.8. Land subsidence in south-central

Arizona (modified from Carpenter7, as present-
ed in Alley et al. 1) . Groundwater development
for agriculture in the basin-fill aquifers of I
Q)
south-central Arizona began in the late 1800s, 0
cQ)
and by the 1940s many of the basins had under- '0
'iii
gone intensive ground-water development. .Q
:J
Ul
Groundwater depletion has been widespread '0
over these basins, and locally, water-level c
j
declines have exceeded 300 feet. These water-
level declines have resulted in regional subsi-
dence, exceeding I 0 feet in some areas. A pro-
file near Luke Air Force Base illustrates that
subsidence is greater near the center of basins,
Figure
where the aggregate thickness of the fine-
San Jo<
grained sediments is generally greater. In con-
(b) Sub
junction with widespread subsidence, numerous
earth fissures have formed at and near the mar-
gins of subsiding basins or near exposed or
shallow buried bedrock.
6.9.3 J
Unconfined
aquifer
aQ)
Migration of
waterfrom
compacting clay
o to confined
t ~--Raiqui~fer~ Figure 6.9.9. Graph of hydraulic and

intergranular pressures as a function
of depth for an unconfined aquifer
overlying a confined aquifer (after
Poland et al.67). (a) Initial condition
with water table and piezometric sur-
face at same elevation (b) Subsequent
0 Pressure- 0 Pressure- condition with piezometric surface
(a) (b) lowered
6.9.4
example, is applied, their internal high void structure collapses, resulting in an erratic subsi-
dence of the land surface.
To define the magnitude of this subsidence, a test pond 30 m by 30 m was constructed on
flat land, and benchmarks anchored at various depths were installed. Water to a depth of 0. m
was admitted in early October 1956. Subsidence of the various benctunarks appears in Figure
6.9.10. As the wetting front moved downward, the bench marks progressively subsided. At
1956 I 1957 I 1958

Oct INoviDecj Jan I Feb! Marl Apr !May[ Jun I Jui!Aug[ Sep I Oct[ Nov[ Dec [Jan IFeb [Mar[ Apr [May I Jun [ JuliAugiSep [Oct INov IDee
r-~~ 1
N Testdryplot~I I~
II
1
Test plot dry
-
0
(a)
oentral
46-mbe~c"h~rl;"--------------
esent-
ment
E
800s,
under-
Ql
u
c
t1)
'0
--------------------
'iii
!. -g 2
ead
l
(/)
'0
c
as
----- ----------------·
vater-
...J
3
----- ------~---------
Jbsi-
\ pro-
that
(b)
•asins, Figure 6.9.10. Hydrocompaction and land subsidence resulting from water application on a test plot containing low-density soil,
San Joaquin Valley, California (after Lofgren55 ; courtesy The Geological Society of America, 1969). (a) Infiltration from test plot
1 COIY-
(b) Subsidence of benchmarks anchored at ground surface and at various depth intervals
merous
te mar-
. or ground surface, the change in level amounted to more than 3 m, while at the 45 m depth no
effect was observed until after 16 months.
Shallow subsidence can influence irrigation, drainage, sewerage, and transportation sys-
tems. • Sprinkler irrigation and pipelines for water conveyance are best suited in these terrains.
6.9.3 Dewatering of Organic Soils

In flat peat or muck land with a shallow water table, lowering of the water table, such as by
drainage, produces land subsidence. Causes include (1) shrinkage due to desiccation, (2) consol-
idation by loss of the buoyant force of groundwater, (3) compaction with tillage, (4) wind erosion,
(5) burning, and (6) biochemical oxidation. 46 Investigations have shown that the rate of subsi-
dence is proportional to the depth to the water table_ To conserve the life of organic soils, the
water table should be maintained as high as crop requirements and field conditions will pennit.
Subsidence of organic soils has been noted in the Netherlands, the Soviet Union, and at
.ulic and various locations in the United States. During the last 70 years extensive drainage for agricul-
mction tural purposes of islands in the Sacramento-San Joaquin Delta, California, has lowered the
uifer
land surface greater than 15 ft over much of the area, as shown in Figure 6.9.11. This has
after
tdition
necessitated construction of a vast network of perimeter levees to prevent inundation of the
tric sur- depressed islands by floods or high tides.
Jsequent
uface 6.9.4 Sinkhole Formation
Catastrophic land subsidence leading to the fonnation of sinkholes can also be associated with
tic subsi- declines in groundwater levels. Soluble rocks such as dolomite and limestone are slowly
ructed on
"More than 100 km along the California Aqueduct, a concrete-lined canal carrying some 25 x 106 m3/day of water
10fO. m to Southern California, cross formations susceptible to shallow land subsidence. To avoid the danger of subsidence
in Figure from canal leakage, large-scale spreading ponds were maintained along the alignment in order to preconsolidate the
sided. At soils before construction of the aqueduct.
. 0
Land subsidence ··;
~KIIomoler$
(feet below sea level) :
[ J 0-1 0 .
1.
• 10-15
• Greater than 15 ·
6.9.5 '
Hat:'lft'r 0. Banks
Della PumJlln,9 P!anl - ...
~·
___.,.. -~liiO(ma AqtJeducl (stete opera/ Figure 6.9.11. Land subsidence in the
Z
- Del!a-IAendof<J Canal
{Fed6rat O{Jerd~
-· Sacramento-San Joaquin Delta,
California (Galloway et alP).
6.9.6 ),\
dissolved locally by groundwater. Eventually the ground surface sinks to form a cup-shaped
depression. Over large areas of this type, a karstic sinkhole plain is fonned with most of the
drainage occurring in the subsurface.
New sinkholes often develop in regions where water tables have been lowered by pump-
ing. For example, groundwater pumping from a dolomite aquifer for mine dewatering in the
Far West Rand, South Africa, led to the fonnation of eight sinkholes larger than SO m in diam-
eter and deeper than 30 m within a period of 39 months. 18 Similarly, the massive Alabama
sinkhole shown in Figure 6.9.12 appeared suddenly after a local decline of the water table. As
Figure 6.9.12. Sinkhole fonned suddenly in

December 1973 with a diameter of 130 m
and a depth of 45 m in Shelby County,
Alabama. This is one of more than 1,000
sinkholes that developed in Shelby County
within a 15-year period. The concentration of
sinkholes is attributed to a local lowering of
the water table {courtesy U.S. Geological
Survey).
a water table is lowered, removal of the buoyant support from the subsurface clay above the
cavern together with increased velocities of infiltrating water are believed to be responsible for
the cave-ins.
6.9.5 Crustal Uplift

The opposite of land subsidence, crustal uplift, can occur over large areas subject to heavy
groundwater pumping. The tectonic uplift of land, involving an elastic expansion of the litho-
sphere, is caused by the removal of large masses of groundwater. The phenomenon has been
noted in parts of Arizona, California, and Texas, where groundwater pumped from aquifers
was removed by evapotranspiration by crops. 36 In the Santa Cruz River Basin, Arizona, a
crustal uplift of 6 em was observed between 1948 and 1967, when 43.5 billion tons of ground-
water was pumped from an 8,070-km2 area. This figure compares favorably with the 17-cm
land surface depression, previously measured, resulting from the filling of Lake Mead on the
Colorado River.
the
6.9.6 Monitoring of Land Subsidence
Several types of data are required in monitoring land subsidence. Wells are required to moni-
tor water levels at multiple levels, and extensometer wells are required to measure sediment
shaped compaction. Borehole data, including inclinometer measurements, lithographic and geophys-
:of the ical data, and geotechnical data., are also collected. The water-level monitoring wells are con-
structed with multiple or nested piezometers at different depths, as illustrated in Figure 6.9.13,
pump- to monitor water levels in the multiple zones in the partially confined aquifer as a function of
~ in the time. Submersible pressure transducers are installed to measure water levels in each of tbe
1 diam- nested piezometers.
labama Compaction is the decrease in the thickness of the subsurface sediments, whereas subsi-
.ble.As dence is the lowering of the land-surface elevation over a long period of time. The various
P1 P2 P3 P4
120feet
6.10
T
gravel pack
640feet 1
T
gravel pack
j_
l------8.75lnch9S - - - -
l - - - - - - - 9.8751nches------1
- - - - - - - - 1 5.Sinches----------l
Figure 6.9.13. Construction of nested

NOT TO SCALE piezometers Pl through P4 (lkehara31).
sediment-sample properties measured in consolidation and permeability tests are listed in

Table 6.9.1. Figure 6.9.14 illustrates the construction of an extensometer well and Figure
6.9.15 illustrates the instrumentation for the extensometer. Figure 6.9.16 shows the cumulative
37
net sediment compaction superimposed on water levels for five water years (from Ikehara ).
For this application in Yolo County, California, five piezometers and an extensometer were
constructed in three boreholes.
Table 6.9.1 Key to Properties Measured in Consolidation and Permeability Tests on Sediment Samples
Time (m) ............... from real time on first day of test (hours)
Sample thickness (c) ...... (LVDT reading)(LVDT slope) +intercept from LVDT calibration
Void ratio, e (c) .......... (height of sample-height of solids)/(height of solids)
Head difference (m) ...... head drop across the sample
Flowrate, Q (m) .... . .... flowrate used to conduct permeability test
Permeability (c) ......... (Qihead difference)(sample thickness/cross-sectional area)
Vertical effective stress
(m) .................. pressure on piston in contact with sample (lb/in2)
(c) ................... effective stress on sample (kPa)
Real vertical effective stress on the sample is about two times the measured effective stress because the
area of the piston in contact with the sample is approximately half the area of the bottom of the loading
chamber.
Compressibility (c) ....... (e 1 - e2)/(effective stress 1- effective stress 2) [equivalent to the slope of the
"void ratio versus effective stress" curve]
Specific storage (c) ....... (cornpressibility)(specific gravity of water)/(1 +e) [see void ratio]
(c), calculated; kPa, kilopascal; 1blin2, pound per square inch; LVDT, linear variable differential transf01mer; (m),
measured
Source: lkeharaJ1
6.10 EFFECTS OF GLOBAL CLIMATE CHANGE ON GROUNDWATER

The earth's temperature is affected by numerous influences, including (a) the incoming solar
radiation that is absorbed by the atmosphere and the earth's surface, (b) the characteristics
(emissivity) of the matter that absorbs the radiation, and (c) the part of the long-wave radiation
emitted by the surface, absorbed by the atmosphere, and then re-emitted as long-wave
radiation either in the upward or downward direction.
The so-called greenhouse effect is caused by the net change of the internal radiation bal-
ance of the atmosphere due to the continued increased emission of greenhouse gases, result-
ing in both the atmosphere and the earth's surface becoming wanner. The magnitude of the
greenhouse effect is dependent on the composition of the atmosphere-with the most impor-
tant factors being the concentrations of water vapor and carbon dioxide-and less importantly
on certain trace gases, such as methane. There is mounting evidence that global wanning is
under way.
In general, the hydrologic effects are likely to influence water storage patterns throughout
the hydrologic cycle and influence the exchange among aquifers, streams, rivers, and lakes.
Chalecki and Glieck8 provide a bibliography of the impacts of climate change on water
resources in the United States.
The Pacific Institute has compiled a comprehensive online bibliography of peer-reviewed
literature dealing with climate change and its effects on water resources and water systems. At
last count, the bibliography included over 3,600 citations. This bibliography can be accessed
at http://www.pacinst.org/resources/index.html. The International Panel on Climate Change
listed in (IPCC) has been active in compiling information concerning future climate change
I Figure (IPCC39-45). These reports can be obtained from http://www.ipcc.ch.
nulative The effects of climate change on groundwater sustainability include (Alley et al. 1):
~hara37 ).
ter were • Changes in groundwater recharge resulting from changes in average precipitation and
temperature or in seasonal distribution of precipitation,
1-12 inches - 1
t-6indles~
2 inches-~
collar, extensometer
reference point -
steel surface
casing
basket welded
to casing
soil
joint casing
cone
lall4
-...,..--- 897 feet

joint casing
_ _,____ S151eet
cement
plug 1,001.5 feet
1,003leet Figm
Figure 6.9.14. Construction of extensometer weU
NOT TO SCALE (Ik.ehara'\
• More severe and longer lasting droughts, with the effects of drought illustrated in
Figure 6.10.1,
• Changes in evapotranspiration resulting from changes in vegetation, and
• Possible increased demands of groundwater as a backup source of water supply.
Surficial aquifers are likely to be part of the groundwater system that is most sensitive to
climate. These aquifers supply much of the flow to streams, lakes, wetlands, and springs.
Because groundwater systems tend to respond more slowly to short-tenn variability in climate
chart recorder
...- counterwelghl
dial
gauge......,_
. ,
~ cardboard pier form
12-inc::h-<liame~r steel
wei casing
cememd In place from 0 to 39 lett
extensomater weM Is
sel1edute 80 steel
2·inch-dlatnettf plpt
NOT TO SCALE
Figure 6.9.15. Instrumentation of ex.tensometer well (Ikehara37).

:ter well
conditions than the response of surface water systems, the assessment of groundwater
resources and related model simulations are based on average conditions, such as annual
:din recharge and/or average annual discharge to streams. The use of average conditions may
underestimate the importance of droughts (Alley et al. 1).
The impacts of climate change on (a) specific groundwater basins, (b) the general ground-
Jly. water recharge characteristics, and (c) groundwater quality have received little attention in
the literature. Vaccaros5 addressed the climate sensitivity of groundwater recharge, finding
:nsitive to that a wanner climate (doubling COz) resulted in a relatively small sensitivity to recharge,
i springs. which depended on land use. Sandstrom72 studied a semiarid basin in Africa and concluded
in climate
that a 15 percent reduction in rainfall could lead to a 45 percent reduction in groundwater
~l
vl \\j)\ 1f
ou
/l
...JU
wu.
mz
a:-
50 .0.05 UJU!
~(..;
~ ~~
t: oa:
1-::l
~
<fJ J:Ul
\\ /'"~/j
-I
'0 1-0
c: o..Z
~
;!: ~:3
0 100
:8 c.,
\t /:
\j
~
:S
li E
~ ~
(f)
(i; I
iii
~
-o.55
150 e / \ z.
Q{J)
\ I • -w
\/ ~ 1-:t
uo
\ '3:z
::2z
0-
u
200
Ill IV I II Ill IV I II Ill IV I II Ill IV I
\/.0.80
II Ill
1988 1989 1990 1991 1992
Figure 6.9.16. Cumulative net sediment
!XPI.ANA110N
Figu11
compaction superimposed on hydrograph from (
- Daily Water level At 0600 Hours of well 11N11E-24Q6 (P3), water years
fractic
--- Daily Sediment Compaction At 0600 Hours 1988-92m in Yolo County, California water-
o Instantaneous Sedimen1 Compaction (Ikehara37).
deposi
Joaqui
water,
recharge. Sharma76 and Green et al.30 reported similar sensitivities of the effects of climate least a
change on groundwater in Australia. Panagoulia and Dimou62 studied the effect of climate deerea
change on groundwater-streamflow interactions in a mountainous basin in central Greece. pactio1
They realized large impacts in the spring and summer months as a result of temperature-
induced changes in snowfall and snowmelt patterns. Oberdorfer61 looked at the impacts of cli-
mate change on groundwater discharge to the ocean using a simple water-balance model to
study the effect of changes in recharge rates and sea level on groundwater resources and flows
in a California coastal watershed. The impacts of sea-level rise on groundwater will include REFJ
increased intrusion of salt water into coastal aquifers.
l. Aile:
WaterR
gov/put
2. Arne
PROBLEMS Practice
3. Aver
6.1.1 Develop a summary on the effects of groundwater develop- effects of the subsidence? Start your search for information
Chicagc
ment on groundwater flow to and from surface-water bodies. with the U. S. Geological Survey Circular 1182, Land Subsidence
4. Bam
Refer to the Web site for the U. S. Geological Survey Circular in the United States, by Galloway et al. 23 and Carpenter7 ,
Amer. G
1186, Sustainability of Ground-Water Resources, by Alley et al., 1 http://water.usgs.gov/pubs/circ/circ 1182/html.
5. Bouv
httpJfwater.usgs.gov/pubsfcirdcirc 1186fhtml. 6.9.2 Write a report on the effect of land subsidence in by phre<
6.1.2 What has caused the reversal of declining groundwater south-central Arirona. What were the major causes and effects of 1975.
levels in the Chicago area? Start with Avery,3 U.S. Geological the subsidence? Start your search for information with the U.S. 6. Bred,
Survey Fact Sheet 222-95. Geological Survey Circular 1182, lAnd Subsidence in the United Jour. Ge
6.9.1 Write a report on the effect of land subsidence in the States, by Galloway et al. 23 and Galloway and Riley24, 7. Carp<
San Joaquin Valley, California. What were the major causes and http://water.usgs.govfpubs/circ/circ 1182/html. UnitedS
References 325
12 -·- -
z
QUl
-w 6
1-;c
oo 0
~z
~:z -6
0-
0
-12
1960 1965 1970 1975 1980 1985 1990 1995
~nt Figure 6.10.1. Effects of drought on groundwater levels and associated subsidence in the San Joaquin Valley, California (modified
aph from Galloway and Riley, 24 as presented in Alley et al. 1). The San Joaquin Valley is a major agricultural area that produces a large
rs fraction of the fruits, nuts, and vegetables in the United States. Groundwater withdrawals during the 1930s to early 1960s caused
water-level declines of tens to hundreds of feet in much of the valley. The water-level declines resulted in compaction of the alluvial
deposits and ex.tensive land subsidence. Subsidence in excess of one foot has affected more than 5,200 square miles in the San
Joaquin Valley, representing perhaps the largest anthropogenic change in land-surface elevation in the world. Importation of surface
water, beginning in the 1960s, led to a decrease in groundwater withdrawals, which in tum led to rising groundwater levels and at
liDate least a temporary end to further subsidence. During severe droughts in 1976-77 and 1987-93, deliveries of imported water were
imate decreased. More groundwater was pumped to meet water demands, resulting in a decline in the water table and a renewal of com-
·eece. paction and land subsidence.
lture-
)f cli-
jel to
flows
dude REFERENCES
1. Alley, W. M., T. E. Reilly, and 0. L. Franke, Sustainability ofGroutui- ical Swvey Circular 1182, httpJ/water.usgs.gov/pubs/circ/circ I I82, Denver,
Warer Resources, U.S. Geological Survey Circular 1186, http://water.usgs. co. 1999.
gov/pubs/circ/circll86, U.S. Geological Survey, Denver CO, 1999. 8. Chalecki, L. H., and P. H. Gleick, A comprehensive bibliography of the
2. Amer. Soc. Civil Engrs., Growul wilier management, Manual Engmg. impacts of climate change and variability on water resources of the United
Practice 40, New York, 216 pp., I972. States, Jour. American Water Resources AsSTL, v. 35, pp. 1657-1665, 1999.
3. Avery, C. F., Reversal of Declining Ground-Water Levels in the 9. Chow, V. T., D. R. Maidment, and L. W. Mays, Applied Hydrology,
nation McGraw-Hill, New York, 1988.
Chicago Area, U.S. Geological Fact Sheet 222-95, 1995.
·idenee
4. Barnes, B. S., The structure of discharge-recession curves, Trans. 10. Clark, W. E., Computing the barometric efficiency of a well, Jour.
enter1 ,
Amer. Geophysical Union, v. 20, pp. 721-725, 1939. Hydraulics Div., Amer. Soc. Civil Engrs., v 93, no. HY4, pp. 93-98, 1967.
5. Bouwer, H., Predicting reduction in water losses from open channels I L Clawges, R. M., and C. V. Price, Digital Data Sets Describing Prin-
nee in by phreatophyte centro!, Water Resources Research, v. II, pp. 96-101, cipal Aquifers, Surficial Geology, and Growul-Water Regions of the Con·
feels of 1975. tenninous United States, U.S. Geological Survey Open-File Report 99-77,
ile U.S. 6. Bredehoeft, J. D., Response of well-aquifer systems to earth tides, http://water.usgs.gov/pubslofr/ofr99-77. 1999.
United low: Geophysical Research, v. 72, pp. 3075-3087, 1967. I 2. Cohen, P., et al., An atlas of Long Island's water resources, Bull. 62,
We/4 , 7. Carpenter, M. C., South-Central Arizona, in Land Subsidence in the New York Water Resources Comm., Albany, 117 pp., 1968.
United States, D. Galloway, D. Jones, and S. Ingebritsen, eds., U.S. Geolog-
13. Cooper, H. H.. Jr., and M. I. Rorabaugh, Ground-Water Movements 32. Hall, F. R., Base-flow recession-A review, Water Resources 51. Kl~
and Bank Storage Due to Flood Stages in Surface Streams, U.S. Geologi- Research, v. 4, pp. 973-983, 1968. in twen
cal Survey Water-Supply Paper 1536-J, pp. 343-366, 1963. 33. Healy, J. H., et al., The Denver earthquakes, Science, v. 161, pp. Natural
14. Cooper, H. H., Jr., et al., The response of well-aquifer systems to 1301-1310, 1968. 52. Kn
seismic waves, Jour. Geophysical Research, v. 70, pp. 3915-3926, 1965. drainag
34. Heath, R. C., Basic Ground-Water Hydrology, US. Geologtcal SuTVey
15. DaCosta, J. A., Effect of Hegben Lake Earthquake on Water Levels in Water Supply Paper 2220, httpJ/water.usgs.gov/pubs/wsp/wsp2220, Den- 3649-31
Wells in the United States, U.S. Geological Survey Professional Paper ver, CO, 1998. 53. Ku1
435,pp. 167-178,1964. during 1
35. Hellwig, D. H. R., Evaporation of water from sand, lour. Hydrology,
16. Domenico, P. A., and M. D. Mifflin, Water from low-permeability V, 18, pp. 93-118, 1973.
cific C01
sediments and land subsidence, Water Resources Research, v. !, pp. 207-2)(
563-576, 1965. 36. Holzer, T. 1., Elastic expansion of the lithosphere caused by ground-
54. Lo(
water depletion, Jour. Geophysical Research, v. 84, pp. 4689-4698, 1979.
17. Farvolden, R_ N., Geologic controls on ground-water storage and base Geologi.
flow, lour. Hydrology, v. I, pp. 219-249, 1963. 37. Ikehara, M. E., Data from the Woodland Land Subsidence Monitoring
55. Lof;
Station, Yolo County, California, Water Years 1988-92, U.S. Geological
18. Foose, R. M., Sinkhole formation by groundwater withdrawal: Far J. Varne
Survey Open·File Report 94-494, Sacramento, CA. 1995.
West Rand, South Africa, Science, v. 157, pp. 1045-1048, 1967. Geo1. Sc
19. Fox, D. J., Man-water relationships in metropolitan Mexico, Geogr. 38. Ineson, J., Form of ground-water fluctuations due to nuclear explo-
56. Mei
Review, v. 55, pp. 523-545, 1965. sions, Nature, v. 198, pp. 22-23, 1963.
Survey 'II
20. Franke, 0. L., Double-Mass-CuTVe Analysis ofthe Effects ofSewering 39. Intergovernmental Panel on Climate Change (lPCC), Climate 57. Me}
on Groundwater Levels on Long Island, New York, U.S. Geological Survey Change 1995: The Science of Climate Change: Contribution of Working graphs, j
Prof Paper 600-B, pp. 205-209, 1968. Group I to the Second Assessment Report of the Intergovernmental
58. Min;
Panel on Climate Change, Cambridge University Press, Cambridge, New
21. Freeze, R. A., Role of subsurface flow in generating surface runoff, Sci. Hyd;
York, 1996.
Water Resources Research, v. 8, pp. 609--{523, 1272-1283; 1972. 59. Non
22. Gabrysch, R. K., and C. W. Bonnet, lAnd-surface subsidence in the 40. Intergovernmental Panel on Climate Change (JPCC), Climate Change
watercou
Houston-Galveston region, Texas, Rept. 188, Texas Water Dev. Board, 1995: Impacts, Adaptations, and Mitigation of Climate Change: Scientific-
pp. 30-4
Austin, 19 pp., 1975. Technical Analysis: Contribution of Working Group II to the Second
Assessment Report of the Intergovernmental Panel on Climate Change, 60. Non
23. Galloway, D., D. Jones, and S. Ingebritsen, eds., Land Subsidence metric fit
Cambridge University Press, Cambridge, New York, 1996.
in the United States, U.S. Geological SuTVey Circular 1182, Research
http://water.usgs.gov/pubs/circ/circll82, Denver, CO, 1999. 41. Intergovernmental Panel on Climate Change (IPCC), Hydrology and
Freshwater Ecology, in Climate Change 1995: Impacts, Adaptations, and 61. Ober
24. Galloway, D., and F. S. Riley, San Joaquin Valley, California, in dieting th
Mitigation of Climate Change: Contribution of Working Group II to the
Land Subsidence in the United States, D. Galloway, D. Jones, and S. Coastal~
Second Assessment Report of the Intergovernmental Panel on Climate
Ingebritsen, eds., U.S. Geological Survey Circular 1182, httpJ/water.usgs. Symposit
gov/pubslcirc/circ1182, Denver, CO, 1999. Change, Cambridge University Press, Cambridge, New York, 1996.
42. Intergovernmental Panel on Climate Change (IPCC), Regional 62. Pana;
25. Gambolati, G., and R. A. Freeze, Mathematical simulation of the sub- flow inter
sidence ofVenice, Water Resources Research, v. 9, pp. 721-733, 1973; v. Impacts of Climate Change: An Assessment of Vulnerability, Cambridge
University Press, Cambridge, New York, 1998. 41, pp. 7~
IO,pp.56>-577, 1974.
63. Peck,
26. Gardner, W. R., and M. Fireman, Laboratory studies of evaporation 43. Intergovernmental Panel on Climate Change (IPCC), Climate Change
Geophysi,
from soil columns in the presence of a water table, Soil Sci., v. 85, pp. 2001: The Scientific Basis, http://www.ipcc.ch, 2001.
244-249, 1958. 64. Pinde
44. Intergovernmental Panel on Climate Change (JPCC), Climate Change
ponent ol
27. Gatewood, J. S., et al., Use of Water by Bottom-Land Vegetation in 2001: Impacts, Adaptation, and Vulnerability,
ResourceJ
Lower Safford Valley, Arizona, U.S. Geological Survey Water-Supply httpJ/www.ipcr.ch/pub.online.htm, 200 L
Paper 1103, 210 pp., 1950. 65. Polan
45. Intergovernmental Panel on Climate Change (IPCC), Climate Change
agement '
28. Gi!Wand, J. A., A rigid plate model of the barometric effect, Jour. 2001: http://www.ipcr.ch, 2001.
Memoir 1:
Hydrology, v. 7, pp. 233-245, 1969. 46. Inti. Assoc. Sci. Hydrology, lAnd subsidence, Publ. nos. 88 and 89 (2 66. Polan
29. Gleick, P. H., et al., Water: The Potential Consequences of Climate vols.), 661 pp., 1969. fluids, in I
Variability and Change for the Water Resources of the United States, A
47. Jacob, C. E., Fluctuations in artesian pressure produced by passing ogy, v. 2, (
Report of the National Water Assessment Group, for the U.S. Global
railroad trains as shown in a well on Long Island, New York, Trans. Amer. 67. Polan•
Change Research Program, Pacific Institute for Studies in Development,
Geophysical Union, v. 20, pp. 666-674, 1939. vey Projes.
Environment, and Security, Oakland, CA. September 2000.
48. Jacob, C. E., On the flow of water in an elastic artesian aquifer, Trans. 68. Rasm1
30. Green, T. R., B. C. Bates, P. M. Fleming, S. P. Charles. and M.
Amer. Geophysical Union, v. 21. pp. 574-586, 1940. Beaverdan
Taniguchi, Simulated Impacts of Climate Change on Groundwater
Recharge in the Subtropics of Queensland, Australia, Subsurface Hydro- 49. Johnston, R. H., Base Flow as an Indicator ofAquifer Characteristics Paper 147:
logical Responses to Land Cover and Land Use Changes, KJuwer Acade- in the Coastal Plain of Delaware, U.S. Geological Survey Professional 69. Ripple
mic Publishers, Norwell, MA, U.S.A., pp. 187-204, 1997. Paper 750-D, pp. 212-215, 1971. Bare Soils
31. Gregg, D. 0., An analysis of ground-water fluctuations caused by 50. Keppel, R. V., and K. G. Renard, Transmission losses in ephemeral Water-SupJ
ocean tides in Glynn County, Georgia, Ground Water, v. 4, no. 3, pp. stream beds, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 88, no. 70. Robim
24-32,1%6. frY3,pp. S9-68, 1962. terns, lour.
References 327
>urces 51. Klein, M., and P. Kaser, A statisrica/ analysis of growui-water levels 71. Robmson, T. W., Earth-tides shown by fluctuations of water-levels in
in twenty selected observation wells in Ohio. Tech. Rept. 5, Ohio Dept. wells in New Mexico and Iowa, Trans. Amer. Geophysical Union , v. 20,
l, pp. Natural Resources, Div. of Water, Columbus, 124 pp., 1963. pp. 656-666, 1939.
52. Knisel, W. G., Jr., Baseflow recession analysis for comparison of 72. Sandstrom, K., Modeling the effects of rainfall variability on ground-
drainage basins and geology, Jour. Geophysical Research, v. 68, pp. water recharge in semi-arid Tanzania, Nordic Hydrology, v. 26, pp.
)urvey
3649-3653, 1963. 313-320, 1995.
,Den-
53. Kunkle, G. R., Computation of Groundwater Discharge to Streams 73. Schicht, R. J., and W. C. Walton, Hydrologic budgets for three sma/1
during Floods, or ro Individual Reaches during Basejlow, by Use of Spe- watersheds in Illinois, Rept. lnv. 40, nlinois State Water Survey, Urbana,
vlogy,
cific Conductance, U.S. Geological Survey Professional Paper 525-D, pp. 40 pp., !961.
207-210, 1965. 74. Schneider, R., Correlation of Ground-Water Levels and Air Tempera-
uund- lures in the Winter and Spring in Minnesota, U.S. Geological Survey
54. Lofgren, B. E., Analysis of Srresses Causing Land Subsidence, U.S.
1979. Water-Supply Paper 1539-D, 14 pp., 1961.
Geological Survey Prof Paper 600-B. pp. 219-225, 1968.
loring 75. Singh , K. P., Some factors affecting baseflow, Water Resources
55. Lofgren, B. E., Land subsidence due to the application of water, in D.
ogica/ Research, v. 4, pp. 985-999, 196&.
J. Varnes, and G. Kiersch, eds., Reviews in Engineering Geology, v. 2,
Geol. Soc. Amer., Boulder, CO, pp. 271-303, 1969. 76. Sharma, M. L, Impact of Climate Change on Groundwater Recharge,
explo- 56. Meinzer, 0. E., Ground Water in the United States, U.S. Geological Proceedings of the Conference on Climate and Water, Volume I, Helsinki,
Survey Water-Supply Paper 836-D, pp. 157-232, 1939. Finland, Val ton Painatuskeskus, pp. 511-520, September 1989.
lim ate 57. Meyboom, P., Estimating ground-water recharge from stream hydro- 77. Shelton, M. J., and L. B. James, Engineer-geologist team investigates
)rking graphs, Jour. Geophysical Research, v. 66, pp. 1203-1214, 1961. subsidence, Jour. Pipeline Div., Amer. Soc. Civil Engrs., v. 85, no. PL2, 18
nental pp., 1959.
58. Minshall, N. E., Precipitation and base flow variability, Inti. Assoc.
,,New 78. Smedema, L. B., and P. J. Zwerman, Fluctuations of the phreatic sur-
Sci. HydrologyPubl. 76,pp. 137-145,1967.
face: I. Role of entrapped air under a temperature gradient, Soil Sci., v.
59. Norris, S. E., and H. B. Eagon, Jr., Recharge characteristics of a
'hange 103,pp.354-359, 1967.
watercourse aquifer system at Springfield, Ohio, Growtd Wazer, v. 9, no. l.
ntific- 79. Stone, M. C., R. H. Hotchkiss, C. M. Hubbard, T. A. Fontaine, L. 0.
pp. 3~1. 1971.
econd Mearns, and J. G. Arnold, Impacts of climate change on Missouri River
1ange, 60. Norom, D. I., and LN. Luthin, The effectS of entrapped air and baro- basin water yield, Jour. of the American Water Resources Assn., v. 37(5):
metric fluctuations in the drainage of porous mediums, Water Resources
pp. 1119-1129, October 2001.
Research, v. 4, pp. 417-424, 1968.
~and 80. Taylor, C. 1., and W M. Alley, Ground-Water-Level Monitoring and
61. Oberdorfer, J. A., Numerical Modeling of Coastal Discharge: Pre-
.s, and the Imporrance of Long-Term Water-Level Data, U.S. Geological Survey
dicting the Effects of Climate Change, in Growuiwater Discharge in the Circular 121 7, http://water.usgs.gov/pubs/circ/circ 1217, Denver, CO,
to the
Coastal- Zone, R. W. Buddemeier, ed., Proceedings of an International 2001.
lim ate
Symposium, Moscow, pp. 85-91, July 1996.
81. Todd, D. K., Ground-water flow in relation to a flooding stream, Proc.
62. Panagoulia, D., and G. Dimou, Sensitivities of groundwater-stream- Amer. Soc. Civil Engrs., v_ 81, sep. 628, 20 pp., 1955.
gional
flow interactions to global climate change, Hydrological Sciences Jour., v.
bridge 82. Troxell, H. C., The diurnal fluctuation in the ground-water and flow of
41, pp. 781-796, 1996.
the Santa Ana River and its meaning, Trans. Amer. Geophysical U~ion,
63. Peck, A. L, The water table as affected by atmospheric pressure, Jour. v. 17,pp.496-504, 1936.
'hange
Geophysical Research, v. 65, pp. 2383-2388, 1960.
83. Tschinkel, H. M., Short-term fluctuation in streamflow as related to
64. Pinder, G. F, and J. F. Jones, Determination of the ground-water com- evaporation and transpiration, Jour. Geophysical Research, v. 68, pp.
ponent of peak discharge from the chemistry of total runoff, Water 6459-6469, 1963.
Resources Research, v. 5, pp. 438-445, 1969.
84. Turk, L. J., Diurnal fluctuations of water tables induced by atmos-
65. Poland, J. F., Subsidence and its control, in Underground Waste Man- pheric pressure changes, Jour. Hydrology, v. 26, pp. 1-16, 1975.
bange
agement and Environmental Implications, Amer. Assoc. Petr. Geologists 85. Vaccaro, J. J., Sensitivity of groundwater recharge estimates to cli-
Memoir 18, pp. 50-71, 1972.
mate variability and change, Columbia Plateau, Washington, Jour. Geo-
189(2 66. Poland, J. F., and G. H. Davis, Land subsidence due to withdrawal of physical Research, v. 97 (03), pp. 2821-2833, 1992.
fluids, in D. J. Varnes and G. Kiersch, eds., Reviews in Engineering Geol- 86. Vacher, H. L., Hydrology of small oceanic islands-Influence of
assing ogy, v. 2, Geol. Soc. Amer., Boulder, CO, pp. 187-269, 1969. atmospheric pressure on the water table, Ground Water, v. 16,
Amer. 67. Poland, J. F., et al., Studies in Land Subsidence, US. Geological Sur- pp. 417-423, 1978.
vey Professional Paper437-A to 437-H, 1964 to 1975. 87. Veihmeyer, E J., and F. A. Brooks, Measurement of cumulative evap-
Trans. 68. Rasmussen, W. C., and G. E. Andreasen, Hydrologic Budget ofrlre oration of bare soil, Trans. Amer. Geophysical Union, v. 35, pp. 601~07,
Beaverdam Creek Basin, Maryland, U.S. Geological Survey Water-Supply 1954.
ristics Paper 1472, 106 pp. , 1959. 88. Visocky, A. P., Estimating the ground-water contribution to storm
sional 69. Ripple. C. D., et al., Estimating Steady-State Evaporation Rates from runoff by the electrical conductance method, Ground Water, v. 8, no. 2,
Bare Soils under Conditions of High Water Table, U.S. Geological Survey pp. 5-10, 1970.
~meral Water-Supply Paper 2019-A, 39 pp., 1972. 89. Visocky, A. P., Water-Level Trends and Pumpage in the Deep Bedrock
:8, no. 70. Robinson, E. S., and R. T. Bell, Tides in confined well-aquifer sys- Aquifers in the Chicago Region, 1991-1995, U.S. Geological Survey
tems, Jour. Geophysical Research, v. 76, pp. 1857-1869, 1971. Circular 182, Illinois State Water Survey, Champaign, IL, 1997_
328 Chapter 6 Groundwater Levels and Envirorunenlal [nfluences
90. Vorhis, R. C., Interpretation of hydrologic data resulting from earth-

quakes, Geologische Rundschau, v. 43, pp. 47-52, 1955.
94. Werner, P. W., and K. J. Sundquist, On the ground-water recession
curve for large watersheds, Inti. Assoc. Sci. Hydrology Pub/. 33, pp.
c
91. Vorhis, R. C., Earthquake-induced water-level fluctuations from a 202-2! 2, 1951.
well in Dawson County, Georgia, Seismological Soc. Amer. Bull., v. 54, 95. White, W. N., A Method of Estimating Ground-Water Supplies Based
pp. 1023-11 33, 1964. on Discharge by Plants and Evaporation from Soil, U.S. Geological Sur-
92. Vorhis, R. C.. Hydrologic Effects of the Earthquake of March 27, vey Water-Supply Paper 659. pp. 1-105, 1932.
1964, Outside Alaska, U.S. Geological Survey Professional Paper 544-C, 96. Winter, T. C., J. W. Harvey, 0. L. Franke, and W. M. Alley, Ground-
54 pp., 1967. water and Surface Water: A Single Source, U.S. Geological Survey
93. Werner. P. W., and D. Noren, Progressive waves in non-artesian Circular 1139, http://water.usgs.gov/pubs/circ/circ1139, Denver, CO,
aquifers, Trans. Amer. Geophysical Union, v. 32, pp. 238-244, 1951. 1998.
7.1 l
:ssion
I, pp.
Chapter7 ·
'lased
I Sur-
vund-
'urvey
·,CO,
Quality of Groundwater
It is now recognized that the quality of groundwater is just as important as its quantity_ All
groundwater contains salts in solution that are derived from the location and past movement
of the water. The quality required of a groundwater supply depends on its purpose; thus, needs
for drinking water, industrial water, and irrigation water vary widely. To establish quality cri-
teria, measures of chemical, physical, biological, and radiological constituents must be speci-
fied, as well as standard methods for reporting and comparing results of water analyses.
Dissolved gases in groundwater can pose hazards if their presence goes unrecognized. The uni-
formity of groundwater temperature is advantageous for water supply and industrial purposes,
and underlying saline groundwaters are important because they offer potential benefits.
7.1 NATURAL GROUNDWATER QUALITY

The groundwater in natural systems generally contains less than 1,000 mg/1 dissolved solids,
unless groundwater has (1) encountered a highly soluble mineral, such as gypsum, (2) been
concentrated by evapotranspiration, or (3) been geothennally heated (Graham et al. 18). Natural
groundwater generally acquires dissolved constituents by dissolution of aquifer gases, miner-
als, and salts. Consequently, soil zone and aquifer gases and the most soluble minerals and
salts in an aquifer generally determine the chemical composition of groundwater in an aquifer.
Table 7.1.1 gives examples of natural groundwater chemistry. Most groundwater is
recharged through a soil zone containing partial pressures of carbon dioxide gas that are higher
than the atmosphere. Recently recharged groundwater therefore generally contains high
inorganic carbon concentrations. Minor and trace element compositions of natural groundwa-
ter depend on the availability of minor and trace elements in easily soluble phases or on
sorption sites, and the redox state of the water in the aquifer (Graham et al. 18). Anthropogenic
sources can create significant levels of trace elements in the atmosphere. Recharge from pre-
cipitation introduces these trace elements into the groundwater.
Knowledge of the natural groundwater quality can provide important insights into the
nature of the resource. Evaluation of the natural chemical and isotopic compositions of
groundwater can provide inferences of the reactions that produce natural water chemistry and
the recharge, movement, mixing, and discharge of groundwater. The chemistry of natural
groundwater flow systems can be used to detennine (1) geochemical reactions that produce
observed water chemistry and changes in observed water chemistry, (2) groundwater flow
paths, (3) groundwater mixing, and (4) groundwater ages and flow rates.
Arsenic, a naturally occurring element in rocks, soils, and water in contact with them, is a
human health concern because it can contribute to skin, bladder, and other cancers (National
Research Council28). Figure 7.1.1 illustrates the levels of arsenic concentrations in groundwa-
329
330 Chapter 7 Quality of Groundwater
Table 7.1.1 Examples of Natural Groundwater Chemistry•

Site Temperature pH Ca Mg Na K HC03 so4 Cl Si02
Carbonate rock aquifer
Cold Creek Springa 10 7.6 1.72 0.64 0.07 0.01 4.79 0.09 0.03 0.11
Com Creek Spring" 21 7.7 1.14 1.34 0.27 0.05 4.67 0.18 0.20 0.28
Indian Springs0 26 7.4 l.l5 0.92 0.18 0.03 3.97 0.16 0.10 0.20 ~~
Carbonate rock aquifer with evaporite mineral I
Lewiston Big Springb 10.6 7.58 1.87 1.15 0.10 0.02 3.31 1.46 0.05
;_l
Hanover Flowing Wellb 20.4 7.63 2.10 l.19 0.12 0.03 3.42 1.77 0.04 ·"
\
Vanek Warm Sprin( 19.6 7.40 3.25 1.65 0.16 0.03 3.53 3.44 O.D7 '·f:
Landusky Springb 20.4 7.24 6.50 4.08 1.79 0.24 3.81 10.11 0.54
..\l,
Granitic rock aquifer
Ephemeral Springs" 6.2 0.08 0.03 0.13 0.03 0.33 0.01 O.Ql 0.27
Perennial Springsc 6.8 0.26 0.08 0.26 0.04 0.90 0.03 0.03 0.41
Shallow Well 2d 8.0 7.89 0.85 0.19 0.54 0.04 2.34 0.09 0.10 0.21
Rhyolitic rock aquifer
Lower Indian Springe 21 7.9 0.15 0.04 2.48 0.04 2.07 0.18 0.42 0.80
Crystal Springse 24 7.7 0.55 0.15 2.18 0.09 2.33 0.23 0.59 0.75
S. Brown Wen' 10.6 6.8 0.50 0.16 1.02 0.10 1.54 0.14 0.23 0.93
N. Brown WeiV 14.7 7.5 0.87 0.23 2.09 0.17 2.46 0.54 0.71 1.03
:<......,
* All analyses are in millimoles per liter (mmolfl)and temperature in • Winograd and Pearson49
degrees Celsius. Note that mmoln times molecular weight equals mg!l. b Plummer et al. 32 -
Source: Graham et al. 18 c Garrels and MacKenzie17
29
d Nordstrom et al. ..,Jt''....
"
t White46
/Thomas et al.43 Figur
ter of the United States (also refer to Welch et al. 45). Generally, arsenic concentrations in
groundwater are the highest in the western part of the United States, with parts of the midwest
and northeast having some locations with arsenic concentrations that exceed 10 f!g/L, which is
the provisional World Health Organization limi~2 for arsenic in drinking water.
7.2 SOURCES OF SALINITY

All groundwater contains salts in solution; reported salt contents range from less than 25 mg/1
in a quartzite spring to more than 300,000 mg/1 in brines47• The type and concentration of salts
depend on the environment, movement, and source of the groundwater. Ordinarily, higher
concentrations of dissolved constituents are found in groundwater than in surface water
because of the greater exposure to soluble materials in geologic strata. Soluble salts in ground-
water originate primarily from solution of rock materials. Bicarbonate, usually the primary
anion in groundwater, is derived from carbon dioxide released by organic decomposition in the
soil. Salinity varies with specific surface area of aquifer materials, solubility of minerals, and
contact time; values tend to be highest where movement of groundwater is least Hence, salin-
ity generally increases with depth. A common geochemical sequence in groundwater includes
bicarbonate waters near ground surface varying to chloride waters in the deepest portions of
formations.
0.11
0.28
0.20
/
0.27
0.41
0.21
0.80
0.75
0.93
1.03 Arsenic, in ug/1
... Greater than 50.0
... 10.0-50.0
250 500 Miles ... 5.0-9.9
0 250 500 Kllcmele<s a Less than 5.0
Figure 7.1.1. Arsenic concentrations in groundwater of the United States (Welch et al.45).
ons in
Precipitation reaching the earth contains only small amounts of dissolved mineral matter.
idwest
Once on earth, the water contacts and reacts with the minerals of the soil and rocks. The quan-
hich is
tity and type of mineral matter dissolved depend on the chemical composition and physical
structure of the rocks as well as the hydrogen-ion concentration (pH) and the redox potential
(Eh) of the water. 5 Carbon dioxide in solution, derived from the atmosphere and from organic
processes in the soil, assists the solvent action of water as it moves underground. 21 The geo-
chemical cycle of surface water and groundwater shown in Fig. 7.2.1 illustrates the principal
~5 mg/1
chemical changes involved in water as it travels through the hydrologic cycle from precipita-
of salts
tion to groundwater.
higher
In areas recharging large volumes of water underground, such as alluvial streams, chan-
~ water
nels, or artificial recharge areas, the quality of the infiltrating surface water can have a marked
~round
effect on that of the groundwater. Locally, absorbed gases of magmatic origin contribute dis-
>rimary solved mineral products to groundwater; mineralized thennal springs are an excellent exam-
n in the
ple. Connate water is usually highly mineralized because it is derived from water originally
1ls, and
entrapped in sedimentary strata since the time of deposition. •
~. salin-
ncludes
lions of 'It should be recognized that most connate water has been altered chemically by various chemical and physical
processes and therefore does not necessarily represent the original water of deposition.
Condensation Atmosphere
Nitrogen, oxygen
carbon dioxide
~
dissolved
1
,------~-P_,reciprtion,-----.
Temporary retention in mountain
areas as soil water 1 r
1. C02 dissolved in soil, Ca, Mg,
rrr Na bicarbonates added to water
2. S04 dissolved in areas where
oxidation of su~ides is occurring
Evaporation
3. Connate water or soluble com-
pounds of marine sediments
added
1
Transpiration;
mineral matter
Chlorides and
sulfates of
sodium
!
Runoff
largely retained
in soil, partly
carried off in
crop plants
magnesium,
calcium, and
potassium
carried with
water vapor 1. C0 2 added, forming carbonic
acid
2. Reaction of soil minerals with Consti
carbonic acid to form soluble
bicarbonates
3. Precipitation of colloidal iron, t Silica (
rrr aluminum, and silica, of car-

bonates as solubility limit is
reached
(/)
Q)
:;,
£
0.
!
0
4. Cation exchange Cii
~
t .c
[l.
Iron (F
Groundwater
c::::==::::J 1. Cation exchange
Outflow to ocean Effluent 2. Sulfate reduction
-+--
Carries mineral matter back seepage by anaerobic bacteria
Ocean substituting bicar-
-+----- Subsurface outflow to ocean bonate for the sulfate
Figure 7.2.1. Geochemical cycle of surface water and groundwater (after USGS Water-Supply Paper 1469).
Mangar
Salts are added to groundwater passing through soils by soluble products of soil weather-
ing and of erosion by rainfall and flowing water. Excess irrigation water percolating to the
water table may contribute substantial quantities of salt. Water passing through the root zone
of cultivated areas usually contains salt concentrations several times that of the applied irriga-
tion water. Increases result primarily from the evapotranspiration process, which tends to con-
centrate salts in drainage waters. In addition, soluble soil materials, fertilizers, and selective
absorption of salts by plants wlll modify salt concentrations of percolating waters. Factors gov-
erning the increase include soil permeability, drainage facilities, amount of water applied,
crops, and climate. Thus, high salinities may be found in soils and groundwater of arid
climates where leaching by rainwater is not effective in diluting the salt solutions. Similarly,
poorly drained areas, particularly basins having interior drainage, often contain high salt con-
centrations. Also, some regions contain remnants of sedimentary deposition under saline
waters; the designation badlands implies in part the lack of productivity resulting from excess
salt contents of the soil and water.
Groundwater passing through igneous rocks dissolves only very small quantities of min-
eral matter because of the relative insolubility of the rock composition. Percolating rainwater
contains carbon dioxide from the atmosphere, which increases the solvent action of the water.
The silicate minerals of igneous rocks result in silica being added to the groundwater.
Sedimentary rocks are more soluble than igneous rocks.· Because of their high solubility,
combined with their great abundance in the earth's crust, they furnish a major portion of the
soluble constituents of groundwater. Sodium and calcium are corrunonly added cations; bicar-
bonate and sulfate are corresponding anions. Chloride occurs to only a limited extent under
normal conditions; important sources of chloride, however, are from sewage, connate water,
and intruded seawater. Occasionally nitrate is an important natural constituent; high concen-
trations may indicate sources of past or present pollution. In limestone terrains, calcium and
bicarbonate ions are added to the groundwater by solution.
A summary of the natural sources and concentrations of the principal chemical con-
stituents found in groundwater is presented in Table 7.2.1 together with their effects on the
usability of the water. A comprehensive discussion of the chemical constituents of groundwa-
ter can be found in Hem. 21
Table 7.2.1 Principal Chemical Constituents in Groundwater-Their Sources, Concentrations, and Effect on Usability
(modified from C. N. Durfer, and E. Baker, USGS Water-Supply Paper 1812, 1964)
Constituent Major natural sources Concentration in natural water Effect on usability of water
Silica (SiOJ Feldspars, ferromagnesiwn Ranges generally from 1.0 to In the presence of calcium and
and clay rni.o.erals, amorphous 30 mg/l, although as much as magnesium, silica fonns a scale in boilers
silica, chert, opal 100 mgll is fairly common; as and on steam turbines that retards heat;
much as 4,000 mgll is found in the scale is difficult to remove. Silica
brines. may be added to soft water to inhibit
corrosion of iron pipes.
Iron (Fe) Igneous rocks: amphiboles, Generally less than 0.50 mgll in More than 0.1 mgll precipitates after
ferromagnesian micas, ferrous fully aerated water. Groundwater exposure to air; causes turbidity, stains
sulfide (FeS), ferric sulfide or having a pH less than 8.0 may plumbing fixtures, laundry, and cooking
iron pyrite (FeS2), magnetite contain 10 mgll; rarely as much utensils; imparts objectionable tastes
(F~OJ as 50 mg/1 may occur. Acid water and colors to foods and drinks. More
Sandstone rocks: oxides, from thermal springs, mine than 0.2 mg/1 is objectionable for most
carbonates, and sulfides or wastes, and industrial wastes may industrial uses.
iron clay minerals contain more than 6,000 mgll.
Manganese (Mn) Manganese in natural water Generally 0.20 mg/1 or less. More than 0.2 mg/1 precipitates upon
probably comes most often Groundwater and acid mine water oxidation; causes undesirable tastes,
·eather- from soils and sediments. Meta- may contain more than 10 mgll. deposits on foods during cooking; stains
~ to the morphic and sedimentary rocks plumbing fixtures and laundry;
otzone and mica biotite and amphibole fosters growths in reservoirs, filters, and
l irriga- hornblende minerals contain distribution systems. Most industrial
te con- large amounts of manganese. users object to water containing more
than 0.2 mg/l.
elective
xs gov- (continues)
applied,
of arid Important mineral sources in sedimentary rocks are feldspar, gypsum, and forms of calcium carbonate.
Table 7.2.1 (continued) Principal Chemical Constituents in Groundwater-Their Sources, Concentrations, and Effect on Usability Tal>
Constituent Major natural sources Concentration in natural water Effect on usability of water Cons
Calcium (Ca) Amphiboles, feldspars, gypsum, Generally less than 100 mg/1; Calcium and magnesium combine with Fluor
pyroxenes, aragonite, calcite, brines may contain as much as bicarbonate, carbonate, sulfate, and silica
dolomite, clay minerals 75,000 mg/1. to fonn a heat-retarding, pipe-clogging
scale in boilers and in other heat-
Magnesium (Mg) Amphiboles, olivine, pyroxenes, Generally less than 50 mg/1; exchange equipment. Calcium and
dolomite, magnesite, clay ocean water contains more than magnesium combine with ions of fatty
minerals 1,000/mgfl, and brines may acid in soaps to form soapsuds; the more
contain as much as 57,000 mg/1. calcium and magnesium, the more soap
required to form suds. A high
concentration of magnesium has a
Nitratt
laxative effect, especially on new users
of the supply.
Sodium(Na) Feldspars (albite); clay minerals; Generally less than 200 mg/1; More than 50 mg/1 sodium and potassium
evaporites, such as halite (NaCl) about 10,000 mg/1 in seawater; in the presence of suspended matter
and mirabilite (Na2 S04 .lOH2 0); about 25,000 mg/1 in brines. causes foaming, which accelerates scale
industrial wastes formation and corrosion in boilers.
Sodium and potassium carbonate in
Potassium (K) Feldspars (orthoclase and micro- Generally less than about 10 mg/1; recirculating cooling water can cause
Dissoh
cline), feldspathoids, some as much as 100 mg/1 in hot deterioration of wood in cooling towers.
micas, clay minerals springs; as much as 25,000 mg/1 More than 65 mg/1 of sodium can cause
in brines. problems in ice manufacture.
Conunonly less than 10 mgfl in Upon heating, bicarbonate is changed
groundwater. Water high in into steam, carbon diox.ide, and carbon-
sodium may contain as much as ate. The carbonate combines with alkaline
50 mgll of carbonate. earths-principally calcium and
magnesium-to fonn a crust-like scale of
_______ Limestone, dolomite calcium carbonate that retards flow of
Bicarbonate (HC03) Commonly less than 500 mgll; heat through pipe walls and restricts flow· '
may exceed 1,000 mg/1 in water of fluids in pipes. Water containing large ·
highly charged with carbon amounts of bicarbonate and alkalinity
dioxide. is undesirable in many industries.
Oxidation of sulfide ores; Commonly less than 300 mgfl Sulfate combines with calcium to form
gypsum; anhydrite except in wells influenced by an adherent, heat-retarding scale. More
acid mine drainage. As much as than 250 mg/1 is objectionable in water in
200,000 mg/1 in some brines. some industries. Water containing about
500 mg/1 of sulfate tastes bitter; water
containing about 1,000 mgll may be
cathartic.
Chloride (Cl) Chief source is sedimentary Commonly less than 10 mg/1 in Chloride in excess of 100 mgll imparts a
rock (evaporites); minor sources humid regions but up to 1,000 salty taste. Concentrations greatly in
are igneous rocks mg/1 in more arid regions. excess of 100 mg!l may cause physiolo- 7.3 MI
gical damage. Food processing industries
About 19,300 mg/1 in seawater; usually require less than 250 mg/1. Some
as much as 200,000 mg/1 industries-textile processing, paper
in brines. manufacturing, and synthetic rubber
manufacturing-desire less than 100 mg/1.
(continues)
7.3 Measures of Water Quality 335
bility Table 7.2.1 (continued) Principal Chemical Constituents in Groundwater-Their Sources, Concentrations, and Effect on Usability
Constituent Major natural sources Concentration in natural water Effect on usability of water
with Ruoride (F) Amphiboles (hornblende), Concentrations generally do not Fluoride concentration between 0.6 and
I silica apatite, fluorite, mica exceed 10 mgll. Concentrations 1.7 mgll in drinking water has a benefi-
ging may be as much as 1,600 mg/l in cial effect on the structure and resistance
brines. to decay of children's teeth. Fluoride in
excess of 1.5 mg/1 in some areas causes
'atty "mottled enamel" in children's teeth.
:more Fluoride in excess of 6.0 mg/1 causes
soap pronounced mottling and disfiguration
of teeth.
Atmosphere; legumes, plant Commonly less than 10 mgll. Water containing large amounts of nitrate
users debris, animal excrement (more than 100 mg/1) is bitter tasting and
may cause physiological distress. Water
•tassium from shallow wells containing more than
.er 45 mgll has been reported to cause
s scale methemoglobinemia in infants. Small
;, amounts of nitrate help reduce cracking
in of high-pressure boiler steel.
mse Dissolved solids The mineral constituents Commonly contains less than More than 500 mgll is undesirable for
towers. dissolved in water constitute 5,000 mg/1; some brines contain drinking and many industrial uses. Less
1 cause
the dissolved solids. as much as 300,000 mg/L than 300 mgll is desirable for dyeing of
textiles and the manufacture of plastics,
1ged pulp paper, and rayon. Dissolved solids
arbon- cause foaming in steam boilers; the maxi-
\ alkaline mum permissible content decreases with
increases in operating pressure.
:scale of
>W of
icts flow An important source of salinity in groundwater in coastal regions is airborne salts origi-
ng large nating from the air-water interface over the sea. Detailed studies on a worldwide basis 13 and
linity for lsrael 25 suggest that salts are deposited on land both by precipitation and by dry fallout.
Chloride deposition in coastal areas has been calculated to range from 4 to 20 kg Cllha. The
:o fonn deposition decreases inland, varying exponentially with distance from the sea. Thus, Israeli
:. More water measurements yielded the relation
11 water in
N = 110 e--O.Ol33d (7.2.1)
ng about
water where N is the annual amount of chloride precipitation in kg/ha and dis the distance from the
1ybe sea in kilometers. In arid regions, where surface runoff is small and evapotranspiration is
large, airborne salt deposition becomes intensified several fold in groundwater.
mparts a
ly in
hysiolo- 7.3 MEASURES OF WATER QUALITY
industries
ifl. Some The chemical characteristics of groundwater are detennined by the chemical and biological
1aper reaction in the zones through which the water moves, as illustrated in Figure 7.3.1. In speci-
.bber fying the quality characteristics of groundwater, chemical, physical, and biological analyses
n 100 mg/1. are normally required. The characteristics of water that affect water quality depend both on
substances dissolved in water and on certain properties. Natural inorganic constituents
(continues) commonly dissolved in water that are most likely to affect water use include bicarbonate,
Atmosphere Land surface and soil zone Shallow aqu.ifers
Chan
HardI
pHfo:
hydr
Oeep aquifers ion'
Specif
elect
cond
Freshwater and saltwater

interfaces
Total
disso
solid!
So.
Figure 7.3.1. The chemical characteristics of groundwater are determined by the chemical and biological reactions in the zones
through which the water moves (Heath20).
7.4 c
carbonate, calcium, magnesium, chloride, flouride, iron, manganese, sodium, and sulfate. See
Heath20 for more details. Table 7.3.1 provides a list of characteristics of water that affect water
quality. A complete chemical analysis of a groundwater sample includes the detennination of
the concentrations of the inorganic constituents present; organic and radiological parameters
are normally of concern only where human-induced pollution affects quality (see Chapter 8).
Dissolved salts in groundwater of normal salinity occur as dissociated ions; in addition, other
minor constituents are present and reported in elemental form. The analysis also includes
measurement of pH and specific electrical conductance. Depending on the purpose of a water
quality investigation, partial analyses of only particular constituents will sometimes suffice. 7.4.1 I
Properties of groundwater evaluated in a physical analysis include temperature, color,
turbidity, odor, and taste. Biological analysis includes tests to detect the presence of coliform
bacteria, which indicate the sanitary quality of water for human consumption. Because certain
coliform organisms are nonnally found in intestines of humans and animals, the presence of
these in groundwater is tantamount to its contact with sewage sources.
Table 7.3.1 Characteristics of Water that Affect Water Quality
Characteristic Principal cause Significance Remarks

Hardness Calcium and magnesium Calcium and magnesium combine with soap USGS classification of hardness
dissolved in the water to fonn an insoluble precipitate (curd) and thus (mg!L as CaC0 3:
hamper the formation of a lather. Hardness 0-60:Soft
also affects the suitability of water for use in 61-120: Moderately hard
the textile and paper industries and certain 121-180: Hard
others, and in steam boilers and water heaters. More than 180: Very hard
pH for Dissociation of water The pH of water is a measure of its reactive pH values: less than 7,
hydrogen- molecules and of acids and characteristics. Low values of pH, particularly water is acidic;
ion activity bases dissolved in water below pH 4, indicate a corrosive water that will value of 7, water is neutral;
tend to dissolve metals and other substances more than 7, water is basic.
that it contacts. High values of H, particularly
above pH 8.5, indicate an alkaline water that, on
heating, will tend to form a scale. The pH signi-
ficantly affects the treatment and use of water.
Specific Substances that fonn ions Most substances dissolved in water dissociate Conductance values indicate
electrical when dissolved in water into ions that can conduct an electrical current. the electrical conductivity, in
conductance Consequently, specific electrical conductance micromhos, of 1 cm3 of water
is a valuable indicator of the amount of material at a temperature of 25°C.
dissolved in water. The larger the conductance,
the more mineralized the water.
Total Mineral substances Total dissolved solids is a measure of the total USGS classification of water
dissolved dissolved in water amount of minerals dissolved in water and is, based on dissolved solids
solids therefore, a very useful parameter in the evalu- (mg/L):
ation of water quality. Water containing less than Less than 1,000: Presh
500 mg!L is preferred for domestic use and for 1,000-3,000: Slightly saline
many industrial processes. 3,000-10,000: Moderately saline
10,000-35,000: Very saline
More than 35,000: Briny
Source: Heath20
oes
7.4 CHEMICAL ANALYSIS
Once a sample of groundwater has been analyzed in a laboratory, methods for reporting water
.te. See analyses must be considered. From an understanding of expressions and units for describing
t water water quality, standards can be established so that analyses can be interpreted in tenns of the
1tion of ultimate purpose of the water supply. In a chemical analysis of groundwater, concentrations of
1meters different ions are expressed by weight or by chemical equivalence. Total dissolved solids can
pter 8). be measured in terms of electrical conductance. These and other measures of chemical qual-
l, other ity are described in the following sections.
1cludes
a water
tffice. 7.4.1 Concentrations by Weight
, color, Concentrations of the common ions found in groundwater are reported by weight-per-volume
<Jlifonn units of milligrams per liter (mg/1) or micrograms per liter (mg/L). The total ionic concentra-
certain tion (or total dissolved solids) is also reported in this manner. The units milligrams per liter
ence of have replaced parts per million; however, they are numerically equivalent up to a concentra-
tion of dissolved solids of about 7,000 mgll.
7.4.2 Chemical Equivalence

Positively charged cations and negative anions combine and dissociate in definite weight
ratios. By expressing ion concentrations in equivalent weights, these ratios are readily deter-
mined because one equivalent weight of a cation will exactly combine with one equivalent
weight of an anion. The combining weight of an ion is equal to its formula weight divided by
its charge. When the concentration in milligrams per liter is divided by the combining weight,
an equivalent concentration expressed in rnilliequivalents per liter (meqll) results. Table 7.4.1 7.4.3
lists the reciprocals of combining weights of cations and anions; concentrations in milligrams
per liter can be converted to milliequivalents per liter by multiplying by the appropriate con-
version factor. For undissociated species with zero charge, of which silica is an example in
groundwater quality, an equivalent weight cannot be computed.
Table 7.4.1 Conversion Factors for Chemical Equivalence

(after Hem21 ) (concentrations in mg/1 times the conversion
factor yields concentration in meq/l)
Chemical Constituent Conversion Factor

Aluminum (AJ+3) 0.11119
Anunoni urn (NH4+) 0.05544
Barium (Ba+2) 0.01456
Beryllium (Be+3) 0.33288
Bicarbonate (HC03-) 0.01639
Bromide (BC) 0.01251
Cadmium (Cd+2) 0.01779 7.4.4
Calcium (Ca.. 2) 0.04990
Carbonate (C03- 2) 0.03333
Chloride CCn 0.02821
Cobalt (Co+2) 0.03394
Copper (Cu+ 2) 0.03148
Fluoride Wl 0.05264
Hydrogen (W) 0.99209
Hydroxide (OW) 0.05880
Iodide Cn 0.00788
Iron (Fe+2) 0.03581
Iron (Fe+3) 0.05372
Lithium (Li+) 0.14411
Magnesium (Mg+2) 0.08226
Manganese (Mn+2) 0.03640
Nitrate (N03-) 0.01613
Nitrite (N02-) 0.02174
Phosphate (P04-J) 0.03159
Phosphate (HP0 4- 2) 0.02084
Phosphate (H 2P04-) 0.01031
Potassium (K+) 0.02557
Rubidium (Rb+) 0.01170
Sodium (Na+) 0.04350
Strontium (Sr+2) 0.02283
Sulfate (S04- 2) 0.02082
Sulfide (S-2) 0.06238
Zinc (Zn+2) 0.03060
In application, therefore, it may be expected that of the total dissolved solids in a ground-
water sample, the sum of the cations and the sum of the anions when expressed in milliequiv-
ight
alents per liter will be equal. If the chemical analysis of the various ionic constituents indicates
~ter
a difference from this balance, it may be concluded either that there are other undetermined
tlem
constituents present or that errors exist in the analysis.
d by
ight,
7.4.1 7.4.3 Total Dissolved Solids by Electrical Conductance
·ams
A rapid determination of total dissolved solids can be made by measuring the electrical con-
COll-
ductance of a groundwater sample. Conductance is preferred rather than its reciprocal, resist-
Ie in
ance, because it increases with salt content. Specific electrical conductance defines the
conductance of a cubic centimeter of water at a standard temperature of 25°C; an increase of
1oc increases conductance by about 2 percent.
Specific conductance is measured in microsiemens/cm (~S/cm). The unit microsiemens/
em is equivalent to micromhos/cm. Because the definition of specific conductance already
specifies the dimensions to which the measurement applies, the length in the units is often
omitted in practice.
Because natural water contains a variety of ionic and undissociated species, conductance
cannot be simply related to total dissolved solids. However, conductance is easily measured
and gives results that are convenient as a general indication of dissolved solids. An approxi-
mate relation for most natural wate~ 6 .34 in the range of 100 to 5,000 ~/em leads to the equiv-
alencies l meq/1 of cations =100 ~Stem and 1 mgll =1.56 ~S/cm.
7.4.4 Hardness
Hardness results from the presence of divalent metallic cations, of which calcium and magne-
sium are the most abundant in groundwater.* These ions react with soap to fonn precipitates
and with certain anions present in the water to fonn a scale. Because of their adverse action
with soap, hard waters are unsatisfactory for household cleansing purposes; hence, water-soft-
ening processes for removal of hardness are needed.
The hardness in water is derived from the solution of carbon dioxide, released by bacter-
ial action in the soil, in percolating rainwater.36 Low pH conditions develop and lead to the
solution of insoluble carbonates in the soil and in limestone formations to convert them into
soluble bicarbonates. Impurities in limestone, such as sulfates, chlorides, and silicates, become
exposed to the solvent action of water as the carbonates are dissolved so that they also pass into
solution. Thus, hard water tends to originate in are:;ts where thick topsoils overlie limestone
formations. A map of hardness of groundwater in the United States is shown in Figure 7.4.1.
Hardness Hr is customarily expressed as the equivalent of calcium carbonate. Thus,
CaC0 3 M CaC03
HT = Ca X -C-a- + g X -C-a---=- (7 .4.1)
where H-, Ca, and Mg are measured in milligrams per liter and the ratios in equivalent weights.
Equation 7.4.1 reduces to
Hr=2.5 Ca+4.1 Mg (7.4.2)
The degree of hardness in water is commonly based on the classification listed in Table 7.4.2.
'The terms hard and soft as applied to water date from Hippocrates (460-354 B.C.), the father of medicine, in his
treatise on public hygiene, Air, Water and Places: "Consider the waters which the inhabitants use, whether they
be marshy and soft,.or hard and running from elevated and rocky situations, and then if saltish and unfit for
cooking ... for water contributes much to health."6
Figure 7.4.1. Hardness of groundwater in the United States. Areas delineated represent average conditions on a generalized basis
(after Ackerman and Lof, Technology in American Water Development, Resources for the Future, copyright © 1959 by The Johns
Hopkins University Press).
Table 7.4.2 Hardness Classification of Water (after Sawyer and McCarty36)
Hardness, mg/1 as CaC03 Water class

0-75 Soft
75-150 Moderately hard
150-300 Hard
Over 300 Very hard
7.5 GRAPHIC REPRESENTATIONS

Tables showing results of analyses of chemical quality of groundwater may be difficult to
interpret, particularly when more than a few analyses are involved. To overcome this, graphic
representations are useful for display purposes, for comparing analyses, and for emphasizing
similarities and differences. Graphs can also aid in detecting the mixing of water of different
compositions and in identifying chemical processes occurring as groundwater moves. A vari-
ety of graphic techniques have been developed for showing the major chemical constituents;
some of the more useful graphs are described and illustrated in the following paragraphs.
Figure 7.5.1 illustrates vertical bar graphs, widely used in the United States for portray-
ing chemical quality. Each analysis appears as a vertical bar having a height proportional to the
total concentration of anions or cations, expressed in milliequivalents per liter. The left half of
a bar represents cations, and the right half anions. These segments are divided horizontally to
show the concentrations of major ions or groups of closely related ions and identified by
D Na+K
D Mg
11
basis
obns Figure 7.5.1. Vertical bar graphs
for representing analyses of
groundwater quality (after Hem21 ).
distinctive shading patterns. • The reference number of the analysis is shown at the top of the
bar. This standard bar graph can be modified to include hardness and silica by additional
segments.21
Another method for plotting chemical quality with radiating vectors is shown in Figure
7.5.2. The lengths of the six vectors represent ionic concentrations in milliequivalents per liter.
Pattern diagrams, first suggested by Stiff,41 for representing chemical analyses by four
parallel axes are illustrated by Figure 7.5.3. Concentrations of cations are plotted to the left of
a vertical zero axis and anions to the right; all values are in rnilliequivalents per liter. The
resulting points, when connected, form an irregular polygonal pattern; waters of a similar
quality define a distinctive shape.
Figure 7.5 .4 indicates circular diagrams of water quality with a special scale for the radii
icult to so that the area of a circle is proportional to the total ionic concentration of the analysis. Sec-
graphic tors within a circle show the fractions of the different ions expressed in milliequivalents per
tasizing liter.
jfferent One of the most useful graphs for representing and comparing water quality analyses is
A vari- the trilinear diagram by Piper, 30 shown in Figure 7.5.5. Here cations, expressed as percentages
jtuents; of total cations in milliequivalents per liter, plot as a single point on the left triangle; anions,
Jhs. similarly expressed as percentages of total anions, appear as a point in the right triangle. These
Jortray- two points are then projected into the central diamond-shaped area parallel to the upper edges
al to the
t half of
ltally to 'It should be noted that the water quality diagrams in Figures 7.5 .l-7 .5 .6 all depict the same four water analyses so
ified by as to facilitate comparisons. Also, equivalents per million is identical to meq/1.
Na + K
10-3 ~
fZ;
Cl
Mg
5 0 5
I, , I I Figure 7.5.2. Vector diagrams for represent-
Equivalents per million ing analyses of groundwater quality (after
Hem21 ).
Na + K L...__ .l..-_...L...L.J...U-l-'-l...U- ' -- -'---' Ci

Ca HC03
Mg so4
Fe C03
30 25 20 15 10 5 0 5 10 15 20 25 Figure 7.5.3. Pattern diagrams for representing

21
Cations (equivalents per million) Anions (equivalents per million) analyses of groundwater quality (after Hem ) .
B-1
Mg
1---___.;;;::==c:=q- Na + K
Cl
10-3
6-6
esent-
't.er
50
100
Figure 7 .5.4. Circular diagrams for repre-
Scale of radii senting analyses of groundwater quality
(total of equivalents per million) (after Hem21 ).
of the central area. This single point is thus uniquely related to the total ionic distribution; a cir-
cle can be drawn at this point with its area proportional to the total dissolved solids. The tri-
linear diagram conveniently reveals similarities and differences among groundwater
samples-those with similar qualities will tend to plot together as groups. Further, simple mix-
tures of two source waters can be identified; for example, an analysis of any mixture of two
waters will plot on the straight line AB on the diagram, where A and B are the positions of the
analyses of the two component waters.
In Europe, the semilogarithmic diagram developed by Schoelle~ 8 is employed widely for
comparing groundwater analyses. Here the principal ionic concentrations, expressed in mil-
liequivalents per liter, are plotted on six equally spaced logarithmic scales in the arrangement
shown by Figure 7.5.6. The points thus plotted are joined by straight lines. This type of graph
shows not only the absolute value of each ion but also the concentration differences among
various groundwater analyses. Because of the logarithmic scale, if a straight line joining the
points A and B of two ions in one water sample is parallel to another straight line joining the
points A' and B' of the same two ions in another water sample, the ratio of the ions in both
analyses is equal. •
*It should also be mentioned that the Schoeller diagram can be adapted to detennine the degree of saturation of
CaCOJ and CaS04 in groundwater.'8
0
0
oo 00
0
0
0
0
7.6 F
0 0
oo -
o..-LO..- u) ..--
Milligrams
per liter
Scale of diameters
7.7 BI
60-+-- '0 ~ ---+rS> Figure 7.5.5. Trilinear diagram

Ca Cl for representing analyses of
Cations Anions groundwater quality (after
Percent of total equivalents per million Hem21 ). 7.8 GR
100.0
<D <D <D <D <D <D
50.0 :§ :§ :§ .~ :§ :§
E E E <D <D <D
::I
·c::;
::I
·u;
::I '0
·g <a
.s::.
<ac
:.0 c. 0
Cii <D 0 :2
() c CIJ ()
:; .c
Cl
ell
CIJ «i0
c ~ iii
~
.E
Cii
c.
~
c
<D
Cii
·s>
0"
w
Figure 7.5.6. Schoeller semi-

logarithmic diagram for represent-
ing analyses of groundwater quality
Ca Mg Na Cl 804 HCOs (after Schoelle~ 8 ).
7.8 Groundwater Samples 345
7.6 PHYSICALANALYSIS
In a physical analysis of groundwater, temperature is reported in degrees Celsius and must be
measured immediately after collecting the sample. Color in groundwater may be due to min-
eral or organic matter in solution and is reported in mgll by comparison with standard solu-
tions. Turbidity is a measure of the suspended and colloidal matter in water, such as clay, silt,
organic matter, and microscopic organisms. Measurements are often based on the length of a
light path through the water which just causes the image of a flame of a standard candle to dis-
appear. The natural filtration produced by unconsolidated aquifers largely eliminates turbidity,
but other types of aquifers can produce turbid groundwater. Tastes and odors may be derived
from bacteria, dissolved gases, mineral matter, or phenols. These characteristics are subjective
sensations that can be defined only in terms of the experience of a human being. Quantitative
determinations of taste and odor have been developed based on the maximum degree of dilu-
tion that can be distinguished from taste free and odor free water. 2
7.7 BIOLOGICAL ANALYSIS

As mentioned before, bacteriological analysis is important for detecting biological pollution of
groundwater. Most pathogenic bacteria found in water are indigenous to the intestinal tract of
animals and humans, but isolating them from natural water is difficult in the laboratory.
Because bacteria of the coliform group are relatively easy to isolate and identify, standard tests
to determine their presence or absence in a water sample are taken as a direct indication of the
safety of the water for drinking purposes. Coliform test results are reported as the most prob-
able number (MPN) of coliform group organisms in a given volume of water. By analysis of a
number of separate portions of a water sample, the MPN is computed from probability tables
iagram for this purpose.
of
:r
7.8 GROUNDWATER SAMPLES
In sampling groundwater for quality analysis, Pyrex glass or polyethylene bottles are generally
satisfactory.33 Volumes of one or two liters are usually sufficient for a normal routine chemi-
cal analysis. After rinsing the bottle with the water being sampled, the sample is then collected
and securely sealed. The water should be stored in a cool place and transferred promptly to a
laboratory for analysis. Samples should be taken from a well only after it has been pumped for
some time; otherwise, nonrepresentative samples of stagnant or polluted water may be
obtained. With each sample, a record should be made of well location, depth of sample, size
of casing, date, water temperature, odor, color, turbidity, and operating conditions of the well
immediately prior to the sampling. To analyze for organic and radiological constituents,
special sampling and storage techniques are required.
The shorter the time that elapses between collection of a sample and its analysis, gener-
ally the more reliable the analytic results will be. For certain constituents and physical values,
immediate analysis in the field is required to obtain dependable results; determinations of tem-
perature, pH, alkalinity, and dissolved gases should always be carried out in the field because
changes are inevitable by the time samples reach a laboratory. Storage of samples prior to
analysis can also affect results. Cations such as Fe, Cu, Al, Mn, Cr, and Zn are subject to loss
by adsorption or ion exchange on the walls of glass containers.33
mi- Finally, it should be noted that samples taken from a well penetrating stratified aquifers
Jresent- can yield solute concentrations that differ significantly from those occurring in the individual
:r quality layers. Under these conditions it is possible to obtain water meeting specified quality criteria,
whereas in individual strata, concentrations could be entirely unacceptable.
7.9 WATER QUALITY CRITERIA 7.9.3 Ir

Whether a groundwater of a given quality is suitable for a particular purpose depends on the
criteria or standards of acceptable quality for that use. Quality limits of water supplies for
drinking water, industrial purposes, and irrigation apply to groundwater because of its exten-
sive development for these purposes. Table 7.9.llists federal laws in the United States that reg-
ulate the quality of water.
7.9.1 Drinking Water Standards

Most drinking water supplies in the United States conform to standards established by the
U.S. Environmental Protection Agency. The National Primary Drinking Water Standards Max-
imum Contaminant Levels (MCLs) and the Secondary Maximum Contaminant Levels
(SMCLs) are provided at the following Internet address: http://www.epa.gov/safewater.
7.9.2 Industrial Water Criteria

It should be apparent that the quality requirements of waters used in different industrial
processes vary widely. Thus, makeup water for high-pressure boilers must meet extremely
exacting criteria, whereas water of as low a quality as seawater can be satisfactorily employed
for cooling of condensers. Even within each industry, fixed criteria cannot be established;
instead, only recommended limiting values or ranges can be stated. 3• 27 Salinity, hardness, and
silica are three parameters that usually are important for industrial water.
Of almost equal importance for industrial purposes as quality of a water supply is the rel-
ative constancy of the various constituents. It is often possible to treat a poor quality water or
adapt to it so that it is suitable for a given process, but if the quality fluctuates widely, contin-
ued attention and expense may be involved. Fluctuations of water temperature can be equally
troublesome. From this standpoint, groundwater supplies are preferred to surface water sup-
plies, which commonly display seasonal variations in chemical and physical quality. As a
result, an adequate groundwater supply of suitable quality often becomes a primary consider-
ation in selecting new industrial plant locations.
Table 7.9.1 Laws Regulating Water Quality
Federal laws in the United States that regulate the quality of water
National Environmental Policy Act of 1969 (P.L. 91-190)
Federal Water Pollution Control Act of 1972 (P.L. 92-500)
Clean Water Act Amendments of 1977 (P.L. 95-217)
Safe Drinking Water Act of 1974 (P.L. 93-523) and amendments
Resource Conservation and Recovery Act of 1976 (RCRA) (P.L. 94-580)
Comprehensive Environmental Response, Compensation and Liability Act of 1980 (CERCLA)
(P.L. 96-510)
Superfund Amendments and Reauthorization Act of 1986 (SARA)
Surface Minlng Control and Reclamation Act (SMCRA) (P.L. 95-87)
Uranium Mill Tailings Radiation and Control Act of 1978 (UMTRCA) (P.L. 95-604 as amended by
P.L. 95-106 and P.L. 97-415)
Toxic Substances Control Act 1988 (TOSCA) (P.L. 94-469 as amended by P.L. 97-129)
Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) (P.L. 92-516 as amended by P.L. 94-140,
P.L. 95-396, P.L. 96-539, and P.L. 98-201)
Oil Pollution Act (OPA) of 1990
7.9 Water Quality Criteria 347
7.9.3 Irrigation Water Criteria

:on the The suitability of groundwater for irrigation is contingent on the effects of the mineral con-
lies for stituents of the water on both the plant and the soil. 34· 48 Salts may harm plant growth physi-
; exten- cally by limiting the uptake of water through modification of osmotic processes, or chemically
hat reg- by metabolic reactions, such as those caused by toxic constituents. Effects of salts on soils,
causing changes in soil structure, permeability, and aeration, indirectly affect plant growth.
Specific limits of permissible salt concentrations for irrigation water cannot be stated because
of the wide variations in salinity tolerance among different plants; however, field-plot studies
of crops grown on soils that are artificially adjusted to various salinity levels provide valuable
. by the information relating to salt tolerance.
is Max- An important factor realting crop growth to water quality is drainage. If a soil is open and
Levels well drained, crops may be grown on it with the application of generous amounts of saline
water; but, on the other hand, a poorly drained area combined with application of good-qual-
ity water may fail to produce as satisfactory a crop. Poor drainage permits salt concentrations
in the root zone to build up to toxic proportions. Today, the necessity of adequate drainage is
clearly recognized in order to maintain a favorable salt balance-where the total dissolved
dustrial solids brought to the land annually by irrigation water is less than the total solids carried away
tremely annually by drainage water. It is believed that this factor accounted for the failure of many of
nployed the elaborate irrigation systems of historical times.
blished; In place of rigid limits of salinity for irrigation water, quality is commonly expressed by
ess, and classes of relative suitability. Most classification systems include limits on specific con-
ductance (expressing total dissolved solids), sodium content, and boron concentration.
; the rei- Sodium concentration is important in classifying an irrigation water because sodium
water or reacts with soil to reduce its permeability (see following section). Soils containing a large pro-
, contin- portion of sodium with carbonate as the predominant anion are termed alkali soils; those with
·equally chloride or sulfate as the predominant anion are saline soils. Ordinarily, either type of sodium-
1ter sup- saturated soil will support little or no plant growth. Sodium content is usually expressed in
.ty. As a terms of percent sodium (also known as sodium percentage and soluble-sodium percentage),
onsider- defined by
%Na= (Na+K)lOO (7.9.1)

Ca+Mg+Na+K
where all ionic concentrations are expressed in milliequivalents per liter. The Salinity Labora-
tory of the U.S. Department of Agriculture34 recommends the sodium adsorption ratio (SAR)
because of its direct relation to the adsorption of sodium by soil. It is defined by
SAR= Na (7.9.2)
~(Ca+Mgj2
where the concentrations of the constituents are expressed in milliequivalents per liter.
,) Boron is necessary in very small quantities for normal growth of all plants, but in larger
concentrations it becomes toxic. Quantities needed vary with the crop type; sensitive crops
require minimum amounts whereas tolerant crops will reach maximum growth on several
times these concentrations.
led by In investigations relating to quality of water for irrigation and the salinity of the soil solu-
tion, particularly regarding sampling programs, it is important to take cognizance of the salt
distribution within the soil. As an illustration, the salt distribution under irrigated cotton plants
. 94-140,
is shown in Figure 7.9.1. It is apparent that the leaching effect of the irrigation water in the fur-
row, together with the movement toward the plants on the ridges, creates wide variations of salt
concentration within short distances.
~
<ll
Q) 25
E
E
~
£c. 50 Figure 7.9.1. Salt distribution
Ql under furrow-irrigated cotton for
Cl
soil initially salinized to 0.2 per-
cent salt (3, 100 jlS/cm) and irri-
Conductivity of extract from saturated soil (p.S/cm) gated with water of medium
salinity (after Wadleigh and
Fireman;44 reproduced by per-
~ Less than 1,000 [ [ ] 2,000 to 5,000 - 6,000 to 10,000 - Above 50,000
mission of the Soil Science
~ 1,000 to 2,000 5,000 to 6,000 - 10,000 to 50,000 Society of America).
7.10 CHANGES IN CHEMICAL COMPOSITION

As groundwater moves underground, it tends to develop a chemical equilibrium by chemical
reactions with its environment Effects of some equilibria in groundwater have important
applications-for example, artificial recharge, movement of pollutants, and clogging of wells.
Chemical precipitation may remove ions in solution by fanning insoluble compounds.
Precipitation of calcium carbonate and release of dissolved carbon dioxide may result from a
decrease in pressure and/or an increase in temperature. Ferrous iron in solution oxidizes on
exposure to air and is deposited as ferric hydroxide.22
Ion exchange involves the replacement of ions adsorbed on the surface of fine-grained
materials in aquifers by ions in solution. Because the exchange involves principally cations 7.11 Dl
(sodium, calcium, and magnesium), the process is known as base, or cation, exchange. The
direction of the exchange is toward an equilibrium of bases present in the water and on the
finer materials of the aquifer. Base exchange is known to soften groundwater naturally and to
produce, in coastal regions where seawater has entered an aquifer, groundwater having a qual-
ity other than a simple mixture of the two source waters (see Chapter 14).
Base exchange causes changes in the physical properties of soils. When high-sodium
water is applied to a soil, the number of sodium ions combined with the soil increases, while
an equivalent quantity of calcium or other ions is displaced. These reactions cause defloccula-
tion and reduction of permeability. In the opposite case, where calcium is the dominant cation,
the exchange occurs in the reverse direction, creating a flocculated and more penneable soil.
The advantage of adding gypsum (CaSOJ to a soil is that by base exchange the soil texture and
drainability can be improved.
Chemical reduction of oxidized sulfur ions to sulfate ions or to the sulfide state occurs fre-
quently in groundwater. The reaction is believed to take place in the presence of certain bac-
teria_21 Typically, waters experiencing sulfate reduction have high bicarbonate and carbon
dioxide content and contain hydrogen sulfide.
The equilibrium achieved by the various chemical reactions described above tends to pro-
duce a quality that remains stable with time because of the slow movement of groundwater and
its long residence time within a given geologic fonnation. In general, quality variations are
more noticeable in shallow aquifers where seasonal variations in recharge and discharge cre-
ate corresponding fluctuations in salinity. Interestingly, freezing of shallow groundwater in
7.!1 Dissolved Gases 349
:=- 120
0>
g
1/1
::g
0
1/1
"0
~
ution 01/1
<ll
tton for '6
l.2 per- ~
nd irri-
{:. 40 Figure 7.10.1. Variation in
lrn
chemical quality of shallow
nd groundwater at a site in
·per- northwestern Alaska due to
:e 0 j j seasonal effects of freezing
1961 1962 (after Fuelner and
Time Schupp'\
arctic regions can also cause seasonal changes in chemical quality, as shown in Figure 7.1 0.1.
Here a marked increase in salinity during winter months results from freezing, which not only
temical
reduces the diluting effect of recharge from precipitation, but also selectively concentrates the
.portant
mineral content in the unfrozen groundwater.
fwells.
Finally, it should be noted that human acts can markedly change the natural quality of
)ounds.
groundwater, usually for the worse. Figures 7.10.2-7 .10.7 provide examples of these changes .
. from a
The resulting degradation of groundwater has become an important international concern;
izes on
Chapter 8 treats the subject of groundwater pollution.
grained
cations 7.11 DISSOLVED GASES
ge. The
Although it is not generally recognized, most groundwater contains dissolved gases derived
l on the
from natural sources. Those involved in the normal geochemical cycle of groundwater include
:y and to
the atmospheric gases: carbon dioxide (C02), oxygen (02), and nitrogen (N2). Others derived
; a qual-
from underground biochemical processes include the flammable gases methane (CH4) and
hydrogen sulfide (H2S). With its distinctive rotten-egg odor, hydrogen sulfide is readily
-sodium
detected at concentrations of less than 1 mg/1 and rarely accumulates to dangerous proportions.
s, while
Methane, which is colorless, tasteless, and odorless, occurs frequently in groundwater and
loccula-
can be a much more serious problem. The gas is a decomposition product of buried plant and
t cation,
animal matter in unconsolidated and geologically young deposits.* The minimum concentration
ble soil.
of methane in water sufficient to produce an explosive methane-air mixture above the water
ture and
from which it bubbles out of solution depends on the temperature, pressure, quantity of water
pumped, and volume of air into which the gas evolves. Theoretically, water containing as little
;urs fre-
as 1 to 2 mg/1 of methane can produce an explosion in a poorly ventilated air space. Fires and
ain bac-
explosions in well pits, basements, and water tanks have occurred from methane emitted by
. carbon
groundwater; safety measures include analyses to detect the presence of the gas, aeration of
water before use, and adequate ventilation where the water is being used. It is possible for peo-
s to pro-
ple to suffocate in dug wells and pump pits where high methane concentrations form.
ater and
ions are
trge ere- 'For example, a buried oak log, only I ft in diameter and 5 ft long, can generate enough methane when it decom-
water in poses underground to fonn an explosive mixture in 400 to 1200 m·' of air.4
(b) NITROGEN BUDGET

Precipitation
Cropland 31%
Cropland Forest Wetland Rhode
• Forest I
fertilizer 46% 1 >52% uptake and River
.Wetland I
and Mudflats Estuary
O subtidal t t
harvest •
I
1
I
denitrification 41°45'
mudflats
Cropland Forest Wetland Rhode

fertilizer 45% Sediment • and River
:
1
Mudflats Estuary
Exp1
transport :
OS
t t ----!--+
I
55%
.
___.__
I Sediment
cteposi~on
Nitn
harvest '
48%
North
Figure 7.10.2. Forests and wetlands separate cropland from streams in the Rhode River watershed in Maryland (a). More than half
of the nitrogen applied to cropland is transported by groundwater toward riparian forests and wetlands (b). More than half of the
total phosphorus applied to cropland is transported by streams to wetlands and mudflats, where most is deposited in sediments (c)
(modified from Correll, 11 as presented in Winter et al.51 ). (Reprinted by perrnissitm of the Estuarine Research Federation)
•.
11,000 r--:-~.....,.-,.--...,...........,....--,.--...,....-~---:"-:---;----, 194,205 Figure 7.

Maryland
Dissolved -Ground-water ~- 158,895 ments ne2
~ 9,900 ~ fromJ. K
/calcium mass inflow
l "c.
I
~ 8,800 (J
Zl
E 88,275 ~
~ 7.12 Tl
~ 7,700 .~
0
0
!l
6,600 ·-·
17,655
i
~
0 Figure 7.10.3. A large input of groundwater dur-
5, 500 ...._~_,_____._.:...._....:...........:..._:.--"-'---'----'----'--'---'---' ~17,655
.... .- ...... ing spring supplies the annual input of calcium to
Nevins Lake in the Upper Peninsula of Michigan
(modified from Krabbenhoft, 24 as presented in
Winter et al. 51 ).
4
• Well and concentration
of dissolved organic
\hode 1 carbon, in milligrams
River • per liter
stuary
4
Tidal • Figure 7.10.4. Seepage from lakes in the sand-
hills of Nebraska causes plumes of dissolved
<change
<1% organic carbon to be present in groundwater on
the downgradient sides of the lakes (modified
0 1 MILE from 1. W. LaBaugh,23 as presented in Winter et
al. 51). (Reprinted with permission of Elsevier
Science-NL, Amsterdam, The Netherlands)
Rhode
River Explanation
Estuary
Nitrate (N03-), In Milligrams Per Liter as N
Greater than 10 Recharge
Sto 10 N03- = 3 to 20 milligrams per liter
Tidal Chesterville Branch
lXChange 2to5 615 N(N03l = 2 to 6 per mil
N03- =9 to 10 milligrams per liter
.. <7% Less than 2 Morgan Creek
N03- = 2 to 3 milligrams per liter 615 N(N03l c 4 to 5 per mil
615 N(NO:Jl ~ 7 to 10 per mil South .
North
than half
of the
ents (c)
NOJ- ~ omtiligrams per titer··

615 N(exc~Ni):.
. ,~
ii~'s
,:
Per mil.
.' . . . .. '· Gr-oUndwater . '
... ' > ..__ •• ~ • ' " •
· N03-: • ·3 to 5 milligrams per liter

· ·. 61.5.,;i~N03l• .4 to 5 per mil
Figure 7.10.5. Denitrification had a greater effect on groundwater discharging to Morgan Creek than to Chesterville Branch in
Maryland because a larger fraction of the local flow system discharging to Morgan Creek penetrated the reduced calcareous sedi-
ments near or below the bottom of the Aquia Formation than the flow system associated with the Chesterville Branch (modified
from J. K. Bolke,7 as presented in Winter et al.51 ).
7.12 TEMPERATURE
Variations in solar energy received at the earth's surface create periodicities, both diurnal and
annual, in temperature below ground surface. The insulating qualities of the earth's crust rap-
idly damp the large temperature range found at ground surface so that only shallow ground-
water dur- water displays any appreciable fluctuation in temperature. 19 In marked contrast to the large
::alcium to seasonal variation of surface water temperatures (except in tropical regions), groundwater
Michigan temperatures tend to remain relatively constant-an important advantage for drinking water
:nted in and industrial uses.
(a) Sampling period February 20 to 22

690
!
~ 660
670 9
:I
s
< 650
~0
630
(b) Sampllng period March 20 to 22
Water !able
Figure 7.10.6. Concentrations of atrazine increased in

the Cedar River in Iowa following applications of the
chemical on agricultural areas upstream from a study
site. During high streamflow (b), the contaminated
<EXPLANATION river water moved into the alluvial aquifer as bank
-=-o. 7- Une of equal atrazlne concentration, • Well location storage, contaminating groundwater. After the river
: ln~lcrograms per liter · level declined (c), part of the contaminated ground
Direction .or
·, 0.21) CoQtentratlon of·alrazine In river at time groundwater water returned to the river (modified from Squillace, 39
of sampling, In mlaograms per liter · \ flow as presented in Winter eta!.5 1) .
An example of the variability of groundwater temperature in relation to that of a surface

water body is shown by Figure 7. 12. I for the Schenectady, New York, well field adjoining the
Mohawk River. The annual variation in groundwater temperature is only a small fraction of
that observed in the river water. Furthermore, the tongue of larger temperature variation point-
ing toward the well field indicates the principal flow path of groundwater from the river to the
Schenectady well field through a permeable zone of the shallow aquifer. •
' It should be noted generally that the velocity of groundwater can have an appreciable influence on the distribution 0 10(
of underground temperatures. Where sufficient field data are available, it may be possible to calculate flow veloci· L..l.
ties from temperature measurements and even to determine aquifer permeability from a combination of water level M
and temperature measurements. 37• 40
!ased in
of th~ Figure 7.10.7. Pumping of municipal water-supply
1 study wells near Lincoln, Nebraska, has induced Platte
ated River water contaminated with atrazine to flow into
•ank the aquifer. Distances shown are from river to moni-
river toring well (modified from Duncan, 12 as presented in
mnd Winter et al51 ). (Used with permission)
illace/9
. surface
ning the
1ction of
m point-
·er to the
Figure 7.12.1. Map of annual variation in

-90c:t groundwater temperature near Mohawk
4-qq
Of River, New York. Numbered lines show
slribution 0 100 200 300 ~1161.
w veloci- I I I I r locations of equal annual groundwater tem-
laler level Meters perature change in •c; annual river water
change was 25°C (after Winslow50).
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I " I II I
Figure 7.12.2. Monthly groundwater

temperatures in two nearby wells on
Long Island, New York (after New York
State Water Resources Comm BulL 55).
Assuming that ground surface temperature is a sinusoidal function of time, the tempera-
ture Tz at a depth z can be given by
rz = to + Ae-z..fnia~ sm
T 27tt- z-v 1t, cx:r
. (-
't
r::;=) (7.12.1)
where T0 is the mean ground surface temperature, A is the amplitude of the surface tempera-
ture variation, 'tis the oscillation period (one day or one year), a. is the thermal diffusivity of
the subsurface material (approximately 0.005 cm2/sec), and tis time. The depth of nearly uni-
fonn temperature occurs at about 10 m in the tropics and increases to about 20m in polar
regions,47 although influences such as rock type, elevation, precipitation, cloudiness, and wind Figun
can produce significant local deviations. illustrative of annual fluctuations and the effect of 25m (
depth are the temperatures in two nearby shallow wells shown in Figure 7.12.2.
Shallow groundwater temperatures can also be influenced by the type of overlying surface
environment. Thus, a study on Long Island, New York, 31 showed that the mean annual tem-
perature and the annual range in temperature of groundwater were larger beneath cleared areas
than beneath wooded areas. The difference can be attributed to the absence of shade and the
lack of an insulating layer of organic material on the ground in the cleared areas.
Below the zone of surface influence, groundwater temperatures increase approximately
2.9°C for each 100m of depth in accordance with the geothennal gradient of the earth's crust.*
Thus, groundwater pumped from deep wells is appreciably warmer than shallow groundwater.
For example, if shallow groundwater has a temperature of l5°C, that obtained from an aquifer
at 500 m might be expected to have a temperature near 30°C.
In the United States it has been found that the temperature of groundwater occurring at a
depth of 10 to 20m will generally be about 1o to 2°C higher than the local mean annual air
temperature. 10 On this basis a groundwater temperature map, as shown in Figure 7.12.3, can
be constructed from climatological data. As extremes, groundwater temperatures may vary
from below freezing to above the boiling point in geothennal areas and in superheated water
emerging from geysers.
7.13 SALINE GROUNDWATER

Many groundwaters contain dissolved salts in such concentrations as to make them unusable
for ordinary water supply purposes. Saline groundwater is a general term referring to any
groundwater containing more than 1,000 mg/1 total dissolved solids. Various classification
·Most geothermal gradients measured in the United States fall within the range 1.8-3.6° C per 100m. The heat flux
from the interior of the earth is estimated to average 1.3 x 10-6 cal/cm 2sec.
7.13 Saline Groundwater 355
•rk
)5).
pera-
12.1)
tpera-
ity of
y um-
polar
~wind Figure 7.12.3. Approximate temperature of groundwater, in degrees Celsius, in the conterminous United States at depths of 10 to
ect of 25 m (Heath20). .
urface schemes based on dissolved solids have been proposed; the simplicity of the one shown in
1 tem- Table 7.13.1 makes it particularly convenient.
1areas
nd the Thble 7.13.1 Classification of Saline Groundwater (after Carroll9)
Total dissolved
.nately
solids (mg/1)
crust. •
Iwater. Fresh water 0-1,000
1quifer Brackish water 1,000-10,000
Saline water 10,000-100,000
ng at a Brine >100,000
mal air
..3, can
A map of the United States showing the minimum depth to aquifers containing ground-
ty vary
water exceeding 1,000 mg/1 is given in Figure 7.13.1. Approximately two-thirds of the coun-
j water
try is known to contain such waters; blank areas indicate areas where either well yields are less
than 40m3/day or no data on saline groundwater are available. Four types of occurrence are
recognized: connate water, intruded seawater (see Chapter 14), water salinized by contact
with soluble salts in the fonnation where it is situated, and water in regions with shallow water
tables where evapotranspiration concentrates the salts in solution.
nusable Although saline groundwater has traditionally been regarded as an undesirable resource,
to any modern technological advances may reverse this role. Rapid advances in desalination
fication techniques suggest that saline groundwater may be a potentially important water supply source
where shortages are imminent. Industrial use of saline groundwater for cooling purposes has yet
to be exploited fully; high costs, involving deep wells, corrosion control, and disposal, have hin-
heat flux
dered such developments. Saline aquifers also serve useful purposes for temporary storage of
fresh water and energy in the form of hot water and for disposal of wastewater (see Chapter 13).
7.10.
usgs.
of th<
7.10.:
Soutl
Prog1
progr
REF
1. AJt.
Water
usgs.g•
2. Am
Poltuti
and~
Depth in meter~
1200p
[=:J Less than 150 3. Am

Waste 1
~150-300 4. Anc
!LY/\)] More th~n 300 413-41
5. Bac.
CJ Inadequate information Hydro~
49-109
14
Figure 7.13.1. Depth to saline groundwater in the United States (after Feth et al. ). 6. Bak·
NewY<
7. BoU
chemic.
contam
Marylru
PROBLEMS 8. CaqJ
United.
7.1.1 Develop a report on the (proposed) standard for arsenic. Dis- the contaminant, and the MCL. The Web site is http://www.epa. logical
Denver,
cuss the technologies and costs for removal of arsenic. What gov/safewater/standards.html.
importance does this have in groundwater studies? You may want 9. Cam
7.9.2 Develop a short report on how the Federal Water Pollution Review,
to start with the following Web site: http://www.epa.gov/safe- Control Act of 1972 relates to the water quality of groundwater. 10. Col
water/arsenic.html.
7.9.3 Develop a short report on how the Resource and Recovery the Unit
7.1.2 Develop a report on the connection between surface-water 97-104,
Act of 1976 relates to the water quality of groundwater.
quality and groundwater quality in a karst aquifer. You might want 11. Con
to start at the Web site for the U.S. Geological Circular 1186. 1 7.9.4 Develop a short report on how the Surface Mining Control scape-1
7.4.1 In U.S. Geological Survey Water Resources Investigations and Reclamation Act relates to the water quality of groundwater. ent tranS)
Report 98-4133, Estimating Ground-Water Exchange with Lakes 7.9.5 Develop a short report on how the Toxic Substance Control 12. Dun
Using Water-Budget and Chemical Mass-Balance Approaches for Act relates to the water quality of groundwater. used as:
Ten Lakes in Ridge Areas of Polk and Highland Counties, Florida, v. 11, no
7.9.6 Develop a short report on how the Superfund Amendments
by Sacks et al,35 a chemical mass balance was used to estimate the 13. Erik
and Reauthorization Act relates to the water quality of ground- circulati<
groundwater exchange with lakes. This report is available at
water. 14. Fe!h
http://fl.water.usgs.gov. Develop a discussion of the method used
to perform the chemical mass balance. How do the results com- 7.9.10 Discuss the laws and regulations that relate to groundwater Showing
pare with using the water-budget analysis? quantity and quality in your state or region. Moretha
veyHydr
7.9.1 Go to the U.S. EPA Web site for Drinking Water Standards 7.10.1 Go to the U.S. Geological Web site http://water.usgs. 15. Feul1
and print out the standards. Discuss one organic and one inor- gov/pubs. Find a water quality study on an area near you and sum· Quality~
ganic chemical, defining the health effects of each, the source of marize the water quality of the area. cal Su111e
References 357
7.10.2 Go to the U.S. Geological Survey Web site http://water. 7.10.4 Develop a report on the water quality factors affecting
usgs.gov/pubs/circ/circl207/index.htrnJ and develop a discussion groundwater sustainability. You might want to start your search
of the water quality in southern Florida. for information in U.S. Geological Survey Circular 1186, Sus-
7.10.3 Discuss the concentrations of total phosphrous (TP) at the tainability of Ground-Water Resources, found at
Southern Florida National Water Quality Assessment (NAWQA) http://water.usgs.gov/pubs/circ/circll86. 1
Program sites. What is the major source(s) of the high TP? The
program Web site is http://water.usgs.gov/nawqa.
REFERENCES
l. Alley, W. M., T. E. Reilly, and 0. L. Franke, Susrainabiliry ofGrowui- 16. Focazio, M. J., A. H. Welch, S. A. Watkins, D. R. Helsel, and M.A.
Water Resources, U.S. Geological Survey Circular 1186, http://water. Hom, A Retrospective Analysis on the Occurrence ofArsenic in Ground
usgs.gov/pubs/circ/circll86, U.S. Geological Survey, Denver, CO, 1999. Water Resources of the United States and Limitations in Drinking-Water
2. Amer. Public Health Assoc., Amer. Water Works Assoc., and Water Resources of the United States and Limitations in Drinking· Water-Supply
Pollution Control Fed., Standard Methods for the Examination of Water Characterizations, U.S. Geological Survey Water-Resources lnvestiga·
and Wastewater, 14th ed., Amer. Public Health Assoc., Washington, DC, rions Report 99-4279, 1999.
1200 pp., 1975. 17. Garrels, R. M. and F. T. MacKenzie, Origin of the chemical composi-
3. Amer. Soc. Testing Matis., Manual on Irulusrrial Warer and Industrial tions of some springs and lakes, Advances in Chemistry Series, No. 67,
Waste Water, 2nd ed., Philadelphia, 992 pp., 1966. American Chemical Society, Chapter 10, pp. 222-242, 1967.
4. Anon., Gas in ground water, Jour. Amer. Water Works Assoc., v. 61, pp. 18. Graham, M. J., J. A. Thomas, and F. B. Melting, Groundwater, in Water
413-414, 1969. Resources Handbook. L. W. Mays, ed., McGraw-Hill, New York. 1996.
5. Back, W., and B. B. Hanshaw, Chemical geohydrology, in Advances in 19. Heath, R. C., Seasonal Temperature Fluctuations in Surficial Sarul
Hydroscience, V. T. Chow, ed., v. 2, Academic Press, New York, pp. near Albany, New York, U.S. Geological Survey Professional Paper415-D,
49-109, 1965. pp. 204-208, 1964.
6. Baker, M. N., The Quest for Pure Water, Amer. Water Works Assoc., 20. Heath, R. C., Basic Ground-Water Hydrology, U.S. Geological Survey
New York, 527 pp., 1948. Water-Supply Paper 2220, http://water.usgs.gov/pubs/wsp/wsp2220, Den-
7. Boike, J. K., and J. M. Denver, Combined use of ground-water dating, ver, CO, 1998.
chemical, and isotopic analyses to resolve the history and fate of nitrate 21. Hem, J. D., Study aru1 Interpretatian ofthe Chemical Characteristics
contamination in two agricultural watersheds, Atlantic coastal plain, of Natural Water, 2nd ed., U.S. Geological Survey Water-Supply Paper
Maryland, Water Resources Research, v. 31, no. 9, pp. 2319-2337, 1995. 1473,363 pp., 1970.
8. Ca.rpentar, M. C., South-Central Ariwna. in Land Subsidence in the 22. Hem, J. D., et al., Chemistry of Iron in Natural Water, U.S. Geologi-
United States, D. Galloway, D. Jones, and S. Ingebritsen, eds., U.S. Geo- cal Survey Water-Supply Paper 1459, 268 pp., 1%2.
logical Survey Circular 1182, http://water.usgs.gov/pubs/circ/circ1182, 23. LaBaugh, J. W., Limnological characteristics of selected lakes in the
NWW.epa. Denver, CO, 1999. Nebraska sandhilis, U.S.A., and their relation to chemical characteristics
9. Carroll, D., Rainwater as a Chemical Agent of Geologic Processes-A of adjacent groundwater, Joumal of Hydrology, v. 86, pp. 279-298, Else-
Pollution Review, U.S. Geological Survey Water-Supply Paper 1535-G, 18 pp., 1962. vier Science-NL, Amsterdam, The Netherlands, 1986.
ndwater. 10. Collins, W. D., Temperature of Water Available for Industrial Use in 24. Krabbenhoft, D. P., and Webster, K. E., Transient hydrogeological
the United Siates, U.S. Geological Survey Water-Supply Paper 520-F, pp. controls on the chemistry of a seepage lake, Water Resources Research.
Recovery ~3l,oo.9,pp. 2295-2305, 1995.
97-104, 1925.
II. Correll, D. L., T. E. Jordan, and D.E. Weller, Nutrient flux in a land- 25. Loewengart, S., Airborne salts-The major source of the salinity of
lg Control scape-Effects of coastal land use and terrestial community mosaic on nutri- waters in Israel, Bull. Research Council Israel, v. JOG, pp. 183-206, 1961.
•undwater. ent transport to coastal waters, Estuaries, v. 15, no. 4, pp. 431-442, 1992. 26. Logan, J., Estimation of electrical conductivity from chemical analy-
12. Duncan, D., D. T. Pederson, T. R. Shepherd, and J.D. Carr, Atrazine sis of natural waters, Jour. Geophysical Research, v. 66, pp. 2479-2483,
,ceControl 1961.
used as a tracer of induced recharge, Grourul Water Monitoring Review,
v. l1,no. 4,pp. 144-150,1991. 27. McKee, J. E., and H. W. Wolf, eds., Water quality criteria, Pub!. no.
nendments 13. Eriksson, E., Atmospheric transport of oceanic constituents in their 3-A, California State Water Resources Control Board, Sacramento, 548
of ground- circulation in nature, Tellus, v. II. pp. 1-72, 1959. pp., 1963.
14. Fetb, J. H., eta!., Preliminary Map ofthe Conterminous United States 28. National Research Council, Arsenic in Drinking Water, National
Showing Depth to and Quality of Shallowest Ground Water Containing Academy Press, Washington, DC, 1999.
roundwater
More than /000 Parts per Million Dissolved Solids, U.S. Geological Sur- 29. Nordstrom, D. K., J. W. Ball, R. J. Donahoe, and D. Whittemore,
vey Hydrologic lrrv. Atlas HA-199, 31 pp., 1965. Groundwater chemistry and water-rock interactions stripa, Geochirn. Cos·
·water. usgs. 15. Feulner, A. J., and R. G. Schupp, Seasonal Changes in the Chemical mochirn. Acta, v. 53, pp. 1727-1740, 1989.
ou and sum- Quality of Shallow Ground Water in Northwestern Alaska, U.S. Geologi- 30. Piper, A.M., A graphic procedure in the geochemical interpretation of
cal Survey Professwnal Paper 475-B, pp. 189-191, I %3. water-analyses, Trans. Amer. Geophysical Unwn, v. 25, pp. 914-928, 1944.
;j
31. Pluhowski, E. J., and I. H. Kantrowitz, Influence of Land-Suiface 42. Taylor, C. J., and W. M. Alley, Ground- Water-Level Monitoring and Cl
Conditions on Ground-Water Temperatures in Southwestern Suffolk rhe lmponance of Long-Tenn Water-Level Data, U.S. Geological Survey
County, Long Island, New York, U.S. Geological Survey Professional Circular 1217. http://water.usgs.gov/pubs/circ/circ 1217, Denver, CO,
Paper475-B, pp. 186-188, 1963. 2001.
32. Plummer, L. N., J. F Busby, R. W. Lee, and B. B. Hanshaw, Geo· 43. Thomas, J. M., A. H. Welch, and A. M. Preissler, Geochemical evo-
chemical modeling of the Madison aquifer in parts of Montana, lution of ground water in Smith Creek Valley-A hydrologic ally closed
Wyoming, and South Dakota, Water Resources Research, v. 26, no. 9, basin in Central Nevada, U.S.A., App/. Geochem, v. 4, pp. 493-510,
pp. 1981-2014, 1990. 1989.
33. Rainwater, F. H., and L. L. Thatcher, Methods for Collection and 44. Wadleigh, C. H., and M. Fireman, Salt distribution under furrow and
Analysis of Water Samples, U.S. Geological Survey Water-Supply f'aper basin inigated cotton and its effect on water removal, Soil Sci. Soc. Amer.
1454, 301 pp., 1960. Proc., v. l3, pp. 527-530, 1948.
34. Richards, L. A., ed., Diagnosis and lmprovemenr ofSaline and Alkali 45. Welch, A. H., S. A. Watkins, D. R. Helsel, and M. J. Focazio, Arsenic
Soils, Agric. Handbook 60, U.S. Dept. Agric., Washington, DC, 160 pp., in Ground-Water Resources ofthe United States, U.S. Geological Survey
1954. Fact Sheet FS-063-00, May 2000.
35. Sacks, L.A., A. Swancar, and T. M. Lee, Estimating Ground-Water 46. White, A. F., Geochemistry of Groundwarer Associated with Tujfa-
Exchange with Lakes Using Warer-Budget and Chemical Mass-Balmtce ceous Rocks, Oasis Valley, Nevada, U.S. Geological Survey Professional
Approaches for Ten Lakes in Ridge Areas of Polk and Highlands Coun- Paper 712-E, Washington, DC, 1979.
ties, Florida, U.S. Geological Survey Water Resources Investigations 47. White, D. E., et al., Data of Geochemistry-Chemical Composition
Report 98-4133, Tallahassee, FL, 1998. ofSubsurface Waters, 6th ed., U.S. Geological Survey Professional Paper
36. Sawyer, C. N., and P. L. McCarty, Chemistry for Sanitary Engineers, 440-F, 67 pp., 1963.
2nd ed., McGraw-Hill, New York, 518 pp., 1967. 48. Wilcox, L. V., Classification and Use of Irrigation Waters, U.S. Dept.
37. Schneider, R., An Applicatian ofThermometry to the Study ofGround Agric. Circular 969, Washington, DC, 19 pp., 1955.
Water, U.S. Geological Survey Water-Supply Paper 1544-B, 16 pp., 1962. 49. Winograd, I. J., and F. J. Pearson, Jr., Major Carbon-14 anomaly in a
38. Schoeller, H., Les eaux souterraines, Masson & Cie, Paris, 642 pp., region carbonate aquifer: Possible evidence for megascale channeling,
1962. south Central Great Basin, Water Resources Research. v. 12, no. 6,
39. Squillace, P. J., E. M. Thurman, and E. T. Furlong, Groundwater as a pp. 1125-1143, 1976.
oonpoint source of atrazine and deethylatrazine in a river during base 50. Winslow, J. D., Effect of Stream Infiltration on Ground-Water Tem-
flow conditions, Water Resources Research, v. 29, no. 6, pp. 1719-1729, peratures near Schenectady, N.Y., U.S. Geological Survey Professional
1993. Paper 450-C, pp. 125-128, 1962.
40. Stallman, R. W., Computation of Ground-Water Velocity from Tem- 51. Winter, T. C., J. W. Harvey, 0. L. Franke, and W. M. Alley, Ground- 8.1 PC
perature Data, U.S. Geological Survey Water-Supply Paper 1544-H, pp. water and Suiface Water: A Single Source, US. Geological Survey Cir-
36-46, 1963. cular 1139, http://water.usgs.gov/pubs/circ/circll39, Denver, CO, 1998.
41. Stiff, H. A. Jr., The interpretation of chemical water analysis by 52. World Health Organization, Arsertic in drinking water, Fact sheet
means of patterns, Jour. Petr. Technology, v. 3, no. 10, pp. 15-17, 1951. 210, http://www. who.intli.nf-fs/en/fact21 O.htrii.l, 1999.
gand
ChapterS
;urvey
·,CO,
ll eva-
closed
3-510,
)Wand
.Amer. Pollution of Groundwater
~rsenic
Survey
i Tuffa-
!ssionLll
Groundwater pollution may be defined as the artificially induced degradation of natural
oosition groundwater quality. Pollution can impair the use of water and can create hazards to public
11 Paper health through toxicity or the spread of disease. • Most pollution originates from the disposal
of wastewater following the use of water for any of a wide variety of purposes. Thus, a large
.S. Dept. number of sources and causes can modify groundwater quality, ranging from septic tanks to
irrigated agriculture. In contrast with surface water pollution, subsurface pollution is difficult
naly in a to detect, and is even more difficult to control, and may persist for decades. With the growing
mneling,
l. no. 6,
recognition of the importance of underground water resources, efforts are increasing to pre-
vent, reduce, and eliminate groundwater pollution.
1/er Tem-
-{essional
Ground· 8.1 POLLUTION IN RELATION TO WATER USE

!rvey Cir-
The possible pollutants in groundwater are virtually limitless. The sources and causes of
~0. 1998.
<act sheet
groundwater pollution are closely associated with human use of water. A complex an.d inter-
related series of modifications to natural water quality is created by the diversity of human
activities impinging on the hydrologic cycle.
The principal sources and causes of groundwater pollution can be categorized-
municipal, industrial, agricultural, and miscellaneous. Most pollution stems from disposal of
wastes on or into the ground. Methods of disposal include placing wastes in percolation ponds,
on the ground surface (spreading or irrigation), in seepage pits or trenches, in dry streambeds,
in landfills, into disposal wells, and into injection wells.
All sources and causes of pollution can be classified as to their geometry. A point source
originates from a singular location, a line source has a predominantly linear alignment, and a
diffuse source occupies an extensive area that may or may not be clearly defined.
Groundwater contamination scenarios can be segregated into two categories: point
sources and nonpoint (area, or distributed) sources. Point sources include storage tanks (both
·one need not go far back in history to find instances of major diseases transmitted by groundwater. A classic
example was the outbreak of cholera in London, England, in 1854. Dr. John Snow, while investigating the epidemic,
noted with considerable astuteness that more than 500 persons died from cholera in ten days within 250 yards of a
public water supply well; this led to his conclusion:
The result of the inquiry then was, that there had been no particular outbreak or increase of cholera, in this part
of London, except among the persons who were in the habit of drinking the water of the above-mentioned
pump-well. I had an interview with the Board of Guardians of St. James parish on the evening of Thursday, 7th
September, and represented the above circumstances to them. In consequence of what I said, the handle of the
pump was removed on the following day (On the Mode of Communication of Cholera, 1855).
359
360 Chapter 8 Pollution of Groundwater
residual (immobile) hydrocarbon Liquid
vadose tone
8.2 l\
8.2.1
(a) (b)
land fill
----- leachate cont.ainiog

dissolved iOOtganic
ch~cals
vadosezo""
8.2.2
(c)
Figure 8.1.1. Example contaminant release scenarios: (a) LNAPL release, (b) DNAPL release, (c) dissolved contaminant release
(Johnson51 ).
underground and above-ground), landfills, pipeline releases, chemical manufacturing

locations, petroleum refining locations, wood treating facilities, and many others are consid-
ered to be point sources. Nonpoint (area or distributed) sources include agricultural activities.
Point-source groundwater contamination problems can be divided further into three main
categories: (1) light nonaqueous phase liquids (LNAPLs), (2) dense nonaqueous phase liquids
(DNAPLs), and (3) inorganics and other dissolved constituents. Example release scenarios for
LNAPLs, DNAPLS, and dissolved inorganics are illustrated in Figure 8.1.1.
ln their pure liquid fonn, LNAPLs are less dense than water and DNAPLs are denser than
water. LNAPL sites are caused by the release of petroleum products or crude oil (e.g. service
stations, refineries, and pipeline spills). DNAPL sites have been caused by dry cleaning, avia-
tion, automobile, and electric circuit board operations, which historically have used chlori-
nated solvents, such as trichloroethylene (TCE) and perchloroethylene (PCE). Inorganics and
other dissolved constituents include metals and salts. Sources that have added contaminants to
groundwater in dissolved form include but are not limited to mining operations, electroplating
operations, leaking wastewater treatment facilities, and landfills.
In the following sections, the principal sources and causes of pollution are briefly
described with regard to their occurrence and their effects on groundwater quality. Consider-
able confusion exists in the literature over the distinction between pollution and contamina-
tion. Here pollution shall signify any degradation of natural water quality, while contamination
shall be reserved for pollution that constitutes a hazard to human health.
8.2 MUNICIPAL SOURCES AND CAUSES

8.2.1 Sewer Leakage
Sanitary sewers are intended to be watertight; however, in reality, leakage of sewage into the
ground is a common occurrence, especially from old sewers. Leakage may result from poor
workmanship, defective sewer pipe, breakage by tree roots, ruptures from heavy loads or soil
slippage, fractures from seismic activity, loss of foundation support, shearing due to differen-
tial settlement at manholes, and infiltration causing sewage flow into abandoned sewer later-
als.60 Because suspended solids in sewage tend to clog sewer cracks and because the
surrounding soil tends to become clogged due to anaerobic conditions, leakage from minor
sewer openings is often small.
Sewer leakage can introduce high concentrations of BOD (biochemical oxygen demand),
COD (chemical oxygen demand), nitrate, organic chemicals, and possibly bacteria into
groundwater. Where sewers serve industrial areas, heavy metals such as arsenic, cadmium,
chromium, cobalt, copper, iron, lead, manganese, and mercury may enter the wastewater.
8.2.2 Liquid Wastes

Wastewater in an urban area may originate from domestic uses (see Figure 8.2.1), industries,
or storm runoff. Most of this highly variable mix of waters receives some degree of treatment
and is then discharged into surface waters. There is an increasing trend for treated wastewater
to be recharged into the ground where it mingles with naturally occurring groundwater and
subsequently becomes available for reuse.
Land application of municipal effluent is accomplished by one of three methods (see Fig-
ure 8.2.2): irrigation, infiltration-percolation, or overland flow. The selection of a method at a
~lease given site is primarily governed by the drainability of the soil, because this property deter-
mines the allowable liquid loading rate. In inigation systems, wastewater is applied by spray-
ing, ridge and furrow (see Chapter 13), and flooding; some water is lost by evapotranspiration.
tcturing For the infiltration-percolation method, effluent is applied by spreading in basins or by spray-
consid- ing; almost all of the water so applied reaches the groundwater. In the overland flow technique,
:tivities. wastewater is sprayed over the upper reaches of sloped terraces and allowed to flow across a
:eemain vegetated surface to runoff collection ditches; percolation to groundwater is minor in overland
! liquids tlow because surface runoff and evapotranspiration account for most of the applied water.
arios for Municipal wastewaters can introduce bacteria, viruses, and inorganic and organic chemi-
cals into groundwater. Where the recharged water is later extracted for potable use, concerns
1ser than exist regarding health aspects of this reclaimed water, particularly involving viruses, trace ele-
. service ments and heavy metals, and stable organics. Furthennore, chlorination of wastewater effluent
ng, avia- can produce additional potential pollutants.
d chlori- Shallow wells are widely employed to place surface runoff and sometimes treated
mics and municipal wastewater underground in freshwater aquifers (see Chapter 13). Such disposal
Domestic uses I
Evapotranspiration Spn
surl
appli<
l
Inside use Outside use
Root;
(about 50%) (about 50%)
Pick up of TDS, N, P, Lawn watering,
disinfectants, bacteria, fertilizers, pesticides, Sut
viruses, toxic materials, soil amendments,
brine from softening, carwash, swimming pools
MBAS, etc.
~
I Wastewater
I I Solid wastes
I
Storm drains
I
~owTDS,
high organic Zon1
content. dust an<
Septic tanks Treatment plants
and leach Rech
fields Primary, secondary,
or tertiary (quality Sanitary r Recharge
of effluent depends landfills of ground
on treatment) water basins
High TDS, N
Sludges
I I
Recharge of Reuse of water Surface water streams
groundwater
Irrigated agriculture, High BOD, nutrients,
basins
industrial and/or MBAS, algae, etc.
High TDS, N03, spreading for ground-
CI,S04 water recharge
Ocean
outfall
Slope 2-
Figure 8.2.1. Domestic uses
Ocean Sanitary Sludge of water and their effects on
outfall landfills lagoons
water quality (after Hassan45).
wells have been criticized from a health standpoint because of the potential for pollutants to be
released directly into an aquifer. The problem is most critical where disposal wells are near
pumping wells (Figure 8.2.3) and where the beneficial effects of water passing through fine-
grained materials may be absent, such as in basalt and limestone aquifers.
8.2.3 Solid Wastes

The land disposal of solid wastes creates a potential important source of groundwater pollu-
tion. A landfill may be defined as any land area serving as a depository of urban, or municipal,
solid waste. Many landfills are simply refuse dumps; only a fraction can be regarded as sani-
tary landfills, indicating that they were designed and constructed according to engineering
specifications. Leachate from a landfill can pollute groundwater if water moves through the fill
material. 110 Possible sources of water include precipitation, surface water infiltration,
Evapotranspiration
Crop
Spray or
surface ---+-
application
Slope
variable
..--Deep
Subsoil---+- (a) percolation
Evaporation
Spray or surface
application
Zone of aeration
and treatment ---- Percolation through
unsaturated zone
Original water table

(b)
Spray
application
Grass and vegetative litter

Runoff
Slope 2-4%
I collection
I Figure 8.2.2. Methods of land application

of municipal wastewaters. (a) Irrigation
t---- -- - - - 30-100 m -------·~1 (b) Infiltration-percolation (c) Overland
flow (after U.S. Environmental Protection
(c) Agency96)
; to be
enear
1 fine- percolating water from adjacent land, and groundwater in contact with the fill. Ordinary mixed
refuse usually has a moisture content less than that of field capacity; therefore, leachate from
a landfill can be minimized if water from the above sources can be kept from the fill material.
In a properly constructed sanitary landfill, any leachate generated can be controlled and pre-
vented from polluting groundwater. In addition, it is assumed that the landfill is properly
pollu- located, operated, and monitored.
1icipal, The problem of pollution from landfills is greatest where high rainfall and shallow water
s saru- tables occur. Important pollutants frequently found in leachate include BOD, COD, iron, man-
1eering ganese, chloride, nitrate, hardness, and trace elements. Hardness, alkalinity, and total dis-
the fill solved solids are often increased, while generation of gases, such as methane, carbon diox.ide,
ration, ammonia, and hydrogen sulfide, are further by-products of landfills.
Pollutants
..--- __--.4ij; introduced
Aquifer
t
Figure 8.2.3. Diagram showing movement of

pollutants from a disposal well to a nearby
Aquiclude pumping well (after Deutsch22).
8.3 1NDUSTRIAL SOURCES AND CAUSES

8.3.1 Liquid Wastes
The major uses of water in industrial plants are for cooling, sanitation, and manufacturing and
processing. The quality of the wastewater varies with type of industry and type of use. A gen-
eralized flow diagram of industrial water use and its effects on water quality is shown in Fig-
ure 8.3 .1. Cooling water that is softened before use to inhibit scale formation produces Ocear
wastewaters with salts and heat as important pollutants. Groundwater pollution can occur (toxic
where industrial wastewaters are discharged into pits, ponds, or lagoons, thereby enabling the on rr
wastes to migrate down to the water table. hat
Cooling water is sometimes recharged underground through shallow disposal wells
because its quality, except for the addition of heat, may be unimpaired. 83 The disposal of haz-
ardous and toxic industrial wastes is sometimes accomplished by means of deep injection wells
that place the fluids into saline water formations far below developed freshwater aquifers.53· 107
8.3.2 Tank and Pipeline Leakage

Underground storage and transmission of a wide variety of fuels and chemicals are common
practices for industrial and commercial installations. These tanks and pipelines are subject to
structural failures so that subsequent leakage becomes a source of groundwater pollution.
Petroleum and petroleum products are responsible for much of the pollution. Leakage is par-
ticularly frequent from gasoline station and home fuel oil tartks.51 · 63· 64• 76· 108 An immiscible Figure~
liquid, such as oil, leaking underground moves downward through penneable soils until it
reaches the water table. Thereafter it spreads to form a layer on top of the water table and
migrates laterally with groundwater flow (see Figure 8.3.2a). Liquid radioactive wastes are
sometimes stored in underground tanks; leakage from such installations, which has occurred,
can cause serious pollution problems in local groundwater. Pumping wells can be used to cre-
ate drawdown cones to contain the leakage, as shown in Figure 8.3.2b, and to remove a mix-
ture of gasoline and water.
8.3.3 Mining Activities

Mines can produce a variety of groundwater pollution problems. 7· 26· 68 Pollution depends on
the material being extracted and the milling process. Coal, phosphate, and uranium mines are
8.3 Industrial Sources and Causes 365
I Industrial uses j
Evaporation
Sanitation use I Manufacturing and processing use Cool uses

l
Effects: Pickup of TDS, Manufacturing: toxic heavy metals Algae inhibitors and pH
bacteria and viruses, B. and chemicals from industries such adjustors are used
MBAS, nitrogen and as mining, metal plating, iron. Effects: high T (thermal
posphorus compounds, fertilizers, pesticides. and paper I pollution), TDS, acidity
high BOD, toxic Processing: such as picking, food, alkalinity, etc.
disinfectants, etc. fruit packing, and washing
I Effects: high TDS, acidity or alka-
linity, heavy metals such as Cr+ 6,
of Cd, As, Hg, Ni, Fe, Mn, pesticides,
high COD, brines from softening
I
t
1 Industrial wastes 1
f
Nonreclaimable wastewater Solid wastes ! Reclaimable wastewater
I
High TDS, toxicity Low TDS, free of toxicity
and
gen-
Fig- t t
uces Ocean outfall I Sanitary
landfill
I Injection into
deep saline
I industrial
Recycled for I
use
To municipal
wastewater
Disposal in
streams or
ccur (toxic effects aquifers treatment injection
5 the on marine plant into
habitat) aquifers
(Primary,
vells secondary,
haz- or tertiary)
.vells
\3,107
t
I Sludge
I
I Reuse Recharge of
groundwater
Surface
water
(back to basins streams
unon f t_ ~ resource
II pool)
~ct to
1tion.
; par-.
I
Sanitary
landfill
I Sludge
lagoons
Ocean
outfall
I
;cible Figure 8.3.1. Industrial uses of water and their effects on water quality (after Hassan45).
ntil it
e and
~s are
urred, major contributors; metallic ores for production of iron, copper, zinc, and lead are also impor-
o cre- tant; stone, sand, and gravel quarries, although numerous, are chemically much less important.
t mix- Both surface and underground mines invariably extend below the water table so that dewater-
ing to expand mining is common. Water so pumped may be highly mineralized and is fre-
quently referred to as acid mine draiMge. Typical characteristics include low pH and high
iron, aluminum, and sulfate. 67
Coal deposits are often associated with pyrite (FeS2). This is stable for conditions below
1ds on the water table, but if the water table is lowered, oxidation occurs. Oxidation of pyrite followed
tes are by contact with water produces ferrous sulfate (FeS04) and sulfuric acid (H2SOJ in solution;
Spill area
c
/ Ground surface
"";d"~.f -
~.
\
~
I saturation '§' 5
/Oil
I Q)
Q)
.s
~
- - J
Water table
.c
E.
Q)
-
0 10
- Water movement -
15
4(
(a)
8.4 A
8.4.1
-
- Figure 8.3.2. Oil pollution of groundwater
-
and control by a pumping well. (a) Fluid oil
-
Water movement
floating on water table and elongating in
direction of groundwater flow (b) Oil on
water table contained in cone of depression
(b) created by pumping well (after Engmg. and
Tech. Research Comm?7)
8.4.2 .
groundwater intermingling with this water will have a reduced pH and an increase in iron and
sulfate contents. Pollution of groundwater can also result from the leaching of old mine tail-
ings and settling ponds; therefore, pollution problems can be associated with both active and
abandoned mines.
8.3.4 Oil~Field Brines 8.4.3 )

The production of oil and gas is usually accompanied by substantial discharges of wastewater
in the form of brine. 52 Constituents of brine include sodium, calcium, ammonia, boron, chlo-
ride, sulfate, trace metals, and high total dissolved solids. In the past, oil-field brine disposal
was handled by discharge to streams or "evaporation ponds." In both instances, brine-polluted
aquifers became commonplace in oil production areas as the infiltrating water reached the
underlying groundwater. Today, such disposal methods are prohibited by most regulatory
agencies; however, regulation is often ineffective, so that many brine-affected areas remain and
will persist for years into the future. Figure 8.3.3 provides a graphic example of aquifer pollu-
tion from an oil-field brine disposal pit.
8.4 Agricultural Sources and Causes 367
.r::
a.
<ll
0 10
Figure 8.3.3. Distribution of saline
water in confined aquifer resulting
from an oil-field brine disposal pit in
southwestern Arkansas. Numbered
200 0 lines are isochlors in mg/1 (after
Distance from disposal pit (meters) Frybergeil6).
8.4 AGRICULTURAL SOURCES AND CAUSES

8.4.1 Irrigation Return Flows
Approximately one-half to two-thirds of the water applied for irrigation of crops is consumed
by evapotranspiration; the remainder, termed irrigation return flow, drains to surface channels
or joins the underlying groundwater. Irrigation increases the salinity of irrigation return flow
from three to ten times that of the applied water. The degradation results from the addition of
salts by dissolution during the irrigation process, from salts added as fertilizers or soil amend-
ments, and from the concentration of salts by evapotranspiration (see Figure 8.4.1). Principal
~ater
cations include calcium, magnesium, and sodium; major anions include bicarbonate, sulfate,
.id oil
chloride, and nitrate. Because irrigation is the primary use for water in arid and semiarid
in
regions, irrigation return flow can be the major cause of groundwater pollution in such regions.
m
ssion
:. and 8.4.2 Animal Wastes
Where animals are confined within a limited area, as for beef or milk production, large amounts
of wastes are deposited on the ground. Thus, for the 120 to 150 days that a beef animal remains
on and in a feedlot, it will produce over a half-ton of manure on a dry-weight basis. With thousands of
animals in a single feedlot, the natural assimilative capacity of the soil can become overtaxed.
ne tail-
Stonn runoff in contact with the manure carries highly concentrated pollutants to surface and
ive and
subsurface waters. Animal wastes may transport salts, organic loads, and bacteria into the soil.
Nitrate-nitrogen is the most important persistent pollutant that may reach the water table. 1• 39• 79
8.4.3 Fertilizers and Soil Amendments

tewater When fertilizers are applied to agricultural land, a portion usually leaches through the soil and
n, chlo- to the water table. The primary fertilizers are compounds of nitrogen, phosphorus, and potas-
lisposal sium. Phosphate and potassium fertilizers are readily adsorbed on soil particles and seldom
>Oiluted constitute a pollution problem. But nitrogen in solution is only partially used by plants or
hed the adsorbed by the soils, and it is the primary fertilizer pollutant. Fertilizers are extensively used
~ulatory and will undoubtedly increase in the future.
1ain and Soil amendments are applied to irrigated lands to alter the physical or chemical properties
!r poilu- of the soil. Lime, gypsum, and sulfur are widely used for this purpose; substantial amounts of
these soil amendments may eventually leach to the groundwater, thereby increasing its salinity.
Agricultural uses
(including feed lots and diaries)
r--
t t t
Agricultural solid wastes Evapotranspiration Application of additives Tail water
(source ot nitrates) (consumptive use) (organic and inorganic) Seepage off land
a. Plant & crop residues salts in appl"led water pesticides, soil surface, high TDS,
b. Manure are concentrated conditioners and organics, BOD, silt,
c. Dead animals (2 or 3 times. amendments, pesticides, etc.
depending on and fertilizers
irrigation efficiency)
I
t t ~
Sanitary landfill Soil reactions Changes in soil zone Surface water
Effects on groundwater Include deposition Organics change to Effects: low DO,

depend on control and of salts, ion ex· inorganics, nitrogenous high BOD, high
depth to water table, rain· change, oxidation, compounds change to TDS, high nutrient
fall and moisture content reduction, etc. f- nitrate, phosporus levels, algae, bacteria
in waste, hydraulic gradient compounds change to fish kills,
of water table, geometry of phosphate, nutrient stream cycle
site, etc. uptake by plant roots,
t
denitrification, nitrogen
fixation, ion exchange
affects pesticides,
t I
To ocean
Leachates Gases biochemical reactions, I
and degradation of
(If water is excessive in fill.) CH4, H2S, NH3, C02 organic compounds
Effects: organic Effect of C02 on
acids, alcohols, other groundwater: alkalinity
volatiles, taste, color, (scale-forming), acidity
odor, high BOD, low DO, (corrosivity),lncrease '+- Percolation to groundwater ~
high TDS, Fe, and other in hardness, TDS, Fe and Effect: high TDS, N03,
minerals. Mn, color, turbidity, etc. Cl, 804, Na
Figure 8.4.1. Agricultural uses of water and their effects on water quality (after Hassan45).
8.4.4 Pesticides, Insecticides, and Herbicides

Pesticides can be significant in agricultural areas as a diffuse source of groundwater pollu-
tion.57· 94• 104 The terms pesticide, insecticide, and herbicide are broadly interpreted here to
embrace any chemical applied to control, destroy, or mitigate pests. The presence of these
materials in groundwater, even in minute concentrations, can have serious consequences in
relation to the potability of the water. The impact of these chemicals on groundwater quality
depends on the properties of these chemicals' residue, rainfall or irrigation rates, and soil char-
acteristics.91 Most pesticides are relatively insoluble in water, while others are readily adsorbed
by soil particles or are subject to microbial degradation.
8.5 MISCELLANEOUS SOURCES AND CAUSES

8.5.1 Urbanization
Shallow groundwater quality can be affected by residential and commercial development as
illustrated in Figure 8.5.1. The U.S. Geological Survey's National Water Quality Assessment
(NAWQA) program (http://water.usgs.gov/nawqal) seeks to determ.ine how shallow groundwa- Figm
ter quality is affected by development (Squillace and Price8\ Residential developments have biolo.
Urban Areas Aural Areas
_w_ater
Iable
~roUnd waun pollu!l:d by lndusfrrilllfld

nllmfr.lpdi'Wa!ii!S,ll!aiflr~g WMrs and lewtt
rettlll:zeJW pesticides, slid he~
pollu-
tere to Density Effects
f these
tees in
~uality
il char-
sorbed
nent as
:ssment
undwa-
Figure 8.5.1. Groundwater pollution occurs in both urban and rural areas and is affected by differences in chemical composition,
tts have biological and chemical reactions, density, and distance from discharge areas (Heath46).
taken up large tracts of land, and as a consequence, have widespread influence on the quality of
water that recharges aquifers and discharges into streams, lakes, and wetlands. Studies of vari-
ous urbanizing areas have been performed in the past by Mullaney and Grady,70 Stackelberg,86
and Thomas. 89 The study by Thomas89 of residential areas near Detroit, Michigan, concluded
that the young and shallow waters have significantly higher median concentrations of nitrate,
chloride, dissolved solids, sodium, calcium, and potassium than old and deep waters.
8.5.2 SpiUs and Surface Discharges

Liquids discharged onto the ground surface in an uncontrolled manner can migrate downward
to degrade groundwater quality. At industrial sites, causal activities may include boilovers,
losses during transfers of liquids, leaks from pipes and valves, and inadequate control of
wastes and storm runoff. Washing aircraft with solvents and spills of fuel at airports can form
a layer of hydrocarbons floating on the water table. Pollution can also occur from the inter-
mittent dumping of fluids on the ground, especially near gasoline stations, small commercial
establishments, and construction sites. It has been estimated that millions of gallons of auto-
mobile waste oil are discharged on the ground surface annually. 67 Finally, accidents involving
above-ground pipelines, storage tanks, railroad cars, and trucks can release large quantities of
a pollutant at a particular site. Hazardous and flammable liquids are often flushed by water
from highways; this action may actually aid in transporting the pollutant to the water table.
8.5.3 Stockpiles
Solid materials are frequently stockpiled near industrial plants, construction sites, and large
agricultural operations. These may be raw materials awaiting use, or they may be solid wastes
placed for temporary or permanent storage. Precipitation falling on unsheltered stockpiles 8.5.6
causes leaching into the soil to occur; this may transport heavy metals, salts, and other inor-
ganic and organic constituents as pollutants to the groundwater.
8.5.4 Septic Tanks and Cesspools

Another potential source of groundwater pollution is septic tanks and cesspools. 13 Commercial
establishments, hospitals, industrial plants, and resorts employ septic tanks in areas where com-
munity sewer systems are not available. A septic tank is a watertight basin intended to separate
floating and settleable solids from the liquid fraction of domestic sewage and to discharge this 30(
liquid with its dissolved and suspended solids into the biologically active zone of the soil man-
tle through a subsurface percolation system such as a tile field, a seepage bed, or a sand-covered
sand filter (see Figure 8.5.2). A cesspool is a large buried chamber with porous walls designed
to receive and percolate raw sewage. Domestic sewage adds minerals to groundwater. Bacteria
and viruses are normally removed by the soil system; phosphorus is generally retained by the 200
'§,
soil, but significant quantities of nitrogen can be added to groundwater.78 E
~
(tJ
'0
't:
0
8.5.5 Roadway Deicing :c
0 100
Pollution of groundwater also results from the application of deicing salts to streets and high-
ways in winter.31 • 88 Urban areas with winter temperatures below freezing are affected the
most Salt reaches the groundwater in solution after spreading on roadways and also from
stockpiles. Salt application rates range from 2 to 11 m-tons per single-lane kilometer per win-
ter season. Sodium chloride accounts for most applications, with calcium chloride the remain- 0
der. A steady growth in roadway deicing has stemmed from the demand to maintain streets and
y of Production Pretreatment Disposal

an-
·g,&6
ded
:ate,
Unsaturated
vard zone
ters,
>I of
:onn
nter-
rcial
lUtO-
tving
-- Figure 8.5.2. Disposal of household
wastes through a conventional septic tank
system (after Miller67).
es of
lv'ater
,Je. highways in safe driving condition throughout the winter. The salts have produced widespread
degradation of groundwater quality and have also hastened the corrosion of wells.
Illustrative of the effect is the dramatic increase in chloride measured in a well in Burling-
ton, Massachusetts, shown in Figure 8.5.3. When chloride exceeded the 250 mg/1 drinking
large water standard in 1970, local use of deicing chemicals on city streets was banned.
'astes
;:piles 8.5.6 Saline Water Intrusion
inor-
Salt water may invade freshwater aquifers to create point or diffuse pollution sources. In
coastal aquifers, seawater is the pollutant, while in inland aquifers underlying saline water may
be responsible. Because of the worldwide importance of saline water intrusion, its occurrence,
mechanisms, and control methods are discussed more fully in Chapter 14.
1ercial
:com-
parate
;e this
lman-
)Vered
signed
acteria
~ 200
by the 'a,
.s
Q)
'0
·;:
..Q
.I;
CJ 100
j high- Figure 8.5.3. Time variation of chloride in
ted the groundwater at Burlington, Massachusetts,
o from due to local roadway deicing (after Terry,88
erwin- reprinted with pennission from Road Salt,
0
emain- 1955 1960 1965 1970 Drinking Water. and Safety, copyright 1974,
:ets and Years Ballinger Publishing Company).
8.5.7 Interchange through Wells

Because wells form highly pem1eable vertical connections between aquifers, they can serve as
avenues for groundwater pollution where inadequate attention is given to the proper construc-
tion, sealing, or abandonment of wells. Pollution occurs where there is incomplete hydraulic
separation within a well and where a vertical difference in hydraulic head exists between two
aquifers. Two flow conditions with a well connecting an unconfined aquifer with a confined
aquifer are illustrated in Figure 8.5.4. In addition, if the top of a well is not constructed so as
to divert surface water away from the well, it can admit a wide variety of surface pollutants.
Although governmental regulations in most locations require plugging of abandoned wells,
many such wells remain unplugged and serve as ongoing interchanges between aquifers.
8.5.8 Surface Water

Polluted surface water bodies that contribute to groundwater recharge become sources of
groundwater pollution. The recharge may occur naturally from a losing stream, or it may be
induced by a nearby pumping well, as indicated in Figure 8.5.5. Because many municipal
water supply wells are located adjacent to rivers to ensure adequate flows, they can serve as
important groundwater pollution mechanisms. Figure 8.5.6 illustrates the various pathways for
the movement of contaminated groundwater into a well when the source of contaminants is at
or near the water table.
...,.....,
....---:~-=--.-~~.--.~::::.-_ _ - -- - - Ground surface t; --- I - - - - - Ground surtace
----------- ·- ----------- Piezometric
----~-..._J[JZ.l '\7------- Water table surface
--
--+ I ~ Unconfined lsz.
------
Water table
_t _ _:'-
--~! -
- - +_ p· aquifer .
---------- --------- 1ezometnc
surtace f Unconfined
aquifer
~--: -::: ~
t ~·
.. c::::
Clay layer
~-~ ~' .. ~
-- ! ----- -----
----- --
- ----- -
Confined Confined
aquifer 1 aquifer 8.6 A~
----- -----
(a) (b)
Figure 8.5.4. Diagrams showing aquifer leakage by vertical movement of water through a non pumping well. (a) Water table above
the piezometric surface (b) Piezometric surface above the water table (after Todd et a1 93)
8.6.1 I
Polluted ater table
---V:------
originaI
surface water
I Pumping water t
-L-~---:;0:::::::=-=-----~
----- -- level
"'-..'-..... ---___ _-
---
......._ --....
Aquifer ..__. - - -
...........
~
Figure 8.5.5. Diagram
showing how polluted
water can be induced to
flow from a surface
8.6.2 s
-....._
------ ---- __...--- stream to a pumping well
(after Deutsch22).
8.6 Attenuatuon of Pollution 373
e as
ruc-
.ulic
two
ined
.o as
mts.
·ells,
Figure 8.5.6. Pathways for movement of
contaminated ground~ater into a well, when
the source of contaminants is at or near the
water table.32
Analytical techniques for estimating
;s of
areas contributing recharge to discharging
ty be wells, including numerical simulation, consid-
cipal er only pathway (1), which is the normal flow
ve as of groundwater from the aquifer to the dis-
{S for charging well. Pathways (2) {flow through a
; iS"'"'"at hole, crack, or casing joint) and (3) (flow
down the annular space) can permit entry of
adjacent ground water into the well from any
Jrface depth above the well screen. The prevention
of contamination along these pathways
ric
depends on the design and construction of the
well and on maintenance of the well through
tie
time. Unfortunately, no analytical tools or
ad simple predictors are available to identify if or
when these pathways are or will become
active (Franke et al. 32).
8.6 ATTENUATION OF POLLUTION

Pollutants in groundwater tend to be removed or reduced in concentration with time and with
distance traveled. Mechanisms involved include filtration, sorption, chemical processes,
microbiological decomposition, and dilution. The rate of pollution attenuation depends on the
above type of pollutant and on the local hydrogeologic situation. Attenuation mechanisms tend to
localize groundwater pollution near its source; they also are responsible for the interest in
groundwater recharge as a water reclamation technique.
8.6.1 Filtration
Filtration removes suspended materials; hence, this action is most important at ground surface
where polluted surface water is infiltrating into the ground. During groundwater flow, filtration
can remove particulate fonns of iron and manganese as well as precipitates fonned by chem-
ical reactions.
•gram
luted
teed to 8.6.2 Sorption
.ce
·ing well Sorption serves as a major mechanism for attenuating groundwater pollution. Clays, metallic
oxides and hydroxides, and organic matter function as sorptive materials. Most pollutants can
be sorbed under favorable conditions, with the general exception of chloride, and to a lesser 8.7
extent, nitrate and sulfate. The sorption process depends on the type of pollutant and the phys-
ical and chemical properties of both the solution and the subsurface materials; a substantial
clay content in the strata above the water table is a key factor. The sorptive capacity of geologic
materials is finite for most inorganic substances; however, for biodegradable substances such
as bacteria and ammonia, the sorptive capacity may be renewed indefinitely.
8.6.3 Chemical Processes

Precipitation in groundwater can occur where appropriate ions are in solution in sufficient
quantities. The most important precipitation reactions for the major constituents involve cal-
cium, magnesium, bicarbonate, and sulfate. Trace elements having important precipitation
potential include arsenic, barium, cadmium, copper, cyanide, fluoride, iron, lead, mercury,
molybdenum, radium, and zinc. In arid regions, where moisture in the nearsurface zone may
be minimal, chemical precipitation becomes a major attenuation mechanism.
In the zone above the water table, oxidation of organic matter acts as an important atten-
uation mechanism. Complex organic compounds are oxidized stepwise to more simple organic
compounds until C02 and H20 are formed along with numerous inorganic ions and com-
pounds. Both oxidation and reduction reactions can occur underground in conjunction with
other mechanisms, leading to precipitates, deposits of insoluble trace metals, and gases.
Volatilization and loss as a gas apply particularly to reactions involving nitrate and sulfate.
Radioactive decay, based on the half-life of a radioisotope, acts as an attenuation mechanism
for radioactive pollutants.
8.6.4 Microbiological Decomposition

Most pathogenic microorganisms in the soil do not flourish in the soil and hence are subject
to ultimate destruction, the timing of which depends on different species and environmental
conditions. Bacteria and viruses as particulate matter suspended in water tendto move slower
through a porous ·media than water. Field studies indicate that these pathogens are largely
removed by passage through as little as one meter of soil, provided reasonable amounts of silt
and ciay are present. 25• 81
8.6.5 Dilution
Pollutants in groundwater flowing through porous media tend to become diluted in concen-
tration due to hydrodynamic dispersion occurring on both microscopic and macroscopic scales
(see Chapter 3). These mixing mechanisms produce a longitudinal and lateral spreading of a
pollutant within the groundwater so that the volume affected increases and the concentration
decreases with distance traveled. Dilution is the most important attenuation mechanism for
pollutants after they reach the water table.
In summary, attenuation above the water table is generally effective for most pollutants
with the exception of the major inorganic constituents. Other exceptions include boron and tri-
tium, as well as some trace elements and organic chemicals. Maximum attenuation requires an
adequate distance to the water table and the presence of fine-grained geologic materials, such
as silt and clay. Without these conditions, almost any pollutant can be introduced directly into
the saturated zone. Pollutant attenuation below the water table occurs more slowly with dilu-
tion serving as the principal mechanism.
esser 8.7 DISTRIBUTION OF POLLUTION UNDERGROUND

)hys- In a large region or basin, bodies of polluted groundwater can be visualized as innumerable
mtial scattered dots on a map, while in some areas diffuse sources, such as from irrigation return
,Jogic
flows, would create areas with appreciable horizontal extent. Entry of pollutants into shallow
such aquifers occurs by percolation from ground surface, through wells, from surface waters, and
by saline water intrusion. The extent of pollution in groundwater from a point source decreases
as pollutants move away from the source until a harmless or very low concentration level is
reached. Because each constituent of a pollution source may have a different attenuation rate,
the distance to which pollutants travel will vary with each quality component.
iicient A hypothetical example of a waste-disposal site is shown in Figure 8.7.1 with groundwa-
'e cal- ter flowing toward a river. Zones A, B, C, D, and E represent essentially stable limits for dif-
itation ferent contaminants resulting from the steady release of wastes of unchanging composition.
:rcury, Pollutants, once entrained in the saturated groundwater flow, tend to form plumes (analogous
,emay to smoke from a smokestack as it drifts downwind in the atmosphere) of polluted water extend-
ing downstream from the pollution source until they attenuate to a minimum quality level.
: atten- Only zone E reaches the river in Figure 8.7.1 and is subsequently diluted by surface water.
'rganic The shape and size of a plume depend on the local geology, the groundwater flow, the type
i com- and concentration of pollutant, the continuity of waste disposal, and any human modifications
m with of the groundwater system, such as pumping wells.61 Where groundwater is moving relatively
gases. rapidly, a plume from a point source tends to be long and thin, such as that shown in Figure
sulfate. 8.7.2, but where the flow rate is low, the pollutant tends to spread more laterally to form a
hanism wider plume. Irregularly shaped plumes can be created by local influences such as pumping
wells and nonuniformities in penneability.
Plumes tend to become stable areas if there is a constant input of waste into the ground.
This occurs for two reasons: enlargement as pollutants continue to be added at a point source
is counterbalanced by attenuation mechanisms, or the pollutant reaches a location of ground-
subject water discharge, such as a stream, and emerges from the underground. When a waste is first
omental released into groundwater, the plume expands until a quasi-equilibrium stage is reached.
! slower
. largely
ts of silt
concen-
,jc scales
ding of a
entration
mism for
Jollutants
mand tri-
:quires an Figure 8.7.1. Plan view of an unconfined
ials, such aquifer showing areal extent to which various
-ectly into pollutants of mixed wastes at a disposal site
with dilu- ...___ Downstream limit disperse and move to insignificant levels (after
of pollutant E LeGrand61 ).
100
/ 200 200100
50
100
~ 8.7
Scale
0 500 1000 m Figure 8.7.2. Plume of
groundwater pollution
from a landflll near
Munich, West Germany.
Numbered lines are
isochlors in mgll (after
Landfill Cole 18).
If sorption is important, a steady inflow of pollution will cause a slow expansion of the plume
as the earth materials within it reach a sorption capability limit.
An approximately stable plume will expand or contract generally in response to changes
in the rate of waste discharge. Figure 8.7.3 shows changes in plumes that can be anticipated
from variations in waste inputs.
Waste site Waste site Waste site / Waste site
I \
I I
1(4_1
-~
o-
._
I
1·
I •
I
I
Present
c: Cll 1 boundary
.Q(lj Present 1
I I
t>:t
~"0
boundary I I
. I I I
·- c:
Q:J I 1 ~"+-Former
e01 I Present \. / boundary
,4--Former boundary
boundary
Enlarging plume Reducing plume Nearly stable plume Shrunken plume

1. Increase in rate of 1. Reduction in wastes 1. Essentially same waste 1. Waste no longer disposed
discharge wastes 2. Effects of changes in input and no longer leached at
2. Sorption activity used up water table 2. Sorption capacity not fully abandoned waste site
3. Effects of changes in 3. More effective sorption utilized
water table 4. More effective dilution 3. Dilution effect falrty stable
5. Slower movement and 4. Slight water table fluctuation
more tlme for decay or effects of water-table
fluctuation not important
Figur1
Figure 8.7.3. Changes in groundwater pollution plumes and causal factors (after LeGrand61 ).
An important aspect of groundwater pollution is the fact that it may persist underground
for years, decades, or even centuries. This is in marked contrast to surface water pollution.
Reclaiming polluted groundwater is usually much more difficult, time consuming, and expen-
sive than reclaiming polluted surface water. Underground pollution control is achieved prima-
rily by regulating the pollution source, and secondarily by physically entrapping and, when
feasible, removing the polluted water from the underground.
8.7.1 Hanford Site in Richland, Washington

The Hanford site (see Figure 8.7.4) is considered to be one of the largest groundwater con-
tamination problems in the United States by the Department of Energy (DOE) (see Graham et
41
al. ) . Tritium and nitrates from the reprocessing of fuels have contaminated tens of square
miles of the aquifer underlying the site (Freshley and Thome35 ). The spread of the tritium
plume with time is shown in Figure 8.7.5. Other chemicals and radionuclides have also been
disposed at the Hanford site and are found in the groundwater (Dirkes et al.23). Figure 8.7.6
ne of shows the distribution of carbon tetrachloride in the groundwater under one area of the site.
tion
I
many.
-e
after
plume
hanges
cipated
o Slllfaee water body localloM

- - - Ephemfral and dlsconl\ooous
drainages
e _ _ Pareoolal or
l~posed lrrte!Tnlnant ser:11ollS
~hed at em Basal ~crop
site
SCala
?H H Mles
Figure 8.7.4. Location of the Hanford site in Richland, Washington (Freshley and Graham34).
1 1974
8(
7(
6C
:J" 50
tll
,.~·"~-? Benton .:!:40
• City .:t
(.)
(.)
30
_, '
Kennewick · ~-- 20
10
1/1
1,()()(
90<
8()(
~ 70<
~ 6()(
~ 5()(
G40C
(.)300
200
100
0
1/"
~ Tritium contamination
300
• Chemical processing and
.__.,.._,..._..;;;;;;:;.._ _ _ _ _ _ _ _..,.. high-level waste tank areas r - - - - - - - - -- - - - - - - '
o 2 4 6 8 Miles
I I II I
I I
I I • Production reactor areas
o 2 4 6 a Kilometer> - Hanford site boundary
Figure 8.7.5. Spreading tritium contamination at the Hanford Site in Washington. The shaded areas on these maps show how
tritium contamination in concentrations above safe drinking-water standards has spread over time (DOE95).
0
1/8
8.8 MASS TRANSPORT OF POLLUTANTS
8.8.1 Transport Processes
The law of conservation (advective-dispersive equation) for solute transport in saturated media Figure 8.
is derived following Ogata,74 Bear,8 and Freeze and Cherry. 33 The conservation of mass for
solutes in porous media considers the flux of solute into and out of a fixed elemental volume
(see Figure 8.8.1) as follows:
180,-----------------------,
160
140
:::J 120
~100
d 80 Well
(.) 60 299-WB-2
8oo.---------------------------------. Collectlon date 59402003.77
700
600
~ 500
.:1:400
..3
8300
200
100
o~L_ _____ L_ _ _ _ _ _~----~------~~
1/1/85 1/1/86 1/1/88 1/1/90 1/1/92 1/1/94
Collectlon date S9402063. 75
-N-
!
i
1/1/87 1/1/89 1/1/91 1/1/93 1/1195

Collection date S9402063.n
......
,REDOX
plant
..,...
:.
59402063.16
•
300 '
Weii299-W19-16 A
~200 A
~ ... "
.,:!;
g1oo
WELL 299-W19-15
0 10
1/85 1/86 1187 2188 4/89 5/90 6/91 7192 1/93
Collection date SfM(J2063.78 0
1/1185 1/1186 1/1188 1/1/90 1/1/92 1/1/94
Collection date S940206:l.74
media Figure 8.7.6. Distribution of carbon tetrachloride in groundwater under the 200-West area of the Hanford site (Dirkes et al. 23).
ass for
tolume
Figure 8.8.1. Elemental control volume

X
for derivation of the conservation of mass c
y showing the flux (Freeze and Cherry 33).
net rate of flux of flux of loss or gain

change of mass solute out solute into of solute mass
:::: ± (8.8.1)
of solute within of the the due to
the element element element reactions
Solutes can be considered in two broad classes: conservative and reactive. Conservative solutes
are nonreactive with the soil and/or native groundwater and do not undergo biological or
radiaoactive decay. Chloride ion is an example of a nonreactive or conservative solute.
Advection and hydrodynamic dispersion are the physical processes that control flux into
-v
and out of the elemental volume of porous media. Advection is the transport of solute by the
flowing groundwater (bulk groundwater flow), expressed by Darcy's law. Hydrodynamic
dispersion results from mechanical mixing (dispersivity) and molecular diffusion.
Mechanical dispersion occurs when contaminated groundwater mixes with noncontami-
nated groundwater, resulting in a dilution of the contaminate, which is referred to as
dispersion. Figure 8.8.2 illustrates longitudinal and transverse spreading due to mechanical
dispersion. Mixing that occurs along the streamline is longitudinal dispersion. Dispersion nor-
mal to the longitudinal dispersion is lateral (transverse) dispersion. Mechanical dispersion in
the transverse direction is a much weaker process than in the longitudinal direction; however,
the coefficients of longitudinal and transverse dispersion are nearly equal at low velocities
when molecular diffusion is the dominant dispersive mechanism.
Figure 8.8.3 compares streamlines and groundwater age distribution with and without dis- 5
c.
persion in a regional layered aquifer system. The effect of not considering dispersion on Q)
0
groundwater age is clearly indicated by the darker areas as compared to the same areas that are
much lighter for advection-dispersion.
Diffusion is the flux of solute from a zone of higher concentration to a zone of lower con-
centration as a result of Brownian motion of ionic and molecular species. Pick's law describes
i
the diffusion of a solute for a steady-state condition:
F= -DdC (8.8.2)
dx
where F is the mass flux of solute per unit area per unit time (.MIL2!f); Dis the diffusion coef-
ficient (L2!f); Cis the solute concentration (M!L3); and dC!dx is the concentration gradient
(.M/L3/L). The negative sign indicates the movement from greater to lesser concentration. Dif-
fusion coefficients for the major ions in groundwater (Na+, K+, Mg2+, Ca2+, Cl-, Hco-, SO/-)
8. 8 Mass Transport of Pollutants 381
- Abrupt
front at
t= 0
-C=1 I
C=O-
Transition zona
-
v
I
I
I
l
I
I
C L =vt -------1•1 Breakthrough curve
me
mass
l3).
I
Ti~e 11.8+_1..:.-0_ _ _______
--1.0-<--------
TI_m_e_t_::+C~-+-f------
-
X
:8.8.1)
-
X
;olutes v
-
Tracer
Injected I I
ical or
llX
by the
into - at t= 0 Contours
lf+\1
of
at I= t 1
C =const.
~
Contours of C =canst.
Figure 8.8.2. Longitudinal
and transverse spreading due
to mechanical dispersion
(from Bear and VerruUt9 with
pennission).
ynamic
mtami-
:1 to as
hanical
ton nor-
rsion in
owever,
~loci ties
lOUt dis-
:sion on
; that are
Figure 8.8.3. Streamlines and
Ner con- groundwater age distributions with
lescribes and without dispersion in a
regional layered aquifer system
(Goode40).
(8.8.2)
ion coef- w-
are in the range of 1 X 9 to 2 X ro-9 m2/s at 25°C (Robinson and Stokes80) . The coefficients
gradient are temperature dependent and at 5°C the coefficients are about 50% smaller.
tion. Dif- Fick's second law describes the change of concentration over time inside the element
,-, so42-) (control volume) subject to diffusion flux,
(8.8.3)
where oC!ot is the change in concentration with time. The above expressions ofFick's first and
second laws are one-dimensional, but can be expressed more generally for three dimensions.
Diffusion in porous media does not proceed at the rate it can in water because ions follow
longer paths caused by the presence of particles in the solid matrix and because of adsorption
on the solids. To account for these phenomena, an apparent (or effective) diffusion coefficient
for nonabsorbed species in porous media flow is defined as
D*=wD (8.8.4)
where w is an empirical coefficient (<1) that takes into account the effect of the solid phase of
the porous media on the diffusion. Freeze and Cherry 33 suggest using the above effective
diffusion coefficient with w ranging from 0.5 to 0.01, to account for the tortuosity of the flow
path.
Because the processes of molecular diffusion and mechanical dispersivity cannot be sep-
arated in groundwater flow, the coefficient of hydrodynamic dispersion is used to account for
both. The longitudinal coefficient of hydrodynamic dispersion, Dv is expressed as
(8.8.5)
where CtL is the dynamic longitudinal dispersivity, a characteristic property of the porous
medium; v is the average linear groundwater velocity; and D* is the molecular diffusion
(referred to as the effective diffusion coefficient or the apparent diffusion coefficient).
A rough approximation of aL based on averaging published date (Gelhar et aJ. 38) is
(8.8.6)
where Lis the length of the flow path (m). For lengths less than 3,500 m, Neuman72 gave
aL =0.0175LI.46 (8.8.7)
For transverse (lateral) dispersivity, Uris typically l/10 to 11100 of the longitudinal dispersiv-
ity aL. Xu and Eckstein 109 used a statistical study to develop the following relationship
aL =0.83(log L)2.4 14 (8.8.8)
where L is in ft or m and CtL is in ft or m.
8.8.2 Advection-Dispersion Equation for Solute Transport in Saturated Porous Media

Consider the elemental volume (Figure 8.8.1) of homogeneous porous media of porosity n. F,
represents the total mass of solute per unit cross-sectional area transported in the x-direction
per unit time. The mass of solute is transported in the x-direction by advection and by disper-
sion expressed respectively as
Mass transported by advection =vJ1-C dA (8.8.9)
Mass transported by dispersion= nD,_2C/2x dA (8.8.10)
where dAis the elemental cross-sectional area of the cubic element and Dx is the dispersion
coefficient in the x-direction defined by
Dx = axvx + D* (8.8.11)
where ax is the dynamic dispersivity and axvx is the mechanical dispersion.
Fx is now represented as
3.8.3)
(8.8.12)
stand
;ions. with the negative sign for the dispersive tenn indicating that the contaminant (solute) moves
·oilow toward the zone of lower concentration. In a similar manner, F,. and Fz are respectively
rption
}cient FY = -
v,.nc -nDYac
- (8.8.13)
CJy
8.8.4)
(8.8.14)
1ase of
:ective The total solute entering (flux entering) the cubic element is
e flow
Fenrering = F,, dz dy + F,. dz dx + Fz dx dy (8.8.15)
Je sep- and the total solute leaving (flux leaving) the cubic element is
unt for
Fieaving = ( Fx + a;: dx }y dz +[ Fy + a~ dy }z dx +( ~ + ~ dz)dx dy (8. 8.16)
(8.8.5)
porous Partial tenns indicate the spatial change of the solute mass in the respective direction.
ffusion For a nonreactive dissolved substance, the flux into the element and the flux out of the
element is equal to the net rate of change of mass of solute within the element, defined as
IS
!J.F = -n-
ac dx dy dz (8.8.17)
(8.8.6) dl
ave Combining the above three expressions (Equations 8.8.15-8.8.17) and simplifying gives
(8.8.7) dF i1F,. i1F dC
_ x +-+.:.:.1.=-n- (8.8.18)
;persiv- dX oy az at
p Substituting Equations 8.8.12-S.8.14 into Equation 8.8.18 gives
(8.8.8)
[-dxa (Dx-
ac) +-a [Dac) a (Dac)]
- +-
dx dy y dy dz z dz
- - [-(vxC)+-
a _ a (-v C)+-(v
ax oy y
a _C)]=ac-
oz z dt
(8.8.19)
For a homogeneous medium with v steady and uniform in space and time, in which the dis-
persion coefficients DX' D,., and Dz do not vary through space, then the above Equation 8.8.19
.ity n. Fx.
lirection simplifies to
1 disper-
(8.8.20)
(8.8.9)
For one dimension, the conservation of mass (advection-dispersion equation) is
(8.8.10)
ospersion 0
a2c _v ac =ac (8.8.21)
x dX 2 x. dX dt
which can also be expressed along a flowline by using L for x where Lis the coordinate direc-
(8.8.11)
tion along the flow line. DL is the longitudinal coefficient of hydrodynamic dispersion and vL
is the average line~ velocity along the flowline.
8.8.3 Analytical Procedure

The analytical solution to Equation 8.8.21 is (see Ogata74 for details)
Initial condition C(x, 0) =0 x:::: 0
Boundary condition C(O, t) = C0 t;::: 0
Boundary condition C(oo, t) =: 0 t~ 0
0 [
C{x,t) =C-
2
erfc( x ~
2-v Dr)
l (l (
- vt +exp -vx erfc x +
~
DL
vt ]]
2-v DLt
(8.8.22) 8.8.4
where xis the distance from the injection point.

The argument of the exponential (v Lx!DJ is the Peclet number, Pe =vLx!Dv which is a
measure of the ratio of the rate of transport by advection to the rate of transport by diffusion.
Large Peclet numbers (Pe > 100) indicate that advection dominates. When advection domi-
nates, the second term (with the exponent) on the right-band side becomes negligible. Other
books that discuss contaminant transport include: Bear, 8 Bear and Verruijt, 9 Bedient et al., 10
Charbeneau/4 Delleur, 19 de Marsily,21 Domenico and Schwartz, 24 Fetter, 3 Freeze and °
Cherry, 33 Javandel et al.,49 and Palmer?7
The objective of this example is to illustrate the use of Equation 8.8.22 to compute the concentration of
a pollutant as a function of time and distance from a point or line source in an aquifer with known prop-
erties. The aquifer properties are: hydraulic conductivity= 2.5 X 10-5 rn/s; hydraulic gradient= 0.001;
effective porosity= 0.25; and an effective diffusion coefficient= 0.75 x 10-9 m2/s. A chloride solution
with a concentration of 600 mg!l penetrates (enters) the aquifer along a line source. Detennine the chlo-
ride concentration at a distance of 25 m from the source after one year, two years, and four years.
SOLUTION Step 1: Compute the pore velocity (average linear velocity) using Darcy's law:
5
_ Ki (2.5x1o- )0.001
VL. = - ;;;;; 1 X 10-? rn/s
n, 0.25
Step 2: Compute the longitudinal dispersivity using the approximation of Neuman,72 Equation 8.8.7:
aL. =0.0175LI.4Q
=0.0175(25)1.46
=1.92m
Using the approximation by Xu and Eckstein, 109 Equation 8.8.8, we have
aL =0.83(1og L)2.414
= 0.83(log 25)2·414
= 1.86m
Step 3: Compute the coefficient of longitudinal mechanical dispersion-diffusion (coefficient of longitu-

dinal hydrodynamic dispersion) using Equation 8.8.11 where al. = 1.86 m:
DL =(J.L.VL + D*
= 1.86 X (1 X 10-7) + 0.75 X 10-9
= 1.9 x 10-7 m2/s
Step 4: Use Equation 8.8.22 to compute the concentration for times of t = 1 year = 60 s/min x l ,440
min/day x 365 dayslyr = 3.15 x 107 s; t =2 years= 6.31 x 107 s; and t= 4years =12.6 x 107 s; vL =1 x
w- 7
rnls; Co= 600 mg/1; X= L:: 25m; DL = 1.9 X w- 7 2
m /s.
Fort= 1 year: C(25m, 1 yr) = 0.0 mg/1
Fort= 2 years: C(25m, 2 yr) = 0.037 rng/1
Fort= 4 years: C(25m, 4 yr) =21.6 mg/1
•
.8.22) 8.8.4 Transport of Reactive Pollutants
This section considers the transport of solutes that behave as nonreactive, but also have the
added influence of chemical reactions. Sorption is the exchange of molecules and ions between
:his a the solid phase and the liquid phase, including both adsorption and desorption. Adsorption is
USlOn. the attachment of molecules and ions from the solute to the rock material, causing a decrease
dorni- of concentration of the solute. This is also referred to as causing a retardation of the contam-
Other inant transport. Desorption is the release of molecules and ions from the solid phase to the
t al., 10 solute. A sorption isothenn is the relationship between the solute concentration in the adsorbed
~e and phase and in the water phase.
The one-dimensional transport, Equation 8.8.16, for advection-dispersion can be
extended to include the effects of retardation of solute transportation through sorption, chem-
ical reaction, biological transfonnations, or radioactive decay by including source-sink term.
ation of The form of the one-dimensional equation that includes retardation, in a homogeneous satu-
'll prop- rated media due to adsorption, adds a source-sink term:
: 0.001;
~ a c _v ac + Pb as = ac
2
solution (8.8.23)
he chlo- d/ 2 L d[ n dt dt
·s.
(dispersion term) (advection term) (reaction term)
where Pb is the bulk mass density of the porous medium, n is the porosity, and Sis the mass of
chemical constituent adsorbed on a unit mass of the solid part of the porous medium. ()S/dt is
the rate at which the constituent is adsorbed [MIMI] and (pbln)(dSidt) is the change in con-
centration in the groundwater caused by adsorption or desorption [M/(L3T)].
Adsorption relationships that can be plotted as straight-line graphical representations on
8.8.7: log-log paper are referred to as Freundlich isotherm, expressed as
log S =b log C + log Kd
(8.8.24)
S=Kdc"
where Sis the mass of solute species adsorbed or precipitated on the solids per unit bulk dry
mass of the porous medium, Cis the solute concentration, and Kd and b are coefficients. These
coefficients depend on the solute species, nature of the porous medium, and other conditions
of the system. b is the slope of the log-log relationship, so if it has a value of 1, then the
isotherm is a linear isotherm, so that Kd = dS/dC and is referred to as the distribution coeffi-
cient. Kd is a representation of the partitioning between liquid and solids only if the reactions
that cause the partitioning are fast and reversible, if the isotherm is linear. Many contaminants
that are of interest in groundwater studies meet these requirements.
>f longitu- The advance of the contaminant front is retarded as a result of the transfer by adsorption
(or other chemical processes) of the contaminant mass from the pore water to the solid part of
the porous medium. For a fast and reversible adsorption with linear isotherm, the partitioning
of the contaminant can be described by the distribution coefficient and the retardation of the
front relative to the bulk mass of water as it passes through the soil as described by the retar-
dation factor, R0 , expressed as the retardation equation,
(8.8.25)
where n is the porosity and Pb is the bulk mass density of the soil, grnlcm3; and Kd is the dis-
tribution coefficient for the solute with the soil. Also 118 = (n- 1)/n where 8 is the volumetric
content of the soil, which is dimensionless. The retardation factor ranges from l to 10,000. A
reactive solute will travel at a slower rate than the groundwater flow because of the adsorption.
The velocity of the solute front vc (where the concentration is 0.5 of the original concentration,
C/C0 = 0.5) is
(8.8.26)
so that the reciprocal of the retardation factor is referred to as the relative velocity.
The amount (or degree) of contaminant adsorbed by the solids is a function of the con-
centration in solution, S =f(C), so that
as as ac
--=-·- (8.8.27)
at dC dt
and
_h . as= Pb. as. ac (8.8.28)
n dt n ac dt
Using Kd =dS/dC and the above relation, the governing equation (8.8.16) can be expressed in
terms of the retardation factor to express the one-dimensional advection-<lispersion equation
as
ac v ac DL a2 c
- = ----+-- (8.8.29)
dt R11 ax R11 Cix 2 Figure
which defines the change in storage of contaminant in the control volume. The first term on the
right-hand side represents the retarded advective inflow-outflow and the second term repre-
sents the retarded diffusion and dispersion. Van Genuchten 106 present~d solutions of the advec-
tion/dispersion/adsorption equation.
8.9 MONITORING GROUNDWATER QUALITY

To protect a groundwater resource against pollution, a water quality monitoring program-
defined as a scientifically designed surveillance system of continuing measurements, obser-
vation, and evaluations-is necessary. Monitoring methods may include not only sampling and
analyses of groundwater quality, but also determination of groundwater levels and flow direc-
tions, measurements of moisture in the unsaturated zone, geophysical surveys, evaluations of
wastes and other materials contributing to subsurface pollution, testing of pipelines and tanks,
and aerial surveillance. 28 •69• 73
Facilities that are regulated under the Resource Conservation and Recovery Act (RCRA)
commonly require four types of monitoring programs (U.S. EPArn. 100• 101 ). (1) Detection mon-
itoring programs detect an impact to groundwater quality. (2) Assessment monitoring pro-
grams assess the nature and extent of contarn.inants and collect data for designing remediation.
(3) Corrective action monitoring programs assess the impact of a particular remediation
design. (4) Peiformance monitoring programs evaluate the effectiveness of an element of a
groundwater remediation system. As pointed out by Houlihan and Lucia,48 these four moni-
toring programs may also be useful in non-RCRA programs such as at Superfund sites, at sites
being voluntarily remediated, and at legally required remediation sites. Figure 8.9.1 shows the
sequence of activities for groundwater monitoring.
.8.25) I Prepare Groundwater Monitoring Plan Install Groundwater Monitoring Features

I
Step 1. Understand hydrogeological/ Step 1. Select appropriate installation
e dis- geochemical setting technique
Step 2. Evaluate background conditions Step 2. Procure appropriate monitoring
netTie Step 3. Design groundwater monitoring feature construction materials
I
OO.A system Step 3. Install monitoring features r-
ption. Step 4. Establish sampling and analysis according to design presented in
methods and data quality objectives Groundwater Monitoring Plan
·ation, Step 5. Establish data evaluation Step 4. Update conceptual model of site
methods hydrology based on findings
Step 6. Develop response criteria of installation program
.8.26)
I
~con-
Sample/Analyze Groundwater Analyze Data; Perfonn Contingency Actions
Step 1. Inspect wells and area Step 1. Select appropriate statistical

.. 8.27) surrounding wells analysis method
:.8.28)
- Step 2. Measure depth to groundwater
and total depth of well
Step 3. Evacuate groundwater from the
well
Step 4. Collect, preserve and ship
groundwater samples
Step 2. Perform statistical analysis of
data
Step 3. If statistically significant increase
is detected, reevaluate validity of
analysis
I Step 4. If statistically significant increase
Step 5. Complete documentation of is confirmed, then implement
;sed in sampling event appropriate response actions
uation Step 6. Analyze groundwater samples
for presence of contaminants
!.8.29) Figure 8.9.1. Sequence of activities for groundwater monitoring (Houlihan and Lucia48).
on the
repre- In developing a groundwater-monitoring program, federal and state regulations must be
advec- observed. Houlihan and Lucia48 provided the following questions as a guide to identifying
applicable groundwater monitoring regulations for a particular site or operation:
• What is the activity that could have an adverse impact on groundwater quality? Are
there related activities that could also have an adverse impact on groundwater quality?
• What specific requirements do the regulations contain for planning groundwater mon-
:ram- itoring, sampling groundwater, analyzing groundwater samples, and evaluating data
obser- from groundwater sampling?
.ng and
, direc- • Do the regulations contain specific requirements for action in the event that impacts to
ions of groundwater quality are identified during the monitoring program?
I tanks, Once the applicable regulatory requirements have been identified, the overall strategy of
the groundwater monitoring program should be defined prior to preparing the groundwater
{CRA) monitoring plan. This will help in guiding the development of the monitoring plan. Figures
nmon- 8.9.2 and 8.9.3 illustrate the development of a groundwater monitoring strategy. Potential
tg pro- sources of contamination and the aquifers of concern should be characterized before develop-
liation. ing a monitoring strategy. The selection of target monitoring zones cannot be made until the
diation source and the aquifer of concern have been evaluated through a hydrogeologic evaluation.
:nt of a A groundwater-monitoring plan is required on a site-by-site basis under the RCRA and the
:mom- Comprehensive Emergency Response and Compensation Liability Act (CERCLA), and for other
at sites regulated facilities under various state and local programs. These plans are developed to describe
JWS the each aspect of monitoring and to control the monitoring activities so that the overall goals of the
strategy are fulfilled. Table 8.9.1 is an example outline for a groundwater monitoring plan.
Generalized
groundwater
flow direction
Potiental source
Upgradient well Downgradlent well
0
- - ~---- -- -.- -
-~-$-~--) Shallow
) ~ Downgradient
monitoring wells
·Generalized groundwater
How direction
· unconfined
aquifer
Table
Thr
Water
Figure 8.9.2. Plan view of typical unconfined aquifer Figure 8.9.3. Section of typical unconfined aquifer groundwater monitor-
or·
groundwater monitoring system (Houlihan and Lucia48). ing system (Houlihan and Lucia48).
Fieldm
-Ten
Table 8.9.1 Sample Outline-Groundwater Monitoring Plan -Spe
1. Introduction -pH
2. Site Conditions -Dis:
2.1 Site Topography and Surface-Water Hydrology -Aci,
2.2 Site Geology (unl
2.3 Site Hydrogeology -Aik
3. Groundwater Monitoring System -Turt
3.I Introduction Major in
3.2 Existing Groundwater Monitoring Wells Nutrient
3.3 Construction of New Groundwater Monitoring Wells Filtered.
3.4 Development of New Groundwater Monitoring Wells Pesticidt
4. Groundwater Sampling Procedures
Volatile 1
4.l lnttoduction
Radon
4.2 Health and Safety Procedures
Trace ele
4.3 Record.keeping
Radium
4.4 Equipment Use, Cleaning, Decontamination
Uranium
4.5 Preparation for Sampling
Tritium, 1
4.6 Groundwater Sample Collection Procedures
chlor
4.7 Groundwater Sample Handling and Preservation
Environrr
4.8 Shipment of Groundwater Samples
5. Laboratory Groundwater Analytical Program [Required·
6. Quality Assurance and Quality Control Procedures of national
pounds. Th
6.1 Data Quality Objectives nent Req,:
6.2 Field Sampling QNQC Requirements 1
Selection .
6.3 Laboratory QA/QC Requirements piing, howe
7. Evaluation of Groundwater Data 2
Schedules
7.1 Introduction National PI
7.2 Groundwater Quality Data Validation approval. A
3
7.3 Evaluation of Static Groundwater Elevations ANC (for(
7.4 Statistical Analysis Procedures for Groundwater Quality Data could have
7.5 Response Actions 'Turbidity r
centrations
7.6 Reporting Requirements for Groundwater Quality Data 5
VOCs are 1
8. References land-use stu
Source: Houlihan and Lucia48
The National Water-Quality Assessment (NAWQA) Program has developed groundwater

data-collection protocols and procedures for the selection, installation, and documentation of
related data (Lapham et al.59) and also for the collection and documentation of water quality sam-
ples and related data (Koterba et al. 58). Three types of studies are defined in the NAWQA pro-
gram: (1) study-unit (or subunit) surveys, designed to obtain occurrence and distribution data on
lll a variety of analytes; (2) land-use studies; and (3) flowpath studies, which assess spatial differ-
ences and possibly temporal variability in each of a selected number of analytes among wells
located in different parts of a local groundwater flow system. Table 8.9.2 provides a summary of
required, recorrunended, and optional water-quality constituents measured in the three studies.
II ow
tfined Table 8.9.2 Summary of Current (1995) Required, Recommended, and Optional Water-Quality Constituents to Be Measured in the
1ifer Three National Water-Quality Assessment Program Groundwater Components of the Occurrence and Distribution Assessment
Water-quality constituent
nitor- or constituent class Study-unit survey Land-use studies Flowpath studies 1 Method2
Field measurements
-Temperature Req Req Req Field
-Specific electrical conductance Req Req Req Field
-pH Req Req Req Field
-Dissolved oxygen Req Req Req Field
-Acid neutralizing capacity (ANC) Rec Rec Rec Field incremental
(unfiltered sample)3
-Alkalinity (ftltered sampld Req Req Req Field incremental
-Turbidit/ Rec Rec Rec Field
Major inorganics Req Req Req NWQLSC2750
Nutrients Req Req Req NWQLSC2752
Filtered organic carbon Req Req Opt NWQL SC2085
Pesticides Req Req Opt NWQL SC200l/2010
NWQL SC2050/2051
Volatile organic compounds (VOCs) Req Req or Opt5 Req or Opt6 NWQLSC2090
Radon Req Req or Rec7 Req or Rec6 NWQLLC 1369
Trace elements4 Opt Opt Opt NWQLSC2703
Radium Opt Opt Opt NWQL-Opt
Uranium Opt Opt Opt NWQL-Opt
Tritium, tritium-helium, Rec Rec Rec NWQL LC1565 (tritium)
chlorofluorocarbons (CFCs) 8
Environmental isotopes9 Rec Rec Rec NWQL-Opt
[Required water-quality constituents to be measured for the Occurrence and Distribution Assessment are determined partly by the water-quality topics
of national interest selected for National Assessment. Topics selected for National Assessment (1994) are nutrients, pesticides, and volatile organic com-
pounds. The topics selected can change over time. Quality-control samples also are required-types of quality CQntrol samples depend on study compo-
nenl Req, Required; Rec, Recorrunended; Opt, Optional; NWQL, National Water-Quality Laboratory; SC, Schedule; LC, Laboratory Code]
1
Selection of constituents for measurement in flowpath studies is determined by flowpath study objectives. During at least the first round of sam-
pling, however, the broad range of constituents measured in study-unit surveys and land-use studies would be measured.
2
Schedules and laboratory codes listed are required for study units that began their intensive phase in !991 or 1994, and apply until changed by
National Program directive. Schedules for radium and uranium can be selected by the study unit, but require NAWQA Quality-Assurance Specialist
approval. A detailed discussion is found in Koterba et at.58
1
ANC (fonnerly referred to as unfiltered alkalinity) is measured on an unfiltered sample. Alkalinity is measured on a filtered sample. Study unit
could have collected ANC, alkalinity, or both to date.
~urbidity measurements are required whenever trace-element samples are collected to evaluate potential colloidal contributions to measured con-
centrations of iron, manganese, and other elements.
5
VOCs are required at all urban land-use study wells, but optional in agricultural land-use studies. IfVOCs are chosen as part of an agricultural
land·use study, then they should be measured in at least 20 of the land-use study wells.
(continues)
Thble 8.9.2 (continued) Summary of Current (1995) Required, Recommended, and Optional Water-Quality Constituents to Be Measured
in the Three National Water-Quality Assessment Program Groundwater Components of the Occurrence and Distribution Assessment
6
VOCs are required at all urban flowpath wells for at least the ftrst round of sampling. If VOCs are measured in an agricultural land-use study, then ... -~ .
they should be measured at all flowpath-study wells within that land-use study for at least the ftrst round of sampling.
7
Radon is required at any land-use or flowpath study well if that well also is part of a study-unit survey; otherwise. radon collection is recommended
for land-use or flowpath study wells located in likely source areas.
8Collection of tritium, tritium-helium, chlorofluorocarbons (CFCs), and/or other samples for dating groundwater is recommended, depending on
hydrogeologic setting. For tritium methods, see NWQL catalog; for CFCs, see Office of Water Quality Technical Memorandum No. 95.02 (unpub-
lished document located in the USGS Offtce of Water Quality, MS 412, Reston, VA 22092).
For a general discussion of the use of environmental isotopes in groundwater studies, see Alley:1
9
The study-unit survey can involve data collection from as many as 100 to 120 wells asso-
ciated with multiple land uses, with wells distributed among several subunit surveys, each of
about 30 wells (Koterba et al.58 and Lapham et al. 59). The flowpath studies used for water-qual-
ity data collection involve about 20 wells with most of the wells in a single aquifer that under-
lies a single land use. Figure 8.9.4 shows the general design of a monitoring well in
Vented locking
protective cover
Open
boreho
(a)
···..··
......._.·.:.·· Figure 8.9.4. General design of a
monitoring well in unconsolidated
deposits for National Water-Quality (a) Monltc
scree1
-.. Assessment Program Land-Use and initsc
Not to scale Aowpath Studies (Koterba et at. 58).
sured
!Dt
then
:-:::.-: ·-: :~ :-:\/:_':"(-:;
· .. ··.· · ·· . . .
ended
~Well casin g/
and annular
1
seal 1
I
I
I
asso-
.ch of Semiconsolidated I
deposits or rock I J--Pipe assembly
-qual-
nder- I 1 for packers
ell in I I
I I
Filter
packs
Open-!._
borehole I Packer
I
L_____j
Figure 8.9.5. Examples of monitor-
L....__.._j ing-well designs in semiconsolidated
(a) Open borehole (b) Packers In bofehole (c) Well screen in bofehole deposits and in rock (Lapham et al. 5~.
unconsolidated deposits for the land-use and flowpath studies. Figure 8.9.5 shows examples of
monitoring-well designs in semiconsolidated deposits and in rock. Figure 8.9.6 shows exam-
ples of three well-cluster designs. Figure 8.9.7 shows the schematic setup for well purge and
sample collection. The American Society for Testing Materials (ASTM5) and the U.S. Envi-
ronmental Protect.ionAgency (EPA)97 also provide standard approaches for design of ground-
water monitoring wells.
:.··.
Anmila~- : :·
seal · ·>
. . ·. · ·'.: / Annular
·. •, ·/._.· seal .·.·· Multiport··.-:_.
sampler.:_::
_.. ··. ·
Well
Filter
of a pack
!ated Figure 8.9.6. Examples of
(a) Monitoring wells YMh short (b) Multiple monltonng wells (c) A series of multipart samplers
!uality screens, each Installed YMh short screens Installed installed in a single borehole three well-cluster designs
se and In Its own borehole in a single borehole (Koterba eta!.58) .
al.ss).
From the portable pump used lor a monitoring

wei, or Iran the garden-hos.lhreaded
connedion for a water-supply well
Extension
lkle
Extension
line
SarnpiKOIIedion
chamber
- - Rlgld-wall Teflon tubln11 @ Antllqcbiphon
8 Quick eonnedlon <B Three-way Teflon flow valve
Flow dlreetion,at di1ferent times Field ten•ors (tlowthrough dlamber)

(1) DIMg lnlllal PII'Q8 staoe DO Dissolved-oxygen sensor
(2) During lntennedlate and final stages T Temperature sensor
(2a) To obtain most fteld measll'emenls pH pH sensor

Figure 8.9.7. Schematic of equipment
(2b) To oblail turbidity samples, and sc Specific electrical conductance sell.lp for well purge and sample collec-
at end of purge to route !low sensor
for oolledion tion {Koterba et al.58).
8.10 REMEDIATION OF CONTAMINATED GROUNDWATER

8.10.1 Remediation Goals
Needs for remediation of contaminated groundwater can be established using the results of a
groundwater assessment or monitoring program. Remediation is a broad tenn that refers to the
reduction of risk caused by exposure to contaminated groundwater. Alternatively, remediation
seeks to protect human health and the environment and to restore groundwater to beneficial Adapted
8.10 Remediation of Contaminated Groundwater 393
uses where practicable, ideally to restore groundwater to drinking water standards. In the
United States, cleanup goals under CERCLA and RCRA generally are set at drinking water
standards. Maximum Contaminant Levels (MCLs) have been established under the Safe
Drinking Water Act. Cleanup requirements may vary for groundwater that is not used for
drinking water.
To determine an appropriate strategy to manage contaminated groundwater, site condi-
tions must be evaluated and remediation goals must be defined. One of the most important
steps in a remediation program is to define goals, such as (NRC71 ):
• Complete restoration, which involves removal of all contaminants from the contami-
nated aquifer;
• Nondegradation, which involves removal of contaminants that exceed either the detec-
tion limits of available analytical equipment or background concentrations;
• Remediation to health-based standards, which involves removal of contaminants that
are present at a concentration that could cause adverse health effects (some examples
of health-based standards are maximum contaminant levels (MCLs), alternate concen-
tration limits (ACLs), and local or federal drinking water standards);
• Remediation to the limits of technology-based standards, which involves use of the best
available technology to remove as much of the contaminants as possible;
• Partial-use restrictions (or institutional controls), such as legal restrictions on the use
of groundwater in areas where groundwater has been contaminated, or physical barri-
ers (e.g., fences) to prevent access to contaminated media;
• Containment, which involves the use of engineered systems for preventing migration of
the contaminants to locations where receptors could be exposed to the contaminants.
Table 8.1 0.1 lists advantages and disadvantages of each of the above remediation goals.
Once a remediation goal is selected, a remedy must be implemented to achieve the goal. Three
major remedies are (1) containment of groundwater, (2) groundwater extraction, and (3) treat-
ment of contaminated groundwater.
Table 8.10.1 Advantages and Disadvantages of Cleanup Goals
Goal Advantages Disadvantages

Complete restoration Eliminates all risk Likely impossible
ment Nondegradation Reduction of contaminants to lowest Extremely difficult, expensive, and
ollec- level measurable time-consuming for many contaminants
and hydrogeologic settings
Health-based standards Designed to prevent measurable Difficult to defineand may not accurately
impacts to human health or environment address all possible health impacts of
exposure to contaminated groundwater
Technology-based Allows treatment to the best capabilities May not reduce risk to a level that is protec-
standards of current technology tive of human health and the environment
Partial-use restrictions Prevents contact between contaminants Leaves contaminants that could cause risk if
and receptors in a cost-effective manner partial use restrictions are ineffective
llts of a Containment Relatively predictable and reliable; Leaves contamination that could migrate if
rs to the typically less costly than other containment system fails
~diation
remediation approaches
:neficial Adapted from National Research CounciJ1 1 with permission
8.10.2 System Design

Design of an efficient, cost-effective groundwater remediation system requires (Houlihan and
Lucia47 ):
• A comprehensive understanding of the nature and extent of contamination;
• Remediation goals;
• A careful evaluation of remedial technologies and their abilities to meet the objectives. Conv·
Vacut
The steps in a groundwater remediation system design are as follows (Houlihan and Lucia47): Airs~
• Step 1-Define the problem. or h

In-sin
• Step 2-Define the goal of the groundwater remediation (see Table 8.10.2). hydJ
• Step 3-Screen candidate remedies (see Table 8.10.3) for a summary of commonly In-sitt
used groundwater remediation technologies. chlo
• Prepare preliminary conceptual design for each candidate remedy. Coso]'
Steam
• Evaluate effectiveness of conceptual designs.
In-situ
• Evaluate cost of remedies. In-situ
• Step 4--Prepare detailed design. In-situ
• Step 5-Implement the design. Intrins:
• Step 6--Confirm the effectiveness of the design.

Conve1
Air spa
8.10.3 Hydraulic Containment of Groundwater hori.zt
In-situ
Hydraulic containment can be used at sites where the contaminant source cannot be removed,
hydro
such as at landfills or in bedrock with DNAPLs. The most widely used types of hydraulic con-
ln-situ 1
tainment are physical barriers and hydraulic barriers. Physical barriers are vertical features
chlori
In-situ r
Table 8.10.2 Candidate Groundwater Remediation Goals and Corresponding Remediation Approaches Intrinsi<
Note: A·
Remediation goal Examples of candidate remediation approaches
"long" cJ
Groundwater restoration Natural (intrinsic) remediation nology a1
Bioremediation 'Becaus(
Combinations of conventional remediation approaches b "Limite
Nondegradation Source removal availabili:

Conventional pump-and-treat Adapted 1
Bioremediation SVE!AS
Return aquifer to health-based standards Source removal
Conventional pump-and-treat
Bioremediation SVEJ AS
Apply technology-based standards Pbytoremediation
Electrokinetics
Solvent extraction
Thermal desorption
Implement restricted-use policies Deed restrictions
Local ordinances
Containment Physical barriers
Hydraulic barriers
Capping
Adapted from National Research Council71 with permission
Table 8.10.3 Summary of Commonly Used Groundwater Remediation Technologies
tand Residual Residual sorbed Number of

Alternative groundwater concentration peer-reviewed
technology concentration In source area Cleanup timea publicationsb
Source Remediation
Conventional pump-and-treat Low to medium Medium to high Long Some

jves.
Vacuum extraction and bioventing NA Low to medium Short Some
ia47): Air sparging (vertical Low to medium Low to medium Short to medium Limited
or horizontal wells)
In-situ bioremediation- Low to medium Low to h.igh Short to medium Some
hydrocarbons
1only In-situ bioremediation- Low to medium Low to high Medium to long Some
chlorinated solvents
Cosolvent and surfactant flushing Low to medium Low to medium Short to medium Some
Steam stripping Low to medium Low to medium Short Some
In-situ thermal desorption Low to medium Low to medi urn Short Some
In-situ chemical oxidation Medium(?) Medium to high (?) Medium(?) Limited
In-situ bioremediation-metals Low to medium Low to high Medium to high Limited
Intrinsic bioremediation Low to medium Low to high Long Limited
• :1
Plume Remediation
Conventional pump-and-treat Low Medium to high Long Many
Air sparging (vertical or Low to medium Low to medium Medium to long Limited
horizontal wells)
In-situ bioremediation- Low to medium Low to high Medium to long Some
1oved, hydrocarbons
.c con- In-situ bioremediation- Low to medium Low to high Medium to long Some
:atures chlorinated solvents
In-situ reactive barriers Low NA Long Limited
~oaches
Intrinsic bioremediation Low to medium Low to medium Long Limited
Note: A "low" residual concentration and ''medium" cleanup time reflect relatively good performance, while a "high" residual concentration and
tes "long" cleanup time reflect much less effective performance. "NA" denotes that the technology is not applicable to this situation. Cost of the tech-
nology and feasibility for different situations are not addressed in this table but should be carefully evaluated before selecting a technology.
• Because few cases of achieving cleanup goals have been reported, these qualitative assessments reflect the judgment of National Research Council.
·oaches b "Limited" indicates that very little information about this technology is available in peer-reviewed publications, while "some" indicates a greater
availability of information.
Adapted from National Research Council71 with permission, as presented in Houlihan and Lucia47
(consisting of a variety of different materials, as listed in Table 8.10.4) in the ground that are
barriers to the flow of groundwater. They can be nonselective physical barriers, which obstruct
the flow of all groundwater, or selective physical barriers, which obstruct only the migration
of target contaminants. Cutoff walls, which limit migration of groundwater by forming a phys-
ical barrier to the flow, are the oldest and most commonly used. Soil-bentonite slurry walls are
the most commonly used in geotechnical and environmental remediation projects (Houlihan
and Lucia47).
Hydraulic barriers create a depression in the piezometric surface of groundwater, acting as
a barrier beyond which groundwater within the zone of influence of the barrier should not flow.
They can be fonned using trenches, wells, or other methods that remove groundwater. Other
options for hydraulic containment include horizontal drains, geomembranes, and well points.
Table 8.10.4 Types of Physical Barriers
Barrier Width Maximum depth Unit cost Production rate

type (ft) (apx) (ft) ($/SF) per I0 hrs (SF}
Soil bentonite 2-3 80 2-8 2,500-15,000
Cement bentonite 2-3 80 5-18 1,000--8,000
Biopolymer drain 2-3 70 7-25 1,500-5,000 Surfa
Deep mixing 2.5 90 6-15 1,000-8,000 Static
;.l;;''o.._:_
DM structural 2.5 90 15-30 ] ,000-3,000

Jet grouting 1.5-3 200 30-80 300-2,500
Grout curtain One row 200 40-100 200-1,000
Sheet piling One sheet 150 15-40 2,000-15,000
Geochemical barrier Varies Hundreds 8-400 100-1,000
Source: Rumer and Mitchell82
8.10.4 Groundwater Extraction Systems

Groundwater extraction systems, the most common being extraction well systems and extrac-
tion trench systems, are a combination of subsurface components and above-ground compo-
nents. The subsurface components are the wells or trenches, which provide access to the
contaminated water, and the above-ground components are used to regulate and monitor the
extraction process. A typical groundwater extraction well system is shown is Figure 8.10.1 and Static
a typical groundwater extraction trench system is shown in Figure 8.10.2.
8.10.5 Treatment of Contaminated Groundwater

A large number of groundwater treatment technologies have been developed as described in
U.S. EPA. 103 These technologies can be classified as in-situ or ex-situ, as summarized in Table
8.10.5./n-situ treatment techniques either render a contaminant nonto?Cic through treatment or
enhance extraction of the contaminant from the aquifer. Ex-situ ·treatment techniques treat
groundwater that has been extracted from the aquifer. These are summarized in Table 8.10.6.
In-situ treatment technologies are generally designed to perform one or more of the fol-
lowing functions (Johnson51 ):
• Remove contaminant source zone
• Restore aquifer
• Prevent or minimize continued contaminant migration
A single technology rarely accomplishes all three of these goals. Contar
sou
Houlihan and Lucia47 discuss in-situ treatment techniques as subdivided into biological
remediation (bioremediation) techniques, volatization processes, and chemical and physical
processes. Bioremediation is the biological degradation of contaminants using naturally occur-
ring microbes in the soil. Volatization processes transfer the chemical from the liquid state to
the gaseous state (vapor phase), because vapor phase contaminants are typically easier to
remediate: Aquifer materials are significantly more permeable to vapors than to liquids, mak-
. ing vapors easier to remove from the aquifer. Johnson51 subdivides in-situ techniques into the
following three categories (see Figure 8.10.3 for most common technologies used);
• Source zone treatment technologies :
• Free-product recovery
• Excavation and disposal or above-ground treatment Gravel t
back
• In-situ soil venting
8. I 0 Remediation of Contaminated Groundwater 397
Contaminated
Treatment .....____~----... groundwater Storage
e system tank
)
Pump Pump
LNAPL pump
Surface of LNAPL LNAPL product
- ·-·- ·--·- ·-·- ·
$tatlc ~nct.vater level
-·- ·-·-·- ·- ·--.......
/ f- - · - · -·- - · -
- -------- - ~ - -- - -·--
Conductance Aquifer
sensor
0 Large-diameter Groundwater pump
borehole
(a)
Contaminated
Treatment L..o.-.....----... groundwater Storage
xtrac- system tank
lmpo- Pump Pump
to the
or the
.land Static groundwater level
--·- ·--·--·-·-·--." - - -- ~----- - --
Aquifer
Groundwater pump
bed in . Static DNAPL level
1Table
nent or
~s treat
U0.6. . .· .·
:he fol- Figure 8.10.1. (a) LNAPL dual-extraction sys-
tem (b) DNAPL dual-extraction system (from
(b) Houlihan and Lucia47).
ElectricaVmechanical
control panel
(1 per sump)
Jlogical
Soil trench
thysical backfill
1 occur-
state to
asier to Removal pipe
is, mak-
Manhole/sump
into the {typical)
Pump (typical)
Figure 8.10.2. Typical groundwater

extraction trench system (Houlihan and
Lucia47).
Table 8.10.5 Summary of Candidate Treatment Technologies
In situ Process description Groun

treat
Bioremediation Biological degradation of contaminants using naturally occurring microbes in soil
techn
Soil vapor extraction Volatization of contaminants that are present in the vadose zone
Air sparging Volatization of contaminants in the saturated zone
Permeable reaction barriers Physical or chemical treatment in a trench Air str
Vacuum vapor extraction Volatization, within a well, of contaminants from saturated zone
Density driven convection Enhanced bioremediation using single-well driven conv~;;ction system in a4uifer. Liquid
Ex situ Process description Steam
Bioreactor Biological degradation of contaminants (activated sludge, fixed-film biological reactor, Memb1
biophysical treatment)
Oxidati
Slurry-phase biological treatment Variation on bioreactor in which contaminants are treated in a slurry form
Air stripping Volatization of contaminants
Activat
Carbon adsorption Adsorption of contaminants to activated carbon
Ion exchange Exchange-type attachment of contaminants to ion-exchange resin Fixed-f
Alkaline precipitation Alteration of water quality (usually pH adjustment) such that concentration exceeds the biolof
compound's solubility limit, causing precipitation Biophy:
Membrane Separation of solids from water using membranes (reverse osmosis, ultrafiltration)
Wetlands treatment Uptake of contaminants by wetland features
Electrokinetic decontamination Desorption of contaminants by "acidic front" of groundwater caused by hydrolysis of
the groundwater Alkalint
47
Source: Houlihan and Lucia Coagula
Ion exd
Adsorpt
Filtrati01
• Bioventing
Reducti<
• In-situ air sparging Membra
• Enhanced in-situ soil venting with soil heating and/or soil fracturing
• In-situ vitrification
Oxidatio
• Phytoremediation
Source: U
• Groundwater pump and treatment systems
• Aquifer restoration techniques:
• Groundwater pump-and-treat systems
• Natural attenuation
(a)
discharge 1
• In-situ air sparging
• Enhanced bioremediation .....
reinjectio1
• Contaminant migration prevention (see Figure 8.10.4):

• Natural attenuation
• Groundwater pump-and-treat systems
• In-situ reaction walls
• In-situ air sparging curtains
• Infiltration barriers
• In-situ contaminant options (grout walls, sheet piling walls, etc.)
Source zone treatment technologies treat the cause of the groundwater contamination so that Figure 8.1
they target the removal and destruction of the residual contaminants in the soil that are the (Johnson51
Table 8.10.6 Summary of Selected Ex-Situ Groundwater Treatment Technologies
Groundwater
treatment Status of
technology Representative examples Residual streams technology
Organic contaminants
Air stripping Packed towers, surface or diffused aeration Air stream with VOCs Commercial
removal of volatile compounds; soil venting
Liquid-phase GAC removal of broad spectrum of VOCs GAC for regeneration or disposal Commercial
Steam stripping Packed tower with steam stripping, removal Recovered solvent Some commercial
of low volatile organics
Membranes Ultraft.!tration for removal of selected organics Concentrated brine side stream Commercial
Oxidation Ozone/UV, or ozone/H20 2, destruction of None Some commercial in
chlorinated organics development stages
Activated sludge Oxygen or air biological oxidations for Sludge Commercial
removal/destruction of degradable organics
Fixed-film Fixed-film fluidized bed, for oxidation of less Sludge Commercial
biological reactors degradable organics
Biophysical Powdered carbon, with activated sludge, Powdered carbon and bacterial Commercial, PACT
treatment of high strength waste waters process
Inorganic contaminants
Alkaline precipitation Heavy metals removal Hazardous sludge Commercial
Coagulation Ferric sulfate or alum for heavy metals removal Hazardous sludge Conunercial
Ion exchange Heavy metals; nitrate Regeneration stream Commercial
Adsorption Selenium removal on activated alumina Regeneration stream Commercial
Filtration Removal of clays, other particulates Backwash wastes Commercial
Reduction S02 reduction of CR(VJ) Sludge Commercial
Membranes Reverse osmosis, ultrafiltration for removal Concentrated liquid waste Commercial, new
of metals, other ions membranes are
under development
O;Udation Fe{II) and Mn(ll) Sludge Commercial
102
Source: USEPA
(a) (b) blower compressor

dischargc or air
.._.
reinjection water
treatment
vapor
treatment
diSSOlved
plume
t so that Figure 8.10.3. Example aquifer restoration process schematics. (a) Groundwater pump-and-treat (b) In-situ air sparging
'are the (Johnson51)
(c) monitoring wells

[
source zone
Figure 8.10.3 (continued). Example

aquifer restoration process schematics.
(c) Natural attenuation (Johnson51 )
(a) (b)
compressor
sheet piling sheet piling
source zone
Figul
8.11
8.11.
(c)
Figure 8.10.4. Example contaminant migration technology processes schematics. (a) Physical barriers {b) In-situ air sparging cur-
tain {c) Reaction barrier (Johnson51 )
source. Aquifer restoration technologies are employed before, during, or after source zone
treatment to target the treatment of dissolved contaminant plumes. Contaminant migration pre-
vention technologies are employed at sites where the source zone location is unknown and/or
there are no practicable source zone and aquifer restoration options in order to minimize future
impacts of contaminants on groundwater.
Site assessment
and conceptual model
development
Select target treatment levels .
No
Remediation necessary?
ple Yes
•tics. Identify potential technologies
Yes
Screening level calculations
ssor No
Technology appropriate?
Yes
Select alternate
Pilot test technology or change
remedial goals
No
Technology appropriate
Yes
Figure 8.1().5. Generalized technology selection and design flowchart (Johnson51 ).
8.11 CONVENTIONAL PUMP-AND-TREATMENT SYSTEMS

8.11.1 Pump-and-Treat Remediation Strategies
Groundwater pump-and-treatment systems are most often used for hydraulic containment pur-
poses. Conventional pump-and-treat systems (see Figure 8.11.1) extract the contaminated
groundwater and treat it on the surface before discharging or reinjection. As described in Sec-
tion 8.10, to determine an appropriate strategy, first an evaluation of site conditions followed
by a definition of the remediation goals must be perfonned. Conflicting objectives for restora-
tion may generally include: (1) reduce contaminant concentrations to cleanup standards, (2)
maximize mass removal, (3) minimize cleanup time, and (4) minimize costs.
The strategies for managing groundwater contamination using pump-and-treat technology
include (Cohen et al. 17): (1) hydraulic/physical containment, (2) groundwater quality restora-
tion, and (3) mixed objective strategies. The selected management strategy depends on site-
specific hydrogeology and contaminant conditions, and remediation goals. Figure 8.11.2
gcur- illustrates several groundwater contamination management strategies using pump-and-treat
technology.
The use of pump-and-treatment for the hydraulic containment of dissolved containment
by pumping groundwater from wells or drains is iJJustrated in Figure 8.11.3. Groundwater
:e zone cleanup is typically more difficult to achieve than hydraulic containment; however, for sites
.on pre- where the contaminative source has been removed or contained, pump-and-treat technology
• and/or may be used for cleanup of a dissolved plume. The use of pump-and-treat technology for
e future aquifer restoration generally combines hydraulic containment with more aggressive manipu-
lation of the groundwater.
- -Semi-confined
-
\I
\ ' .' ~ -· ' - I
\ '
Overburden .. , - - Bedrock
Figure 8.11.1. Example of
~1§:: Sand --....Flow line
a pump-and-treat system
//J Silt ~ 'f _ Water table under
(after Mercer et al.,65 as pre-
. Clay pumping conditions
sented in Cohen et aL 1\
Today Future
Plume containment
Plume cut-olf
Downgradlent equllef retloratlon

Figur«
well w
8.11,;
Downplont aquifer rfltOfatlon
Aqv!Mr Mlllratlon
Pumping wells lot ctee.nup
Restored aqu~"'
Figure 8.11.2. Several groundwater contamination
. . - .. management strategies using pump-and-treat technol-
ogy (after NRC,71 Cherry et al., 15 as presented in
Cohen et al. 1\ ·
Static water table
>leof
as pre-
1\
...._ Upgradient
barrier wall
Figure 8.11.3. Examples of hydraulic containment in plan view and cross section using an extraction well (a), a drain (b), and a
well within a barrier wall (c) (Cohen et al. 1\
8.11.2 Characterizing Sites for Pump and Treatment Design

Site characterization should have as the main goal to obtain sufficient data to select and design
a remedy (NRC71 ). Investigations are made of (Cohen et al. 17) the following:
• Nature, extent, and distribution of contaminants in source areas and down-gradient plumes;
• Potential receptors and risks posed by contaminated groundwater; and
• Hydrogeologic and contaminant properties that affect containment, restoration, and
system design in different site areas.
Figure 8.11.4 illustrates the types of data used to develop a site conceptual model for rem-
nation edy assessment. hnportant goals for contaminant characterization data include (Cohen et al. 1\
technol-
din 1. Delineating contaminant source areas and release characteristics;
2. Defining the nature and extent (horizontal and vertical) of contamination;
Contaminant Sources Nature and Extent of

Contamination
• Location and characteristics of • Spatial distribution of

continuing near-surface releases subsurface contaminants
and sources
• Types and concentrations
• Location of subsurface sources of contaminants
(NAPL pools, residual NAPL,
metal precipitates, etc.)
. ~
c
•(
d
•1\i
Ct
rE
Hydrogeologic Setting
• Contaminant receptors
• Description of regional and site Restoration Potential • 81
geology so
• Estimates of contaminant mass co
• Stratigraphy (thickness and
lateral extent of units, • Temporal trend in contaminant • Pi
preferential pathways, etc.) concentration
• Depth to ground water • Sorption data
• Hydraulic gradients • Contaminant transformation
processes and rates
• Hydraulic conductivity, storage
coefficient, porosity distribution • Contaminant migration rates
• Temporal variation in water levels • NAPL properties Figure 8.11.4. Types of data used to develop a
site conceptual model for remedy assessment
• Groundwater recharge and discharge • Other characterls~cs that affect
transport and fate (modified from U.S. EPA99 as presented in
• Re
· • Groundwater/surface water Interactions Cohen, et al. 17). eve
sut
• Re
3. Characterizing contaminant transport pathways, processes, and rates; de~
4. Estimating risks associated with contaminant transport; and • Ret

per
5. Assessing aquifer restoration potential.
• Sy~
A site conceptual model is used to formulate remedial strategies, restoration and/or nee
containment.
Many systems will operate for decades in order to clean up and contain contaminated
groundwater, because of the slow contaminant transport and interphase transfer. This results in
the use of a phased and integrated approach to site characterization as illustrated in Figure
8.11.5. The performance of the pump-and-treat systems is assessed by measuring hydraulic
• Site background and

history review
• Preliminarv site
conceptual model
• Preliminary site
investigation
Drum and soil Excavation and

capping of lagoon
• Nature and extent of
contamination defined
• Containment systems
designed
• Near-surface
contaminant sources
removed
• Subsurface contaminant Monitoring

sources Identified and systems
contained
Source and plume
• Pilot studies conducted containment
pumping
Cleanup pilot
!lop a
1ent
n
• Restoration potential
evaluated for site
subareas
• Remediation systems
designed and installed
• Remediation system
performance monitored Figure 8.11.5. Iterative phases of site
characterization and remediation (modi-
• System adjusted as
and/or necessary Full-scale pump and treat fied from U.S. EPA,99 NRC,71 as presented
implementation in Cohen et al. 1\
ninated
!SUits in heads and gradients, groundwater flow directions and rates, pumping rates, pumped water and
, Figure treatment system effluent quality, and contaminated distributions in groundwater and porous
rdraulic media (see Cohen et al. 16• 17).
8.11.3 Capture Zone Analysis

The capture zone of an extraction well or drain is the portion of the subsurface containing
groundwater that ultimately will discharge to the well or drain (see Figures 8.11.3 and 8.11.6).
The capture zone of a well is not coincident with the drawndown zone of influence, as illus-
SOl
trated in Figure 8.11.6. The shape of the capture zone depends on the natural hydraulic gradi-
ent as well as the pumping rate and transmissivity, whereas the drawdown zone of influence
depends largely on the pumping rate and transmissivity alone. There have been many models
developed to detennine capture zones, ranging from simple graphical and semianalytical
methods (Javandel and Tsang,50 Blanford et aJ., 11 Strack et al.,87 and Haitjema et a1.43) to
numerical models for groundwater flow and contaminant transport (see Chapter 9) to opti-
mization models that link numerical models (see Chapter 10). Refer to van der Heijde and
Elnawawy 105 for a summary of models. 8.11
.. .......::;:..
(d) Cross section

Partially-penetratong
trench drain
PRO:
8.1.1 (
tory (T
your zi
8.1.2 (
amina11
discuss
the che
OGWI
Figure 8.11.6. (a) Illustration of drawdown contours (i.e., zone of influence) and the capture zone of a single pumping well in a 8.1.3 c
unifonn medium. Equations for the dividing streamlines (w = Q/211) that separate the capture zone of a single well from the rest of sheet o
an isotropic, confined aquifer with a uniform regional hydraulic gradient are given in (b), where T =transmissivity, Q =pumping site, idt
rate, and i = initial uniform hydraulic gradient. Simplified capture zone analysis methods may provide misleading results when contam
applied to more complex problems, such as those dealing with heterogeneous media, as depicted in (c), where K = relative status o
hydraulic conductivity, and tluee-dimensional flow ((/)(Cohen et al.1\ gov/suJ:
Problems 407
EXAMPLE 8.11.1 Determine the maximum width of the capture zone for a well that is pumping 2,200 m3/day from a con-
fined aquifer with a hydraulic conductivity of l ,200 m!day, an initial hydraulic gradient of 0.0008, and a
mg saturated thickness of 35 m. What is the distance from the well to the stagnation point?
.6).
lus- SOLUTION Maximum width of the capture zone (w) is computed using the following equation from Figure 8.11.6:
adi-
:nee w =Q/(2Ti) =Q/(2Kbi) = 2200 m3/day/[2( 1200 m/day)(35 m)(0.0008)) = 32.7 m
dels Distance from the well to the stagnation point (refer to Figure 8.11.6) is computed using
:ical
') to
>pti-
X= Q/(2rtKbi) = w/rt = 32.7 m/3.14 = 10.4 m
•
and
8.11.4 Extraction/Injection Scheme Design
The design of pump-and-treat systems involves optimizing well locations, depths, and
injection/extraction rates to maintain an effective hydraulic sweep through the contamination
zone, minimize stagnation zones, flush pore volumes through the system, and contain contam-
inated groundwater. Figure 8.11.7 illustrates a conceptual modeling analysis of an idealized site
using three pumping strategies for plume management. The medium is uniform and has a sin-
gle nondegrading contaminant with continuing release and linear equilibrium sorption.
The three strategies are down-gradient pumping, source control with down-gradient
pumping, and source control with mid-plume and down-gradient pumping. The down-gradient
.. pumping results in expansion of the plume, making it more difficult to achieve cleanup. The
other two strategies are more effective, with the source control clearly being important in pre-
venting continued off-site migration. The third strategy with source control, mid-plume, and
down-gradient pumping reduces the flow path and travel time of the contaminants to extrac-
tion wells and diminishes the impact of processes that cause tailing (Faust et al. 29). Harvey et
al. 44 have modeled the effects of pulsed pumping, namely alternating pumping and resting
periods, as shown in Figure 8.11.8. Pulsed pumping has been studied as a means to address
tailing and flush stagnation zones. Various pumping schemes have been reported in the litera-
ture (Gailey and Gorelick,37 Haggerty and Gorelick,42 Satkin and Bedient84). Marquis and
Dineen62 studied the relative merits of various methods, including pump-and-treat, in-situ
bioremediation, and others. Bumb et al. 12 discuss the design of groundwater extraction/rein-
jection systems using the MODFLOW model (see Chapter 9).
PROBLEMS
8.1.1 Go to the U.S. EPA Web site for the Toxic Release Inven- 8.1.4 Go to the U.S. EPA Web site for the Integrated Risk Infor-
tory (TRI) program and find out about toxic chemical releases in mation System (IRIS) and develop a report on what can be found
your zip code. The Web site is http://www.epagov/tri. at the Web site. What is news on the site? The Web site is
8.1.2 Go to the U.S. EPA Web site for the Drinking Water Cont- http://www.epa.gov/iris.
aminant Candidate List (CCL) and print out the list. Write a short 8.1.5 Go to the U.S. EPA Web site for the National Drinking
discussion on one of the microbiological contaminants and one of Water Contaminant Occurrence Database (NCOD) and perform a
the chemical contaminants. The Web site is http://www.epa.gov/ search to see if there have been any violations by your local water
OGWDW/ccl!cclfs.html. source and what contaminants were found in the water source.
ina 8.1.3 Go to the U.S. EPA Web site for Superfund and find a fact The Web site is http://www.epa.gov/safewater/data!ncod.html.
rest of sheet on a Superfund site in your state. Write a summary of the 8.8.1 Resolve Example 8.8.1 with a hydraulic gradient of 0.003
tping site, identifying the location, contaminants of concern, source of
rnlm. All other data remain the same.
ten contamination, techniques being used for remediation, and the
status of the remediation process. The Web site is http://www.epa. 8.8.2 Resolve Example 8.8.1 with a hydraulic conductivity of 5.0
gov/superfund. x 10-5 mls. All other data remain the same.
~
oc
Di$$0ived
Contamfnant
Concentrahon
Continuing Conl3minont Pre-Pumping Dissolved Conta mlnanl

Source Area Coneenltation Distribulion
Simulated Potentiometric Surface [after 5 years or plume growl/>)
tor each PumpinQ Alternative
Downgradlent Pumping
Source and Downgradient Pumping
Source, Mid-Plume, and

9 Years
Downgradient Pumping 5Years 19Years
Duration of Groundwater Extraction

29 Years 39 Years
......
Figure 8.11.7. Results of FrwORK (Faust et al. 29) simulation analysis of three pump-and-treat alternatives for an idealized site (with uniform media, linear equilibrium
sorption, and a single nondegrading contaminant) showing dissolved contaminant concentrations with time of pumping (as presented in Cohen et al.i 7).
r ··~
"<:l
'"0
c: N
"' •
.. --
~..,.
::8 ~ S ?O ~ ?0 m ::r ,. . .
~ ~ o ?O >o o ~. ?O
;.,. ;· ;J>
I
::s .....
0.. .- trl g ~ l!->= ~o 8::roo~?<> !l
g goo c............. . Coo IOelllbaliOit Mass fraction remo ved
0 (lq
-..JV>
:::r
~poe- l'%j ~ N c;; - ~ ~ ~~ .::1 B· ~ -=· ~- .,.... 1...1 F? p p p 0 ...,
References 409
Sensitivity to pumping rate

1.0
"0
Q)
~o.a
E
Q)
~ 0.6
0
t~ 0.4
~ 02 .
:::; (a) (d)
0
11111111
0.6
High pumping rat& Short rest period
IM
02
0
0.6
Figure 8.11.8. Effects of varying pulsed
pumping parameters (after Harvey et a1. 44).
1. 02
The fraction of total mass removed with time
is shown in (a) and (d); pumping well con-
centrations are shown in (b), (c), (e), and (f).
Dashed lines represent equivalent constant
pumping rates. Black bars atop of figures
0
represent pumping periods and white bars
0 ro 100 1!m :ro 1000 1!m
represent rest periods (as presented in Cohen
Thle(ci¥} et al. 17).
8.8.3 Develop the relationship (table and graph) of the dimen- 8.9.3 What are the selection, design, and documentation criteria
sionless concentration (C/C0) versus distance (0 m to 600 rn) and for weJls used in the NAWQA Program?
consider times ranging from I to 60 years. v = 0.04 mid and dis- 8.10.1 Write a description of the RCRA and CERCLA programs.
persion coefficient D =0.4 m2/d.
8.10.2 Write a description of the cleanup goals of the RCRA and
8.8.4 An effi uent with a concentration of a solute of I,250 mg/L is CERCLA programs.
leaking from a landfill into an aquifer with a hydraulic gradient of
8.11.1 Determine the maximum width of the capture zone for a
9.8 m/day, a gradient of 0.004, and an effective porosity of 0.15.
well that is pumping 2,000 m3/day from a confined aquifer with
What are the solute concentrations in monitoring wells 25 m and
hydraulic conductivity of 1,500 m/day, an initial hydraulic gradi-
100m from the Jandfill300 days after the leak began? Assume that
ent of 0.0009, and a saturated thickness of 35 m. What is the dis-
the effective diffusion coefficient D* is 0.0 and use the Xu and
tance from the well to the stagnation point?
Eckstein 109 method to compute the longitudinal dispersivity.
8.11.2 Determine the maximum width of the capture zone for a
8.9.1 Find the ASTM 5 standards for groundwater monitoring
well that is pumping 30,000 ft3/day from a confined aquifer with
wells and write a brief description of the designs.
a hydraulic conductivity of 95 ft/day, an initial hydraulic gradient
8.9.2 Write a description of the National Water-Quality Assess- of 0.004, and a saturated thickness of 50 ft. What is the distance
ment (NAWQA) Program. from the well to the stagnation point?
REFERENCES
1. Adriano, D. C., et al., Nitrate and salt in soils and ground waters from 3. Alley, W. M., ed., Regional Ground-Water Quality, Van Nostrand
land disposal of dairy manure, Soil Sci. Soc. Amer. Proc., v. 35, pp. Reinhold, New York, 1993.
759-762, 1971. 4. Alley, W. M., T. E. Reilly, and 0. L. Franke, Sustainabi/ity of Ground-
2. Ahlfeld. D.P.. and C. S Sawyer. Well location in capture 7.one design Water Resources. U.S. Geological Survey Circular 1186, U.S. Geological
using simulation and optimization techniques, Ground Water, v. 28, no. 4, Survey, Denver, CO, 1999.
pp. 507-512, 1990.
5. American Society for Testing and Materials (ASTM), Standard prac- 26. Emrich, G. H., and G. L. Merritt, Effects of mine drainage on ground
46. H
tice for design and installation of ground water monitoring wells in water, Ground Water, v. 7, no. 3, pp. 27-32, 1969.
Water·
aquifers, Standard D5092-90, Philadelphia, 1990. 27. Engineering and Technical Research Committee, The migration of
47. H
6. American Society for Testing and Materials (ASTM), Standard guide petroleum products in soil and ground water, Pub!. 4149, Amer. Petroleum
Hand/
for sampling groundwater monitoring wells, Standard K4448-45a; ASTM Inst., 35 pp., 1972.
Boca I
Standards on Environmental Sampling, Philadelphia, 1995. 28. Everett, L. G., et al., Monitoring Groundwater Quality: Methods and
48. H·
7. Barnes, I., and F. E. Clarke, Geochemistry of Ground Water in Mine Costs, Rept. EPA-600/4-76-023, U.S. Environmental Protection Agency,
Handt
Drainage Problems, U.S. Geological Survey Professional Paper 473-A, 6 Las Vegas, 140 pp. 1976.
Boca I
pp., 1964. 29. Faust, C. R., P. N. Sims, C. P. Spalding, P. F. Anderson, B. H. Lester, M.
49. Ja
8. Bear, J., Dynamics of Fluids in Porous Media, American Elsevier, New G. Shupe, and A. Harrover, FTWORK: Groundwater flow and solute trans-
Handh
York, 1972. pan in three dimensions, Version 2.8, Geo Trans, Inc., Sterling, VA, 1993.
Ameri'
9. Bear, J., and A. Verruijt, Modeling Groundwater Flow and Pollution, 30. Fetter, C. W., Contaminant Hydrogeology, 2nd ed., Prentice Hall,
50. Ja•
Reidel, Dordrecht, The Netherlands, 1987. Upper Saddle River, NJ, 1999.
aquife!
10. Bedient, P. B., H. S., Rifai, and C. J. Newell, Groundwater Contami- 31. Field, R., et al., Water Pollution and Associated Effects from Street
51. Jo!
nalion: Transport and Remediation, Prentice Hall, Upper Saddle River, Salting, Repl EPA-R2-73-257, U.S. Environmental Protection Agency,
Hydra~
NJ, 1994. Cincinnati, 48 pp., 1973.
52. K1
II. Blandford, T. N., P. S. Huyakom, and Y. Wu, WHPA-A Modular 32. Franke, 0. L., T. E. Reilly, D. W. Pollock, and J. W. LaBaugh, Esti-
Man's,
Semi-Analytical Mode/ for the Delineation of Wellhead Protection Areas, mating Areas Contributing Recharge to Wells: Lessons from Previous
Agene)
U.S. EPA Office of Drinking Water and Ground Water. Washington, DC, Studies, U.S. Geological Survey Circular 1174, 1998.
1993. 53. Ka
33. Freeze, R. A., and J. A. Cherry, Groundwater, Prentice Hall, Engle-
Drain.
12. Bumb, A. C., J. T. Mitchell, and S. K. Gifford, Design of a ground- wood Cliffs, NJ, 1979.
water extraction/reinjection system at a Superfund site using MOD FLOW, 54. Keo
34. Freshley, M.D., and M. J. Graham, Estimation of ground-water travel
Ground Water. v. 35, no. 3, pp. 400-408, 1997. tions, I
time at the Hanford site: Description, past work, andjurure need, PNL-
Researc
13. Canter, L. W., and R. C. Knox, Septic Tank System Effects on Ground 6328, Pacific Northwest Laboratory, Richland, WA, 1988.
Water Quality, Lewis Publishers, Chelsea, MI, 1986. 55. Ko
35. Freshley, M.D., and P. D. Thome, Ground-water contribution to DOE
and sar
14. Charbeneau, R. J., Groundwater Hydraulics and Pollutant Transport, from past Hanford operations, PNWD-1974 HEDR, Battelle Pacific
v. 25, ill
Prentice Hall, Upper Saddle River, NJ, 2000. Northwest Laboratories, Richland, WA, 1992.
56. Ke~
15. Cherry, J. A., S. Feenstra, and D. M. Mackay, Developing rational 36. Fryberger, J. S.. Investigation and rehabilitation of a brine-contami-
sarnplin
goals for in situ remedial technologies, Subsurface Restoration, Confer- nated aquifer, Ground Water, v. 13, no. 2, pp. 155-160, 1975.
pp. 300
ence Proceedings, Third International Conference on Ground Water Qual- 37. Gailey, R. M., and S.M. Gorelick, Design of optimal, reliable, plume
ity Research. National Center for Ground Water Research, Dallas, TX, 57. Kol
capture schemes: Application to the Gloucester landfill ground-water con-
June, 1992. water ol
tamination problem, Ground Water, v. 31, no. I, pp. 107-114, 1993.
858-86:
16. Cohen, R. M., A. H. Vincent, J. W. Mercer, C. R. Faust, and C. P. 38. Gelhar, L. W. , C. Welty, and K. R. Rehfeldt, A critical review of data
Spalding, Methods for Monitoring Pump-and-Treat Performance, 58. Kot·
on field-scale dispersion in aquifers, Water Resources Research, v. 28,
EPA/600/R-94/123, U.S. EPA, ORD, R. S. Kerr Environmental Research Collecti
no. 7, pp. 1955-1974, 1992.
Laboratory, Ada, OK, 1994. Assessm
39. Gilham, R. W., and L. R. Webber, Nitrogen contamination of ground- Samples
17. Cohen, R. M., J. W. Mercer, R. M. Greenwald, and M.S. Beljin, water by barnyard leachates, Jour. Water PolL Control Fed., v. 41. no. 10, 95-399,.
Design Guidelines for Conventional Pump-and-Treatment Systems, pp. 1752-762, 1969.
EPA/540/S-97/504, U.S. Environmental Protection Agency, 1997. 59. Lapl
40. Goode, D. J., Simulating Contaminant Alternatives, Double-Porosity Collecti.
18. Cole, J. A., ed., Groundwater Pollution in Europe. Water Information Exchange, and Water Age in Aquifers Using MOC3D, U.S. Geological Assessm,
Center, Port Washington, NY, 347 pp., 1975. Survey Fact Sheet 086-99, 1999. and Coli
19. Delleur, J. W. , ed., The Handbook ofGroundwater Engineering, CRC 41. Graham, M. J., J. A. Thomas, and F. B. Metting, Groundwater, in 95-398, J
Press, Boca Raton, FL, 1999. Water Resources Handbook, L. W. Mays, ed., McGraw-Hill, New York, 60. LeG
20. Delleur, J. W., Elementary groundwater flow and transport process, in 1996. ination, c
The Handbook of Groundwater Engineering, J. W. Delleur, ed., CRC 42. Haggerty, R., and S. M. Gorelick, Design of multiple contaminant 61. LeG
Press, Boca Raton, FL, 1999. remediation: Sensitivity to rate-limited mass transport, Water Resources ground, 1
21. de Marsily, G., Quantitative Hydrogeology-Groundwater Hydrology Research, v. 30, pp. 435-446, 1994.
for Engineers, Academic Press, Orlando, FL, 1986. 62. Marc
43. Haitjema, H. M., J. Wittman, V. Kelson, and N. Bauch, WhAEM: Pro- treat, biOI
22. Deutsch, M ., Groundwater Contamination and Legal Controls in gram Documentation for the Wellhead Analytical Element Model, from con
Michigan, U.S. Geological Survey Water-Supply Paper 1691, 79 pp., 1963. EPA/600/R-941210, U.S. EPA, ORD, R. S. Kerr Environmental Research v. 14, no.
23. Dirkes, R. L., R. W. Hanf, and R. K. Woodruff, Hanford site environ- Laboratory, Ada, OK, 1994.
63. Mati:
metal report for calendar year I 993, PNL-9823, Batelle Pacific Northwest 44. Harvey, C. F., R. Haggerty, and S.M. Gorelick, Aquifer remediation: GroundJ.\
Laboratory, Richland, WA, 1994. A method for estimating mass transfer rate coefficients and an evalu-
64. Me~
24. Domenico, P. A., and F. W. Schwartz, Physical and Chemical Hydro- ation of pulsed pumping, Water Resources Research, v. 30, no. 7,
Fed., v. 4<
geology, John Wtley & Sons, NY, 1990. pp. 1979-1991, 1994.
65. Mero
25. Drewry, W. A., and R. Eliassen, Virus movement in ground water, 45. Hassan, A. A., Water quality cycle--Reflection of activities of nature
Groundwt
Jour. Water Poll. ContrOl Fed., v. 40, pp. 257-272, 1968. and man, Ground Water, v. 12, no. I, pp. 16-21, 1974.
tionAgem
References 411
uund 46. Heath, R. C., Basic Ground-Water Hydrology, U.S. Geological Survey 66. Meyer, C. F , ed., Polluted Groundwater: Some Causes, Effects. Con-
Wa1er-Supply Paper 2220, 1998. trols, and Monitoring, Rept. EPA-600/4--73-0016, U.S. Environmental
on of 47. Houlihan, M. F . and P. C. Lucia, Remediation of groundwater, in The Protection Agency, Washington, DC. 282 pp., 1973.
•leum Handbook of Groundwater Engineering, J. W. Delleur, ed., CRC Press, 67. Miller. D. W., ed., Waste Disposal Effects on Ground Water, Premier
Boca Raton, FL. 1999. Press, Berkeley, CA, 512 pp., 1980.
'sand 48. Houlihan, M. F., and P. C. Lucia, Groundwater monitoring, in The 68. Mink. L. L. et al., Effect of early day mining operations on present
:ency, Haruibook of Groundwater Engineering, l W. Dclleur, ed .. CRC Press, day water quality, Ground Water, v. 10, no. 1. pp. 17-26, 1972.
Boca Raton, FL, 1999. 69. Morrison, R. D., Ground Water Monitoring Technology: Procedures,
er, M. 49. Javandel, I., C. Doughty, and C. F. Tsang, Groundwater Transport: Equipment, and Applications, Timco Manufacturing, Praire du Sac, WI,
trans- Handbook of Mathematical Models, Water Resources Monograph 10, 1983.
993. American Geophysical Union, Washington, DC, 1984. 70. Mullaney, J. R., and S. J. Grady, Hydrogeology and Water Quality in a
Hall, 50. Javandel, I.. and C. F. Tsang, Capture-zone type curves: A tool for Surficial Aquifer Underlying an Urban Area, Manchester, Connecticut, US.
aquifer cleanup, Growui Water. v. 24, pp. 616-625, 1986. Geological Sr1rvey Water Resources Investigation Report 97-4195, 1997.
Street 51. Johnson, P. C., Hydraulic design for groundwater contamination, in 71. National Research Council (NRC), Water Science and Technology
~ency,
Hydraulic Design Handbook, L W. Mays, ed .. McGraw-Hill, New York.1999. Board, Alternatives for Ground Water Cleanup, Washington, DC, 1994.
52. Karubian, J. F., Polluted Groundwater: Estimating the Effects of 72. Neuman, S. P., Universal scaling of hydraulic conductivities and dis-
'Esti- Man 's Activities, Rept. EPA 600/4-74-002, U.S. Environmental Protection persivities in geologic media, Water Resources Research, v. 26, no. 8,
~ious
Agency, Washington, DC, 1974. pp. 1749-1758, 1990.
53. Kaufman, M. 1., Subsurface wastewater injection, Florida, Jour. lrrig. 73. Nielson, D. M., ed., Practical Handbook ofGrowui-Water Monitor-
Engle- Drain. Div., Amer. Soc. Civil Engrs., v. 99, no. IRI, pp. 53-70, 1973. ing, Lewis Publishers, Chelsea, MI. 1991.
54. Keely, J. F., Performnnce Evaluation of Pump-and-Treat Remedia- 74. Ogata, A., Theory of Dispersion in a Granular Medium, U.S. Geolog-
·travel tions, EPA/540/4-89/005, U.S. EPA, ORD, R. S. Kerr Environmental ical Survey Professional Paper 411-1, U.S. Government Printing Office,
PNL- Research Laboratory. Ada, OK, 1989. Washington, DC, 1970.
55. Keely, J. F., and K. Boateng, Monitoring well installation, purging, 75. Ogata, A., and R. B. Banks, A Solution ofthe Differential Equation of
?DOE and sampling techniques-Part I: Conceptualizations, Ground Water, Longitudinal Dispersion in Porous Media, U.S. Geological Survey Pro-
>acific v. 25,no.3,pp.300-313, 1987. fessional Paper 411-A, U.S. Government Printing Office, Washington,
56. Keely, J. F. and K. Boateng, Monitoring well installation, purging, and DC, 1961.
ntami- sampling techniques-Part 2: Case Histories, Ground Water, v 25, no. 4, 76. Osgood, J. L., Hydrocarbon dispersion in ground water: Significance
pp.300-313, 1987. and characteristics, Ground Water. v. 12, no. 6, pp. 427-438, 1974.
plume 57. Kolpin, D. W., I. E. Barbash, and R. J. Gilliom, Pesticides in ground- 77. Palmer, C. M. , Contaminant Hydrogeology, Lewis Publishers, Boca
ercon- water of the United States, 1992-1996, Ground Water, v. 38, no. 6, pp. Raton, FL. 1996.
l. 858-863, 2000. 78. Quan, E. L., el a!., Subsurface sewage disposal and contamination of
of data 58. Koterba, M. T., F. D. Wilde, and W. W. Lapham, Ground-Water Data- ground water in East Portland, Oregon, Ground Water, v. 12, no. 6, pp.
v. 28, Collection Protocols and Procedures for the National Water-Qualicy 356-368, 1974.
Assessment Program: Collection and Documentation of Water-Quality 79. Robbins, J. W. D., and G. 1. Kriz., Relation of agriculture to ground-
:round- Samples and Related Data, U.S. Geological Survey Open-File Report water pollution: a review, Trans. Amer. Soc. Agric. Engrs., v. 12. pp.
no. 10, 95-399, Reston, VA, 1995. 397-403, 1969.
59. Lapham, W. W., F. D. Wilde, and M. T. Koterba, Ground-Water Data- 80. Robinson, R. A., and R. H. Stokes, Electrolyte Solutions, 2nd ed.,
'orosicy Collection Protocols and Procedures for the National Water-Qua/icy Butterworth. London, 1965.
•logical Assessment Program: Selection, Installation, and Documentation ofWells, 81. Romero, J. C., The movement of bacteria and virus through porous
arui Collection ofRelated Data, U.S. Geological Survey Open-File Report media, Ground Water, v. 8, no. 2, pp. 37-48, 1970.
·ater, in 95-398, Reston, VA, 1995.
82. Rume4 R. R., and J. K.. Mitchell, Assessment of barrier containmettt
N York, 60. LeGrand, H. E., Environmental framework of ground-water contam- technologies -A comprehensive treatment for environmental remediation
ination, Ground Water, v. 3, no. 2, pp. 11-15, 1965. applications, National Technical Information Services, Springfield, VA,
uninanl 61. LeGrand, H. E., Patterns of contaminated zones of water in the 1995.
sources ground, Water Resources Research, v. I , pp. 83-95, 1965. 83. Sasman, R. T., Thermal pollution of ground water by artificial
62. Marquis, Jr., S. A., and D. Dineen, Comparison between pump and recharge, Water and Sewage Works, v. 119, no. 12, pp. 52-55, 1972.
M:Pro- treat, biorestoration, and biorestoration/pump and treat combined: Lessons 84. Salkin, R. L., and P. B. Bedient, Effectiveness of various aquifer
Model, from computer modeling, Ground Water Monitoring and Remediation, restoration schemes under variable hydrogeologic conditions, Ground
.esearch v. 14, no. 2, pp. 105-119, 1994. M/ater.~26.no.4,pp. 488-498, 1988.
63. Matis, J. R., Petroleum contamination of ground water in Maryland, 85. Squillace, P. J., and C. V. Price, Urban lAnd-Use Study Plan for the
:diation: Ground Water, v. 9, no. 6, pp. 57-61, 1971. National Water-Qua/icy Assessment Program, U.S. Geological Survey
1 evalu-
64. McKee, J. E. eta!., Gasoline in groundwater, Jour. Water Poll. Control Open-File Report%-211, 1996.
• no. 7, Fed., v. 44, no. 2, pp. 293-302, I 972. 86. Stackelberg, P. E., Relation Between lAnd Use and Qualicy of Shal-
65. Mercer, J. W., D. C. Skipp, and Griffin, Basics of Pump-and-TrearrMnt low, lntermediaie, arui Deep Ground ¥Vater in Nassau and Suffolk Coun-
Jf nature Groundwater Remediation, EPA/60018-90/003, U.S. Environmental Protec- ties, Long Island, New York, U.S. Geological Survey Water Resources
tion Agency, R. S. Kerr Environmental Research Laboratory, Ada, OK, 1990. lnvestigatil!ns Report94-4080, 1995.
87. Strack, 0. D. L., E. I. Anderson, M. Bakker, W. C. Olsen, J. C. Panda,

R. W. Pennings, and D. R. Steward, CZAEM User's Guide: Modeling
99. U.S. Environmental Protection Agency (EPA). Guidance for Evaluat-
ing the Technicallmpracticabiliry of Ground- Water Restoration, Directive
c
Capture Zones of Ground- Water Using Analytic Elements, EPN600/R- 9234.2-25, Interim Final, U.S. EPA, Office of Solid Waste and Emergency
94/174, Office of Research and Development, Cincinnati, OH, 1994. Response, Washington, DC, 1993.
88. Terry, R. C .. Jr., Road Salt, Drinking Water. and Safety. Ballinger, 100. U.S. Environmental Protection Agency (EPA), Subswface Charac-
Cambridge, MA, 161 pp., 1974. terization and Monitoring Techniques: A Desk Reference Guide, Volume 1:
89. Thomas, M. A., The effect of residential development on ground- Solids and Ground Water, Appendices A and 8, EPA/625/R-93-003a,
water quality near Detroit, Michigan, Jour. American Water Resources Washington, DC. 1993.
Assn., v. 36, no. 5, pp. 1023-1038, October 2000. IOL U.S. Environmental Protection Agency (EPA), S(tbsuiface Charac-
90. Tinlin, R. M .. ed., Monitoring Groundwater Quality: ll/ustrative terization and Monitoring Techniques: A Desk Reference Guide, Volume
Examples, Rept. EPA-600/4-76-036, U.S. Environmental Protection II: The Vadose Zone, Field Screening and Analytical Methods, Appen-
Agency, Las Vegas, 81 pp., 1976. dices C and D, EPA/625/R-93-003b, Washington, DC. 1993.
91. Todd, D. K.. and D. E. 0. McNulty, Polluted Groundwater, Water 102. U.S. Environmental Protection Agency (EPA), Supeifund Innovative
Information Center, Por1 Washington, NY, 179 pp .. 1976. Technology Evaluation Program, Technology Profiles. 7th ed.,
EPN540/R-94/526, Washington, DC, 1994.
92. Todd, D. K et al., A groundwater quality monitoring methodology,
Jour. Amer. Water Works Assoc., v. 68, pp. 586-593, 1976. 103. U.S. Environmental Protection Agency (EPA), Ground- Water and
93. Todd, D. K. et al., Monitoring Groundwater Quality: Monitoring Leachate Treatmenr Syste1ns, EPN625/R-941005, Washington. DC, 1995.
Merhodology, Rept. EPA-600/4-76-026, U.S. Environmental Protection 104. U.S. Geological Survey. The Quality of Our Nation's Warers-Nlltri-
Agency, Las Vegas, 154 pp., 1976. ents and Pesticides, U.S. Geological Survey Circular 1225, 1999.
94. U.S. Department of Agriculture Agricultural Research Service, Pesti- 105. Van der Heijde, P. K. M., and 0. A. Elnawawy, Compilation of
cide properties database (httpJ/www.arsusda.gov/acsl/ppbd.html), 1995. ground-water models, EPA/600/R.-93/118, U.S. EPA, ORD, R. S. Kerr
Environmental Research Laboratory, Ada, OK, 1993.
95. U.S. Department of Energy (DOE). Closing the Circle on the Splitting
of the Atom, The Environmental Legacy of Nuclear Weapons Productions 106. van Genuchten, M. T., Analytical solutions for chemical transpor1
in the United States and What the Department of Energy is Doing About with simultaneous adsorption, zero-order production and first order decay,
It, U.S. Department of Energy, Office of Environmental Management, Journal of Hydrology, v. 49, pp. 213-233, 1981.
Washington, DC, 1995. 107. Warner, D. L., and J. H. Lehr, An Introduction to the Technology of
96. U.S. Environmental Protection Agency, Land Application of Waste- Subsuiface Wastewater lnjec/ion, Rept. EPA-60012-77-240, U.S. Environ-
9.1 ,
water, Rept. EPA 903/9-75-017, Philadelphia, 94 pp., 1975. mental Protection Agency, Ada, OK, 344 pp., 1977.
97. U.S. Environmental Protection Agency (EPA), RCRA Ground-Water 108. Williams, D. E., and D. G. Wilder, Gasoline pollution of a ground-
Monitoring, technical enforcement guidance document, Office of Waste water reservoir-A case history, Ground Water, v. 9, no. 6, pp. 50-56, 1971.
Programs Enforcement and Office of Solid Waste and Emergency 109. Xu, M., andY Eckstein, Use of weighted least squares method in
Response, OSWER-9950.1, Washington, DC, 1986. evaluation of the relationship between dispersion and field scale, Ground
98. U.S. Environmental Protection Agency (EPA), Handbook of Water, v. 33, no. 6, pp. 905~908, 1995.
Suggested Practices for the Design and Installation of Ground-Water 110. Zanoni, A. E., Ground water pollution and sanitary landfills-A crit-
Monitoring Wells, PB90-159-807, Washington, DC, 1989. ical review, Ground Water, v. 10, no. I, pp. 3-16, 1972.
'uat-
:tive
Chapter9 ·,...
ency
lmc-
ll!e 1:
l03a,
arac-
llwne
ppen-
Grou11dwater Flow Modeling
votive
ed.,
Techniques
?1'and
l99S.
-Nutri-
Groundwater is essentially a hidden resource; therefore, studies of groundwater under both
tiOIIof natural and artificial boundary conditions have employed modeling techniques. Fortunately,
). Kerr during the past several decades, computer simulation models for analyzing flow and solute
transport in groundwater systems have played an increasing role in the evaluation approaches
an sport
rdecay,
to groundwater development and management.
Jlogyof
:nviron- 9.1 WHY DEVELOP GROUNDWATER MODELS?
ground- Alley et al. 1 define groundwater models:
;6, l97l.
Groundwater models are an attempt to represent the essential features of the actual ground-
ethod in
Ground water system by means of a mathematical counterpart. The underlying philosophy is that an
understanding of the basic laws of physics, chemistry, and biology that describe groundwater
-Acrit- flow and transport and an accurate description of the specific system under study will enable
a quantitative representation of the cause and effect relationships for that system. Quantita-
tive understanding of cause and effect relationships enables forecasts to be made for any
defined set of conditions.
Konikow and Bredehoeft31 point out that such forecasts, which are usually outside the
range of observed conditions, typically are limited by uncertainties due to sparse and inaccu-
rate data, poor definition of stresses acting on the system, and errors in system conceptualiza-
tion. Even though forecasts of future events using models may be imprecise, they do represent
the best available information at a given time that can be used in decision making. Simulation
models can be used as learning tools to identify additional data needed in order to better define
and understand groundwater systems. These models have the capability to test and quantify the
consequences of various errors and uncertainties in the information needed to determine the
cause and effect relationships and related model-based forecasts. In the process of forecasting,
this model capability may be the most important aspect to define the uncertainties of the fore-
casts, allowing water managers to evaluate the significance of their decisions. Also this process
may allow the evaluation of unexpected consequences of their decisions.
Groundwater modeling has become a very important process in managing groundwater
resources. It is interesting to note that each of the classic texts by Walton,59 Freeze and
Cherry, 11 and Todd,57 dated respectively 1970, 1979, and 1980, had very little material dealing
with groundwater modeling using the computer. The past two decades have resulted in tremen-
dous changes in our use of the computer for groundwater management.
413
414 Chapter 9 Groundwater Flow Modeling Techniques
9.2 TYPES OF GROUNDWATER MODELS

Todd57 classified groundwater models as porous media models (sand tank models and trans-
parent models), analog models (viscous fluid models, membrane models, Moire pattern mod-
els, thermal models, and blotting paper models), electric analog models (conductive solid
models, resistance-capacitance networks, and resistance networks), and digital computer mod-
els (finite difference models, finite element models, and hybrid computer models). Prickett49· 50
provided a detailed list of references on the various models that ellisted during that time, listing
applications of various types of groundwater models, including aquifer features, purpose of the 9.3
models, and schematic sketches of the models.
A great deal has changed in groundwater modeling since the 1970s, when the numerical
(or mathematical) modeling approach was a relatively new field. It was not extensively pur-
sued until the mid-l960s, when digital computers with adequate capacity became generally
available (Appel and Bredehoeft).5 During the 1970s, many articles in the literature, govern-
ment publications, and books, such as Peaceman,43 Pinder and Gray, 45 Prickett and
Lonnquist, 51 Remson et al.,54 Rushton and Redshaw, 55 Texas Water Development Board,56 and
Trescott et al.,58 appeared. Presently, the use of mathematical models is the state-of-the-art in
practice for groundwater modeling.
Groundwater models (for the purposes of this book) are physically based mathematical
models derived from Darcy's law and the law of conservation of mass.64 Various established
solution techniques based upon either finite difference or finite element approximations, or a
combination of both, are available for solving the governing equations of the model. The accu-
racy of the solutions (model predictions) is dependent upon the reliability of the estimated
model parameters and the accuracy of the prescribed boundary conditions.
Groundwater models can be classified in several ways: steady state or transient; confined,
unconfined, or a combination of confined and unconfined; one-dimensional, two-dimensional,
quasi-three-dimensional, or three-dimensional. They can be solved using finite difference
methods or finite element methods or a combination of these. The major emphasis in this chap-
ter is on the finite difference method.
The finite difference method requires a rectangular element shaped discretization of the
aquifer and the finite element method consists of a triangular discretization as illustrated in Fig-
ure 9.2.1. The size and shape of the finite elements are arbitrary, typically being triangular or 9.4 s
quadrilateral. In fact, the elements can be disordered and nonuniform and should be the small- GRO
est where flow is concentrated, such as near a well. Also, it is easy to define the boundaries of
9.4.1 I
Figure 9.2.1. Example

of a triangular finite ele-
Aquifer ment within an aquifer
region (after Prickett49).
9.4 Simulation of Two-Dimensional Groundwater Systems 415
irregularly shaped aquifers and to ensure that node points coincide with monitoring wells or var-
ious types of geographic features. The mathematical basis for finite element methods is more
ms- complex than for the finite difference method. Galerkin's method is the most commonly used
lod- solution method for the finite element method. Detailed descriptions, while not discussed in this
olid book, of the finite element method can be found elsewhere in the literature. 23 · 54· 62
:IOd-
49, 50
;ting
9.3 STEPS IN THE DEVELOPMENT OF A GROUNDWATER MODEL
fthe
Anderson and Woessner4 presented the following modeling protocol for groundwater modeling:
:rical
l. Determine the purpose of the model.
pur-
:rally 2. Develop a conceptual model.
vern- 3. Select the governing equations and computer code.
• and 4. Model design-put the conceptual model into a form suitable for modeling.
;6 and
5. Calibration-establish that the model can reproduce field-measured heads and flows.
art in
6. Calibration sensitivity analysis-determine the effect of uncertainty on the calibrated
tatical model.
lished 7. Model verification-use the calibrated parameter values and stresses to reproduce a
s, or a second set of field data in order to establish greater confidence in the model.
accu- 8. Prediction to quantify the response of the system to future events.
mated 9. Predictive sensitivity analysis to quantify the effect of parameter uncertainty.
1fined, 10. Presentation of modeling design and results.

sional, 11. Post-audit analysis in the future with new field data.
erence 12. Model redesign using insight from the post-audit analysis.
; chap-
See Hil125 for more insight on effective model calibration.
. of the
in Fig-
~ular or
9.4 SIMULATION OF TWO-DIMENSIONAL
~small GROUNDWATER SYSTEMS
aries of The first finite difference groundwater models that found fairly widespread usage were devel-
oped for two-dimensional (horizontal) flow. These included models such the ones developed
by Prickett and Lonnquist,51 Trescott et al.,58 and the Texas Water Development Board.56 Even
though the use of these models has been superceded by the U.S. Geological Survey
MODFLOW model (which can be used for either two- or three-dimensional modeling), it is
worthwhile to describe the two-dimensional framework as a precursor to a description of the
more complicated three-dimensional models.
9.4.1 Governing Equations

Darcy's law relates the Darcy flux v with dimension (LfT) to the rate of headloss per unit
length of porous medium ah I at, as discussed in Chapter 3. The negative sign indicates that the
total head is decreasing in the direction of flow because of friction. This law applies to a cross
section of porous medium, which is large compared to the cross section of individual pores and
grains of the medium. At this scale, Darcy's law describes a steady uniform flow of constant
~xample
inite ele- velocity, in which the net force on any fluid element is zero. For unconfined saturated flow, the
aquifer two forces are gravity and friction. Darcy's law (Equation 3.1 .5) can also be expressed in terms
). of the transmissivity for confined conditions as
T dh (9.4.1)
v=---
b dl
or for unconfined conditions as
v=---
r oh (9 .4.2)
h az
where b is the thickness of the confined aquifer and h is the saturated thickness of the uncon-
fined aquifer.
Considering two-dimensional (horizontal) flow, a general flow equation can be derived by
considering flow through a rectangular element (control volume), as shown in Figure 9.4.1.
The flow components (q =Av) for the four sides of the element are expressed using Darcy's
law where A =&. · h for unconfined conditions and A =&. · b for confined conditions so that
ql (dh)
=-Tx i-l .j !!.Y,. -OX I
(9.4.3a)
(9.4.3b)
(9.4.3c)
(9.4.3d)
where Tx· is the transmissivity in the x flow direction from element (i, J) to element (i + l , j).
'J
The terms (ohlox) 1, (ohlox)2 , • .• , define the hydraulic gradients at the element sides 1, 2, ....
The rate at which water is stored or released in the element over time is
oh
q5 =st,j.&_.f!.y.
I
-
I Ot
(9.4.4)
in which SiJ is the storage coefficient for element (i, J). In addition, the flow rate q6 for constant
net withdrawal or recharge from the element over time interval !lt is considered as
(9.4.5)
/-1 i+1
Node
j- 1 • • • 9.4.2
q6 (Net constant
~q4
withdrawal rate
.- q,
I q2
f+- • ---+...- •
tq3
from cell i, j)
Figure 9.4.1. Finite difference grid

I showing the control volume for an
q5 (Released or
j +1 • • • taken into storage) element of a two-dimensional (hori-
zontal) flow. The node within each
!lx1 grid (cell} is the point where the head
Cell-to-cell water transfers is calculated.
in which q1J.r has a positive value for pumping, whereas it has a negative value for recharge.
U) By continuity the flow into and out of a grid or cell is
(9.4.6)
Substituting in Equations 9.4.3 and 9.4.5 gives

4.2)
;on- -T
X;.l ,j Y; (axah) +T y) (ah)
6. · -
I ax -Txi.j
tJ. . -
2 Yi.j•l
tJ.x. -
I
[ah)
3 Y;j
[ah)
dy +T tJ.x. -ay
l 4
(9.4.7)
d by ah
= s. .tJ.x.f1y .- + q. .
.4.1. 1,) I I Ot 1,),1
rcy's
•that Dividing Equation 9.4.7 by tJ.x tJ.y and simplifying for constant traosmissivities in the x andy
directions yields
4.3a)
4.3b) (9.4.8)
.4.3c)
For Llx and tJ.y infinitesimally small, the terms in brackets [ ] become second derivatives of h.
Then Equation 9.4.8 reduces to
.4.3d)
(9.4.9)
- I ,j).
which is the general partial differential equation for unsteady flow in the horizontal direction
in which W = qij,ltJ.xi tJ.yi is a sink term with dimensions (Uf).
In the more general case for unsteady, two-dimensional heterogeneous anisotropic flow,
Equation 9.4.9 is expressed as
(9.4.4)
mstant (9.4.10a)
or more simply
(9.4.5)
.i.(r;
ax. ,).j!:_J
ox. =s at
I J
i)h + w i,j =1,2 (9.4.10b)
9.4.2 Finite Difference Equations
The partial derivative expressions for Darcy's law, (Equations 9.4.3a-d), can be expressed in
finite difference form for timet in Equation 9.4.7 using
ah) =(hi-l,j,l- hi.j,, J

:e grid
or an
(OX I fui
(9.4.1la)
I (hori-
1 each
the head (9.4.l lb)
dh ) = ( hi,j+
( 3dy
I ,I -
flyj
h;,j,l l (9.4.llc)
(dh) =(hi,j ,l - hl,j-l,tl (9.4. lld)

9.4
dy 4 fly j
and the time derivative in Equation 9.4.10 is
~~ =( hi,j.l ~~i,j,l-1 ) (9.4. 12)
Substituting Equations 9.4.11 and 9.4.12 into Equation 9.4.7 yields
-T fl . h. · -h.·
1- 1.J.I ) +T fl . (h·· -h· I + 1.j.l
· )
X;-l.j yJ ( /).):
I,J.I
X;.j yJ
1,).1
/).):
l
i ;
-T /).): . h· I,J + 1. I · ] + T /).):.

-h.1.).1 (h·· -h. ·
I.J.I 1,)- 1,I (9.4.13)
Yi,jtl I( fly j Y•.j I fly j 9.4
-s../).xI.flY,. h.·
I,J .'
-h.l,;.r
111
·- t) _
q,,,,l
. . =0
1,)
(
which can be further simplified to
(9 .4.14)
where
A. = T
fly .
- 1 +T -
fly ·
1 +T
h
_'-"_; +T
A-
_; -S. -- -
_t.l..\.
/).):.fly·]
1 1
(9.4.15a)
I, J [ ,_xi-1.) llx; x,,j !J.x; Yi,j+l f:.yj Yi .J fly j I,J flt
fly .
B-I,J. =-Txl-1 ,, -!:u.1 (9.4. 15b)
I
t::.y .
C.1,)-= -TX;.j -/).):'. (9.4.15c)
I
/:u.
D· · =-T -' (9.4.15d)
1
·1 Yi.J • I fly j
/:u. (9.4.15e)
£ 1. . =-T -~
·1 Yi.j !::.yj
z.. . ~ h.'
F;,1,r = - si.J ( ''i,J.I 1,),1-1
)- qi.J.r
(9.4.150
!::.t 0 1(
~
The coefficients A;j• B;j• CiJ, D;j• EiJ, and F;J.r are linear functions of the thickness of cell (i,
0
j) and the thickness of one of the adjacent cells. For artesian conditions, this thickness is a
~
known constant, so if cell (i, J) and its neighbors are artesian, Equation 9.4. 14 is linear for all
t. For unconfined (water table) conditions, the thickness of cell (i, j) is h;J,t- BOTij• where Figw
BOT;,j is the average elevation of the bottom of the aquifer for cell(~ j). Then for unconfined
. lc) conditions, Equation 9.4.14 involves products of heads and is nonlinear in tenns of the heads .
9.4.3 Solution
lld)
An iterative alternating direction implicit (lAD/) procedure can be used to solve the set of
equations. The IADI procedure involves reducing a large set of equations to several smaller
sets of equations. A smaller set of equations is generated by writing Equation 9.4.14 for each
cell or element in a column but assuming that the heads for the nodes on the adjacent colunms
·.12) are known. The unknowns in this set of equations are the heads for the nodes along the column.
The heads for the nodes along adjoining columns are not considered unknowns. This set of
equations is solved by Gauss elimination and the process is repeated until each column is
treated. The next step is to develop a set of equations along each row, assuming the heads for
the nodes along adjoining rows are known. The set of equations for each row is solved and the
process is repeated for each row in the finite difference grid.
U3)
9.4.4 Case Study
Once the sets of equations for the columns and the sets of equations for the row have been
solved, one "iteration" has been completed. The iteration process is repeated until the procedure
converges. Once convergence is accomplished, the tenns hiJ represent the heads at the end of the
time step. These heads are used as the beginning heads for the following time step. For a more
detailed discussion of the lADI procedure, see references 44, 51, or 62. Two-dimensional finite
difference models for groundwater flow are by Prickett and Lonnquisii 1and by Trescott et al. 58
.4.14) Two studies utilizing a two-dimensional finite difference groundwater model have been
made of the Edwards (Balcones Fault Zone) Aquifer. First, this aquifer has been modeled using
the GWSIM groundwater simulation model developed by the Texas Water Development
Board. 56 GWSIM is a finite difference simulation model that uses the IADI method similar to
USa) the model by Prickett and Lonnquist.51 The finite difference grid for the Edwards Aquifer is
shown in Figure 9.4.2, which bas 856 active cells to describe the aquifer. Second, Wanakule61
has modeled a smaller portion of the Edwards aquifer called the Barton Springs-Edwards
USb) Aquifer in Austin, Texas. This application used a finite difference grid system (Figure 9.4.3)
containing 330 cells whose dimensions varied from 0.379 x 0.283 mi2 to 0.95 x 1.51 mi2.
t 21
4.15c)
4.15d) N EXPLANATION
Study ar/:'\ ~
Spring
·,-. (· · ·. _.-- - - I
__ ...
.4.15~) Approximate upper boundary
. ---- ·. ---._--- x~ :-~ of aquifer recharge area
I ' ' I
_/ # Approximate lower boundary

. . . - - -.......... . . __ , -.... . . ~ 1'_.,' of aquifer recharge area
~~
'..._ ___ , ~ r, ,../
t.4.15f) " ~- ' , -' ', - --- - San Antonio Approximate western and
o 10 20 30 Kilometers , _ \ - ---
eastern limits of aquifer
kJ-;;1-J ---\
·cell (i,
0 10 20 30 Miles Artesian Approximate location of
ess is a
k;;~ area bad water line
r for all
, where ). Figure 9.4.2a. Edwards (Balcones Fault Zone) Aquifer, San Antonio, TX, region.56
I'
I
2
3
4
5
~
7
10
11
12
m
Outcrop cell
[!]
Af1esian cell
Figure 9.4.2b. Cell map used for the digital computer model of the Edwards (Balcones Fault Zone) Aquifer (after Klempt et al.30).
The above two case stuq_\es are for the purposes of illustrating the two-dimensional finite dif-
ference method for horizontal flows. The present day modeling of the Edwards Aquifer is being
done using the MODFLOW model that is introduced in the following sections of this chapter.
9.5 THREE-DI.MENSIONAL GROUNDWATER FLOW MODEL

9.5.1 Derivation of Finite Difference Equations
The partial differential equation for transient three-dimensional groundwater flow in a hetero-
geneous and anisotropic medium (provided the principal axes of hydraulic conductivity are
aligned with the coordinate directions), for a confined or unconfined aquifer, is expressed as
j_(K ax ay ayah)+~(K
ax ah)+j_(K
XX az zz ah)
az _w= s ahat
yy S
(9.5.1)
where Kxx, KYY' and Kzz are the hydraulic conductivities along the x, y, and z coordinate axes
parallel to the major axes of hydraulic conductivities; h is the potentiometric head; Wis a vol-
umetric flux per unit volume representing sources (W is negative) and/or sinks (W is positive)
of water; Ss is the specific storage of the porous medium; and t is time. Kxx, KYY' Kzz, and Ss
are functions of space (x, y, z) and Wis a function of space and time (x, y, z, t).
Equation 9.5.1, together with specification of flow and/or head conditions at the bound-
aries of an aquifer system and specification of initial head conditions, constitutes a mathemat-
ical representation of a groundwater flow system. The solution of Equation 9.5.1 requires the
use of a numerical method such as the finite difference method. The application of finite dif~
ferences replaces the continuous system described by Equation 9.5.1 with a finite set of dis-
crete points in space and time, such that the partial derivatives are replaced by terms calculated
from the head differences calculated at these points. A system of simultaneous linear algebraic
difference equations results from this process, which is solved for the heads at specific points
and times that constitute the approximation to the time-varying head distribution.
Figure 9.5.1 illustrates a three-dimensional hypothetical aquifer system discretized into a
mesh of blocks called cells, described by rows, columns, and layers. The hypothetical system
is discretized into five rows, nine columns, and five layers. An i, j , k indexing system is used
I 2 3 < S 6 7 ! 9 10 I I 12 13 I~ 1; 16 17 18 19 Z0 21
I
v:
2
I I
"
I !
10 I '
11 I
I V Aqvifer boundary
12 3
\ / I I I
13
r
\
l~itlal ~ater 1
table and
ex artesian zones dilider
IS
I ''
I \
16 ® \
17 ®® I
\
0 @ @
18
~\
19 I
.... .. 1·1·· ..... -~- ..... .., ... ........... 0 ....... ·-···· ...... ..®.. .... \ !.~~~!.~~~ . HoysCoun\y
I I
:edif- I \ I
being 21
r ® @ @ \
I I
)ter.
22 e 1
I
\
23 @ \
\.
~\
24
., I
~ m
1:
.etero- 25
.... !11 0
N
I
.ty are
1--
.Q
~
... 'iii"'
c: (
~ I
ed as I ~ @ \
3:"'
28
<t \
, "' - r-- -- _r--, .... _

..... \
-'
27
(9.5.1) I I
I " ...
28 I I ·'
te axes ~i
; a vol-
I
I
- ·"
>sitive)
and S5 ® Pumping node ~ Springs' Cell
bound-
hemat-
ires the
0
-----
0 1 2 3 4 5 6 7
2 3 4 5 Miles
a Kilometers
Figure 9-4.3. Pumping locations used in Barton
Springs-Edwards Aquifer model (after Wanalcule61).
lite dif-
: of dis-
lculated to reference the cells. Layers generally refer to horizontal geohydrologic units or intervals, so
lgebraic that the k index denotes changes along the vertical (z-direction in Cartesian coordinates). The
c points layers are numbered from the top down so that an increment in the k index corresponds to a
decrease in elevation. Rows are considered parallel to the x axis and columns are parallel to
:d into a the x axis.
l system A Mde is a point within each cell (grid) where the head is calculated. Figure 9.5.2 illus-
1 is used trates block-centered and point-centered grid systems. Whichever grid system is used, the
Columns {J)
2 3 4 5 s 7 8 9
- - - - Aquifer boundary
t Active cell
o Inactive cell
6rJ Dimension of cell along the row direction. Subscript (J) indicates the number of the column.
6c1 Dimension of cell along the column direction. Subscript (I) indicates the number of the row.
Figure 9.5.1. A discretized hypo-
thetical aquifer system (McDonald
AvK Dimension of cell along the vertical direction. Subscript (K) indicates the number of the layer. and Harbaugh39).
spacing of nodes should be selected so that the hydraulic properties are generally uniform over
the extent of the cell. The finite difference equations developed herein are applicable for either
grid system. Figure 9.5 .3 illustrates cell (i, j, k) and the six adjacent cells that are used to derive
the finite difference equation for the cell.
Figure 9.5.4 illustrates the flow into cell (i, j, k) from cell (i, j - 1, k). The effective
hydraulic conductivity for the entire region between the nodes is denoted as KRiJ-l/2.k' The
subscript j - 1/2 does not relate to a specific point between the nodes but is used to relate to
the region from cell (i, j- 1, k) to cell (i, j, k). The effective hydraulic conductivity is computed
as the harmonic mean, so that for flow into cell (i, j, k) from cell (i, j - l, k), the effective
hydraulic conductivity is
KY _ (I:J.Y--t
1
+!:J.y.1 Y
''2 _ KY_I . kKY.I,J..k (!:J.y._t
1 ,J, r + I:J.y1.)
i-112,j,k - ( !:J.yi_tf2 + I:J.y/2]- KYi.j.k6.Yi-! + KY;-t,j.ki:J.Y; (9.5.2)
KY1- I .}.'k KY.l,j,'k
If !:J.y is constant, then Equation 9.5.2 is expressed as

2KY 'kKY. I 1. k
KY . = '·'· ,_ ' ' (9.5.3)
r-ti2,J,k KY. . +KY. .
l,j,k l-l,j,k
Flows are considered positive if they are entering cell (i, j, k) and the negative sign
incorporated into Darcy's law is dropped from all terms. The flow into cell (i, j, k) from cell
(i, j- 1, k) is expressed by Darcy's law as (McDonald and Harbaugh39)
Block-Centered Grid System
ypo-
:mald
1 over
Point-centered grid system
either
jerive
·ective @ Nodes
- - Gridlines
k· The Figure 9.5.2. Grids showing
late to - - - Cell boundaries for point-
the difference between block-
centered formulation
tputed centered and point-centered
rective W- ~Cell associated with
~//~ selected nodes
grids (McDonald and
Harbaugh3~.
:9.5.2)
(9.5.4)
where qiJ-II2.k is the volumetric fluid discharge through the face between cell (i,j -1, k) and cell
(9.5.3) (i, j, k); h1J,k and hiJ-I,k are the heads at the respective nodes; tll./lzk is the area of the cell face
normal to the row direction; and ~Yj-l/2 is the distance between nodes (i, j, k) and (i, j- 1, k).
Equation 9.5.4 defines the flow for a one-dimensional steady-state flow through a block of
•e sign aquifer extending from node (i,j -1, k) to node (i, j, k), and having cross-sectional area &i&k.
>ID cell Darcy's law is applied in a similar manner to describe the flow through the remaining five
sides of cell (i, j, k) to obtain the following (McDonald and Harbaugh 3 ~:
Figure 9.5.3. Cell (i, j , k) and indices

for the six adjacent cells (McDonald
and Harbaugh39).
Cell i,fl,k Cell i,f,k
I:
I:
• _ _,f---•1 :.
I:
q i.j-f ,k ••• ,..
•. •. • I
.•..•·• I
""--------~.A.._
v - ___v- ~---"
tJ. 'J-1
\. Figure 9.5.4. Flow into cell (i, j, k) from cell
39
(i, j - l, k) (McDonald and Harbaugh ).
Flow cell (i, j , k) to cell (i, j + 1, k)
q i, j+ll2.k =KYi,j+ll2,ktJ.xi~k (
..
hl,J+
A
1-h. 'k)
I k l,J,
(9.5.5)
· 0 Yj+l/2
Flow cell (i, j, k) to cell (i + l,j, k)
(9.5.6)
Flow cell (i - l ,j, k) to cell (i, j, k)
(9.5.7)
Flow cell (i. j, k) to cell (i, j, k + 1)
qi,j,k+l/2 -
_
KZi,j.k+li2.6.Y/~.xi [ A~
1-h·k]
h.I.),"k + l,j,
(9.5.8)
~k+L/2
Flow cell (i, j, k- 1) to cell (i, j, k)
A A- h.kl-h'k]
1,), - 1,],
qi,j,k-L/2 = KZi,j,k-112uYj'lii [ ~.:...."---~ (9.5.9)
~k-112
Equations 9.5.4 through 9.5.9 describe the steady-state, one-dimensional flow through each
side of cell (i, j, k) in terms of the heads, grid dimensions, and hydraulic conductivity. The
notation can be simplified by defining a constant referred to as the hydraulic conductance or
the conductance by combining the grid dimensions and hydraulic conductivity. Conductance
is the product of the effective hydraulic conductivity and cross-sectional area of flow divided
by the length of the flow path (distance between nodes). The hydraulic conductance CYiJ-lf2.k
indices
in row i and layer k between nodes (i,j- 1, k) and (i,j, k) is
·onald
CYiJ-112.k = KYi.J-ll2.k 11x;fu.J!.6.yj-112 (9.5.10)
where CYiJ-LI2.k is the conductance in row i and layer k between nodes (i,j -1, k) and (i,j, k) [L2!f].
Equations 9.5.4 through 9.5.9 are expressed with the conductance as
=CYiJ-t12,k (hiJ-l.k- h;J,V
qiJ-tl2.k (9.5.11)
qiJ+112.k =CYiJ+LI2,k (hiJ+l.k- hiJ,k) (9.5.12)

qi-LI2J.k =cxi-112J,k (h,_JJ,k- hiJ.k) (9.5.13)
qi+112J,k. =cxi+112J.k (h,+lj,k- hiJJc> (9.5.14)
qiJ,k-112 = cziJ,k-l/2 ChiJ,k-1- h,J,k) (9.5.15)
qiJ.k+l/2 = cziJ.k+I/2 (hiJ,k+l- hiJ.k) (9.5.16)
The flows from features or processes external to the aquifer, such as streams, drains, areal
recharge, evapotranspiration, or wells, are described through additional terms. These flows
m cell may be dependent on the head in the receiving cell but independent of all other heads in the
I. aquifer, or they may be entirely independent of head in the receiving cell.
The rate at which water is stored or released in the cell, t::.V/.6.t over time .6.t, is
bJr. ·k
-t::.V
!::.t
=SS .. kfu./ly1.D.zk _z_.J._
l,), l !JJ (9.5.17)
(9.5.5) where SSij,k is the specific storage of cell (i, j, k); MiJ,/t::.t is the finite difference approxima-
tion for the derivative of head with respect to time; and &i t::.yj !::.zk is the volume of cell (i, j,
k). The time derivative of head is expressed in terms of specific heads and time by defining the
head difference as M =hTj.k- hTJJ and the time interval as t::.t = tm- tm-1:
z.m hm-1
(9.5.6) M ''i.j,k - i,J,k
-=.....;.;..:....___;=..__ (9.5.18)
bJ tm- tm-L
The rate at which water is stored in or released from cell (i,j, k) is expressed as
/lV s,m -hm-L
(9.5.7) .. (&z..6,Y;.fu.k ) '>j,j,k
-"t =ssz.;.k t _t
i.j,k (9.5.19)
Ll m m-1
External flows (sources and stresses) from outside the aquifer into each cell, such as
recharge, streams, and flow out of each cell-such as evapotranspiration and well pumpage for
each individual cell-must also be taken into account. The total external flow for cell (i, j, k)
is designated W ;J·,k> which is the combination of source and/or stress terms for an individual
cell expressed as
N N
"'f.j,k =I Pi,j,k,nhi,j,k +I qi,j,k.n = P;, j.k hi.j.k +Qi,j,k
n=I n=I
(9.5.20)
where N is the number of external sources or stresses affecting cell (i,j, k), and PiJ.k.n and qiJ.k.n
are constants that describe the individual external sources or stresses. These constants for the
MODFLOW model are described in more detail in subsection 9.6.3.
The continuity equation for cell (i,j, k) is expressed as
qi,j-II2,k +qi,j+II2,k + qi-I /2,j,k + qi+I12.j,k +qi,j,k-I /2 9.5.2

(9.5.21)
+q·1,),' k+I/2 + Wl , j,'it. =SS.1,),'it. llh;,j,k
/j, [ (&.b.y./:~zk)
I J
Equations 9.5.11-9.5.16, 9.5.l9, and 9.5.20 are substituted into 9.5.21, resulting in the fol-
lowing finite difference approximation for cell (i, j, k):
.. 112,~c(hlJ-
crlJ- .. 1.... - hIJjc/
... \ + cr!J+
.. 1n ....(hIJ+ .. *)+ ex.1- 1n J·.~c<h1·- 1J·.~c - hIJ..,~c)
.. I.k- hIJ,
+ cxi+l/2j,k<h;+tj,k- h;j,Jc) + czij.k-ln<hij.Jc-•- h;,j,Jc) + czij.k+LI2(hi,j.k+l- hi,j.k) (9.5.22)

+ Pi.J.khi.J.J: + Qi.J.k = SS;J,K(&;flyil:lzk)(llh;.i./tli)
The flow terms can be expressed in terms of hm at time tm as
. 1/2, k(h~.
CYI,J- I ,J- I, k- h~. L) +CY1,;+
1,).~ 1.;+ I, k- h~.
. 1/2,k(hm. 1- 1/2,),. k(h~I
1,). k) +ex. 1- ,J,. k- h~.
1,), k)
+CX.1+ l/2,J,·k (h~l'k-h~·

I+ ,j, I,J , k)+CZ.1,)·.~- I,J. - I-h~·
•· 112(h':'.k I,J, k)+CZ.1,)',k+lt2(h~.
1, ) , k +I-hm.,.)
I,), ..
(h~. k-h~.-~) 9.5.3

+ pI,J,. khm. k. + Q .. k =SS .. k (11x .fly ./j,zk) _,____I,J;_•---''':'.._
' -'-
I,J, t,j. 1,). JI I -I
m m-L (9.5.23)
This equation can be rearranged so that all terms of hmcontaining heads at the current time step
are grouped on the left-hand side of the equation and aJI remaining known terms independent
of heads at the current time step are on the right-hand side. The resulting equation is
czi.j.k-llz (hrj.k- d + cxi- l l2.j.k ( ht'~l.j.k) + C1J.j-I/2,k (ht}-J. k)
+(-czi.j,k-112 - cxi-ll2.j.k- CJi.j-l/2,k - CJi,J+112.k

(9.5.24)
-CZi +lt2.J,k -CZ;,J,k+lt2 + HCOfi.J,k ) h;,J-t,k
+C.Y;,j+ttz.k (hP,j+l.k) + cxi+II2.j.k (h;:I .j,k) + czi,J,k+l/2 (h:J.k+t) =RHsi.j.k
where
HCOF. = P . - SCI.I,J.· k [ 2 ]
L /T
t.; .k l,j,k (t - t ) m m- 1
RHSl,j,k
. =-Q.l,j.k. - SCI . · kh':'.1-ki
'·'·
(tm- tm- 1)'
1
'
[ 3 ]
L IT
9.5 Tlu:ee-Dimensional Groundwater Flow Model 427
and
1ch as
1ge for
i, j, k)
vidual Expressing Equation 9.6.24 for each variable-head cell in the aquifer results in a system
of equations of the following matrix fonn:
[A]{h}={q} (9.5 .25)
.5.20)
1
where [A] is a matrix of the coefficients of heads from the left side of Equation 9.5.24 for all
active nodes in the mesh; {h} is the vector of unknown heads at the end of time step m for all
lqiJ,k.,n
for the nodes in the mesh; and {q} is the vector of constant terms, the right side of Equation 9.5.24,
for all nodes of the mesh.
9.5.2 Simulation of Boundaries

1.5.21) In order to simulate boundary conditions, cells are grouped as constant-head cells and inactive
(or no-flow) cells. Constant-head cells have a specified value through all time steps of the sim-
ulation. Inactive cells do not allow flow into or out of the cell. The remaining cells of a finite
1e fol- difference grid system (mesh) are variable head cells. The finite difference Equation 9.5.24 is
formulated for each variable-head cell in the mesh. The system of finite difference equations
is solved simultaneously for each time step in the simulation.
The constant-head and no-flow cells are used to represent condition.s of various hydro-
).5.22) logic conditions, as illustrated in Fig. 9.5.5. Even though the aquifer is irregular in shape the
mesh defining the aquifer is always rectangular in outline. No-flow cells are used to delete the
portion of the array of cell beyond the aquifer boundary. Figure 9.5.5 shows constant-head
cells along one section of the boundary to simulate direct contact with a major surface water
body. Boundary conditions, such as areas of constant flow or areas where inflow varies with
head, can be simulated as an external source or through a combination of external source and
a no-flow cell. ·
9.5.3 Vertical Discretization

).5.23) The range of vertical discretization of an aquifer cross section (see Fig. 9.5.6) can be visualized
simply as an extension of areal discretization in an effort to represent individual aquifers or
oe step
Aquifer boundary
endent
Model impermeable boundary
~ Inactive cell
mrwt C<M~stant·head c~l

9.5.24)
D Variable-head celt
Figure 9.5.5. Discretized aquifer

showing boundaries and cell designa-
tions (McDonald and Harbaugh39).
(a) Aquifer cross section
Grid layer 1 9.5.4
Cell contains material

from three stratigraphic
units. All faces are
Grid layer 3 rectangles
(b) Aquifer cross section with

rectilinear grid superimposed
....
Ji. .
Grid layer 1 ~~~

=-=~
·---4
Grid layer2
Cell contains material

Grid layer 3 from only one stratigraphic
unit. Faces are not rectangles
Figure 9.5.6. Schemes of verti- layer

(c) Aquifer cross section with
deformed grid superimposed cal discretization?9
permeable zones by individual layers of the model. The simpler approach (Figure 9.5.6b) leads
to a rigid superposition of an orthogonal tluee-dimensional mesh on the geohydrologic system. 2
In the second approach (Figure 9.5.6c), the model layer thickness is considered variable in order 3
4
to simulate the varying thickness of the geohydrologic units leading to a deformed mesh. 5
The advantage of the second approach is that the uniform hydraulic properties of a cell are 6
7
better approximated when the model layers conform to the geohydrologic units. Also, greater
accuracy is expected if model layers correspond to intervals within which vertical head loss is
negligible, which is more likely using the approach shown in Figure 9.5.6c. However, the 8
deformed mesh of Figure 9.5.6c fails to conform to the assumptions of the model equations.
As an example, cells may no longer have rectangular sides and the major axes of hydraulic
conductivity may not be aligned with the model axis, both introducing errors. Figure~
In practice, vertical discretization schemes tum out to be a combination of the viewpoints

illustrated in Figure 9.5.6 (McDonald and Harbaugh39). Figure 9.5.7 illustrates a system con-
sisting of two sand layers separated by a clay layer. Each of the units is of equal thickness so
that each of the units could be represented by a single layer without deformation of the mesh;
however, the flow pattern is much more complicated as illustrated by the flow lines. In this
case, several layers must be used to represent each layer. Figure 9.5.8 illustrates another case
of a sand and clay system in which pumpage from the sands is sustained partially by vertical
flow of water released from storage in the clay. Several model layers would be required to
detennine the pattern of storage release in the clay as shown in the figure. In the case that the
storage is released only in the sands, the flow would be nearly horizontal with the flow in the
clay nearly vertical. A single layer could be used for each sand unit and the clay layer could
be represented by a vertical conductance between sand layers, which is sometimes referred to
as a quasi-three-dimensional approach. Each of these vertical-discretization approaches leads
to a set of model equations of the form of Equation 9.5.24.
9.5.4 Hydraulic Conductance Equations

Conductance has previously been defmed as the product of hydraulic conductivity and cross-
sectional area of flow divided by the length of the flow path. Start with Darcy's law defining
Equipotentials
Sand
Figure 9.5.7. Possible pat-

--------
----------- tern of flow in a cross sec-
7/:ff:S:L7~f~:N~JL-~~:;~:h~<Il(:fj<;~iJ! -
tion consisting of two high
conductivity units separated
by a low conductivity
unit.39
Layer
verti-
7) leads
,ystem. 2
in order 3
4
.h. 5
8
cell are 7
greater
.dloss is
8
ver, the
uations.
ydraulic
Figure 9.5.8. A cross section in which a low conductivity unit is represented by six modellayers.39
one-dimensional flow in a prism of porous medium (Figure 3.1.1) as Q = -KA(h2 - h1)/L;

where Q is the volumetric flow, K is the hydraulic conductivity, A is the cross-sectional area
perpendicular to flow, h 1 - hz is the head difference across the prism parallel to flow, and L is
the length of the prism parallel to the flow path. Then the hydraulic conductance is defined as
C = KAIL and Darcy's law can be written as Q =C(h 1 - h;J. Another form of the hydraulic con-
ductance for horizontal flow in a prism is C = TBIL where T is the transmissivity in the direc-
tion of flow L2rr and B is the width of the prism.
Hydraulic conductance is defmed for a particular prism of material and for a particular
direction of flow. Anisotropic media are characterized by three principal directions of hydraulic
conductivity so that the hydraulic conductances of a prism differ in the three directions.
Now consider a prism of porous medium consisting of two or more subprisms in a series,
as shown in Figure 9.5.9. Knowing the conductance in each subprism, we can compute a
hydraulic conductance representing the entire prism, referred to as the equivalent hydraulic
conductance. The equivalent hydraulic conductance for the entire prism is C = Q!(hA- h8 ). The
total head loss (referring to Figure 9.5.9) is
n
LL1hi = hrhB
i=l (9.5.26)
9.6
hA=ho ha=hn
9.6.;
h, h2 h3 h4
t.h, t, h2 t,h3 t, h4 f:.hn
Q
_,.,I Q
I
c;J Cz c3 c4 Cn
I
I
.__.......__.....___J...__,_..J_------- ....___,
1 1 1 1 1
-=-+-+-+•••+-
c Cr Cz C3 Cn
Q Is the flow rate

Cm Is conductance of prism m
hm Is head at the rtght side of prism m
l::.hm Is the head change across prism m
Figure 9.5.9. Calculation of conductance through several prisms in a series. 39

IlL; And substituting for head change across each section using Darcy's law then leads to
rrea
Lis
ias (9.5.27)
:on-
rec- Because flow is one dimensional and it is assumed that there is no accumulation or depletion
in storage; then all q; are equal to the total flow Q. Then
ular h -h
mlic Q L-cl =
n
hA - hs and A
Q
B =L-
n 1
c
j:( i= (
I I
(9.5.28)
ries,
From C = Q/(hA- h8 ) and Equation 9.5.28, the equivalent hydraulic conductance is
tte a
1ulic I 1
-=I-
II
.The
c C; i=t
(9.5.29)
which states that for a set of hydraulic conductances arranged in series, the inverse of the
equivalent conductance equals the sum of the inverses of the individual conductances. For two
5.26) sections, the equivalent hydraulic conductance is C = C1Ci(C1 + C2) which explains the fonn
of Equation 9.5.3.
9.6 MODFLOW-2000 DESCRIPTION

9.6.1 Model Introduction
MODFLOW-2000 is a computer program that simulates one-, two-, or three-dimensional
groundwater flow using a finite difference solution of the model formulation presented in
Section 9.5 of this chapter, MODFLOW, the original computer program, was first published by
McDonald and Harbaugh.38 This program was developed using a modular structure for ease in
adapting the code for particular applications. The second version of MODFLOW, documented
in McDonald and Harbaugh,39 was referred to as MODFLOW-88. The third version was
documented by Harbaugh and McDonald21 • 22 and was referred to as MODFLOW-96. The
most recent version, MODFLOW-2000, documented in Harbaugh et aL, 20 includes many
capabilities added since the original computer program. This section briefly describes these
capabilities.
MODFLOW-2000 can simulate steady and nonsteady flow in an irregularly shaped flow
system in which aquifer layers can be confined, unconfined, or a combination of confined and
unconfined. Flow from external stresses, such as flow to wells, areal recharge, evapotranspi-
ration, flow to drains, and flow through river beds, can be simulated. The hydraulic conduc-
tivities or transmissivities for any layer may differ spatially and be anisotropic (restricted to
having the principal directions aligned with the grid axes), and the storage coefficient may be
heterogeneous. Specified head and specified flux boundaries can be simulated, as can a head-
dependent flux across the model's outer boundary, which allows water to be supplied by a
boundary block in the modeled area at a rate proportional to the current head difference
between a source of water outside the modeled area and the boundary block. MODFLOW-
2000 has been expanded to simulate solute transport and parameter estimation.
The MODFLOW-2000 computer program is divided into a main program and a series of
independent subroutines called modules. The modules are grouped into packages that deal
with a single aspect of the simulation. The packages are listed and briefly described in Table
9.6.1. Individual packages may or may not be required, depending on the problem being
Table 9.6.1 Processes, Packages, and Additional Capabilities of MODFLOW-2000 (Version 1.1, 1117/200 l)
Processes Packages Additional Capabilities
GWFl-Ground-Water Flow Process BAS6--Basic Package HYDMOD-Hydrograph Option 11
SEB !-Sensitivity Process BCF6--Block-Centered Flow
Package 15· 21 · 22· 40
OBSl-Observation Process LPFl-Layer-Property Flow Package
PES !-Parameter-Estimation Process RJV6--River Package
DRN6-Drain Package
WEL6-Well Package
GHB6--General Head Boundary Package
RCH6-Recharge Package
EVT6--Evapotransipiration Package
CHD6--Time-Variant Specified-Head
Package36
HFB6--Horizontal Flow Barrier Package28
SIPS-Strongly Implicit Procedure Package
SOR5-Siice Successive Over-Relaxation
Package
PCG2-Version 2 of Preconditioned Conjugate
Gradient Package24 Lay
DE45-Direct Solver 19
STR6-Streamflow-Routing Package53 Quasi
ADV2-Advective-Transport Observation confu
Package2
Layer
RESt-Reservoir Package 10
FHB 1-Flow and Head Boundary Pack.age35
IBS6--Interbed Storage (Subsidence) Paclcage36
Layer
HUFl-Hydrogeologic-Unit Flow Package3
LAK3-Lak:e Package41
ETSl-Evapotranspiration with a Segmented
Function Package6
DRTl-Drains with Return Flow Package6
solved. The basic documentation for MODFLOW is contained in Harbaugh and McDon-
ald,21· 22 Harbaugh et al.,20 Hill et al.,26 and McDonald and Harbaugh. 39
9.6.2 Space and Time Discretization

The physical size of the finite difference grid is provided through input to the program. In all
versions of MODFLOW, the finite difference grid is assumed to be rectangular horizontally
and can be distorted vertically as illustrated in Figure 9.6.1. The horizontal grid dimensions are
specified by the cell widths DELR and DELC (Figure 9.6.la). Columns are numbered starting
from the left side of the grid and rows are numbered starting from the upper edge (plan view)
of the grid. All cells in a column have the same width and all cells in a row have the same
width. Layers are numbered starting from the top layer down (see Figure 9.6.lb). The eleva- 9.6.3
tion of the top of Layer 1 and the bottom elevation of each Iayer for each cell are used to deter-
mine the thickness of each cell. A confining bed through which only vertical flow exists can
be simulated below each layer except the bottom layer. This simulation of confining beds is
referred to as the quasi-three-dimensional (quasi-3D) approacb. 39
DELR1 DELR1 DELR3 DELR. DELRs
Row\
l DELC,
Row2 DELC2
Row3
DELCJ
Row4 DELC.
Row 5 DEL(s
(a) Plan View
Figure 9.6.1. Finite difference grid

with (a) plan view and (b) cross-sec·
(b) Cross section along row i.
[cOon- tiona! view (Harbaugh et al. ~.
Time discretization is dealt with through time steps, which are grouped into stress periods.
For each stress period, the user specifies the total length (PERLEN), the number of time steps
(NSTP), and the multiplier for the length of the time step n - 1 times TSMULT. This proce-
1. In all
dure develops a geometric series with the length of the first time step given as
;ontally
ions are !J.t =PERLEN( TSMULT - 1 )
starting (9.6.1)
I TSMULTNSTP -1
n view)
te same
.e eleva- 9.6.3 External Sources and Stresses
to deter- 9.6.3.1 Rivers and Streams
tists can
~beds is The effects of flow between surface water features and groundwater systems are modeled by
adding terms for the seepage to or from the surface water feature to the finite difference model
Equation 9.5.24. Flow between a stream and the groundwater system, WiJ.k =QRIV, is
expressed as
QRIV = CRIV(HRIV- h1J,k), hiJ.k > RBOT
(9.6.2)
QRIV = CRIV(HRIV- RBOT), h1J,k:::; RBOT
where CRIV is the hydraulic conductance defined as KLB/M, in which K is the hydraulic con-
ductance of the streambed layer, Lis the length of the stream reach contained within the cell,
B is the width of the stream reach, and M is the thickness of the streambed material; HRIV is
the head in the stream; h;j,k is the head at the node in the cell underlying the stream reach; and
RBOT is the elevation of the bottom of the streambed layer.
9.6.3.2 Areal Recharge

Areal recharge, which most commonly occurs as a result of precipitation percolating into the
groundwater system, W,J,k = QRiJ• is expressed as
(9.6.3)
where QR;J is the recharge flow rate applied to horizontal cell (i,j), expressed as water volume
per unit time, and I 1J is the recharge flux (length per unit time) applicable to the area
DELRpELCr
9.6.3.3 Drains
Buried and open drains are shown in Figure 9.6.2. The flow into a drain is illustrated in the
cross-section through cell (i, j, k) in Figure 9.6.3. Flow in the drain is assumed to be only par-
tially full, so the head within the drain is approximately equal to the median drain elevation,
diJ.k The head, hiJ.k> is the average head for the cell. There are three processes of flow into the
drain: convergent flow through the drain, flow through inunediately around the drain with a
different hydraulic conductivity, and flow through the wall of the drain. Each of these
processes creates head losses that are assumed to be proportional to the discharge, QD1J.k• with
the total head loss h1J,k- d;J.k· The drain function, WiJ.k =QDiJ,k• is expressed as
Drain
opening
Figure 9.6.2. Factors affect-

ing head loss inunediately
around a drain: (a) buried
drain pipe in backfilled ditch
and (b) open drain
Drain pipe
(a) (McDonald and Harbaugh39).
'·
/, is Open
drain
6.2)
.·~~~·.· Lo~ pe~meabillty
con- . . '-. ·. ·. '
•• • • . • . · . '
•• • •• •.• • • . 'A.
•
. • • • •
drain-bottom
Figure 9.6.2 (continued). Factors affect-
ing head loss immediately around a drain:
1
.-/. . • -.....:.___: • sediments (a) buried drain pipe in backfilled ditch
cell, and (b) open drain (McDonald and
[Vis (b) Harbaugh 39).
; and
o the
1.6.3)
.]ume
Figure 9.6.3. Cross section through cell (i, j, k) illus-
area
trating head loss in convergent flow into drain.
(McDonald and Harbaugh39).
(9.6.4)
in the
(9.6.5)
y par-
·ation, where CDlJ.k is a lumped or equivalent conductance describing all of the head loss between the
itO the drain and the cell (see McDonald and Harbaugh 39). See Banta6 for more infonnation on mod-
with a eling drains with return flows.
these
~·with 9.6.3.4 Evapotranspiration
To simulate the effects of plant transpiration and direct evaporation of water from the saturated
groundwater, the volumetric rate of flow from a given cell is required. The volumetric rate of
evaporation, W;J,k =QETiJ,k• is the product of the loss rate per unit area of the cell and the hor-
izontal surface area, DELRj DELC;, given as
(9.6.6)
V.:here RrnJ is the rate of loss per unit surface area of water table due to evapotranspiration,
given as
Rr:"T':.
'·1
C..'
=RETM' . 1,] h;,j,k > hsi,j
R"""'.
"'' 1,]
=0 hi,j,k < hsi,j - di,j
R .. = R
Efr,J
. . h1,..1,k-(h·s1,;
ETMI,J [ d. .
.-d.·)]
,,,
(9.6.7)
affect- 1,]
ltely
ried in which hsiJ is the water table elevation at which the maximum value of evapotranspiration
~d ditch loss occurs, RETMiJ is the maximum possible value of RE7lJ• and d;J is the cutoff or extinction
depth so that when the distance between hsiJ and h;J,k exceeds diJ' then evapotranspiration
ceases. See Banta6 for modeling evapotranspiration with a segmented function.
436 Chapter 9 Groundwater F1ow Modeling Techniques
9.6.3.5 General Head Boundary

The flow Qbij.k into or out of a cell from an external source with a known head is proportional to
the difference between the head in cell, hiJ.k• and the head assigned to the external source, hbij.k• as
QbIJ,
.. k= cbIJ,
.. k(hb IJ,
.. k-hk)
IJ, (9.6.8)
where CbiJ,k is the conductance between the external source and cell (i,j, k).
DE
9.6.3.6 Wells
For injection wells, the source strength is
W;J,k =QWiJJ< (tm) (9.6.9)
where QWiJitJ is the well injection discharge during the stress period at time tm.
9.6.4 Hydraulic Conductance-Layer-Property Flow Package (LPF)

9.6.4.1 Horizontal Hydraulic Conductance
MODFLOW solves the finite difference Equations 9.5.24 using equivalent hydraulic conduc-
tances between nodes of adjacent cells, as explained in Section 9.5.4. CR (along rows) and CC
(along columns) are the horizontal equivalent hydraulic conductance terms between adjacent
horizontal nodes. The LPF Package in MODFLOW-2000 reads data defining the horizontal
hydraulic conductivity for individual cells and uses this information to compute the equivalent
hydraulic conductance between nodes, using one of three methods.
The first method assumes a constant transmissivity (hydraulic conductivity times thickness)
within a cell, but allows a discrete change in transmissivity at the boundary between the cells.
Because it is at the center of the cells, the hydraulic conductance between nodes is the equivalent
hydraulic conductance of two half cells in series, as shown in Figure 9.6.4. Using C =C1Cf(C1
+ C0, we obtain the conductance CR given in Figure 9.6.4 and the conductance CC as
(9.6.10)
where TR;J,k is the transmissivity in the row direction at cell (i,j, k). This approach for calcu-
lating interblock conductance is the harmonic mean method. CCi+II2,k is computed in the same
manner. The equations for CR and CC are used unless the transmissivities of both cells are
zero, resulting in an equivalent hydraulic conductance of zero.
Goode and Appel 15 describe three alternative approaches for calculating horizontal
hydraulic conductances. One method assumes that the transmissivity varies linearly between
nodes, referred to as the logarithmic-mean interblock transmissivity method. The second
method assumes a flat, homogeneous aquifer with a water table defining the arithmetic-mean
interblock transmissivity method. The third method assumes a flat aquifer with a water table
in which hydraulic conductivity varies linearly between nodes defining the arithmetic-mean
thickness and logarithmic-mean hydraulic conductivity metlwd. The first and third methods are
also included in the LPF package in MODFLOW-2000. Harbaugh 18 describes a generalized
Ito
,as
I~ DELRj ·I~
_ _ _....,.,__ _ _ _ DELRj+l_ _ _ __..,-.jl
T I l
TR .. k TA l,j+
.. 1' k
i.8) I,J,
I I
. k. I I
DT I,J, I
I
I
' l,j+
I
I
I
.. 1' k
I I
6.9)
.duc-
dCC
:1cent TR·/k
1 TR· · 1 k
ontal CR. k= 2 OELC
1 x " I,J+ '
1,}+2. TR I,J,
.. kOELAJ+ + TA l,j+
.. 1,II
nELR·
•alent 1 J
Figure 9.6.4. Calculation of conduc-
TR "jk Is transmissivity in the row direction in cell (i,j,k) tance between nodes using transmis-
1,
:ness) sivity and dimensions of cells
CR . , k Is conductance in the row direction between nodes (i,j,K) and (l.j+ 1,I<)
cells. 1,]+2, (McDonald and Harbaugh39).
valent
~zi(Cl
fmite-difference interblock transmissivity for unconfined aquifers and for aquifers having
·smoothly varying transmissivity.
9.6.4.2 Vertical Hydraulic Conductance

1.6.10)
The vertical hydraulic conductance is calculated assuming that nodes are in the center of cells
and discrete changes in vertical hydraulic conductivity occur at the boundaries of the layers,
as illustrated in Figure 9.6.5. Using Equation 9.5.29 with the hydraulic conductance for each
calcu- half cell results in
esame 1 1 1
~lis are -- -= + -=-==-==--==~=-===---
c~.j,k+ Y2, DELRpELC1VK1,j,k DELRpELC;VKi,j,k+l (9.6.11)
izontal (Y2)TIDCKi,j,k (Y2)TIDCK;,j,k+l
etween
second where VK;j,k is the vertical hydraulic conductivity of cell (iJ,k) and TIITC~j.k is the saturated
c-mean thickness of cell (i,j, k). Simplifying Equation 9.6.11 results in
er table
·c-mean 1 _ DELRjDELC;
10ds are c~.j.k+Y2- ..:;_;::.:...._
(Yz)nncKi.j,k (Y2)nncK;,j,k+l
__...;.,_ +..:;_;::..:...._____ (9.6.12)
eralized
VKI,],. k VKl,}t.. k+l
DELC;
T i,j,k •
VK;,;.k
THlCK;,j,k
I
I
I
I
/tI
I
I
1,f--------------- --- 1 c~.J.k•i
I
I
I
e VKiJ,k•l
i.},k+1
THICKI,j,k+1
~-------------- ------
l
I
I
I
I
I
Figure 9.6.5. Calculation of vertical
/ conductance between two nodes
/I
(Harbaugh et. al. 20).
The calculation of vertical conductance between two nodes with a semiconfining unit in
between (Figure 9.6.6) assumes that the semiconfming layer makes no measurable contribu-
tion to the horizontal conductance or the storage capacity of either model layer. Under such
assumptions, the confining bed restricts vertical flow between the cells. A quasi-three-dimen-
sional approach (McDonald and Harbaugh39) is used. In this approach, three intervals (lower
half of upper aquifer, semiconfining unit, and the upper half of the lower aquifer) are
represented in the summation of conductance between the nodes. Using Equation 9.5.29, we
obtain for the vertical hydraulic conductance
1 1 1 1
+
~=-;:--=-;::,;-=-;=-;-- + -=-;::;-;-;::--;::=-=-:;-;-;-;,---
cv 1
- DELRpELC;YK;,j,k DELR1 DELC.VKCB,,,,k DELRiDELC;VK;,j,k+t
i,j,k+2 (Ji)THICK;,j,k THICKcs (Ji)THICKi.j.k+l
(9.6. 13)
where VKCBij.k is the hydraulic conductivity of the semiconfining unit between cells (i,j, k)
and (i, j, k + 1) and THICKen is the thickness of the serniconfining unit. Equation 9.6.13 can
be simplified to
CV =-'----='-'------
()i)THICK;,j,k + THICKcs + ...:;....;;;;.:..._
(?i)THICK;,j,k+t
____
(9.6.14)
VKi,i.k VKCBi,j,k VK1,j,k+I
9.6.4.3 Vertical Flow Calculation under Dewatered ConditWns

A modification to the vertical flow calculation is required when a cell in a convertible layer is
unconfined and the cell immediately above is fully or partially saturated (see Figure 9.6. 7).
This modification (vertical flow correction) is applied to every iteration of the solution process .
.• P·' -
OELRj
cviJ.k+ t
CBVK
:!,
I
I
.
I,J,k+l
VK;,j,k+1
Figure 9.6.6.
Calculation of vertical
:al l.__,///_,/,_ ___
----_
-----_____.- - conductance between
two nodes with a semi-
confining unit
between.20
tin The actual flow between the nodes in the upper and lower cells is the flow through the con-
tbu- fining bed, q1.J,k+tl2 computed using
uch
len- (9.6.15)
·wer
are
so that the flow is downward from cell (i,j, k) to cell (i,j, k + 1) but is not dependent on hij.k+t·
, we
Equation 9.6.15 could be substituted into the finite difference Equation 9.4.24; however, the
bl,/,k --7 -----------------------

.k+l
VK J.i.k
i. l3) representing
semiconfining unit
j, k)
l can
•
IJ,k
6.14)
TOPI,J,k+1 --7 f---------i
h i,J,k+l ---7
.yer is
lj.k+l
VK ,J,k+1
• Figure 9.6.7. Situation in which a correction is
representing required to limit the downward flow into cell
1.6.7). (i, j, k + 1), as a result of partial desaturation of
aquifer unit
ocess. the ceU.20
matrix A in Equation 9.4.25 would be asymmetric, generating problems in the solution process.
Alternatively, in MOFLOW-2000, the flow is computed as qiJ,k+ll2 =CViJ,k+ttihiJ,k+t•- hiJ,k) so
that this tenn remains on the left side of Equation 9.4.24. Then a correction tenn qc is added to
the right side to compensate for allowing the computed flow to remain on the left side of Equa-
tion 9.4.24,
q; =cv .. k+ 2
l,j,
l (h;J-:-Ll-
. '
TOPi ' }. ' k+l)
(9.6.16)
where q; is the value of qc added to the right side of Equation 9.5.24 in the nth iteration and
hiJ,k+ln-l is the head from the preceding iteration.
9.6.5 Solver Packages

The four solver packages in the MODFLOW-2000 are (1) the strongly implicit procedure
(SIP), 39 (2) the slice successive over-relaxation (SOR), 39 (3) the preconditioned conjugate gra-
dient (PCG), 24 and (4) the direct solution based on alternating diagonal ordering (DE4). 19 A
detailed discussion of these various solvers is beyond the scope of this book. Interested read-
ers should refer to the noted references.
9.6.6 Telescopic Mesh Refinement

Telescopic mesh refinement in groundwater flow modeling is an approach to use a larger
(regional) encompassing model to define the boundary conditions and model parameters for a
smaller (local) embedded model. 34 (Also refer to Buxton and Reilly 8 and Ward et al. 63 for more
detail and applications of the approach.) This method is used when a detailed model is needed
for a relatively small area as compared to the entire aquifer system, such the analysis of well
fields 42 and contaminant plumes. 63 Essentially, the approach can be used to construct a detailed
model from a larger (regional) groundwater flow model and can be used repeatedly to construct
successively smaller (local) embedded models, hence the name "telescopic" mesh refinement.
Leake and Claar34 discuss two basic approaches to construct data for local groundwater
models. The first approach is to use a telescopic mesh refinement program such as MODTMR
(MODFLOW Telescopic Mesh Refinement), which constructs the data set for the perimeter
boundary conditions of a local model using cell-by-cell tenns and head values in a regional
modeL This model can construct embedded-model data sets for most of the MOD FLOW pack-
ages. A second approach is to use a manual-gridding process, a geographic information sys-
tem (GIS), or other data processing programs to construct local-model data sets from
grid-independent data. The first approach can provide maximum consistency between the
local and regional model and would require less effort than the second approach, which has the
main advantage in that local-model data sets are most consistent with locations and distribu-
tions of aquifer properties and sources and sinks contained in the grid-independent data sets.
Two other programs have been developed by the U.S. Geological survey. TMRDIFF (Tele-
scopic Mesh Refinement DIFFerence) constructs files with comparisons of computed heads or
drawdowns in the local model with computed heads or drawdowns in the regional model.
RIVGRD constructs MODFLOW input data files for the River Package, the Drain Package,
the General-Head Boundary Package, and the Stream Package. 53
Figure 9.6.8 illustrates the telescopic mesh refinement modeling approach. Figures 9.6.9
and 9.6.1 0 illustrate the application of this approach to a hazardous waste site in the Miami
River Valley near Dayton, Ohio. Ward et al. 63 used a nested series of three models designed on
the regional, local, and site scales. Note that the grid for the site model is oriented differently
than the local scale and regional models.
ess.
) so
:d to
:jUa·
i.l6)
. and
:dure
:gra-
1. 19 A
read-
......
~
~
M
-o
«i
~
u
larger
z<
"0
;.for a ~
~I
'-'
·more 0
~
;::::
0
«l
.eeded 0
....
.f well
0 0..
0..
~ "'t::
~tailed
lStruCt
:ment.
= ~
"0
0
1)1)
s
iwater i:l
v
nMR s
0
s:<
imeter t<::
:gional e
..c::
r pack· "'
II.)
E
m sys- 0
s from
·s..0
0
:en the
has the
]"'
istribu- .s....
0
1ta sets.
!
; (Tele-
1eads or
model.
ackage,
~
<
u
0
~
\.....~......
:.a"'
~
a0..
\
II.)
0
s:<
0
es 9.6.9 u
: Miami oq
\0
gned on oi
'ferently e
~
iZ
HAMILTON
STREAM NORTH WELLFIELD
REACHES
A
t
Y CONSTANT PRESSURE
BOUNDARY
HAMILTON
E SOUTH WELLFIELD
F
t
1·-+4-++H+H-1-H-
H
l Figure 9.6.9. Finite difference grid

used for the regional scale flow
model of the Great Miami River
Valley-Fill Aquifer. 63
--
,
HALON
• ~·C,RTH ~E~LFIELD
1-·1-- r--- 1- ·
f- .... - '
i.
'='
Ct~
TW O·MIL
ltiAIIILTQN
/J
"'Ill. ~-~
7
CHAM PI 0 P-AP
11
:l'!-DROELECTRI~
.
D M
"
~~
1- l- I~
1-- ·-
- IX"
iiTIE'IIiOpE-L
1--- 1- --·
--- f- 1-1-- 1-
---
I- l"'i
.
;
---
I
~E~hv
I
•
PS~!1AL
~
\BEC:~~T PAPER
I ~--
-.,-
i I - -..
"·"" ..., - -
"
• CONSTANr PRESSURE 0
I
1000 FEET
BOUNDARY I I I I
grid 0 300METERS Figure 9.6.10. Finite difference grid used for
e PUMPING CENTER the local-scale flow model of the Great Miami
:r River Valley-Fill Aquifer. Relationship to
regional-scale model is shown in Figure 9.6.9.63
9.7 CASE STUDY Using MODFLOW: Lake Five-0, Florida
MODFLOW is used widely in the United States and throughout Lake Five-0-represented by a zone of highly conductive
the world by consultants and government agencies. The U.S. Geo- material in layers 1-4 and for transient simulations. The stor-
logical Survey, as well as many other federal and state govern- age properties of the lake zone are identical to those of water.
ment agencies, use this program extensively and have applied it to
The within-layer variations in horizontal hydraulic conductiv-
many modeling applications. A few of the many applications by
ity were accounted for by using equivalent hydraulic conductivity 30'2!
the U.S. Geological Survey include Belitz and Phillips, 7 Grubbs, 16
values computed using a weighted mean algorithm (Freeze and
and Prince et al. 52
Cherry 11 ). The transmissivity, T, of cells in the transition zone
The hydrologic budget for Lake Five-0 in northwestern
were computed using
Florida (location shown in Figure 1.6.4) by Grubbs 16 was intro-
duced in Chapter 1. The data-collection sites for the study area are
shown in Figure 1.6.5. A three-dimensional model was developed
by Grubbs 16 to represent the Lake Five-0 groundwater system
under steady-state and transient conditions.
9.7.1 Finite Difference Grid and Boundary Conditions

A finite difference grid was developed to approximate the geom- (9.7.1)
etry of the major lithologic contacts and aquifer boundaries. The where L is the layer thickness, Kh,eq is the equivalent horizontal
areal discretization (81 rows and 57 columns) and boundary con- hydraulic conductivity, and Kh,SA and Kh,ICI are respectively the
ditions for the models are illustrated in Fig. 9. 7.1. All the rows in horizontal hydraulic conductivities for the surficial aquifer and
the grid have a width of 20m. Columns 4 through 1 progressively intermediate confining unit. The vertical flow between model lay-
increase in size westward from 30 to 100 m (expansion factor of ers was simulated using the leakance parameter (vertical hydraulic
1.5). Colurrms 7 through 57 have a constant width of 20 m. The conductivity divided by flow path distance). In the cases when the
vertical discretization and boundary conditions are illustrated in vertical hydraulic conductivity varied within a model layer or
Fig. 9. 7.2. In the vertical, the sediments from the land surface to across contiguous model layers, the leakance values were calcu-
the uppermost part of the Upper Floridan aquifer were modeled lated with equivalent vertical hydraulic conductivities using the
using seven horizontal layers of varying thickness. Layer 1 was harmonic mean algorithm of Equation 9.5.2.
modeled as a water table layer; layers 2-6 were modeled as con-
fined layers; and layer 7 was modeled as a specified boundary rep- 9.7.2 Model Calibration and Sensitivity Analysis
resenting the Upper Floridan Aquifer.
Figure 9.7.2 presents the calibrated values of the hydraulic Four steady-state models were calibrated to the hydrologic condi-
conductivity for the parameter zones used to represent variations tions on four separate dates in 1988 to 1990 in order to satisfy one
in hydraulic properties. A single value or limited number of values or more of the following objectives (see Grubbs 16): (1) Refine th~
of the hydraulic properties were assigned to parameter zones rep- premodeling or prior estimates of hydraulic properties and their
resenting the lithologic units in order to approximate the surficial spatial distribution, (2) Provide initial conditions for transient sim-
aquifer and intermediate confining unit. The following parameter ulations, and (3) Provide flow fields necessary for evaluation of
zones were used by Grubbs: 16 flow paths and residence times of groundwater inflow into the
Basal clay zone-to represent the dense, low-penneability clay lake. Calibration of the steady-state models was performed by
at the base of the intennediate confining unit; comparing simulated and observed heads, and by comparing sim-
ulated lake seepage with minimum leakage estimates from hydro-
Plateau zone-to represent increased anisotropy in the upper 5 logic budget calculations (see Section 1.6 for the hydrologic
to 10m of the intennediate confining unit in the plateau area; budget and see Grubbs 16 for more detail on the calibration).
Transition zone-to represent the geometry of the contact Calibration-sensitivity analysis of the steady-state model was
between the surficial aquifer and intennediate confining unit perfonned by varying model input variables (within probable
where this contact dips toward the breaches in the intermediate ranges) and comparing model outputs (simulate heads, lake
confining unit; inflow, and lake leakage) to known conditions. The root-mean-
Lakebed sediment zone-to represent lower permeability sed- square error (RMSE) of the simulated heads to changes in hori-
iments (relative to the surficial aquifer) in surficial aquifer cells zontal hydraulic conductivity, leakance, recharge, and anisotropy
contiguous with and beneath the lake in model layers 3 and 4; (continues) Figure~
(Grubbs
85°40' 65'39'30'
/
/
/
Rattlesnake
:tive ,'
/ Pond
stor-
ater. 40
LCtiV-
tivity
! and
zone
'
''
''
''
I
I '
I
:9.7.1)
zontal
~ly the
</)
er and ~
0
lellay- a:
draulic Campers
aen the Pond
tyer or
: calcu-
ing the
30'25'
I
I
I
I
>'
0 ---
60
; condi-
isfy one Big
~fine the Island
nd their Lake
entsim-
.ation of
into the
rmed by
I
'' \
0 1/4Mile \
ingsim- I
\
\
\
I
nhydro- 25 Kiklrnele< \
\
\
drologic
m).
odel was· ITE!Inactive area of model grid
probable EB3l
AcHve area of model grid: aquifer cells
tds, lake • Active area of model grid: lake cells
ot-mean- - - No-flow boundary
s in hori- • Welllocation
nisotropy
ontinues) Figure 9.7.1. Areal discretization and boundary conditions for simulation models of the groundwater flow system near Lalce Five-0
(Grubbs 16).
A A'
Recharge Recharge Recharge Recharge Recharge Recharge Recharge
~~~~~ ~~~~~ ~~~~~ 1~~~~ ~~~~~ ~~~1~ ~~~~~ Meters

~
~
Surf.1caI aqUI'fer { Kv
Kn :- 1B_ mid
1 8 rn!d .§
't:
12.2 ~
J:;
2 ~
9.1 ~
§
3 0"'
4.6 ..g
~
4
""'
0'
Sea
"'c:
level
""'E
00
5 a::
6.1
13.7 0
7 Upper Floridan Aquifer
Vertical exaggeraUon xtO
>>>>?Free-surface boundary (recharge flux specified over the surfical

H rn aquifer: net precipitation flux specified over Lake Five-0).
1 F fl :r
1
I
FEET
1001 ,..,..,RS
•••• I 0 50 "'~'"
- • Specified-head boundary
n
•
Transition { upper halt = surf. aquifer
zone lower half = lnt conf. unit I Plateau zone { Kn =0.61 meters/day (mtd}
Kv = 0.0012 meters/day (m'd)
fS=Sl Lake bed { Kh = 18 meters/day (mid)
U sediment zone K. =0·11 meter;tday (mid)
I Basal clay zone{ Kv = 3 x 10-6 meters/day (mtdl
Figure 9.7.2. Vertical discretization, boundary conditions, and calibrated values of hydraulic conductivity for simulation models of
the groundwater flow system near Lake Five-O(Grubbs 16). Cross-section A-A' location is sho\Vll in Figure 1.6.5.
9.7 CASE STUDY Using MOD FLOW' Lake Five-0, Florida (continued)
are shown in Figures 9.7.3 and 9.7.4. The steep slope of the head- recharge rates, net precipitation rates, and beginning and ending
sensitivity curves (change from calibrated values versus RMSE) heads to each stress period. First the specific yield in layer 1 and
illustrate the sensitivity of the model to these variables. Head the specific storage in layers 2-6 were adjusted to simulate the
response was most sensitive to changes in recharge, horizontal slope of well hydrographs. Recharge was negligible during the
hydraulic conductivity, and anisotropy of the surficial aquifer, and calibration period selected because of below-nonnal rainfall,
in horizontal hydraulic conductivity of the intennediate confining which made it possible to calibrate specific yield and specific stor-
unit. age independent of recharge. Sensitivity of the transient model to
Transient model calibration determined storage properties and specific yield and specific storage is illustrated in Figure 9.7 .5.
temporal changes in recharge rates. Hydraulic conductivities The model is more sensitive to changes in specific yield than to
determined from the calibrated steady-state model calibrations changes in specific storage.
were used in the transient model. The simulation period was
divided into 43 stress periods. Assigned to each stress period were (continues)
1.4
Horizontal hydraulic
conductivity, Recharge
surficial BQuifer
1.2
..
F
ii
\ I
.5. 1.0 Anisotropy, \ /
..
.,
"D
\ SUrlical I
"'.,
"D \ aquller \ I
~ \.
..."E 0.8 \ \ I
0 Horizontal hydraulic \
"' \ I
i
~
0.6
conductivity,
interme<liate
confining unit
\
\
''-.
AniSotropy, \ '
\ I
\ I I
...
c: \ lake sediments , \ / 1
E
0.-4 \ I
~
\ \ I •••••··
\ I
' \ (..••••.··Anisotropy,
0.2 Leakance, Layer 6 ' '

...................................................................... I
'-
I ................. •• intennediate
confining unit
O L-~~~~~~~~~~~~~~~
0.001 0.01 0.1
Chlllg8 from calibrated value, es a mufliPie ol caibrale<J value
Change from caftbrated valve. as a mulllpfe or calt>tated valve
Figure 9.7.3. Sensitivity of steady-state models of ground-

Figure 9.7.4. Sensitivity of steady-state models of ground-
water flow system near Lake Five-0 to changes in horizon-
tal hydraulic conductivity and leakance of layer 6 water flow system near Lake Five-0 to changes in recharge
and anisotropy (Grubbst6).
(Grubbs 16) .
0.04 r-r"""T"T1m.nr-
..1 T'T'nm,.,,....,.....
. .,,_,...,..,.,,_
, ..or.,rnm.·""'·.
Specific yield
leis of
. ending
!r 1 and
tlate the
ring the Specific
rainfall, storage
ific stor- /
/
nodel to
re 9.7.5. _____ ./ /
/
1 than to
Figure 9.7.5. Sensitivity of transient model of
.wl JJJj the groundwater flow system near Lake Five-0
10 100 1,000 10.000
to changes in aquifer storage properties
mtinues) Change trom calbraled vaAl&, as a lllUiiple
ot caibrated value (Grubbs 16).
9.7 CASE STUDY Using MODFLOW Lake Five-0, Florida (continued)
The calibration of recharge consisted of adjusting the recharge Figure 9.7.6. Simulated and computed net groundwater flows to
function to reproduce well hydrographs and monthly estimates Lake Five-0 are shown in Figure 9.7.7.
of the net groundwater flow. The recharge function used by
Grubbs 16 is 9.7.3 Model Results
Grubbs 16 concluded that the model simulations indicated that the
(9.7.2) groundwater system is the dominant source of water for Lake
Five-0 and that the groundwater system is the dominant sink for
water leaving Lake Five-0. In fact, the simulated groundwater
where Rtu i~ the recharge over time interval !J.t, S>' is the specific inflow and leakage were approximately four and five times larger
yield, and hrecess is the absolute value of the average change in than precipitation inputs and evaporative losses, respectively, dur-
water table during a given hydrograph recession under suitably ing the calendar years 1989-1990. Spatial distributions of simu-
dry conditions. Simulated and measured heads are shown in lated groundwater inflow and leakage showed little variation
between wet and dry periods.
19
18
17
18 16 9.8
15 15
9.8.
14 14
16
9.8.2
15
14
UN JFMAMJJASONDJFMAMJJASON 1
3 N J F MA MJ J A S0 N
1988 1989 1990 1988 1989 Figure 9.7.6. Simulated and
observed heads for Lake Five-0
and adjacent groundwater sys-
Simulated head from transient model
tem, December 12, 1988,
+ X o o Observed head ln Indicated wells
through January 22, 1991
W8 well number
(Grubbs 16).
9.8 Particle Tracking-MODPATH 449
6(),000 Corn~ul&d ne~ ground-.uater flow wilt\

lT99--perCE-nl confidence inteNC~l error bar
to
50,000 D Sim\Jia1ed net groundwater flow
40.000
the 30,000
!Ike
20,000
for
tj
- - - - -r
iter
10,000
-ger
-r- -------- ---~--H--p--~------------ --- ---t-------------

lur-
IUU-
tion -10.000
-20,000
-30.000
H tttt fH Figure 9.7.7. Simulated and com-
puted monthly net groundwater flow
-40.000 J F M A M J J J J to Lake Five-0, 1989-90
1989 1990 (Grubbs 16) .
9.8 PARTICLE TRACKING-MODPATH

9.8.1 What Is Particle Tracking?
Particle tracking is used to trace out flow paths, or path lines, by tracking the movement of
infinitely small imaginary particles placed in the flow field. 4 Particle tracking computer pro-
grams, such as the MODPATH,46-48 are postprocessors to MODFLOW MODPATH accepts
the head distribution from MODFLOW and uses it to compute the velocity distribution, 12
which is then used to trace out path lines. Particle tracking can be used to help visualize the
flow field and to track contaminant paths. Anderson and Woessner4 also point out that particle
tracking analysis should be used routinely with groundwater flow modeling to detect concep-
tual errors that cannot be detected solely by examining the head distribution. Particle tracking
analysis can _be used to evaluate the effects of different boundary conditions and can show the
location of recharge and discharge more clearly than results from groundwater flow models.
9.8.2 Particle Tracking Analysis-An Application

Grubbs 16 used the particle tracking code by Pollock47 to evaluate the groundwater path lines
(flow paths) and residence times for hypothetical parcels of groundwater (particles) that dis-
charge to Lake Five-0 in northwestern Florida. Figures 9.8.1 and 9.8.2 illustrate the results of
nd the particle tracking. Figure 9.8.1 indicates that most of the surrounding groundwater basin
;ive-0
contributes to flow into the lake. According to Grubbs, 16 the residence times (travel time of the
sys-
inflow particles from recharge at the water table to discharge at the lake) generally ranged from
0.6 to 9 years, with mean residence times ranging from 3 to 6 years. Also, there was very lit-
tle variation in these residence times between high and low water conditions.
Inactive area of model glid

c::::::::J Active area of model grid; aquifer cells
- Active area of model grid: lake cells
--Groundwater flow path
1 1 • 1 • Extent of contributing area Lc
- - - - No-flow boundary hy
n€
Vertic
Figure 9.8.1. Contributing area to Lake Five-0, as defined by groundwater flow paths of particles that discharge to the lake
(Grubbs 16). Figu
The flow paths in Figure 9.8.2 illustrate the vertical character of the groundwater flow
within and between the surficial aquifer and intermediate confining unit. Flow lines are
deflected downward at two stagnation points, a very small distance outside the lakebed. These
are locations where the head in the groundwater flow is the same as the altitude of the lake sur-
face. At these two points, groundwater flow into the lake and leakage from the lake to the
groundwater occur below these points. According to Grubbs, 16 the location of these two points
did not change much from low to high water conditions. Flow paths in Figure 9.8.2 indicate
Row
that most all of the groundwater flow near Lake Five-0 occurs within the surficial aquifer. In
addition the groundwater that discharges into the lake does not move through the intennediate
confining unit.
9.9 EXAMPLE APPLICATIONS AND INPUT OF MODFLOW

&UMPJ;J$ 9,9.1 Consider the aquifer (uniformly discretized into cells of 500ft x 500ft) shown in Figure 9.9.1, with spec-
ified head boundaries along row 1 and along column 7. Think of this as two rivers intersecting perpen-
TWO- dicularly in the northeastern corner of the modeled groundwater system. A single aquifer is being
DIMENSIONAL modeled so that only one layer is required. The aquifer is treated as a confined aquifer because of its rel-
STEADY-STATE ative thickness and because it does not experience large changes in saturated thickness. The transmis-
PROBLEM sivity of the aquifer is 500 ft2/d and it recharges at a rate of 0.001 ftld. A well is located at row 5, column
3, which discharges at a rate of 8,000 ft3/d. Heads for all the active cells are set at 10ft. A time step of
365 days is to be used.
9.9 Example Applications and Input of MODFLOW 451
~ru..nalto
Pond
8'
B
819
Location of 'r:::
hydrogeologic section B·B'
near Lake Five-0
B B'
Veltical exaggeration Is xt 0 0 100 200 300 Feet

I I II II
o 50 100 Meters
Figure 9.8.2. Particle traces projected onto hydrogeologic section B-B' near Lake Five-0 (Grubbs 16).
Column
2 3 4 5 6 7
I
10 9 8 6 4 2 0
.er flow 6 6 6 6 6 6 6
nes are
1. These 2 . . . . . . 6
3
ake sur-
e to the 3 . . . . . . 6 6 = constant head cell
o points 6
• = active cell
indicate
1uifer. In
Row
4 . . . . . . 6
8
1o = assigned head (ft)
mediate
5 . . 8 . . . 12 Q =well
6 L_____j
6 . . .
. . . 6
15 500ft
.vith spec-
1g perpen- 7 . . . . . . 20
Figure 9.9.1. Example 9.9.1
aquifer discretization with 500 ft
r is being '6 x 500 ft cells.
:of its rel-
. trarumiS-
The MODFLOW model input data is presented in Table 9.9.1 (Appendix B) with brief explanations
5, column of the input. The strongly implicit procedure (SIP) solution technique is used and a maximum of 50 iter-
me step of ations (MXITER) is specified. The number of iteration parameters (NPARM) is 5, the acceleration
number (ACCL) is 1, the head change criterion is 0.01, IPCALC = 1, WSEED = 0, and IPRSIP = 1. See
the MOD FLOW input manual for detailed descriptions of the input.
SOLUTION Output data are given in Table 9.9.2 (Appendix B).

•
EXA MPLE 9.9.2 This example uses MOD FLOW model for a steady-state problem in a confined aquifer. The aquifer is the
same as described in Example 4.2.2:
STEADY-STATE
K = 16.43 m/day
APPLICATION
COMPARING b =10m
ANALYTICAL
AND NUMERICAL
SOLUTIONS
The radius of influence of pumping is at 300 m from the pumping well. Construct a numerical model E)
under steady-state conditions using MOD FLOW, calculate the drawdown curve from r =0 m tor= 300
m and compare it with the analytical (Thiem) solution. T&
APl
co1
SOLUTION The input data are presented in Table 9.9.3 (Appendix B). To represent the radius of influence given in AN1
the original problem, a 600 m x 600 m one-layer mesh is constructed with 19 rows and 19 columns in AN1
MODFLOW model. The well is placed at the center point (300 m, 300 m). An increasing grid spacing SOl
expansion gives better results than a regular grid spacing model. The grid spacing scheme is given in
Table 9.9.4 (Appendix B) and is shown in Figure 9.9.2. SOL
1·····--·--................ +. . . . . . . . . . . . . . . i ....................... f ..............................j. . . ................ ·+···- . . . l---jtJ:LI. . . . . . . . . +·· . ··-..·-........ +-.........................;...................... ·· . ,............................... 1
·~1~
.: ! ) 3.1
" 1 2
-+--------:---------·-+----------1~----- ............... +---·+H~~J- -- ----·----- ------1--·---+---------..-+------- -1
-+-------+----·--·--j----------+----+----·- +~>·~~---- ...... -------!--·----.............+------- --:----------+--------- -I 2.!
----+---+----1 ----t----;---t--HH-+--t----+--+-----+----~: ---

l i - 2.(
--·------· --------·-:--------j--------------"·-·1- 1
............ ~ ----7-+t--f---+----~-- .............................................. ____ -----..·+----------+-·---·--
!
sc
' I 3:
-·-·----···-·-i---------·-t·---·---+-- . . . . . . . . . +--++++-+--r·----..------- ---·-·--·--·---
0
"C
1.5
i
3:
i ~
Cl
I
1.0
I
0.5
o.olc
I
i
I
I I Figure 9.9.2. Grid-1 scheme used for

l Example 9.9.2.
9.9 Example Applications and Input of MODFLOW 453
:ee The layer type is 0: confined. The Transmissivity flag is set to User-specified The boundaries of the
model are defined asjUed-head boundaries. A homogeneous transmissivity of 164.3 m/day is specified
over the model domain. To calculate the drawdown curve, 27 boreholes are defined along a radial direc-
tion from r =0.5 m to r =300 m. The pumping well is placed at the center of the domain, row 10, col-
• umn 10, with a pumping rate of 425m3/day. Simulation flow type is set to steady-state. See Table 9.9.5
(Appendix B) for output.
the The resulting drawdown curves from the analytical (Thiem) and nwnerical (MODFLOW) solutions
are given in Table 9.9.6 (Appendix B) and plotted in Figure 9.9.3.
The MOD FLOW results compare well to the analytical solution. The numerical results of the given
problem are within 0.05 m of the analytic results except near the pumping well (0.5 m < r < 8 m). An
exact comparison is not attained because of the use of a discrete rather than continuous spatial domain.
It should also be remembered that the radius of influence in the original problem, r:::; 300m, is approx-
imated by setting the boundaries of the square model domain as fixed-head boundaries. •
EXAMPLE 9.9.3 Using the data given in Example 4.4.4, construct a one-layer model in MODFLOW and calculate the
,odeI
drawdown vs. time curve 50 m away from the pumping well fort= 0 to t =1day. Compare the numerical
:300 TRANSIENT results to the analytical solution of Theis.
APPLICATION
COMPARING
en in ANALYTICAL
ms in AND NUMERICAL
acing SOLUTIONS
renin
SOLUTION The aquifer given in Example 4.4.4 is confined with the following supplemental data:
b =12m
K = 25 rnlday
s : :; 0.0001
Q =300 m3/day
The exact analytical solution of the problem (using Theis) is also given below for comparison. Two dif-
ferent grid spacing schemes are used in the MODFLOW model listed in Table 9.9.7 (Appendix B).
3.0.------------------------------------------.
--Thiem
2.5 -- -- MOD FLOW
2.0
E
c:
~
0
'0
~
0
o.oOL__ 50_ _ _10L0_ _1...L5_0--20L0--2....J:5=0=-30.1...0-~350

_l.
Figure 9.9.3. Example 9.9.2: Comparison of dis-
tance versus drawdown for the Thiem equation
Distance from wei\ (m) and MODFLOW.
.. ~.
I_I_! JI .i : I
II
1
1
it
I
!
I
I
! _I
l
I I
+
I il : i i !
I
I l
l II : !
I : i !
I
' l' f
i I
1-
I
f:l i
! . i
!
:
I I
9.1
9.1
. . J-::=:7":-';::_
1-------t-----\- -·- +---- r-·-- ! t=
1--- --- t ---- -- ·- · - -----~f--i--1·-t·-f·+IIHtt-!··-+·i···--;,·- ----+-···---· I_ ·- L
i
· !- --··- f----- --·:--- ' -·+·H HIHH+' -+-+·--11- ··-··+ ·---·--+---·--·+--------
- !-······--+•-- --1----f~-~~- +-f+lll;HI~.L -~~-

11-1- - -1- f ---- -
1--~~'---1--1~1--~-~+~~l~_:r•--+J~;~-~lt-f --
1 : +-----i--+---+ -~j_-
-- -- ·· - ----'---!1-+++;HIItl+i-
r--·· -·.- j - - -at-- ' ----c !-+-
!
I 1
! ' j Figure 9.9.4. Grid-2 model scheme for
i i ! j Example 9.9.3.
0.6
/-
....-:.....:/._..
-- --
0.5 /-
,?,;_...------·---·-·-·-·-·-·
;,
E'
-;; 0.4
,~
~
0
- - Theis solution
- -- MOD FLOW (Grid-1)
0.1 - - - - MOD FLOW (Grid-2)
0.2 0.4 0.6 0.8 1.2 Figure 9.9.5. Example 9.9.3: Comparison of time-
Time (days) drawdown for the Theis equation and MODFLOW
Grid-I covers a square area of 2,000 m x 2,000 m and consists of 19 rows and 19 columns. An increas-
ing grid spacing is used as shown in Table 9.9.7. Grid-2 (Figure 9.9.4) is an expanded version ofGrid-1 with
29 rows and 29 columns covering an area of 5,000 m x 5,000 m. The simulation time is set to one day with
. '\
20 time steps and a time step expansion factor of 1.3. The SIP package is used as the model solver with five
iteration parameters, 0.000 l as the closure criterion, and the maximum number of iterations is set to 50.
The resulting drawdown curves 50 m away from the pumping well using the Theis solution and
MODFLOW model are illustrated in Figure 9.9.5. The artificial fixed head boundaries of the Grid-!
model start to affect the drawdown curve at 50 m approximately 0.1 day after the pumping starts. To meet
the requirement of an infinite aerial extent aquifer, inherent in the Theis solution, the fixed head bound-
aries are moved farther away from the pumping well in the Grid-2 model. With this modifLcation, the
effect of the boundaries are not observed at 50 m for the given time period (0 < t < 1). The MODFLOW
results using Grid-2 compare well to the analytic solution. The numerical results are generally within I
em of the analytic solution. •
9.10 SOLUTE TRANSPORT MODELING-MOC3D

9.10.1 SoJute Transport Equation
The solute transport equation32 is
a(c:c) + a(pbc) +~(c:cv)-~[w.. 1

acJ
Clt 0( ClX; dX; I] ClXj (9.10.1)
- I,.c'W+A.(cC+pbC)=O
(summation over repeated indices is understood), where Cis volumetric concentration (mass
of solute per unit volume of fluid, MV\ Pb is the bulk density of the aquifer material (mass
of solids per unit volume of aquifer, ML-3), C is the mass concentration of solute sorbed on
or contained within the solid aquifer material (mass of solute per unit mass of aquifer material,
for MM- 1), E is the effective porosity (dimensionless), V is a vector of interstitial fluid velocity
components (LT- 1), Dis a second-rank tensor of dispersion coefficients (LT1), W is a volu-
metric fluid sink (W < 0) or fluid source (W> 0) rate per unit volume of aquifer (11), C' is the
volumetric concentration in the sink/source fluid (ML- 3), A. is the decay rate (1 1), tis time (T),
and xi are the Cartesian coordinates (L).
The terms controlling sorption are combined into a single parameter-the retardation
factor (R1), which is assumed to be constant in time because on a linear isotherm, C/ C is con-
stant. The retardation factor is defined as
R
1
=1+ PbC. (9.10.2)
eC
A number of the assumptions that follow are made in the development of the governing
equations. 32
1. Darcy's law is valid, and hydraulic head gradients are the only significant driving
mechanism for fluid flow.
2. The hydraulic conductivity of the aquifer system is constant with time. Also, if
the system is anisotropic, it is assumed that the principal axes of the hydraulic-
conductivity tensor are aligned with the coordinate system of the grid, so that the
cross-derivative terms of the hydraulic-conductivity tensor are eliminated.
time- 3. Gradients of fluid density, viscosity, and temperature do not affect the velocity
LOW. distribution.
4. Chemical or biological reactions do not affect the fluid or aquifer properties.
increas·
d-1 with 5. The dispersivity coefficients are constant over a flow time step, and the aquifer is
lay with isotropic with respect to longitudinal dispersivity.
The solution of the transport equation requires know ledge of the velocity (or specific dis-
charge) field. After the head distribution has been calculated for a given time step or steady-
state tlow condition, the specific discharge across every face of each finite difference cell
within the transport subgrid is calculated using a finite difference approximation. 32
The mass-tracking algorithm requires that the seepage velocity at any point within a cell
be defined to compute advective transport. Seepage velocity is calculated at points within a
finite difference cell based on linearly interpolated estimates of specific discharge at those
points divided by the effective porosity of the cell. 32
Gl
9.10.2 MOC3D Model ~
MOC3D is a general-purpose computer model developed by the U.S. Geological Survey for
simulation of three-dimensional solute transport in groundwater (see references 13, 14, 15, 23,
27, 29, 32, and 41 for more detail). This model computes change~ in concentration of a single
s
p
dissolved chemical constituent over time that are caused by advective transport, hydrody-
namic dispersion (including both mechanical dispersion and diffusion), mixing (or dilution)
from fluid sources, and mathematically simple chemical reactions (including linear sorption,
which is represented by a retardation factor, and decay). MOC3D can be used to simulate
groundwater age transport and the effects of double porosity and zero-order growth/loss. The
model uses the method of characteristics to solve the transport equation on the basis of the
hydraulic gradients that are computed with MODFLOW for a given time step. Particle track-
ing is used to represent advective transport and explicit finite difference methods are used to
calculate the effects of other processes.
The notation and conventions used in MOC3D to describe the grid and to number nodes
are illustrated in Figures 9.1 0.1 and 9.1 0.2. The indexing notation used is consistent with that
used in MODFLOW.
Application of MOC3D is the third step in a logical sequence of (1) flow, (2) advective
transport, and (3) solute transport simulation as illustrated in Figure 9.10.3. After the flow
TR
~
• ,,i-l,k •
(/J
~
0 •j-I,i,k •j,i,k •j+I,i,k
t
6.Yi Be
a: Ct
Di
R1
D!
• •j,i+l,k • /
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
G1
y L
/
/
/
/
/
/
/
/
/
/
/
,/" De
k v
j--------------------•x
Colums
1//
vv
/
1// /
Figure
• Node of finite-difference cell MODP
/v
!uj = llry = cell dimension in row direction /
1/ v/
ll.y; =ll.c; =cell dimension in column direction /
1// /
/v
Figure 9.10.1. Notation used to label rows,
columns, and nodes within one layer (k) of a
z 1//
three-dimensional, block-centered, finite dif- Figure 9.10.2. Representative three-dimensional grid for MOC3D illustrating
ference grid for MOC3D. 32 notation for layers. 32
!S- Concentration or occurence

ly- probability
ell D (}--20
• 21-40
ell • 41-75
na • 76-100 1 I I',
)Se
•,
./ f /
j ... ·-
.• I
GROUNDWATER FLOW
Hydraulic conductivity
Saturated thickness ' '
for No-flow zones
Recharge I ''.
23, Stream elevations . \
tgle Pumping ra!es

>dy-
ion)
Jon,
1late
The
f the
ack-
~ to
FLOWPATHS
todes
1 that
/ : /
~tive
. flow
Background concentrations
Concentration In sources
Dispersivity
Retardation factor
Decay coefficient - \ ......... ./ ! , \
Growth/loss rate .. ,_/ ... /' ~ "' ,_ / - , \ '-

Double porosity , '·
.. -
• I
r . ,
: ,smcl<
.. - j
'' '
' ' ' , ~ I
' _., f "\ i : I - "· . , I
Figure 9.10.3. Typical input and illustration of output for simulation of groundwater flow and solute transport using MODFLOW,
MODPATH, and MOC3D (Goode 14).
model (MODFLOW) is constructed, advective transport can be modeled by particle tracking

using MODPATH.48 Because travel times are needed, the effective porosity must be specified.
The solute transport as a result of advection and dilution (mixing) can be modeled using
Jstrating MOC3D using specified initial and source concentrations. Optional input on dispersion and
reactions allows effects of these processes to be modeled. The simulation of solute transport
with dispersion using MOC3D can enhance particle-tracking analyses by approximating uncer-
tainty in source-area delineation and potential contamination advection. 13· 14
Figure 9.10.3 shows nine closely spaced pathlines of groundwater flow from a continuous
source of contamination after 20 years. These were calculated using MODFLOW and
MODPATH. MOC3D was used to simulate the dispersed plume from the same source loca-
tion. This dispersed plume can be interpreted as the probability that the contaminant will be
detected at each location within 20 years. As pointed out by Goode, 13 the dispersion coefficient
approximately accounts for the uncertainty in predicted flowpaths caused by imperfect knowl-
edge of aquifer properties, recharge rates, stream levels, and other factors. Detailed field char-
acterization can reduce uncertainties and make the contaminant occurrence probability map
9.
more accurate. These approaches can also be used to delineate probabilistic source areas (cap- 9.
ture zones) for pumping wells using MOC3D.
An example of a MOC3D application is illustrated in Figure 9.1 0.4, which uses time- and
space-varying decay coefficients to approximate biodegradation of benzene at a Laurel Bay,
South Caro1ina research site. Benzene attenuation depends on decay rates that are generally
higher in aerobic (oxygen rich) zones and lower in anaerobic (zero oxygen). Anaerobic
biodegradation rates are dependent on the specific terminal-electron-accepting process
(TEAP) controlling the redox state. Figure 9.1 0.4a shows the TEAP zones. Also in the ground-
water at this site is methyl tert-butyl ether (MTBE). This is also a gasoline-derived contami- 9.:
nant, but is essentially nonreactive during transport. Source concentrations are the same for
Berw me fppb) MTBE (ppb}

Reactive 0 Non-reactive
9.1
9.1
- - 1994 MTBE (ppb) contours

------ 1998 MTBE (ppb) contours
0-10,000 parts per billion 0- 10,000 parts per billion
1998 TEAP iron/sulfate zone
Figure 9.10.4. MOC3D simulation of benzene and MfBE at the Laurel Bay, SC, research site, based on field and lab data of
Landmeye~ 3 and others (1998): (a) Location map with anaerobic terminal-electron-accepting-process (TEAP) zones in 1994 .
(hatched) and 1998 (shaded); (b) Benzene in 1994 and 1998; (c) MTBE in 1994 and l998.ln (b) and (c), the 1998 shaded maps
range from 0 to 10,000 parts per billion (ppb); contours show I 00, 1,000, and 10,000 ppb concentration levels (as presented in
Goode 14).
9.11 Groundwater Modeling Software Support 459
er- both contaminants; therefore, differences in simulated concentrations result from differences
in decay coefficients. Outside the TEAP zones fast aerobic degradation is taking place, except
JUS at the early time when microbes are assumed to be in low concentrations. This is illustrated by
md the differences in the 1994 and 1998 zones (see Figure 9.10.4a). Figure 9.l0.4b show the
ca- results of MOC3D simulations of the benzene migration for 1994 to 1998 that has temporally
be and spatially varying decay coefficients. Figure 9.10.4c shows the results of MOC3D simula-
.ent tions of the MBTE for 1994 to 1998 .
wl-
lar- 9.11 GROUNDWATER MODELING SOFTWARE SUPPORT
nap
:ap- 9.11.1 U.S. Geological Survey
The U.S. Geological Survey provides access to public-domain software, such as MODFLOW,
and
MODPATH, MOC3D, and several other simulations models such as SUTRA for two- and
3ay,
three-dimensional variably saturated flow, solute or energy transport and several other com-
mly
puter models. ZONEBUDGET is a program for computing water budgets for MODFLOW.
:Jbic
Access this software at http://water.usgs.gov/softwarelground_water.html.
cess
Llnd-
amt- 9.11.2 U.S. EPA Center for Exposure Assessment Modeling (CEAM)
~ for
The Center for Subsurface Modeling Support (CSMoS) provides public domain groundwater
and vadose zone modeling software and services to public agencies and private companies.
CSMoS is located at the National Risk Management Laboratory (NRML), U.S. EPA Center
for Ground-Water Research, Ada, Oklahoma. Its primary aims are to provide direct technical
support to EPA and state decision makers in subsurface modeling applications and to support
the groundwater models and databases resulting from the research at NRML. A major focus of
CSMoS is the coordination of the use of models for risk assessment, site characterization,
remedial activities, wellhead protection, and Geographic Information Systems (GIS) applica-
tion. The Web address for CSMoS is http://www.epa.gov/ada/csmos.html.
9.1 1.3 International Groundwater Modeling Center (IGWMC)

The International Ground Water Modeling Center (IGWMC) provides infonnation, education,
and research for groundwater modeling. It is located at the Colorado School of Mines, Golden,
Colorado. IGWMC advises on groundwater modeling problems, distributes software, organ-
izes short courses and workshops, and provides technical assistance on problems related to
groundwater modeling. The Web address for IGWMC is http://www.mines.edu/igwmc/about.
9.11.4 Processors for MODFLOW

Because of the wide usage of the MODFLOW model, several pre and post-processing models
have been developed. Some of the processors are PMWIN, Visual MODFLOW, MODFLOW-
GUI PIE, and VIEWLOG. These processors can be located easily on the Internet by perform-
ing a search for their names.
PMWIN (see Chiang and Kinzelbach9) is a simulation package for pre- and post-pro-
cessing that includes MODFLOW, MT3D, MT3DMS, PMPATH, UCODE, and PEST-LITE.
Visual MODFLOW, developed by Waterloo Hydrogeologic, Inc. of Ontario, Canada, is a
pre-and post-processor for the MODFLOW and MODPATH models. This allows the user to
ps solve the flow model and to perform particle tracking and velocity vector analysis. The program
is completely graphical, allowing the user to input all necessary input parameters, run the
program, and visualize the analysis results. The Visual MODFLOW package includes the
MODFLOW and MODPATH models and can import previously created MODFLOW data 9.
files. VIEWLOG is also a pre-and post-processor for the MOD FLOW and MOD PATH models. st
MODFLOW-GUI PIE is a U.S. Geological Survey public domain GIS pre-and post- ra
processor graphical-user interface for preparing MODFLOW, MOC3D, MODPATH, and 7.
til
ZONEBDGT input and viewing model output for use within Argus ONE. MODFLOW GUI
fll
PIE is used to graphically input GIS data, run MODFLOW, and visualize the results from
PJ
within Argus ONE, which is an easy-to-use graphical workplace. pc:
PROBLEMS
9.2.1 Explain the concepts of physical scale models, analog mod- grid. Compare the results with those obtained in Example 9.9.1 R
els, and mathematical models. and justify your conclusions. 1.
9.2.2 Discuss the concepts of porous media models, analog mod- 9.9.3 Run the MOD FLOW model for the system in Example 9.9.1 Wt
els, and electric analog models. in a transient mode. Five time steps are to be used with a time step 2.
multiplier of 1.5 and a stress period length of 365 days specified (A
9.2.3 Discuss the theory upon which viscous fluid models are
in the BASIC package. Print the mass balance (budget) and head Ot
based. Gr
distributions for all five time steps using the OUTPUT CON-
9.2.4 Explain how electric analog models are used to model Ge
TROL PACKAGE. Use a specific storage of 0.01.
groundwater flow. 3.
9.9.4 Modify the data set from Problem 9.9.3 to use the OUTPUT icc
9.2.5 Explain how sand tank models are used to model ground-
CONTROL PACKAGE to print out the model-wide mass balance Hy
water flow.
and to save the cell-by-cell budgets for the BCF, WELL, and Op
9.2.6 Explain the concept of membrane models for groundwater RECHARGE packages at time step 1. Use the hydraulic heads 4.
flow. generated for time step 1 to manually compute the model-wide Me
9.2.7 Explain how Hele-Shaw models have been used in ground- rate components into storage, out of storage, well discharge, out Sa1
water flow modeling. of constant heads, into constant heads, and recharge. Use Darcy's 5.
9.2.8 Discuss in general the difference between the finite law to compute constant head flux. Also use the definition of the the
difference and finite element methods of groundwater modeling. storage coefficient to determine the rate change in storage. Com- 6.
9.3.1 Discuss in detail the processes of calibration, calibration pare your results to those computed by the model. Gn
sensitivity analysis, and model verification and how they work 9.9.5 Use the input file developed for Problem 9.9.4 and run it to trai
together to develop a model. obtain the cell-by-cell budgets. Compare the model-generated val- Fie
pp,
9.3.2 Discuss in detail the processes of prediction and predictive ues to the Problem 9.9.4 hand calculations.
7.
sensitivity analysis in model development. 9.9.6 Use the model developed in Problem 9.9.1 and eliminate the
Flo
9.3.3 Develop a brief discussion of the various ASTM standards well in row 5, column 3. Perform a steady-state simulation for a U.5
for groundwater flow modeling. recharge of 0.001 ft/d and a transmissivity of 500 ft 2/d. 8..
9.4.1 Show the development of Equation 9.4.10a. 9.9. 7 Use the model developed in Problem 9. 9.1 and eliminate the Sys
well in row 5, column 3. Perform a steady-state simulation for a Yor
9.5.1 Derive Equation 9.5.2 for the effective hydraulic conductiv-
recharge of 0.001 ft/d and a transmissivity of 50 ft2/d. Compare 198
ity.
the hydraulic heads to those generated in Problem 9.9.6. 9. l
9.5.2 Derive Equation 9.5.22, which is the finite difference PM
9.9.8 Use the model developed in Problem 9.9.1 and eliminate the
a,pproximation of the continuity for cell (i,j, k). !uti,
well in row 5, column 3. Perform a steady-state simulation for a
9.6.1 Derive Equation 9.6.10 for the calculation of conductance recharge of 0.0001 ft/d and a transmissivity of 50 ft2/d. Compare 10.
between nodes. the hydraulic heads to those generated in Problem 9.9.6. putt
9.6.2 Derive Equation 9.6.12 for the calculation of vertical con- ulm
9.9.9 Rerun the steady-state model (one stress period and one time Gee
ductance between nodes. step of one day) developed in Problem 9.9.6 with the option in the
11.
9.9.1 Run the MODFLOW model for the problem in Example BCF package invoked to print out the individual specified head woe
9.9.1 using a time step of one day and a time step multiplier of 1.0. fluxes.
12.
Compare the results with those obtained in Example 9.9.1 and 9.9.10 Run the steady-state model (one stress period and one time difft
justify your conclusions. step of one day) developed in Problem 9.9.9 with a well in row 5, 5, PI
9.9.2 Run the MODFLOW model for the problem in Example column 3 pumping at a rate of -8,000 ft 3/d. Print out the individ- 13.
9.9.1 using a time step of one day and a time step multiplier of 1.0. ual specified head fluxes and compare these to those generated in tion.
Use an initial head condition of 1,000 ft in the active part of the Problem 9.9.9. Sun
References 461
'
·.. ~
ata 9.9.11 Develop a drawdown model using the parameters and 9.9.12 Resolve Problem 9. 9.11 with the well rate doubled.
~Is.
stresses of Problem 9.9.10. Use an initial head ofO.O, a recharge Compare the heads of Problem 9.9.ll with those generated in
rate of 0.0, and specified heads of 0.0 along row one and column this problem.
!St-
7. Run the steady-state model with one stress period and one 9.9.13 Resolve Example 9.9.2, doubling the transmissivity. All
tnd
time step of one day, printing out the individual specified head other data remain the same. Compare the results with those
lUI fluxes. On a node-by-node basis, add the heads generated in reported in Example 9.9.2.
om Problem 9.9.9 and the solution generated in this problem. Com- 9.9.14 Resolve Example 9.9.2 using a 30 by 30 grid. Compare
pare these heads to those generated in Problem 9.9.1 0. the results with those presented in Example 9.9.2.
1.9.1
REFERENCES
I. Alley, W. M., T. E. Reilly, and 0 . L. Franke, Sustainability ofGrottnd- 14. Goode, D. J.. Simulating Contaminant Auenuation, Double-Porosity
1.9.1 Water Resources, U.S. Geological Survey Circular 1186, 1999. &change, and Water Age in Aquifers Using MOC3D, U.S. Geological
2. Anderman, E. R., and M. C. Hill, Advective-Transport Observation Survey face Sheet 086-99, April 1999.
step
(ADV) Package, a Computer Program for Adding Advective-Transport 15. Goode, D. 1., and C. A. Appel, Finice Difference lnterblock Transmis-
ified
Observations of Steady-State Flow Fields to the Three-Dimensional sivity for Unconfined Aquifers and for Aquifers Having Smoothly Varying
itead
Growul-Water Flow Parameter-Estimation Node/ MODFLOWP, U.S. Transmissivity, U.S. Geological Survey Warer-Resources Investigations
ON- Geological Survey Open-File Report97-14, 67 pp, 1997. Reporl92-4124, 79 pp, 1992.
3. Anderman, E. R., and M. C. Hill, MODFWW-2000, The U.S. Geolog- 16. Grubbs, J. W., Evaluacion of Ground-Water aJld Hydrologic Bw:Jget for
PUT ical Survey Modular Groundwacer Model- Documentarian of the l..ilke Five-0, a Seepage Lake in Northwestern Florida, U.S. Geological Sur·
lance Hydrologeologic-Unit Flow (HUF) Package, U.S. Geological Survey vey, Water-Resources Investigations Repon 94-4145, Tallahassee, FL. 1995.
, and Open-File Repon 00-342, 89 pp, 2000. 17. Hanson, R. T., and S. A. Leake, Documentation for HYMOD, A Pro-
teads 4. Anderson, M. P., and W. W. Woessner, Applied Groundwater gram for Extracting and Processing Trme-Series Data from the U.S. Geo-
wide Modeling: Simulation of Flow and Advective Transpon, Academic Press, logical Survey's Modular Three-Dimensional Finite Difference
:, out San Diego, CA, 1992. Ground-Water Flow Model, U.S. Geological Survey Open-File Repon
lCcy's 5. Appel, C. A., and J. D. Bredehoeft, Status of Groundwater Modeling in 98-564, 57 pp. 1999.
)f the the U.S. Geological Survey, U.S. Geological Survey Circular 737, 1976. 18. Harbaugh, A. W., A Generalized Finite-Difference /nterblock Trans-
Com- 6. Banta, E. R., MODFLOW-2000, The U.S. Geological Survey Modulilr missivity for Unconfined Aquifers and for Aquifers Having Srrwothly Vary-
Groundwater Model-Documentation of Packages for Simulating Evapo- ing Transmissivicy, U.S. Geological Survey Open-File Report91-494, 60
transpiration with a Segmented Function (ETS I) and Drains with Return pp, 1992.
nit to
Flow (DRTJ ), U.S. Geological Survey Open-File Report 00-466, 127 19. Harbaugh, A. W., Direct Solution Package Based on Alternating
:d val-
pp. 2000. Diagonal Ordering for the U.S. Geological Modular Finite-Difference
7. Belitz, K., and S. P. Phillips, Numerical Simulation of Ground-Water Groundwater Flow Model, U.S. Geological Survey Open-File Report
1te the Flow in the Central Pan of the Western San Joaquin Valley, California, 95-288,46 pp, 1995.
1 for a U.S. Geological Survey Water-Supply Paper 2396, 69 pp, 1993. 20. Harbaugh, A. W., E. R. Banta, M. C. Hill, and M. G. McDonald,
8. Buxton, H., and T. E. Reilly, A Technique for Analysis ofGround-Water MODFLOW-2000: The U.S. Geological Survey Modulor Ground-Water
ate the Systems of Regional and Subregional Scales Applied to Long Island, New Model-User Guide to Modularization Concepts and the Ground-Water
.\ for a York, U.S. Geological Survey Water-Supply Paper 2310, pp. 129-142, Flow Process, U.S. Geological Survey Open-File Report 00-92, 121 pp,
mpare 1986. Reston, VA, 2000.
9. Chiang, W. H., and W. K.inzelbach, 3D-Groundwater Modeling with 21. Harbaugh, A. W., and M. G. McDonald, User's Documentation for
ate the PMW/N, A Simulation System for Modeling Groundwater Flow and Pol- MODFWW-96, An Update to the United States Geological Survey Mod-
n for a lution, Springer, Berlin, 2000.
I 0. Fenske, J.P., S. A. Leake, and D. E. Prudic, Documentation ofa Com-
ular Finite Difference Groundwater Flow Model, U.S. Geological Survey
Open-File Report 96-485, 56 pp, 1996.
>mpare
puter Program (RES I) to Simulate Leakage from Reservoirs Using Mod· 22. Harbaugh, A. W., and M.G. McDonald, Programmer's Documenta-
ular Finite Difference Groundwater Flow Model (MODFLOW), U.S. tion for MODFWW-96, An Updare to the U.S. Geological Modular Finite
11e time Geological Survey Open-File Repon 96-364, 51 pp, 1996. Difference Groundwater Flow Model, U.S. Geological Open File Report
11 in the %-486, 220 pp, 1996.
II. Freeze, R. A., and J. A. Cherry, Groundwater, Prentice Hall, Engle-
:d head wood Cliffs, NJ, 1979. 23. Heberton, C. I., T. F. Russell, L. F. Konikow, and G. Z. Hornberger,
12. Goode, D. J., Particle velocity interpolation in block-centered finite A Three-Dimensional Finite Volume Eulerian-Lagrangian Localized
ne time difference groundwater flow models, Water Resources Research, v. 26, no. Adjoint Method (ELLAM)for Solute Transpon Modeling, U.S. Geological
trow 5, 5,pp. 925-940, 1990. Survey Water Resources Investigations Report 00-4087, Reston, VA, 2000.
tndivid- 13. Goode, D. J., Age, Double Porosity, and Simple Reaction Modifica- 24. Hill, M. C., Preconditioned Conjugate-Gradient 2 (PCG2), a Com-
rated in tions for the MOC3D Ground-Water Transport Model, U.S. Geological puter Program for Solving Ground-Water Flow EquaJions, U.S. Geologi-
Survey Water-Resources Investigations Report 99-4041 , 1999. cal Survey Water-Resources Investigations Repon 90-404S, 43 pp, 1990.
25. Hill, M. C., Methods and Guidelines for Effective Model Calibration, 40. McDonald, M.G., A. W. Harbaugh, B. R. Orr, and D. J. Ackerman,
U.S. Geological Survey Water-Resources Investigations Report 98-4005, A Method of Converting No-Flow Cells to Variable-Head Cells for the
90 pp, 1998. U.S. Geological Survey Finite Difference Groundwater Flow Model, U.S.
26. Hill, M. C., E. R. Banta, A. W. Harbaugh, and E. R. Anderman, Geological Survey Open-File Report 91-536, 99 pp, 1992.
MODFLOW-2000, The U.S. Geological Survey Modular Ground- Water 41. Merritt, M. L., and L. F. Konikow, Documentation of a Computer
Model-User Guide to the Observation, Sensitivity, and Parameter- Program to Simulate Lake-Aquifer Interaction Using MODFLOW
Estimation Processes and Three Post-Processing Programs, U.S. Groundwater Flow Model and the MOC3D Solute-Transport Model, U.S.
Geological Survey Open-File Report 00-184,210 pp, 2000. Geological Survey Water-Resources Investigations Report 00-4167,
27. Hornberger, G. Z., L. F. Konikow, and P. T. Harte, Simulating 146 pp, 2000.
Horizontal-Flow Barriers Using the MOD FLOW Ground- Water Trans- 42. Miller, R. T., and C. I. Voss, Finite difference grid for a doublet well
port Process, U.S. Geological Survey Open-File Report 02-52, 2002. in an anisotropic aquifer, Ground Water, v. 24, no. 4, pp. 490-496, 1987.
28. Hsieh, P. A., and 1. R. Freckleton, Documentation of a Computer 43. Peaceman, D. W., Fundamentals of Numerical Reservoir Simulation,
Program to Simulate Horizontal-Flow Barriers Using the U.S. Geologi- Elsevier Scientific Publishing Company, Oxford, UK, 1977.
cal Survey Modular Three-Dimensional Finite Difference Ground-Water 44. Peaceman, D. W., and H. H. Rachford, Jr., The numerical solution of
Flow Model, U.S. Geological Survey Open-File Report 92-477, 32 pp, parabolic and elliptic differential equations, Jour. Soc. of Industrial and
1993. Applied Mathematics, v. 3, pp. 28-41, 1955.
29. Kipp, K. L., L. F. Konikow, and G. Z. Hornberger, An Implicit Dis- 45. Pinder, G. F., and W. G. Gray, Finite Element Simulation in Surface
persive Transport Algorithm for the U.S. Geological Survey MOC3D and Subsurface Hydrology, Academic Press, San Diego, CA, 295 pp,
Solute-Transport Model, U.S. Geological Survey Water-Resources Inves- 1977.
tigations Report 98-4234, 1998.
46. Pollock, D. W., Sernianalytical computation of path lines for finite
30. Klempt, W. B., T. R. Knowles, G. R. Elder, and T. W. Sieh, Ground- difference models, Ground Water, v. 26, no. 6, pp. 743-750, 1988.
water Resources and Model Applications for the Edwards (Balcones
47. Pollock, D. W., Documentation of Computer Programs to Compute
Fault Zone) Aquifer in the San Antonio Region, Texas, Report 239, Texas
and Display Pathlines Using Results from the U.S. Geological Survey
Department of Water Resources, Austin, TX, October 1979.
Modular Three-Dimensional Finite-Difference Ground-Water Flow
31. Konikow, L. F., and J.D. Bredehoeft, Groundwater models cannot be Model, U.S. Geological Survey Open-File Report 89-381, 188 pp, 1989.
validated, Advances in Water Resources, v. 15, pp. 75-83, 1992.
48. Pollock, D. W., User's guide for MODPATHIMODPATH-PLOT, Ver-
32. Konikow, L. F., D. J. Goode, and G. Z. Hornberger, A Three-Dimen- sion 3: A Particle Tracking Post-Processing Package for MODFLOW, the
sional Method-of-Characteristics Solute-Transport Model (MOC3D), U.S. Geological Survey Finite-Difference Ground-Water Flow Model,
U.S. Geological Survey Water Resources Investigations Report 96-4267, U.S. Geological Survey Open-File Report 94-464, 234 pp, 1994.
87 pp, 1996.
49. Prickett, T. A., Modeling techniques for groundwater evaluation, in
33. Landmeyer, J. E., F. H. Chapelle, P.M. Bradley, J. F. Pankow, C. D. V. T. Chow, ed., Advances in Hydroscience, Academic Press, v. 10, pp.
Church, and P. G. Tratnyek, Fate of MTBE relative to benzene in a
1-143, 1975.
gasoline-contamination aquifer (1993-1998), Ground Water Monitoring
and Remediation, v. 18, No.4, pp. 93-102, 1998. 50. Prickett, T. A., Ground-water cq_l)lputer models-State of the art,
Ground Water, v. 17, pp. 167-173, 1979.
34. Leake, S. A., and D. V. Claar, Procedures and Computer Programs
for Telescopic Mesh Refinement Using MOD FLOW, U.S. Geological Sur- 51. Prickett, T. A., and Lonnquist, C. G., Selected digital computer tech-
vey Open-File Report 99-238, 53 pp, 1999. niques for groundwater resources evaluation, Bulletin No. 55, Illinois
State Water Survey, Urbana, IL, 1971.
35. Leake, S. A., and M. R. Lilly, Documentation of a Computer Pro-
gram (FHB 1) for Assignment of Transient Specified-Flow and Specified- 52. Prince, K. R., 0. L. Franke, and T. E. Reilly, Quantitative Assessment
Head Boundaries in Applications of the Modular Finite Difference of the Shallow Ground-Water Flow System Associated with Conneeetquot
Groundwater Flow Model (MOD FLOW), U.S. Geological Survey Open- Brook, Long Island, New York, U.S. Geological Survey Water-Supply
File Report 97-571,50 pp, 1997. Paper 2309, 28 pp, 1988.
36. Leake, S. A., and D. E. Prudic, Documentation of a Computer Pro- 53. Prudic, D. E., Documentation of a Computer Program to Simulate
gram to Simulate Aquifer-System Compaction Using the Modular Finite Stream-Aquifer Relations Using a Modular, Finite-Difference, Ground-
Difference Ground-Water Flow Model, U.S. Geological Survey Tech- water Flow Model, U.S. Geological Survey Open-File Report 88-729,
niques of Water-Resources Investigations, Book 6, Chap. A2, 68 pp, 113 pp, 1989.
1991. 54. Remson, 1., G. W. Hornberger, and F. J. Molz, Numerical Methods in
37. Liggett, J. A., and P. L-F., Liu, The Boundary Integral Equation Subsurface Hydrology, Wiley-Interscience, New York, 389 pp, 1971.
Method for Porous Media Flow, Allen and Unwin, 255 pp, 1983. 55. Rushton, K. R., and S. C. Redshaw, Seepage and Groundwater
38. McDonald, J. M., and A. W. Harbaugh, A Modular Three- Flow-Numerical Analysis by Analog and Digital Methods, John Wiley
Dimensional Finite Difference Groundwater Flow Model, U.S. Geologi- & Sons, New York, 1979.
cal Survey Open-File Report 83-875,528 pp, 1984. 56. Texas Water Development Board, GWSIM groundwater simulation
39. McDonald, J. M., and A. W. Harbaugh, A modular three- program, program document and user's manual, UM S7405, Austin, TX,
dimensional finite difference groundwater flow model, Techniques of 1974.
Water Resources Investigations of the United States Geological Survey, 57. Todd, D. K., Groundwater Hydrology, 2nd ed., John Wiley & Sons,
Book 6, pp. 586, 1988. NewYork, 1980.
References 463
m, 58. Trescott, P. C.. G. F. Pinder, and S. P Larson, Finite-difference model 62. Wang, H. F., and M. P. Anderson, Introduction to Groundwater
he for aquifer simulation in two dimensions wilh results of numerical exper- Modeling: Finite Difference and Finite Element Models, W. H. Freeman,
:s. iments, in US. Geological Survey Techniques of Water Resources Inves- San Francisco, CA, 1982.
tigations, Book 7, CJ, U.S. Geological Survey, Reston, VA, 1976. 63. Ward, D. S., D. R. Buss. J. W. Mercer, and S. S. Hughes, Evaluation
ter 59. Walton, W.C., Groundwater Resource EvaluaJion. McGraw-Hill, of a groundwater corrective action at the Chem-Dyne hazardous waste
JW New York, 1970. site using telescope mesh refinement modeling approach, Water
I.S. 60. Walton. W. C., Numerical Growtdwater Modeling: Flow and Conr- Resources Research, v. 23, no. 4, pp. 603-617, 1987.
67, amina/11 Migration, Lewis Publishets, Boca Raton, FL. 272 pp. 1989. 64. Yeh, W. W. G., Groundwater Systems, Ch. 16, in Water Resources
6I. Wanakule, N., Optimal Groundwater Management Models for the Handbook, L. W. Mays, ed., McGraw-Hill, New York, 19%.
1ell Barton Springs-Edwards Aquifer, Edwards Aquifer Research and Data
·87. Center, San Marcos, TX. 1989.
~on,
n of
and
face
pp.
inite
pure
:rvey
"low
989.
Ver-
V. the
odel,
>n, in
}, pp.
eart,
tech-
Jinois
sment
etquot
lupply
nu/ate
round- ·
8-729,
'rods in
71.
:iwater
1Wiley
ulation
in,TX,
11:. Sons,
Chapter 10
Management of Groundwater
t
7
8
Maximum development of groundwater resources for beneficial use involves planning in terms
of an entire groundwater basin. Because a basin is a large natural underground reservoir, it
follows that utilization of groundwater by one landowner affects the water supply of all other 2
landowners. Management objectives must be selected in order to develop and operate the basin. 3
These involve not only geologic and hydrologic considerations but also economic, legal, polit- 4.
ical, and financial aspects. Typically, optimum economic development of water resources in an 5.
area requires an integrated approach that coordinates the use of both surface water and ground- 6.
water resources. After evaluation of total water resources and preparation of alternative man- 7.
agement plans, action decisions can then be made by appropriate public bodies or agencies. 8.
9.
10.
10.1 CONCEPTS OF BASIN MANAGEMENT

The management of a groundwater basin implies a program of development and utilization of
subsurface water for some stated purpose, usually of a social or economic nature. In general,
the desired goal is to obtain the maximum quantity of water to meet predetermined quality
requirements at least cost.* Because a groundwater basin can be visualized as a large natural
underground reservoir, it follows that extraction of water by wells at one location influences
the quantity of water available at other locations within the basin.
10.1
Groundwater is extracted from the ground just as are other minerals, such as oil, gas, or
gold. Water typically carries a special constraint: It is regarded as a renewable natural resource.
Thus, when a water well is drilled, people presume that production of water will continue
indefinitely with time. In effect, this can occur only if there exists a balance between water
recharged to the basin from surface sources and water pumped from within the basin by wells.
Development of water supplies from groundwater begins typically with a few pumping
wells scattered over a basin. With time, more wells are drilled and the rate of extraction
increases. As wells become more numerous, development of the basin reaches and exceeds its
natural recharge capability. Continued development thereafter without a management plan
could eventually deplete the groundwater resource.
By regulating inflow to and outflow from the basin, an underground reservoir can be made
to function beneficially and indefinitely just as a surface water reservoir. The increasing demand
for water in the United States and throughout the world has produced the realization that the vast
underground reservoirs formed by aquifers constitute invaluable water storage facilities; proper
management of them, therefore, has become a matter of considerable interest. Some of the pros
and cons of subsurface and surface reservoirs are summarized in Table 10.1.1.
•As Bear and Levin 16 succinctly stated: "The basic idea is to regard the aquifer as a system which has to be operated
in an optimal manner."
464
10.1 Concepts of Basin Management 465
Table 10.1.1 Advantages and Disadvantages of Subsurface and Surface Reservoirs (after U.S. Bureau of Reclamation89)
Subsurface reservoirs Surface reservoirs

Advantages Disadvantages
1. Many large-capacity sites available 1. Few new sites available
2. Slight to no evaporation loss 2 High evaporation loss even in humid climate
3. Require little land area 3. Require large land area
4. Slight to no danger of catastrophic structural failure 4. Ever-present danger of catastrophic failure
5. Unifonn water temperature 5. Fluctuating water temperature
6. High biological purity 6. Easily contaminated
7. Safe from immediate radioactive fallout 7. Easily contaminated by radioactive material
8. Serve as conveyance systems-<:anals or pipelines across lands 8. Water must be conveyed
of others unnecessary
Disadvantages Advantages
ms
I. Water must be pumped 1. Water may be available by gravity flow
• it
2. Storage and conveyance use only 2. Multiple use
1er
3. Water may be mineralized 3. Water generally of relatively low mineral content
an.
4. Minor flood control value 4. Maximum flood control value
lit-
5. Limited flow at any point 5. Large flows
an
6. Power head usually not available 6. Power head available
nd-
7. Difficult and costly to investigate, evaluate, and manage 7. Relatively easy to evaluate, investigate, and manage
an-
8. Recharge opportunity usually dependent on surplus surface flows 8. Recharge dependent on annual precipitation
9. Recharge water may require expensive treatment 9. No treatment required of recharge water
10. Continuous expensive maintenance of recharge areas or wells 10. Little maintenance required of facilities
Forecasts of future water demand suggest that mismanagement-or lack of management-

1 of of major groundwater basins cannot be pennitted if adequate ongoing water supplies are to be
:rat, provided. The management objective consists of providing an economic and continuous water
Jity supply to meet a usually growing demand from an underground water resource of which only
ural a small portion is perennially renewable.
tees
10.1.1 Managing Groundwater Resources
;, or
.rce. The demand for groundwater in recent decades has led to development of a variety of strate-
inue gies for managing subsurface water and making efficient use of the available underground stor-
·ater age space. These may involve shlfting of local water sources, changing pumping patterns,
ells. limiting pumpage, artificial recharge, conjunctive use of groundwater and surface water, and
ping reuse of wastewater (Galloway34). The key concern is to maintain a sustainable long-tenn yield
:tion from aquifers (Alley et al.9). The consequences of overdraft in terms of adverse impacts on
ls its water quantity, quality, land subsidence, and water rights have been mentioned already. What
plan is often overlooked in large-scale groundwater pumping is the fact that surface water deple-
tions can occur far removed from the responsible wells. This situation, usually unbeknownst
nade to pumpers and unrecognized by water managers, has become commonplace in the United
nand States and poses an ongoing threat to our nation's rivers (Glennon36).
. vast It is important to recognize on a regional basis that groundwater and surface water con-
uper stitute a single unified source. What is surface water today can be groundwater tomorrow and
pros vice versa. It follows that a depletion of one can affect the other. Failure to realize this basic
concept regionally has served as the cause of several interstate water rights litigations.
Illustrative is the basin of the Republican River, draining from Colorado to Nebraska and into
erated Kansas. The area, centered in the com and wheat belt of the Midwest, derives almost all of its
water for irrigation from the extensive and underlying High Plains Aquifer. Since World War
466 Chapter 10 Management of Groundwater
11, the availability of deep-well turbine pumps, and later the innovation of center-pivot irriga-
tion systems, stimulated groundwater production. In the drier western portion of the basin,
water levels declined substantially, thereby reducing base flow in all tributaries. At the same
time, conservation practices such as terracing of fields and tail-water ponds reduced summer
thunderstorm runoff. The cumulative effect has been to deplete a majority of the river flow
reaching Kansas. Thus, although all water was used beneficially, a maldistribution has resulted
in the absence of a management approach that considered regional needs.
10.2 GROUNDWATER BASIN INVESTIGATIONS AND DATA COLLECTION [

Ideally, before groundwater is developed in a basin, an investigation of the underground water
resources should be made. In practice this rarely occurs; instead, a study is usually initiated
either after extensive development with a view toward further development or after overde-
velopment when a problem threatening the water supply appears imminent. Investigations are
seldom concerned with simply locating groundwater supplies. More commonly the concerns
involve evaluating the quantity and quality of groundwater resources already known to exist or
determining the impact of human plans or activities on the quantity and quality of groundwa-
ter. Figure 10.2.1 illustrates the sequence of activities preceding the start of a groundwater
management investigation.
Groundwater management studies are usually undertaken by local government agencies. Four
levels of study are generally recognized, although not all are required. 10 In brief, these include:
1. Preliminary Examination-Based largely on judgment by experienced personnel,
this study identifies the management possibilities of meeting a defined need for a
Fig
specified area.
Identify
problem
Lowering water levels,

- Formulate alternative
management objectives
and select objective
Water seNice at least cost,

f-+ Define elements of
management plan
Pumping locations and

f-+
Obtain authorization and
financing of water
resources management
investigation
f-+ Develop goals for
Investigation
Prepare report discussing

f-
waterlogging, subsidence, basin versus region, social, schedule, artificial recharge alternate management plans,
seawater Intrusion legal, and economic con- locations and schedule, Including nonstructural
through offshore out- straints, aquifer management boundary conditions, legal alternatives, and recommend
crops and from excava- as part of the total water structure, monHoring, organ- a single plan. It is recognized
tion of harbors and canals, supply and disposal system izational, and financial that Implementation of a
water quality degradation suitable plan is ultimate goal
as a result of improperly
constructed and destroyed
wells, percolation of waste-
waters, Increasing salinity ~
from adverse salt balance,
and other water quality
Select scope of Develop work program Create planning Commence
problems
investigation r+ for investigation 1-+ organization
4
Investigation
Deterrmlne study area, Identify what must be Engineers, economists,

study schedule, and level accomplished to achieve geologists, attomeys, and
of study-preliminary planning goals, with emphasis others
examination, reconnaiss- on need tor historic basic
ance, or feasibility data, continued data col-
lection, and EDP techniques. 10.2
Include diagrams showing
work elements, sequence,
and schedule
Figure 10.2.1. Sequence of activities preceding start of a groundwater management investigation (after Amer. Soc. Civil Engrs. 10).
10.2 Groundwater Basin Investigations and Data Collection 467
a- Projecting future Surface and subsurface Assessment of Capability or extraction

n, water demands exploration water resources and recharge facilities
ne
ter Present and future applied water Geophysical techniques and Yields of local water resources Inventory of existing extraction
and consumptive use demands test wells and availability of imported and recharge facilities and cap-
lW
for irrigation, domestic, and ind- supplies ability of modifying existing system
ed ustrial uses
Aquifer boundary Legal and organizational Alternate plans for water Report on the
conditions considerations resources management investigation
tter
ted Quantity and quality of subsurface Water rights, rights to use of Formulate plans considering Report presenting study results,
inflow and outflow storage, damages due to ground- physical and economic aspects with emphasis on alternative
dewater level lowering and water- plans
are logging, legal Implications of water Physical factors: patterns and
quality degradation, aulhority schedules of recharge and extrac- Most economical physically
~rns required by management agency, tioos, use of imported water, under- feasible plan is desired
;tor and types of agency required ground storage for infiltration of
stomnflows, control of groundwater
wa- levels lo avoid severe lowering
ater and waterlogging, water quality,
seawater Intrusion controls,
subsidence, and integration into
four total water resources system
e: Economic factors: costs and

benefits, financial feasibility
mel,
for a Figure 10.2.2. Sequence of activities during a feasibility investigation for groundwater management (after Amer. Soc. Civil Engrs. 1 ~.
2. Reconnaissance-This study considers possible alternatives in the formulation of a

water management plan to meet a defmed need for an area, including estimates of
benefits and costs. The investigation draws on available data and generally necessi-
tates a minimum of new data collection.
~g
llans, 3. Feasibility-This study requires detailed engineering, hydrogeologic, and economic

analyses together with cost and benefit estimates to ensure that the selected project
,end
lized is an optimum development. The sequence of activities normally involved in a feas-
a ibility investigation is outlined in Figure 10.2.2. Typically, the investigation concludes
goal
with a report recommending approval and funding for the project.
4. Define Project-This investigation involves planning studies necessary for defining
specific features of the selected project. The completed report forms the basis for
J starting final design and preparation of plans and specifications.
The types of data and the tasks involved in the physical portion of a reconnaissance or fea-
sibility study for groundwater management are outlined in the sections below. Table 1.8.1 pres-
ents the principal types of data and data compilations.
10.2.1 Topographic Data

Contour maps, aerial photographs, and benchmarks related to a leveling network are basic
requirements. They are directly applicable for locating and identifying wells, measuring
1grS.10)· groundwater levels, conducting crop and land use surveys, and plotting areal data.
10.2.2 Geologic Data

Surface and subsurface geologic mapping provides the framework for the occurrence and
movement of groundwater and hence is essential for feasibility studies. Subsurface informa-
tion is gained from a drilling program, including classification and analysis of well logs, and
geophysical surveys (see Chapter 12). As part of the drilling program, pumping tests of wells
are conducted to evaluate storage coefficients and transmissivities of aquifers, while samples
of groundwater are collected and analyzed for quality. From interpretation of subsurface geo-
logic data, principal aquifers and their extent are mapped together with regions of confined and
unconfined groundwater. Location of faults, dikes, and other structures that may significantly
affect groundwater is also a part of the geologic program.
10.2.3 Hydrologic Data

The principal purpose of hydrologic data collection is to evaluate the equation of hydrologic
equilibrium. The following outline summarizes types of basic data required and methods of
their analysis:
Surface Inflow and Outflow; Imported and Exported Water. These quantities are measurable
ll
by standard hydrographic and hydraulic procedures. Where complete data on surface flows to and
from..the basin are not available, supplemental stream gauging stations should be installed.
Precipitation. Records of precipitation in the area should be assembled. Gauges should be well
distributed over the basin to provide a good estimate of the annual precipitation from the isohyetal
or Thiessen methods. If gauges are not so located, supplemental stations should be established.
Consumptive Use. All water, surface and subsurface, released into the atmosphere by processes of
evaporation and transpiration is consumptive use, or evapotranspiration. To compute this discharge
from a given basin, it is first necessary to make a land use, or cultural, survey to yield the amount of
each type of water-consuming area. Aerial photographs are helpful for this task. Unit values of con-
sumptive use must then be determined. For crops and native vegetation, methods based on available
heat (such as the Thornthwaite or Blaney-Criddle method) are generally satisfactory. For water
surfaces, local evaporation records should be employed. Urban and industrial areas require careful
estimates from samples of representative areas using metered deliveries and sewage outflows. Mul-
tiplying the unit value of consumptive use by the corresponding acreage gives the water consump-
tion for each area; the sum of these products yields the total consumptive use over the basin.
Changes in Surface Storage. These can be computed directly from changes in water levels of sur-
face reservoirs and lakes.
Changes in Soil Moisture. The moisture content of the soil can be measured by devices embed-
ded in the soil or by a neutron probe (see Chapter 2). In practice, however, the variability of soil
moisture both in time and place makes it difficult to obtain an accurate basin-wide measurement.
The problem can be minimized by selecting periods of storage change in which the amount of
water in unsaturated storage at the beginning and end of the period is nearly equal. In irrigated
areas period limits should correspond to the beginning or end of the irrigation season.
Changes in Groundwater Storage. Changes in groundwater storage can be determined from
geologic data on aquifers and measurement of groundwater levels. Antecedent information on
groundwater levels, pumping records, pumping tests, and artificial recharge should be collected.
Specific yields of unconfined aquifers are determined by laboratory tests of samples and/or by
classifications of well logs; storage coefficients are best determined from pumping tests of wells.
Select a grid of measuring wells distributed over the basin. Supplement with test holes where
required. Water levels in these wells should be measured under conditions as nearly static as pos-
sible, preferably after the season of heavy draft and again after the season of recharge. A few
10.3 Yield 469
control wells should be equipped with automatic water-level recorders or have their water levels
measured monthly to facilitate detailed study of groundwater fluctuations. A basin map showing
lines of equal change in groundwater level is then prepared. The product of change in water level
times storage coefficient times area gives the change of groundwater storage for each aquifer
d within the basin .
.S
Subsurface Inflow and Outflow. These items of the equation are the most difficult to evaluate
!S
because they cannot be directly measured. Often one of them, or the difference, is calculated by
)-
being the only unknown in the equation. From geologic investigation it may be found that either
td
subsurface inflow or outflow is lacking, or both. Many times after study, subsurface inflow may be
ly
estimated to equal that of subsurface outflow so that the items cancel.
Difficulties arise in the case of underground flows from one basin to another. The direction of flow
can be established from water table or piezometric gradients. Knowing groundwater slopes and
transmissivities, subsurface flows can be computed from Darcy's law. Where surface streams and
~ic
subsurface drainage systems control groundwater levels, better estimates of subsurface flow are
of usually possible because more data are available.
.ble 10.3 YIELD

and
10.3.1 Alternative Basin Yields
11ell The maximum quantity of water that is actually available from a groundwater basin on a
etal perennial basis is limited by the possible deleterious side effects that can be caused by pump-
d. ing and by the operation of the basin. As a result, several concepts of basin yield are generally
:s of recognized. 10 These are briefly defined in the following subsections together with conunents
arge as to their consequences.
ntof
con- Mining Yuld
lable
Nater If groundwater is withdrawn at a rate exceeding the recharge, a mining yield exists.29 As a con-
treful sequence, this yield must be limited in time until the aquifer storage is depleted. Many ground-
Mul- water basins today are being mined; if mining continues, the local economy served by this
ump- pumping may change, evolving into other fonns that use less water or involve importations of
water into the basin. The Salt River Valley of southern Arizona and the High Plains of western
Texas are classic examples of such situations.
1f sur-
Various valid arguments, economic and other, have been advanced to justify mining of
groundwater. One is that water in storage is of no value unless it is used.75 In arid areas, such as
nbed- · the Sahara Desert, where groundwater represents the only available water resource, almost any
'f soil· development of groundwater constitutes a mining yield. But the needs are there and the benefits
~ment are great, and so such exploitation will continue. With proper management plus water conserva-
.unt of tion, such groundwater resources can be made to last from several decades to a few centuries.
igated
Perennial Yuld
~from
ion on The perennial yield of a groundwater basin defines the rate at which water can be withdrawn
Uected. perennially under specified operating conditions without producing an undesired result.* An
dlor by
f wells.
*In the past the term safe yield, implying a fixed quantity of extractable water basically limited to the average
; where annual basin recharge, has been widely used. The term has now fallen into disfavor because a never-changing
as pos- quantity of available water depending solely on natural water sources and a specified configuration of wells is
. A few essentially meaningless from a hydrologic standpoint.
undesired result is an adverse situation, such as (1) progressive reduction of the water resource,
(2) development of uneconomic pumping conditions, (3) degradation of groundwater quality,
(4) interference with prior water rights, or (5) land subsidence caused by lowered groundwater
levels. 56• 105 (Evaluation of perennial yield is discussed in a subsequent section.) Any draft in
excess of perennial yield is referred to as overdraft. Existence of overdraft implies that con-
tinuation of present water management practices will result in significant negative impacts on
environmental, social, and/or economic conditions.
A schematic diagram of a groundwater basin developed to less than perennial yield is
shown in Figure 10.3.1a. Here a portion of the natural recharge is lost by subsurface outflow
from the basin. But Figure 10.3.lb suggests a minimum perennial yield situation in which
extractions balance recharge so that no groundwater is lost.
Deferred Perennial Ywld

The concept of a deferred perennial yield consists of two different pumping rates. The initial
rate is larger and exceeds the perennial yield, thereby reducing the groundwater level. This
planned overdraft furnishes water from storage at low cost and without creating any undesirable
effects. In fact, reducing storage eliminates wasteful subsurface outflow of groundwater and
losses to the atmosphere by evapotranspiration from high wate( table areas. After the ground-
water level has been lowered to a predetennined depth, a second rate, comparable to that of
Recharge
Extraction 10.
Ground surt•~ac~e~~-----tTtT-----::;7
(a) (b)
Extraction Recharge
Legend
~ Storage capacity
~ Storage needed to
llllliM1Iil regulate recharge
j.-.:----.--1 Water in storage
(c)
Figure 10.3.1. Schematic diagram showing storage relations in a groundwater basin for three stages of development (a) Less than
perennial yield (b) Minimum perennial yield (c) Increased perennial yield (after Peters66)
I0.3 Yield 471
I
perennial yield, is established so that a balance of water entering and leaving the basin is main-
' tained thereafter. With a larger available storage volume, more water can be recharged and a
r
larger perennial yield can be obtained. Figure 10.3.lc indicates this situation schematically.
n
1-
n Maximum PerenniAl Yield

The maximum perennial yield, as the name suggests, means the maximum quantity of ground-
IS
water perennially available if all possible methods and sources are developed for recharging
w the basin. In effect, this quantity depends on the amount of water economically, legally, and
:h politically available to the organization or agency managing the basin. Clearly, the more water
that can be recharged both naturally and artificially to a basin, the greater the yield.
To achieve the maximum perennial yield, the aquifer should be managed as a unit Thus,
efficient and economic production of water requires that all pumping, importations, and dis-
al tributions of water be done for the benefit of the largest manageable system. Where surface
liS water is available in addition to groundwater, these two sources are operated conjunctively.
ole Such a conjunctive use scheme provides a larger and more economic yield of water than can
ild be obtained from the two sources operated independently. The limit to such an operation is
d- governed by the ability to import and distribute water and also by the storage available for sur-
of face water and groundwater.
10.3.2 Evaluation of Perennial Yield

Consideration of the above definitions of perennial yield reveals that there can be more than
one "undesired result" from pumping a groundwater basin, that perennial yield may be limited
to an amount less than the net amount of water supplied to the basin, and that perennial yield
can vary with different patterns of recharge, development, and use of water in a basin.
If groundwater is regarded as a renewable natural resource, then only a certain quantity of
water may be withdrawn annually from a groundwater basin. The maximum quantity of water
that can be extracted from an underground reservoir, and still maintain that supply unim-
paired, depends on the perennial yield. Overdraft areas constitute the largest potential ground-
water problem in the United States. 88 Until overdrafts are reduced to perennial yields in these
basins, permanent damage or depletion of groundwater supplies must be anticipated.
Factors Governing Perennial Yield

Determination of the perennial yield of a groundwater basin requires analysis of the undesired
results that may accrue if the extraction rate is exceeded. The recharge• criterion (progressive
reduction of the water resource) is the most important because exceeding this factor is normally
responsible for introducing other undesired results. Water supplied to a basin may be limited
either by the storage volume of the underground basin or by the rate of water movement through
the basin from the recharge area to the withdrawal area. The quantity concept is usually applica-
ble to unconfined aquifers where supply and disposal areas are close together, whereas the rate
concept applies more to confined aquifers where supply and disposal areas are widely separated.
Economic considerations can govern perennial yield in basins where the cost of pumping
groundwater becomes excessive. Excessive costs may be associated with lowered ground-
water levels, necessitating deepening wells, lowering pump bowls, and installing larger pumps.
Where pumpage is largely for irrigation, power costs, crop prices, or government farm subsi-
dies may establish an economic limit for pumping groundwater; alternatively, other uses that
can support higher pumping costs may evolve.
han
'Here, recharge refers to water reaching the saturated zone of an aquifer, where it is available for extraction.
472 Chapter lO Management of Groundwater
Water quality can govern perennial yield if draft on a basin produces groundwater of infe-
rior quality. Possibilities include: (1) pumping in a coastal aquifer could induce seawater intru-
sion into the basin (see Chapter 14); (2) lowered groundwater levels could lead to pumping of
underlying connate brines; (3) polluted water from nearby areas might be drawn into a pumped
aquifer. A quality limitation on perennial yield depends on the minimum acceptable standard
of water quafity, which in tum depends on the use made of the pumped water. Therefore, by
lowering the quality requirement, the perennial yield can be increased.
Legal considerations affect perennial yield if pumpagc interferes with prior water rights.6 l
Finally, if pumpage is responsible for land subsidence, a (imitation on perennial yield can result.
Calculation of Perennial Yield

Fit
In general, the basin recharge criterion will govern perennial yield because, as mentioned SOt
earlier, one or more of the other undesired results will often be induced by pumpage exceed-
ing this rate. Quantitative determination of perennial yield where recharge is the limiting fac-
tor can be made under specified conditions if adequate knowledge of the hydrology of the
basin is available. Methods are based on the equation of hydrologic equilibrium or approxi-
mations thereto.s7 Basically, this implies that perennial yield is defined in terms of a rate at
which groundwater can be withdrawn from a basin over a representative time period without
producing a significant change in groundwater storage.
Variability of Perennial Yield

It is important to recognize that perennial yield of a groundwater basin tends to vary with time.
Any quantitative determination is based on specified conditions, either existing or assumed,
and any changes in these conditions will modify the perennial yield. This fact applies to the
degree and pattern of groundwater development within a basin as well as to the other factors
that govern safe yield.
Investigations of the availability of groundwater within a basin are typically not initiated
until basin development has produced an overdraft. Yet this is almost necessary in order to
obtain a reasonable estimate for perennial yield. In a virgin basin, where a balance exists 10.·
between natural inflow and outflow and there is no pumping, the absence of hydrogeologic
data may not justify the cost of a management investigation. Similarly, estimating future peren-
nial yield of a basin under greater development than at present requires careful evaluation of
all items in the equation of hydrologic equilibrium.
Perennial yield may vary with the level of groundwater within a basin. Thus, if levels are
lowered, subsurface inflow will be increased and subsurface outflow will be decreased,
recharge from losing streams will be increased and discharge from gaining streams will be
decreased, and uneconomic evapotranspiration losses will be reduced. Conversely, a rise in
water levels will have the opposite effects. Therefore, where recharge is sufficient, the greater
the utilization of underground water, the larger the perennial yield. The maximum perennial
yield will be controlled by economic or legal constraints.
An unconfined basin fed by an adequate recharge source can increase its perennial yield,
not only by increasing pumpage but also by rearrangement of the pumping pattern. If the con-
centration of wells is shifted to near the recharge source, greater inflow can be induced. The
rearrangement has the additional advantage that a greater supply may be obtained without nec-
essarily increasing pumping lifts. For example, in the cross section shown in Figure 10.3.2a,
it is assumed that the stream is the principal recharge source. By moving the well field nearer
to the stream as in Figure 10.3.2b, the water table slope is increased and a greater yield for
equal pumping depths results.
For a confined aquifer with its recharge area located some distance from the pumping
area, the rate of flow through the aquifer will govern the perennial yield. In large confined
:e- Old well field

u-
of
•ed
l.fd Waterlable
by Unconfined
aquifer
s.6'
u!L Impermeable Impermeable
(a) (b)
Figure 10.3.2. Example of increased groundwater yield for same pumping depths obtained by shifting wells nearer to a recharge
ned source.
~ed
fac- aquifers, pumpage of water from storage can be carried on for many years without establish-
the ing an equilibrium with basin recharge. Although the slope of the piezometric surface will
axi- increase, the penneability of the aquifer is seldom sufficient to maintain a compensating flow
te at into the basin. 88
hout Besides operational changes, perennial yield can also vary due to gradual and subtle mod-
ifications occurring within a basin. Changes in vegetation and even in crops, particularly where
root depth is affected, may influence surface infiltration and subsequent percolation to the water
table. Urbanization of an area, accompanied by greater surface runoff and installation of sewer
time. systems, can be expected to reduce recharge. Changes in the purpose of pumping groundwater,
med, such as from irrigation to municipal or industrial use, may-from an economic viewpoint-per-
.o the mit greater pumping lifts; consequently, perennial yield can be increased. Other economic fac-
tctors tors include changes in value of irrigated crops, increased efficiency of new wells and pumps,
treatment to meet revised water quality standards, and power costs.
tiated
:ler to 10.4 CONJUNCTIVE USE AND WATERSHED MANAGEMENT
exists
::>logic In basins approaching full development of water resources, optimal beneficial use can be
peren- obtained by conjunctive use, which involves the coordinated and planned operation of both
ion of surface water and groundwater resources to meet water requirements in a manner whereby
water is conserved. • The basic difference between the usual surface water development with
els are its associated groundwater development and a conjunctive operation of surface water and
·eased, groundwater resources is that the separate flfill yields of the fonner can be replaced by the
.vill be larger and more economic joint yields of the latter.
rise in The concept of conjunctive use of surface water and groundwater is predicated on surface
greater reservoirs impounding streamflow, which is then transferred at an optimum rate to ground-
rennial water storage. Surface storage in reservoirs behind dams supplies most annual water require-
ments, while the groundwater storage can be retained primarily for cyclic storage to cover
1! yield, years of subnonnal precipitation. Thus, groundwater levels would fluctuate, being lowered
he con- during a cycle of dry years and being raised during an ensuing wet period. Figure 10.4.1
ed. The depicts how groundwater levels might vary under such a system of conjunctive use.
)Ut nec- During periods of above-normal precipitation, surface water is utilized to the maximum
10.3.2a, extent possible and also artificially recharged into the ground to augment groundwater storage
.:1 nearer and raise groundwater levels (see Chapter 13). Conversely, during drought periods, limited
1ield for
*coordinated use of surface water and groundwater does not preclude importing water, as required, to meet growing
Jumping needs. In fact, to store and distribute additional water economical!y may require more intensive use of groundwater
~onfined storage space.
474 Chapter I0 Management of Groundwater
0 0 5 10 15 20
Time (years} Time (years)
Figure 10.4.1. Illustrative example of variation in groundwater levels in relation to annual precipitation under conjunctive use
management.
surface water resources are supplemented by pumping groundwater, thereby lowering ground-
water levels. The feasibility of the conjunctive-use approach depends on operating a ground-
water basin over a range of water levels; that is, there must be space to store recharged water,
and, in addition, there must be water in storage for pumping when needed.
Management by conjunctive use requires physical facilities for water distribution, for
artificial recharge, and for pumping. The procedure does require careful planning to optimize
use of available surlace water and groundwater resources. Such operations can be complex and
highly technical; they require competent personnel, detailed knowledge of the hydrogeology
of the basin, records of pumping and recharge rates, and continually updated infonnation on
groundwater levels and quality. A schematic diagram of a systematic approach for a conjunc-
tive use analysis is illustrated in Figure 10.4.2.
A conjunctive use management study requires data on surface water resources, ground-
water resources, and geologic conditions; data on water distribution systems, water use, and
wastewater disposal are also necessary.45• 67 Figure 10.4.3 shows a simplified flowchart of the
various phases and steps involved for a basin management study in California. This suggests
the diversity of data and effort required in order to determine an optimal basin management
plan. It should be noted from Figures 10.4.2 and 10.4.3 that mathematical models are usually
incorporated in such studies (see Chapter 9). A basin model simulates the responses of a basin
to variations in variables such as natural and artificial recharge and pumping so that the best
operating procedures for basin management can be practiced. In effect, this will optimize the
water supply obtained from the basin. 28· 40· 72
Because every water development project is unique, it is impossible to generally present
economic considerations for conjunctive operations and have them apply specifically to any
given situation. Nevertheless, the advantages and disadvantages, mostly economic, are sum-
marized in Table 10.4.1. The tabulation compares a conjunctive use operation relative to devel-
opment of surface-water resources only, assuming irrigation to be the principat water use in a
semiarid region.
Total usable water supply can be increased by coordinated operation of surface and under-
ground water resources. With an optimum coordinated operation, the unit cost of water supply
storage and distribution can be minimized. The basic principles of groundwater basin opera-
tion that will produce an optimum water resources management scheme include the following,
as reported by Fowler: 32
L The surface and underground storage capacities must be integrated to obtain the most
economical utilization of the local storage resources and the optimum amount of
water conservation.
Identify nature of the problem
.~
Identify level of the problem I
Identify all physical, economic,

and legal variables
t
Determine the significant
elements of the system
Define the objective(s)

md-
md-
ner,
. for
nize
System dynamics;
mathematical modeling H Data I
and
logy
Model verification Criteria
non include social
unc- preferences
t
und-
.and r-1 I
Model use
h
f the
gests
I Decision making I Learning
I Figure 10.4.2. Schematic diagram of
ment
I I End I a systematic approach for studying
I
ually conjunctive use problems (after
)asin ----1 Optimal policy
I Maknoon and Burges,55 reprinted
from Journal American Water Works
. best
e the Association, Vol. 70, by permission of
Optimal policy Implementation I the Association; copyright© 1978 by
American Water Works Association,
esent I 6666 West Quincy Avenue, Denver,
HOY I End I co 80235).
sum-
level-
2. The surface distribution system must be integrated with the groundwater basin trans-
e in a
mission characteristics to provide the minimum cost distribution system.
tnder- 3. An operating agency must be available with adequate power to manage surface-water
upply resources, groundwater recharge sites, surface-water distribution facilities, and
,pera- groundwater extractions.
wing, The procedure for developing a sound conjunctive-use operation within a basin requires
estimation of the various elements of water supply and distribution. The optimum use of sur-
:most face-water and groundwater resources is detennined for assumed conditions, usually those
Jnt of during the most critical drought period of record. There are many article in the literature
addressing conjunctive management (particularly modeling), including: Basagaoglu and
Hydrologic phase
Survey of
Development Analyses of capacities
Hydrologic o'f primary primary and
analyses ~ pipeline r----+ pipeline locations
network 1\Jumber of
networks of existing facilities
models facilities and quantity
~
of water
t t Simulation of
supply of
each plan of
operation
Historical Future water Formulation of coordinated
__..... water supply,
use, and
supply
~ demand and
supply
~
alternative
plans of
operation
---+-
operation of
surface and
subsurface
facilities
!
Cost of
facilities,
~ ~ t energy, and
~
water supply
Analyses of of each plan
Development plans of of operation
Unit cost Cost
and test of operation on analyses of
Geologic mathematical r------+ groundwater comparison
analyses ~ facilities, of alternative
model of basin
groundwater energy, and plans
mathematical water supply
basin model
Geologic phase Operational-
economic phase
Figure 10.4.3. Flow diagram of a management study for the San Gabriel Valley, California, groundwater basin (after Amer. Soc.
Civil Engrs. 10).
Marifio; 15 Bredehoeft and Young; 19 Galloway et al.; 34 Haimes and Dreizen; 39 LaBolle, Ahmed,
and Fogg; 47 Reichard; 68 Willis and Yeh; 102 Winter et al.; 103 and Young and Bredehoeft. 106
Table 10.4.1 Conjunctive Use of Surface Water and Groundwater Resources

(after Clendenen24 )
Advantages Disadvantages
1. Greater water conservation 1. Less hydroelectric power H

2. Smaller surface storage 2. Greater power consumption
3. Smaller surface distribution system 3. Decreased pumping efficiency
4. Smaller drainage system 4. Greater water salination
5. Reduced canal lining 5. More complex project operation
6. Greater flood control 6. More difficult cost allocation
7. Ready integration with existing 7. Artificial recharge is required
development 8. Danger of land subsidence
8. Stage development facilitated
9. Smaller evapotranspiration losses
10. Greater control over outflow
11. Improvement of power load and pumping
plant use factors
12. Less danger from dam failure
13. Reduction in weed seed distribution
14. Better timing of water distribution
10.5 Groundwater Management: Water Laws and Policies 477
10.5 GROUNDWATER MANAGEMENT: WATER LAWS AND POLICIES

10.5.1 Water Law and Policy
Water-management decisions and policies should have water laws as the fundamental basis. In
the United States, water law has two basic functions: (1) creation of supplemental private
property rights in scarce resources and (2) imposition of public interest limitation on private
use. For our purposes water law is divided into surface water law and groundwater law.
Surface water law is further categorized into riparian law and appropriation law. Riparian law
is based on the riparian doctrine, which states that the right to use water is considered real
property, although the water its~lf is not properw of the landowner (Wehmhoefer97). Appro-
priation law states that the allocation of water rests on the proposition that the beneficial use
of water is the basis, measure and limit of the appropriative right: the first in time is prior in
right. See Tarlock83 for further details on concepts of water law.
In the western United States, surface water policy generally follows this doctrine of "fust
in time, first in right." In order to appropriate water, the user need only demonstrate availabil-
ity of water in the source of supply, show an intent to put the water to beneficial use, and give
priority to more senior permit holders during times of shortage (Schmandt et al.76). Beneficial
use of water under the law includes domestic consumption, livestock watering, irrigation,
mining, power generation, municipal use, and other circumstances. The states of Arizona and
New Mexico follow the appropriation law of surface water, and in California and Texas the
appropriation doctrine and the riparian doctrine coexist.
Groundwater allocation is handled quite differently and is typically divided into common
law or statutory law. Conunon law doctrines include the overlaying rights doctrines of absolute
ownership, reasonable use, and correlative rights. These doctrines give equal rights to aJl
>C.
landowners overlaying an aquifer. Arizona, California, and Texas have adopted these principles
for groundwater allocation.
The above surface and groundwater laws serve as the basis for individual state water
1IDed, policies. The burden of developing water policies lies upon the state. This is often achieved by
16
the state proposing a water project and securing federal funds for the construction. It is also up
to the states to agree on apportionment in interstate waters; if states cannot agree, then the
courts will intervene and settle the dispute by decree. The federal government gets involved in
such disputes only where interstate waters, federal lands, or Indian reservations are concerned.
10.5.2 Arizona's Groundwater Management Code

Throughout Arizona's history, water policy has been directed at supporting growth of its popu-
lation and major revenue-producing activities. Starting with mining, ranching, and farming, with
the gradual shift to municipal and industrial uses, the water policy of the state has been directed
at obtaining imported supplies. This bas been an effort to augment what has appeared to be an
insufficient and indigenous resource. Waterstone96 points out that the "state's water policies have
Jed to the protracted exercise to capture and secure the Central Arizona Project (CAP), the ongo-
ing infatuation with weather and watershed manipulation, the current experimentation with
groundwater recharge and effluent use, and the recent spate of purchases of remote water fanns."
In Arizona, the state's water policy and management focused more on surface water than
groundwater prior to 1980, when the Groundwater Management Code was developed; there-
after, the emphasis has been on groundwater. In regards to surface water, Arizona law defines
surface water as "the waters of all sources, flowing in streams, canyons, ravines or other natu-
ral channels, or in definite underground channels, whether perennial or intemtittent, flood,
waste, or surplus water, and of lakes, ponds and springs on the surface." These surface waters
are subject to the doctrine ofprior appropriation (ADWR 12). In Arizona, surface water rights
are obtained by filing an application with the Department of Water Resources for a permit to
appropriate surface water. Once the permit is issued and the water is actually put to beneficial
use, proof of that use is made to the Department and a certificate of water right is issued to the
applicant. Once a certificate is issued, the use of the water is subject to all prior appropriations.
Since water law in the state of Arizona has changed substantially over the years, Arizona
is now conducting a general adjudication of water rights in certain parts of the state. Adjudi-
cations are court determinations of the status of all state law rights to surface water and all
claims based upon federal law within the river systems. These adjudications will provide a
comprehensive way to identify and rank the rights to the use of water in some areas. The adju-
dications will also quantify the water rights of the federal government and the Indian reserva-
1
tions within Arizona.
In Arizona, groundwater problems arise from the overdrafting of water from the aquifers. 1
Groundwater overdrafts cause many problems, such as increased well pumping costs and water H
quality issues. In areas of severe groundwater depletion, the earth's surface may also subside, il
IT
causing cracks or fissures that can damage roads or building foundations. In order to manage
T
groundwater pumping in Arizona, the Arizona Groundwater Management Code was developed
a
in 1980 as state legislation. The Arizona Groundwater Management Code was named one of the m
nation's ten most innovative programs in state and local government by the Ford Foundation in (1
1986. This achievement came from the cooperation of Arizonans working together and com- ar
promising when necessary in order to protect the future of the state's water supply.
The Groundwater Management Code has three primary goals (ADWR 12): at
e~
1. control the severe overdraft currently occuuing in many parts of the state; Li
2. provide a means to allocate the state's limited groundwater resources to most effec- th
tively meet the changing needs of the state; and taJ
qu
3. augment Arizona's groundwater through water supply development.
tol
In order to achieve these goals, the code set up a comprehensive management environment and aq
established the Arizona Department of Water Resources. an
There are three levels of water management outlined by the code. Each level is based on inc
different groundwater conditions. The lowest level applies statewide, and includes general
the
groundwater provisions. The next level applies to Irrigation Non-Expansion Areas (INAs), and
Aq
the highest level applies to Active Management Areas (AMAs), where groundwater depletion
Co
is the highest. The boundaries that divide the INAs and AMAs are determined by groundwater Ed
basins and not by political jurisdiction. The main purpose of groundwater management is to An
determine who may pump groundwater and how much may be pumped. This includes identi- ob1
fying existing water rights and providing new ways for nonirrigation water users to initiate new om
withdrawals. In an AMA or INA, new irrigation users are not allowed. Even with the original wa·
publicity and enthusiasm, many people now feel that the efforts under the Groundwater Man- Wal
agement Code have been very costly with very little savings in water, rrtaking the success ques- ton
tionable. For more information, refer to the Arizona Department of Water Resources Web site rna:
at http://www.water.az.gov/adwr. dev
ing
tori
10.5.3 Texas Groundwater Law reg1
of g
As a comparison to Arizona's water law, it is worthwhile to look at Texas. A large amount
of Texas' water supply is from groundwater resources, which has resulted in a severe depletion regi
of some of the major aquifers in the state. It is surprising that a state such as Texas that depends drm
on groundwater so much has such an ill-defined groundwater plan. The Texas groundwater law Cor
is based on the "absolute ownership rule," which states, "percolating waters are the private caw
property of the landowner" (Schmandt et al. 76). This means that a landowner has the right to Wat
pump water from a groundwater deposit beneath his land at any rate, as long as the withdrawal affe,
10.6 Groundwater Management: Examples 479
does not maliciously hann his neighbor. The groundwater law excludes underground streams,
al
where the classification of an underground stream is still a matter of debate. This lack of reg-
he
lS.
ulation has caused many problems in the state of Texas, such as overdrafting, subsidence, and
water quality issues. The Texas Water Development Board has conducted efforts to develop the
na
best management techniques for regional groundwater planning.
ii-
al1
~a 10.6 CASE STUDY Groundwater Management: Examples
JU-
va- 10.6.1 Edwards Aquifer Ma11agement recharging, as well as prevent waste and pollution of the aquifer.
However, management efforts failed as bitter conflicts sparked
The Edwards Aquifer of Texas extends in a narrow band from
~rs. between the various users of the aquifer. In 1993, for fear of fed-
Hays to Kinney Counties in the south-central part of Texas, as
tier eral intervention, !he Texas legislature enacted a management plan
illustrated in Figures 2.6.2 and 9.4.2a. The aquifer is approxi-
ide, for the aquifer. The plan abolished the district and created the
mately 175 miles in length and ranges in width from 5 to 30 miles.
age Edwards Aquifer Authority. It called for the Authority to imple-
The aquifer has a shallow outcrop area where recharge occurs, and
ped ment a comprehensive plan for the aquifer that regulated pumpage
a deeper (800-1 ,200 feet) confined or artesian area through which
·the while taking into consideration the needs of all entities involved
most of the flow moves in an east-northeasterly direction
that rely on the aquifer. It was also important for the plan to ensure
n in (TWBD87). Figure 9.4.2a illustrates the outcrop area (recharge
a successful balance to maintain the delicate relationship between
om- area) in the aquifer.
spring flows and the environment. In essence, the Authority was
This aquifer is highly transmissive and accepts an average of
created to regulate the withdrawal of groundwater from the
about 642,000 acre-feet of recharge annually, and discharges an
aquifer and ensure !hat the spring flows do not drop below the
equivalent amount of water through various springs and wells.
jeopardy level. In order to achieve this, the Edwards Aquifer
Large fissures and cracks decrease the amount of filtration Authority changed the rules from "rule of capture" to a system of
ffec- through the aquifer, which provides a continuous threat of con-
permits to be awarded and managed by the Authority. The objec-
tamination of the aquifer through the recharge process. The water
tive was to maintain a sufficient level of spring flows at Comal and
quality in the aquifer generally contains less than 500 mg/L of
San Marcos Springs for the purpose of protecting endangered
total dissolved solids, with salinity increasing with depth. This
species.
tand aquifer is the source of 246,000 acre-feet for municipal supplies Some important management issues for the Edwards Aquifer
annually, 79,000 acre-feet of irrigation water, 29,000 acre-feet of include the following:
industrial water, and 13,000 acre-feet of water for other purposes.
~don
The Edwards Aquifer is managed by three different entities: 1. Establishing a level of groundwater withdrawals to ensure
:neral
the Edwards Aquifer Authority, the Barton Springs/Edwards adequate water levels and at least minimum spring flows;
), and Aquifer Conservation District, and the Texas Natural Resource
(etion 2. Developing a plan for the acceptable and equitable distribu-
Conservation Commission. The principal uses of water from the tion of pumping among the numerous users involved;
water Edwards Aquifer are for municipal and irrigation purposes. San
tis to 3. Securing additional supplies to meet projected demands in
Antonio, Texas, with a population of over 1.5 million people,
excess of the allowable pumping limits.
denti- obtains its entire municipal water supply from the aquifer and is
:enew one of the largest cities in the world that relies solely on ground-
water for its supply. The aquifer also supplies industrial users with 10.6.2 High Plains Aquifer: Conjunctive Water Use on the
i.ginal
water in both the San Antonio and Austin, Texas regions. The Bar- High Plains
·Man-
; ques- ton Springs/Edwards Aquifer Conservation District is the aquifer The High Plains Aquifer (see Section 2.9) is a 174,000-square-
management entity for the Austin region. A comprehensive plan. mile-area of flat to gently rolling terrain including parts of Col-
eb site
developed by the district, outlines programs and activities cover- orado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota,
ing areas of concern, such as water level and water quality moni- Texas, and Wyoming (see Figures 2.9.2, 2.9.3, and 2.9.5). The
toring, well construction and spacing standards, production High Plains have moderate precipitation with a low natural
regulations, and public education in conservation and protection recharge rate to the groundwater system. The region is underlain
of groundwater. by the High Plains Aquifer, a water table aquifer of unconsoli-
tmount The management of the Edwards Aquifer in the San Antonio dated alluvial deposits that form the water table aquifer. The High
pletion region bas been under controversy for many years. After a severe Plains Aquifer consists largely of the Ogallala Aquifer, which cov-
epends drought from 1950 to 1957, the water levels lowered enough that ers more than 35,000 square miles of the High Plains.
tter law Comal Springs located in Comal County went dry. This event Irrigation water pumped from the aquifer has made the High
private caused the Texas legislature to create the Edwards Underground Plains one of the United States' most important agricultural areas.
right to Water District. The district, including five counties that are
1drawal affected by the aquifer, was to conserve, protect, and monitor (continues)
· }i0:6 cASE S'fUPY Groundwater Management: Examples(continued)
In 1949, the groundwater pumpage for trrigation was approxi- tection Agency and the Texas Department of Health primary stan-
mately 480 million cubic feet per day. By 1980 the pumpage had dards for fluoride and nitrate.
reached approximately 2,150 million cubic feet per day (U.S. More water is pumped from the Ogallala Aquifer in Texas than
Geological Survel 1). Because of major water level declines, by from any other aquifer in the region. The total pumpage from the
1990 the pumpage declined to about 1,870 million cubic feet per aquifer in 1994 was approximately 5.9 million acre feet, and 96%
1.
day (McGuire and Sharpe62). According to Alley et al.,9 pumping of the water pumped was used for irrigation. In addition to irriga-
from this aquifer has resulted in the largest decrease in storage of tion, many communities use the Ogallala Aquifer as their only
any major aquifer in the United States. Much of the predevelop- source of drinking water. Recha~ge rates cannot keep up with the 2.
ment saturated thickness has been dewatered (refer to Figure large amounts of withdrawal. The combination of these activities
2.9.5). The water table continues to decline in much of the High has significantly reduced the water level in the aquifer. The water
PlainsAquifer; however, monitoring of water levels in wells indi- level is so low that the cost of pumping has become increasingly
cates an overall reduced rate of decline (McGuire and Sharpe62). high. The previous lack of water management in the region has 3.
The decline is the result of improved irrigation and cultivation added to this problem of groundwater depletion.
practices, decreases in irrigated acreage, and above-normal In order to beuer manage the aquifer, the High Plains Ogallala
precipitation. Area Water Management Plan (HPOAWMP) was developed.
The Ogallala Aquifer is primarily composed of sand, gravel, From 1990 to 2005, the HPOA\VMP will conduct a study of the
clay, and silt deposits. The groundwater moves slowly through aquifer and will reflect the analyses and provide management rec-
the aquifer in a southeastward direction. The water quality is gen- ommendations for the 47 -county area in multiple states. The study oftl
erally fresh; however, both dissolved solids and chloride concen- consists of individual water assessments by each munici~alit)r that essa
dro~
trations in the aquifer increase from north to south. These examine the present and potential water needs. It will also include
ter r
dissolved solids' concentrations range from about 300 to more water conservation measures, reuse, new project development,
nom
than 3,000 mgfL. In some areas the aquifer contains fluoride and and improved system operations.
nitrate concentrations which ex:ceed the U.S. Environmental Pro- (continues)
Figure 10.6.1. Playa Lakes on the

Texas High Plains
Playa lakes like these can be found lO.i
throughout the High Plains between
Lubbock and Amarillo. Studies suggest 10.7
that playa lakebeds can be modified to
increase recharge to the Ogallala Aquifer.
The Texas High Plains are covered with
more than 17,000 shallow, naturally
occurring depressions called playa lakes.
The lakes capture an average of 1.8 to
5.7 million AF of water annually. The
characteristics of individual playas vary
tremendously. Some cover only a small
area while others can be as large as 200
acres. Most of the playas are 2 to 10 feet
deep when full of water. There are an
average of one to two playas per square
mile in the Southern High Plains.
In the spring and fall, intense thun-
derstorms frequently fill the playas.
However, little infiltration occurs, 50% to
80% of the water is lost to evaporation
and winds, and minimal groundwater
recharge occurs. Almost all the lakes
have clay bottoms (photo courtesy of
Lloyd Urban, Tex.as Tech University).
l 0.6 CASE STIJDY Groundwater Management: Examples(continued)
an- According to the Ogallala plan, for the sake of consistency, and evaluating new management strategies. The interconnections
examination of the full range of alternatives, and taking account of between surface water and groundwater and the relationship to
un individual area differences, it is a joint goal of the regional man- watersheds are much less direct and obvious in the High Plains
the agement plan to: Aquifer as compared to the Edwards Aquifer in the San Ar1tonio
6% region. Texas law is clear concerning the use and ownership of
1. have some commonality in the range of various management
tga- percolating groundwater under the absolute ownership doctrine as
tools examined for each user class and geographical area;
mly established in a 1904lawsuit (Houston and TC. Ry. Co. v. East).
2. identify some common regional management proposals Landowners can pump and use groundwater beneath their prop-
the
(such as conservation plans) and best-use techniques that erty with little regulation. According to Templer and Urban, 84 con-
ities
would be appropriate for all user classes and areas to imple- junctive water use is widely practiced concerning water from (I)
·ater
ment; and groundwater in the dwindling Ogallala Aquifer, (2) runoff water
flgly
has 3. consider each area's uniqueness in selecting an appropriate accumulating in the thousands of small playa lakes that dot the
array of these tools for developing a customized plan for each High Plains (see Figure 10.6.!), and (3) the meager flow of
Hala management area or municipality that is consistent with the streams in the area. Water from the playa lakes is used mainly to
·ped. planning methodologies and goals used for the overall region. supplement irrigation water pumped from the Ogallala.
f the Further infonuation on the High Plains aquifer is found in
In order to execute the plan, a management team is in charge Alley and Schefter,8 Grubb,38 Longley and Jordan,51 and Smerdon
t rec-
;tudy
of the above tasks and identifies any special studies that are nec-
essary. These studies may include identifying and evaluating
°
and Jordan. 8 For more general discussions of water development
y that in the southwestern United States, refer to Baker et a!., 13 Bow-
drought management aspects, developing appropriate groundwa- den,18 Espeland,31 Fradkin, 33 Ingram,44 Killgore and Eaton,46
elude
ter models, addressing water quality issues, performing an eco- Reisner,69 Shupe and Folk-Williams,77 and Smith. 81
ment,
nomic analysis of water management alternatives, and developing
'nues)
md 10.7 GROUNDWATER MANAGEMENT USING MODELS

n
10.7.1 What Are Groundwater Management Models?
;gest
d to Aquifer simulation models (discussed in Chapter 9) have been used to examine the effects of
.quifer. various groundwater management strategies. Use has been primarily of the "case study" or
. with "what if' type. The analyst specifies certain quantities and the model predicts the technical and
y perhaps economic consequences of this choice. The analyst evaluates these consequences and
•lakes. uses judgment and intuition to specify the next case. Groundwater management models may
8 to
be used for optimization or as combined simulation-optimization. Reviews of groundwater
The
; vary management models and applications can be found in Ahfeld and Heidari,4 Ahfeld and
small. Mulligan,5 Gorelick,37 Wagner,92 Willis and Yeh, 102 and Yeh. 104
IS 200 Yeh104 reviewed inverse solution procedures for parameter identification, groundwater-
10 feet management models, and optimal experimental design methodologies. Optimization methods
~an have been developed extensively over the past three decades for inverse solution procedures,
;quare but used only sparingly in applications. Groundwater management applications have focused
on explicitly combining simulation and optimization, resulting in so-called simulation
thun- management models. Groundwater management models can be developed and used very effec-
lS. tively for groundwater management purposes.
, 50% to
An example of a groundwater management model is one that is used to identify the least-
ration
cost management strategy to meet specified hydraulic and water quality restrictions in an
vater
~es
aquifer. Typically, total pumping is used as a surrogate for total cost. The restrictions dictate
;y of the allowable heads, gradients, velocities, and concentrations through space and time. The
·sity). management strategy defines the well locations and rates to minimize pumping.
482 Chapter l 0 Management of Groundwater
10.7.2 Optimization Methods

Optimization procedures use mathematical expressions to describe a system and its response
to the system inputs for various parameters. Constraints are used to define the limits of the
design variables, and performance is evaluated through the use of objective functions. An
optimization problem may be formulated in a general framework as
Optimize .f{x) (10.7.1)
subject to the constraints
G(x) = 0 (10.7.2)
and bound constraints on the decision variables

(10.7.3)
where x1 and x11 are the vectors of lower and upper bounds, respectively, on the decision vari-
ables x. A set of values of the decision variables that simultaneously satisfies the constraints
is a feasible solution. A feasible region is the region of feasible solutions defined by the
constraints. An optimal solution is a set of values of the decision variables that satisfies the
constraints and provides an optimal value of the objective function.
Depending upon the nature of the objective function and the constraints, an optimization
problem can be classified as linear or nonlinear. Linear programming (LP) problems consist
of a linear objective function and all the constraints are linear. Nonlinear programming (NLP)
problems have a nonlinear objective function and/or nonlinear constraints. Linear program-
ming (LP) models have been extensively applied to optimal resource allocation problems.
The general form of an LP model can be expressed as
Max (or Min) X0 = uJX1 (10.7.4)
which is the linear objective function subject to the linear constraints
ra;x1 < b; (10.7.5)
X1 <0 (10.7.6)
The c's are the objective function coefficients, the a's are the technological coefficients, and
the b's are the right-hand side (RHS) coefficients. The well-known algorithm for solving LP
problems is the simplex method. References for the simplex method include Hillier and Lieber-
man,42 Mays and Tung,60 and Taha.82
10.7.3 Types of Groundwater Management Models

A classification of groundwater management models is presented in Figure 10.7 .1. TWo basic
categories are (a) hydraulic management models that are aimed at managing pumping and
recharge and (b) policy evaluation models that also can consider the economics of water allo-
cations. Both categories of models have been developed ~ased upon four major approaches: (1)
embedding approach, (2) optimal control approach, (3) response matrix approach, and (4)
heuristic optimization-simulation approach. In addition, the policy evaluation models have
been developed using hierarchical approaches.
The embedding approach incorporates the equations of the simulation model (represented
as a set of difference equations) directly into the optimization problem to be solved. This
method has limited applications and is mostly used in groundwater hydraulic management.
The optimization problem quickly becomes too large to solve by available algorithms when a
large-scale aquifer (especially an unconfined aquifer) is considered. Unconfined aquifers result
in nonlinear programming problems. Previous work based on this approach includes Aguado
Embedding
Hydraulic
tse management
Heuristic optimization-simulation
Jle
Groundwater
An Optimal control
'.1) Response matrix

'------~ Policy evaluation
and allocation
Hierarchial models
1.2) Figure 10.7 .1. Classification of optimization models for groundwater management.
7.3)
et al} Aguado et al} Willis and Newman, 101 Aguado and Remson, 1 Remson and Gorelick, 71
and Willis and Liu. 100
ran- The optimal control approach, based upon concepts from optimal control (Mays 59), cou-
tints ples optimization techniques with a groundwater simulator to implicitly, repeatedly solve the
the governing equations of groundwater flow. This metilodology could be thought of as a variation
; the of the embedding approach. The state variables that represent heads and the control variables
that represent pumpages are implicitly related through tile simulator. The simulator equations
1tion are used to express the state variables in terms of the control variables, yielding a much smaller
nsist reduced optimization problem that must be solved many times. Wanakule, Mays, and Lasdon95
IJLP) presented a general groundwater management model, based upon a nonlinear programming
.ram- code GRG2 (Lasdon and Warren48 ) interfaced with a groundwater simulation model GWSIM
s. (Texas Water Development Board 8\ that can be used to solve both hydraulic management
problems and groundwater policy evaluation (allocation) problems. Other applications of opti-
).7.4) mal control include Andricevic; 11 Chang, Shoemaker, and Liu; 23 Culver and Shoemaker;25• 26
Georgakakos and Vlatsa; 35 Lee and Kitanidis;50 and Whiffen and Shoemaker. 98
The response matrix approach generates a unit response matrix by solving the simulation
0.7 .5) model several times, each with unit pumpage at a single pumping node. Superposition is used
to determine total drawdowns. This yields a smaller optimization problem, but tile method has
0.7.6) two major limitations. It is exact only for a confined aquifer but has good accuracy for an
unconfined aquifer with relatively small drawdowns compared to the aquifer thickness. A
s, and
ngLP drawdown correction method may be used to improve accuracy for an unconfined aquifer with
,ieber- larger drawdowns, but acceptable accuracy cannot be guaranteed (Heidari41 ). In addition, the
response matrix must be recomputed when exogeneous factors such as aquifer boundary con-
ditions or potential well locations change. An alternative is to treat these factors as decision
variables, but then more variables and constraints are included in the optimization problem.
Work stemming from this approach includes that by Ahfeld and Sawyer6, Ahfeld et al.7,
) basic Barlow et al. 14 , Maddock,52· 53 Maddock and Haimes,54 Morel-Seytoux and Daly,64 Morel-
ng and Seytoux et al.,65 Heidari,41 lllangasekare and Morel-Seytoux,43 Reichard,68 and Willis. 99
!f allo- More recently, less conventional heuristic optimization methods have been developed,
les:(l) including neural networks (Rogers and Dowla74 ), genetic algorithms (McKinney and Lin63,
md (4) Ritzel et al.73), and simulated annealing (Dougherty and Marryott,30 Marryott, Dougherty, and
is have Stollar,58 and Skaggs et a1. 78 ·79). These methods use a simulation model interfaced with a
heuristic optimization procedure such as genetic algorithms or simulated annealing
esented approaches. Figure 10.7.2 illustrates the interfacing of such a procedure. These procedures
:d. This have the advantage of being able to handle nonlinearities, but can be computationally time con-
~ement. suming because of the many times that the simulation model must be solved during tile heuris-
when a tic search procedure.
rs result Groundwater policy evaluation and allocation models are used for water allocation plans
Aguado involving economic management objectives subject to institutional policies as constraints in
Initialize Simulator/Optimizer
• Pumping rates
• Sorbtlon barrier dimensions
• Contaminant source(s)ldlstrlbutlon(s)
• Etc.
Flow and Transport Simulator
• Pumping rates
• Sofbtlon barrier dimensions
• Potantlalsurflctlveloclty fttk:l
• Contaminant concentratfona
Evaluate Objective Function

(e.g., minimize pumping, gradient
control, achieve MCLs, etc.)
• Pumping nd8l
• Sorbtlon barrier dimensions
• Contamlnatit removal
• Etc.
Figure 10.7.2.
Optimizer Flowchart of ground-
(Simulated Annealing) water management
methodology.
E
addition to the hydraulic management constraints. These types of models have been applied to
transient aquifer problems that consider the agricultural economy in response to institutional
policies and to conjunctive use of surface water-groundwater problems. Figure 10.7.1 illus-
trates five types of approaches for policy evaluation and allocation problems. The response so
matrix approach is used for problems considering hydraulic-economic response (Gorelick37).
Linked simulation-optimization models use the results of an external groundwater simulation
model as input to a series of subarea economic optimization models (Gorelick37). Examples of
the linked simulation-optimization models include Young and Bredehoeft, 106 Daubert and
Young, 27 and Bredehoeft and Young. 19 Hierarchical models use subarea decomposition and a
response matrix approach (Haimes and Dreizen, 39 Bisschop et al. 17 ).
10.8 GROUNDWATER MANAGEMENT MODELING:

HYDRAULIC MANAGEMENT MODELS
10.8.1 Steady-State One-Dimensional Problems for Confined Aquifers
Consider a confined aquifer with flow in one dimension and fixed-head boundaries as shown
in Figure 10.8.1 with pumping wells that are fully penetrating. The governing equation for
steady-state flow is
(10.8.1)
Using a central differencing scheme, we can write Equation 10.8.1 in finite difference fonn as
hi+l- 2h; +hi- I W;
(10.8.2)
(Ax) 2 = Tx
Aguado et al. 2 formulated the following type of linear programming model for determining the
optimal steady-state pumpage from a one-dimensional confined aquifer with fixed head
boundaries. The objective function for optimization problem can be stated as
Maximize Z = Lh 1 (10.8.3)
ie/
subject to Equation 10.8.2 for each well and the following additional constraints
I w; ~ wmin
ie/ (10.8.4)
h; ~ 0 i E I (10.8.5a)
W;~ 0 iEI (10.8.5b)
where I is the set of wells and Wmin is the minimum total production rate for the wells. The
unknowns in this problem are hand W. Once the model is solved the pumpage can be deter-
mined from W = q/&?. The head maintenance objective of Equation 10.8.3 is practical for
managing some aquifers; however, other types of objective functions could be used, such as
minimizing pumping costs. The above model formulation considers negligible well diameters
and negligible well losses.
; EXAMPLE 10.8.1 Develop an LP model for determining the optimal (maximize heads) steady-state pumpage of the one-
dimensional confined aquifer shown in Figure 10.8.1. The wells are equally spaced at a distance of tJ.x
~d to apart with constant head boundaries ho and h 5•
1onal
tllus-
SOLUTION The objective function is simply
'onse
;k37). Maximize Z = h1 + h2 + h3 + h4
.ation subject to the following finite difference equations
lesof
t and
and a
shown
ion for
--::-:-:-:-:-:-:-:-:-r-:-:-:-
~
10.8.1)
Figure 10.8.1. Confined one-dimensional aquifer.
and the constraint on the production rate:

WI + W2 + WJ + W4 2: Wmin
=<
. i
h; 2: 0 i = I, . .. ,4
W; 2: 0 i = l, ... ,4
The unknowns in this LP model are h1, ... ,h4 and W1, ... , W4. Additional constraints can be used to force
the heads to be decreasing in the direction of flow, which are
•
10.8.2 Steady-State One·Dimensional Problems for Unconfined Aquifers
A one-dimensional unconfined aquifer is shown in Figure 10.8.2 with constant head bound-
aries and fully penetrating wells that are equally spaced. The governing equation for steady- so
state flow is
~(r ah}= w (10.8.6)

dX X dX
where Tx =Kh so that

(10.8.7)
In order to simplify the notation, the substitution w = h2 can be made to linearize the problem
so that the finite difference expression can be written as
d2w wi+l- 2w; + wi-1 2W; (10.8.8)
dx2 = (Llx)2 = 7
assuming that the hydraulic conductivity K is constant throughout the aquifer. The governing
Equation 10.8.8 for each well is now linear. Aguado et a1.2 fonnulated the following linear pro-
gramming model for determining the optimal steady-state pumpage from a one-dimensional
unconfined aquifer:
Maximize Z = w; L
ie/
(10.8.9)
subject to Equation 10.8.8 for each well and

10.8.
(10.8.10)
W; ~ 0 iE I (10.8.1la)
W; ~ 0 i E I (10.8.11b)
l0.8 Groundwater Management Modeling: Hydraulic Management Models 487
q2 q3 q4
t .. t >·t
Figure 10.8.2. Unconfined one-dimensional

aquifer.
lrce
The unknowns in the LP model are w; and W;. The heads h; can be determined from h; =..,J;;
once the LP model has been solved for the unknowns .
•
Develop an LP model for detennining the optimal steady-state pumpage of the one-dimensional uncon-
fined aquifer in Figure 10.8.2 to maximize heads. The wells are equally spaced at a distance of Ax. apart
with constant head boundaries h0 and h5.
md-
ady- SOLUTION The objective function is simply
Maximize Z = w1 + w2 + w3 + w4
.8.6) subject to the following finite difference equations:
2(&)2
-2w 1 +w2 - - - W1 = -w0
K
).8.7) w1 -2w2 +w3 - -
2(axi
-W = 0 2
K
lblem 2{Ax.}2
w2 - 2w3 + w4 - - -W3 = 0
K
2(a.t)2
w3 -2w4 - - - W4 =-w5
0.8.8) K
and the constraint on the production rate:
~rning
w, + w2 + w3 + w4 ~ wmin
upro-
lSional W; ~ 0 i =I, ... ,4
W; ~ 0 i = 1, ... ,4
l0.8.9) The unknowns in this LP model are w1, ... ,w4 and W1, ... ,W •
4
•
0.8.10) 10.8.3 Steady-State Two-Dimensional Model for Confined Aquifers
The governing steady-state two-dimensional equation for a homogeneous confined aquifer is
>.8.lla)
o2h a2h w
-+- - -
).8.llb) (I 0.8.12)
dx2 di - T
for which Tx = TY = T. Using central differences, we can express Equation 10.8.12 in finite
difference form as
hi+l,j - 2hi,j + hi-l,j + hi,j+l - 2hi,j + hi,j-1 = lti,j
(10.8.13)
(Lix)2 (~y)z T
which can be reduced for Lix = ~y to
(10.8.14)
An LP model for the optimal steady state pumpage from a two-dimensional confined
aquifer can be formulated as
~ h.1,).
Maximize Z = £.. (10.8.15)
i,je/
subject to Equation 10.8.14 for each cell and
I
i,je1
lti.j 2:: wmin (10.8.16)
hi,j2:: 0 (10.8.17a)
wi,j2:: o (10.8.17b)
where I represents the set of pumping wells. The unknowns in the LP model are hi.) for all the
cells and Wi,J for the pumping cells.
1
Develop an LP model for determining the optimal steady-state pumping from the two-dimensional con-
fined aquifer shown in Figure 10.8.3. This aquifer has constant (fixed) heads along the aquifer bound-
aries. It has three pumping cells, (2, 2), (3, 2) and (3, 3), as shown in Figure 10.8.3.

Maximize Z = h 2,2 + h3,2 + h 3,3
subject to the finite difference Equation 10.8.14 written for each cell in the aquifer. The finite difference
equations for cell (1, 1) is
h2,I- 4hl,l + ho,l + hl,2 + hi,O =0

W1•1 = 0 because there is no pumping and heads h0,1 and h1,0 are known constant heads, so this constraint
can be written as
h2,l- 4hl,l + hl,2 =-ho,I- h!,O

with the known values on the RHS. The finite difference equations for pumping cell (2, 2) is
(ill)2
h3,2 -4~.2 +h1.2 +~.3 +~.~--r-w2,2 =0
The finite difference equation can be written for each of the remaining cells in the aquifer. The pumpage
constraint Equation 10.8.16 is simply
•
ite X
I
I
(0, 0) Yr:. 0) : (2, 0) I (3, 0) (4, 0) (5, 0)
I
13) I f+- M---+1
-----
(0, 1) (1' 1) (2, 1) (3, 1) (4, 1) (5, 1)
• • • •
14) -----
(0, 2) {1, 2) (2, 2) (3, 2) (4, 2) (5, 2)
ned
j+ 1 • 8 8 • ~ Aquifer boundary
-----
(0, 3) (1, 3) (2, 3) (3, 3) (4, 3) f(~.~~-
.lS)
• • 8 • Ay
-----
t___ _
(0, 4) [1, 4) (2, 4) (3, 4) (4, 4) (5, 4)
U6) j-1 • • • • 8 Pumping cells
___ _..._
.17a) (0, 5) (1, 5) (2, 5) (3, 5) (4, 5) (5, 5)
.17b) j- 1
ll the
Figure 10.8.3. Confined two-dimensional aquifer (plan view).
10.8.4 Transient One-Dimensional Problem for Confined Aquifers

U. con-
The governing equation for transient, one-dimensional confined problems has the form
lound·
To1h=Sdh+W (10.8.18)
ax 2 dt
Using the Crank-Nicholson scheme (Remson et a1.70), we can express the finite difference
approximation of the second-order derivative in Equation 10.8.18 as
fereoce
~:~ =W'H,< -(:;2+h;-1,< + h,.o,H - ~~)i +hi-1,<-1 l (10.8.19)
The finite difference equation for Equation 10.8.18 is determined using Equation 10.8.19 and
nstraint the finite difference approximation for dh!dt = (h;,,- h~ 1_ 1 )/Ar,
· '2 · + hi+lr-1-2hir-l+h;-tr-1]1hil-h;r-l]
T hi+lt-2h;t+h;-Lr · · 2 · - ' ' - W;r+lf;,_l
· · =0
[ 2(Ax) 2(Ax) A1 2
(10.8.20)
which can be simplified to
>Urnpage
• (10.8.21)
Aguado et al.2 proposed an LP model for transient one-dimensional flow with an objective of
maximizing the sum of heads in the last time step, '!, that is,
Maximize Z = LA 1 (10.8.22)
The constraints are Equation 10.8.21 written for each cell along with
t = 1, ... , '! (10.8.23)
EXAMPLE 10.8.4 Develop an LP model for the transient, one-dimensional confined aquifer in Figure 10.8.1 considering
two time periods. The head boundaries at h0 and h5 are known for timet"' 0, t"' 1, and t"' 2.

Maximize Z =h 1,2 + ~. 2 + h3,2 + h4, 2
and the constraint for well i "' 1 and time period t = 1 is
in which h0,1 is the constant boundary and W1,0, h2,0, h1•0, and h0,0 represent the known initial conditions.
This constraint can be rearranged with only the unknowns on the left-hand side:
2{!U)2 (&)2 2{6..x)2 {&)2

z. -2h1,1 - -T(Llt)
"2.1
- S h1,1 ---W.
T --h0,I _z."2,0 +2h1,0 _z."0,0 - -T(Llt)
1.1 -
- S hI.0 +--W.
T I, 0
The constraint for well i =2 and time period t = 2 is
The finite difference equations are written for each well at each time period.
The pumpage constraints are
wl,l + w2,1 + w),l + w4,1 ~ wmin,l
and
10.8.5 Steady-State Two-Dimensional Problem for Unconfined Aquifers

This type of problem may be typical for dewatering of a construction or mining site (see Fig-
ure 10.8.4). Aguado et a1. 2 presented an LP model for solving this problem. The governing
equation for a homogeneous and isotropic aquifer is
a2h2 a2h2 zw
--+--=- (10.8.24)
ax2 ol K
or
(10.8.25)
Wells
.)
I)
Wells
Figure 10.8.4. Excavation site for two-dimensional unconfined aquifer.
where w =h2. With the central differencing scheme of Equation 10.8.2, the finite difference
representation of Equation 10.8.25 is
wi+l.j -2wi.j +wi-J,j + wi.j+l -2wi,j + wi.j-l = 2Wi.j
(10.8.26)
(t1x)2 (L1y)2 K
For the purposes of this modeling effort, both the heads and the pumpages are to be detennined
so that the unknowns are wand W, then Equation 10.8.26 is
[ (~)2]wi+l,j +[ (~)2]wi-l,j- [ (~)2 + (A:)2 ]wi,j +[ (A:)2 ]wi,j+l +[ (A~)2 ]wi,j-1 - ; ~.j = 0
(10.8.27)
An LP model for the dewatering problem could be stated to minimize the amount of steady-
state pumpage required to maintain head levels below specified levels. The LP model state-
ment is
• Minimize Z = .~
( . W1,). (10.8.28)
i,jel
subject to Equation 10.8.27 written for each cell, and the maximum allowable head require-
Fig- ment in the excavation site
ning
wij$ w, (10.8.29)
for c,j e excavation site with w, = h; and nonnegativity constraints
8.24)
w;,j ~ 0 (10.8.30a)
wi.j ~ o iJ e I (10.8.30b)
The number of wells and their locations are affected by the well spacing in the two-
dimensional grid.
10.9 POLICY EVALUATION AND ALLOCATION MODELS:

RESPONSE MATRIX APPROACH
The response matrix approach uses a groundwater simulation model external to the optimiza-
tion model to develop unit responses. A unit response describes the influence of a pulse stim-
ulus (such as a unit pumpage or injection over a time period) at a selected location (well) or
cell upon the hydraulic heads for the other locations (wells) or cells throughout an aquifer. The
response matrix consists of all the unit responses. Maddock52 derived a function relating the
drawdown in a confined aquifer to pumpages through the use of a unit response function, ~.
This function, also referred to as an algebraic technological junction, defines the drawdown
sk.n in the kth cell at the end of the nth time period:
n J
sk,n = LL
p=lj=l
Pk,j,n,p qj,p (10.9.1)
where the unit response function Pk,j,n,p is the change in drawdown (unit drawdown) in the kth
cell at the end of the nth time period due to a unit pumpage from the jth cell (j may equal k)
during the pth time period; qj,p is the quantity pumped from j during the pth time period; and
1 is the number of cells.
Equation 10.9.1 is based on the assumption that (1) flow is horizontal only, (2) wells are
fully penetrating, (3) the transmissivity and storage coefficient may be nonhomogeneous, and
(4) the pumpage is constant over a time period but can vary for different time periods. The unit
response function, p, can be calculated analytically or numerically. To numerically determine
~' a separate groundwater simulation is required for each pumped well or cell. For each sim-
ulation, unit pumpage over a time period is assumed and the simulation determines the
response or drawdown at the other well locations or cells, for this unit pulse. Maddock53
extended the idea of the algebraic technological function to unconfined aquifers.
A groundwater management model by Heidari41 was formulated based upon the response
function approach as follows:
J N
Maximize Z = L Lqj,n (10.9.2)
j=ln=l
subject to:
a. satisfying the governing equation of flow through the response matrix
n J
sk,n = I:L~k.j,n,pqj,p k=1, ... ,J
p=lj=l
n=l, ... ,N (10.9.3)
b. pumpage cannot exceed qj,n' which is the smaller of the appropriated right or its
capacity
j =1, ... ,]
(10.9.4)
n = 1, ... ,N
c. drawdown at each well or cell cannot exceed an upper limit, sk,n
0<
- <-sk,n
sj,n- k = 1, ... ,]
(10.9.5)
n = 1, ... ,N
I 0.9 Policy Evaluation and Allocation Models: Response Matrix Approach 493
d. demand, Qn, for each time period n should be satisfied

1
1- I.qJ,n ~ Qn n =i, .. .,N (10.9.6)
l- j=l
)[
1e The upper limit on drawdown sk.n can be defined as a fraction (y) of the saturated thickness
1e bk of the aquifer at the kth well (or cell), so that
p.
(l0.9.7)
Actually, constraint Equations 10.9.3 and 10.9.5 can be combined. Heidari41 applied the above
LP model to the Pawnee Valley in Kansas to detennine optimal pumpage for policies with and
.1) without net appropriations as constraints.
tlh
EXAMPLE .J0.9.1 Develop a policy evaluation model using LP for a simple confined aquifer defined by four cells with a
k) pumping well in each cell. A total of four time periods are to be considered. The unit response function
md ~kj,n.p is known and the objective is to maximize pumpage.
are
SOLUTION 4 4
md
MaximizeZ= I.~>J.n
mit j=l 11~1
Line
im-
the subject to:
;k53
a. satisfying governing equations of flow through the response matrix
Well! (k =I )
mse
(n =I)
n=l 4
.9.2) s1.1 L
= I.~k.j.n,p qj,p
p=! j=l
=~ 1,1,1,1 ql,l + ~ 1,2.1,1 q2,1 + ~1.3.1.1 q3,! + ~1.4.1.1 q4,1

(n =2)
n..2 4
s,,z =I,.L,~kJ,n,p qj.p

P"'l j "l
).9.3) .
= Pl.l,2.Jql,l +~1,2.2,1q2,l + ~ l.3.2.lq3.l + Pl.4,2,1q4.1
or its +P1.1.2.2q1.2 + ~1.2.2.2q2.2 + Pt.3,2.2q3.2 + P t.4,2,2q4.2
(n = 3)
n=J 4
0.9.4)
sl,3 =I.I.~kj.n.pqj.p
p•l j=l
=Pt.t,3.tql,1 + ~~.2.3,Jq2,1 + ~~.3.3,tq3,1 + Pl.4.3.tq4.t

+~1.r.3.zqr.2 + ~1.2.3.2q2.z +P,.3.3.2q3.2 + Pt.o.2q4.2
0.9.5)
+P 1,1.3.3qr.3 +P,.z.3.Jq2.3 + p,.3.3.3q3.3 +Pl.4.3.3q4,3
(n = 4)
n=4 4
St,4 =L L~k.j,n,p qj,p

p=l j=i
=~1.1.4,1 ql,l + ~1,2,4,1 q2,1 + ~1,3.4,1 q3,1 + p1,4,4,1 q4,1

+~ 1,1,4,2q1,2 + ~ 1,2,4,2q2,2 + ~1 .3,4,2q3,2 + ~ 1,4,4,2q4,2
+Pt .l.4.3qt.3 + ~1 .2,4,3q2,J + ~t.3,4.3 q3.3 + ~t.4 ,4,3q4,3
+~1,1.4.4ql.4 + ~1.2,4,4q2,4 + ~1.3.4.4q3,4 + ~1,4,4,4q4,4
1
l\
We112 (k =2)
(n = 1)
n=l 4
S2.1 = LL~k.j,n,p
p=! }=1
qj,p
= ~2,1,\,lql.l + ~2.2,l,lq2,1 + ~2.3,l.lq3,1 + ~2.4,1,1q4,1
(n=2)
n=2 4
s2,2 = LL~k.j,n,p
p=! j=l
qj,p
= ~2.1,2,1 q\,1 + ~2.2,2,1 q2,1 + ~2.3,2,1 q3,1 + ~2.4,2,1 q4.1

+~2.1,2,2ql,2 + ~2.2,2,2q2,2 + ~2.3,2,2q3,2 + ~2.4,2,2q4,2
The remaining governing constraints are written in a similar manner.
b. pumpage cannot be greater than the appropriated right

O~ql,l ~qu
0 ~ q1,2 ~ (/1,2
c. drawdown at each well cannot exceed an upper limit

0 ~ sl,l ~ sl.l
0 ~ sl,2 ~ s\.2
10.10 Optimal Control Groundwater Management Modeling 495
d. demand requirement
ql,l + q2,1 + qJ,I + q4,l ~ Q,

q1.2 + q2.2 + q3.1 + q4.2 ~ Qz
qi,J + q2,3 + q3,3 + q4.3 ~ QJ
ql.4 + q2,4 + q3,4 + q4,4 ~ Q4
•
10.10 OPTIMAL CONTROL GROUNDWATER
MANAGEMENT MODELING
The general groundwater management problem (GGMP) can be expressed mathematically as
follows:
Optimize Z == f(h, q) (10.10.1)
subject to
a. the general groundwater flow constraints Equation 9.4.10a orb or 9.5.1
g(h, q) =0 (10. 10.2)
b. the simple bounds
(10.10.3)
(10.10.4)
c. other constraint sets, such as demand constraints

w(h, q) :5 0 (10. 10.5)
where \) represents an upper bound and (_) represents a lower bound.
Both head h and pumpage or recharge q are vectors of decision variables that have maxi-
mum dimensions equal to the product of the number of active nodes within the aquifer bound-
ary and time steps. Fixed pumpages or recharges are considered to be constants. By
convention, available pumpages have a positive value and the elements of q have a negative
value where there is available recharge. Usually the number of variable pumpages and/or
recharges (hereafter the terms pumpages that refer to q will imply both pumpages and/or
recharges) is small and results in a much smaller dimension of q than h.
The objective function Equation 10.10.1, which may be either maximization (e.g., sum of
heads) or minimization (e.g., minimize pumpage), can be a linear or nonlinear function. Also,
it may be nonseparable or contain only tenus of pumpages or heads, but has to be differen- ·
tiable. Constraint Equation 10.10.2 represents a system of equations governing groundwater
flow, which are finite difference or simulator equations when q is unknown. The upper ( ij) and
lower (~)bounds on pumpages physically may or may not exist. Unlike pumpage, the lower
bound on heads ( !D can be viewed as the bottom elevation of the aquifer while the upper head
bound ( h) can be regarded as ground surlace elevations for the unconfined cells. In addition to
constraint Equations 10.10.2-10.10.4, constraint Equation 10.10.5 may be included to impose
restrictions such as water demands, operating rules, budgetary limitations, and so forth.
The above general optimization model, Equations 10.10.1-10.10.5, has been solved
through the development of several simulation optimization groundwater management models
based upon optimal control approaches (Mays59). These include models developed by Culver
and Shoemaket-S and Wanakule, Mays, and Lasdon.95
LO.I 1 CASE STUDIES Groundwater Management Modeling Applications
10.11.1 Optimal-Control Modelfor Barton Springs- Maximize total pumpage of all existing wells in 1981 subject
Edwards Aquifer, Texas to these constraints: (
93
Wanakute developed a groundwater management (simula- 1) The groundwater hydraulics of the Edwards Aquifer for the 1
tion-optimization) model for determining the optimal yield under 1981 hydrologic conditions including recharge.
various management policies of the Barton Springs-Edwards 2) Recreation constraints for maintaining the spring flows at Bar-
n
Aquifer, Austin, Texas (see Figure 9.4.3). The model (Wanakule et ton Springs. l;
al.94• 95) used the Texas Water Development Board (TWDB) finite ti
Dewatered areas in the eastern part of the aquifer had to be
difference GWSIM groundwater simulation model (TWDB 86) n
considered due to the high aquifer bottom elevations. Two differ-
interfaced with the GRG2 nonlinear programming optimizer v
ent cases were presented based upon the size of the dewatered
(Lasdon et al.49). Figure 9.4.3 shows the finite difference grid sys-
zones, as shown in Figure 10.11.2. The case 002, which allows for
tem containing 330 cells whose dimensions varied from 0.379 x
a larger dewatered zone (5.095 mi 2), resulted in an optimal yield
0.283 mi2 to 0.95 x 1.51 m? Figure 10.11.1 shows the 1981 water
of 8,999 acre-feet. Case 003 with the smaller dewatered zone
level contours.
(4.233 mi2) resulted in a smaller yield of 4,928 acre-feet as com-
The objective of the management model was to maximize the
pared to the 1981 pumpage of 2,666 acre-feet. This example
pumpage of all existing wells in 1981. This mathematical model
can be stated as follows: (continues)
0 5 10 15 20
::
"
II
Figure 10.11.1. (a) 1981 Water level

contours in the Barton Springs-Edward
Aquifer at 20-ft. intervals.
(b) Perspective block diagram of 1981
water levels viewed from the east side
of the aquifer (Wanakule93 ).
illustrates the use of a simulation-optimization model to define N-Springs is the location of the 1301-N Liquid Waste Disposal
optimal pumpages to maximize yield and maintain spring flows. Facility (LWDF), which was used for disposal of radiological-
contaminated water associated with pass-through cooling, spent
10.11.2 Heuristic Optimization-SimullltWn Model for fuel storage, and other activities related to operation of the N
Groundwater Remediation Design: Reactor (U.S. DOE90).
e N-Springs Site, Hanford, Washington The LWDF is a long, zigzagging trench running parallel to
and about 800ft from the Columbia River. From 1963 untill985,
Skaggs et al. 78• 79 developed a heuristic optimization-simulation the LWDF received discharges of water containing both radiolog-
model for groundwater remediation design, based upon a simu- ical and nonradiological contaminants at a rate of approximately
lated (enhanced) annealing algorithm with a groundwater simula- 2,100 gal/min. Almost immediately after receiving water, mobile
tion model (see Figure 10.7.2). The groundwater management contaminants from the LWDF were observed in springs along the
model was applied to the N-Springs site located at Hanford,
:d Washington, adjacent to the Columbia River (see Figure 10.11.3). (continues)
)f
ld
1e I ! I I I I 7 I I l~ II II II H IS II 17 11 11 111
-
:t1
n- ~~· ,mA Case 002 & 003 Dew.tered c.JIJ

"-
le I "'~ [''Y~
.:··· CsSll 002 Dewllterodc.J/11
___ ._.._
I!!
~s) I! \..
r; IIJIJ Sprlngt' Cell
• lllloo
2
0 l t l 4 I 6 T IIOI:Nin
= ·.
yAflulter boundary
1!
~ ~
~
. I. I I
~
.- ~ l~itiallterlble and -
~~~-~~..
-- ~
X artesian zones divider
~
~
l\
~
[\
::. 1
~
_,
\
~
... ~-- - --1- --·· --·· --· -- ~ ·'liVi'•
r ...;
~;l~-
~~:
-
I
.
~
_,
" R-.
II ~~ :·
.......
.,,
g
;!;
¥1 ~~~
'
•I
'
•(
:•:
,.
~
0
ell
"'2 "'c
.'!!
0
N
c
.!!!
' ~
~ Xl
~
t:
<
II
\
I'; ... ~.... ~
el , ~~ ,d~
1ard
l81
ide
~
I: ,
~
'r#.,,
Figure 10.11.2. Dewatered area
for Cases 002 and 003
(Wanakule93).
c
; I 0.11 CASE STUDIES Groundwater Management Modeling Applications (continued)
v
d
banks of the Columbia River. Beginning in 1980, 90Sr was trations on the order of 4,000 pCi!L. The highest concentrations 0
detected in the groundwater near the river. It is estimated that are shown to be immediately beneath and down gradient (i.e., ti
2,997 Ci of 90Sr was discharged to the facilities at N-Springs, of toward the river) from the LWDF. 6
which -46 Ci was released to the Columbia River and 1,085 Ci The modeled area is that used by the U.S. DOE90 analysis as 9
has decayed away. The majority of the radiological inventory shown in Figure 10.11.3, encompassing the 8,000 ft by 12,000 ft S\
remaining at the N-Springs area is located in the sediments under- study area. The prevailing hydraulic gradient is almost due north, }I
lying the LWDF, including 75.5 Ci within the unconfined aquifer, from the lower right-hand corner of the model area to the upper 1[
of which it is estimated that 75 Ci is adsorbed by the sediments left-hand corner. Over the study area the gradient varies from th
and 0.5 Ci is dissolved in the groundwater. 0.0005 to 0.003, producing average groundwater velocities
Contours depicting the areal concentration distribution of 90Sr between 0.1 to 2 ft/day in the vicinity of the LWDF. The U.S. SJ
in the groundwater at N-Springs are shown in Figure 10.11.4. The DOE90 study was based on a two-dimensional finite-difference w
plume covers approximately 55 acres and has maximum concen- (continues) te
b<
e~
March 1994 N W'
!
e386.4 Wall with water table elevation
/
"JO;;r::, Water table elevation
./
0 ~000 4000 6000 Feet
0 500 1000 1500 Metoo;

''
Lir
''
Lir
. '
''
Lin
'
''
''
''
''
''
'
''
Lirr
lim
Lim
Figure 10.11.3. Plan view of

N-Springs study area (from
U.S. DOE90) .
code with the 8,000 ft by 12,000 ft study area discretized using a injection rate). Twenty-eight candidate well locations with twelve
variable grid spacing. The flow model assumed steady-state con- possible flow rates for each well were considered: ftve injection
ditions and transport was determined using particle tracking based rates, zero, and six extraction rates. Similar to the U.S. DOE
on computed groundwater pathlines and accounting for retarda- study, four possible barrier configurations were considered: no
tion and radioactive decay. The site was discretized using a 35 by barrier and candidate lengths of 2,000 ft, 3.000 ft, and 3.800 ft.
68 element grid. In the vicinity of the 90Sr plume (from 3,000 ft to The optimal solution obtained consisted of a 3,000-ft barrier
IS 9,000 ft in they direction and 0 ft to 3,000 ft in the x direction) a with a single withdrawal well and a single injection well. The com-
fl subgrid, regularly spaced with element dimensions of 100 ft by puted contaminant release to the river is virtually zero for the ten-
h, 100 ft, was used. Outside the plume area, element dimensions year period. The optimal solution had a total cost of $17.6 million
er increased from 250ft to 1,500 ft. An overlay of the subgrid over as compared to the U.S. DOE90 solution costing $23.1 million.
m the model area is shown in Figure 10.11.5.
es The management problem under consideration for the N- 10.11.3 Groundwater Management Model Using Response
S. Springs area is selection of a least-cost containment system that Matrix Approach: Santa Clara-Calleguas Basin, California
ce would prevent any release of 90Sr into the Columbia River for a
ten-year period. Factors evaluated include construction costs for a Reichard68 presented an application of a response matrix model
!S)
for groundwater-surface water management to identify efficient
barrier and/or well(s), treatment costs for any extracted contami-
nated groundwater, plus maintenance and operating costs for both strategies for meeting water demand (with uncertain surface water
extraction and injection wells. In addition, the net groundwater (continues)
withdrawal should be zero (i.e., total extraction rate equals total
9000
0 Extraction well
Injection well
,.., Sr groundwater
0
,..A 50 concentration
'l Monitoring wells
Limit of 3800 It wall---1 N-46 and N-85
7800
Limit of 3000 ft wall---l

LWDF
g
Q)
(.)
c:
~
i:5
>-
N-46 and N-85
Limit of 3800 ft wall 4200
~\
!Wof 3000 -t-~--.------.-'-.---.---,--,-~-....---,

0 600 1200 1BOO 2400 3000 Figure 10.11.4. Areal distribution of 90Sr in the
om
X-Distance (ft) groundwater at the N-Springs area.79
se
JOJ l CASE STU DlES Groundwater Management Modeling Applications (continued)
c~
w
supplies) and controlling regional water quality (seawater i.ntrusion) water. Supplemental water was any additional source of surface ar
in the Santa Clara-Calleguas Basin in southern Califomia, shown in water, including capturing additional surface water in the basin and
Figure 10.11.6. The study focused on the issues relating to ground- importation. Surface water that was not delivered directly to water ar
water use and operation of the facilities under the control of the users could be artificially recharged. The surface water availability th·
United Water Conservation District (UWCD). An efficient strategy from the river diversion is uncertain. sy
was assumed to be one that met demands and water quality con- aq
10.11.3.1 Description of Study Area
straints while minimizing the need for supplemental water, imposed up
water use reductions, or changes from current regional pumping In order to apply the groundwater management model, the Santa thi
patterns. The management agency (Fox Canyon Groundwater Man- Clara-Ca!leguas Basin was subdivided into 13 water demand sec- W!
agement Agency (GMA) created by the state of Califomia to man- tors, with the Oxnard Plain and Las Posas subbasins further sub- on
age endangered groundwater resources) must supply water to divided, as shown in Figure 10.11.6. Surface water is diverted
demand sectors, using a combination of water delivered by from the Santa Clara River, with the goal of capturing as much of
pipelines and distributed pumpage from wells in an upper and lower the river water as possible, subject to the water rights. The major
aquifer. Water delivered by pipelines includes river diversions, regional water quality issue addressed in the modeling effort is
water pumped from central production wells, and supplemental
(continues) 34
:we
Limit of 3800 It wall 420
1200 1800 2400 3000

Figure 10.11.5. Overlay of LOO ft by 100ft
X-Distance (ft) subgrid on the study area. 79 Figu
10. 11 Case Studies: Groundwater Management Modeling Applications 501
seawater intrusion, illustrated in Figure 10.1 1.7 with the high present in the Pleasant Valley, Santa Rosa, Las Posas subbasins
chloride plumes with concentrations greater than 500 mg!L. The (sectors 5, 6, and 7-9, respectively, in Figure 10.1 1.6}.
water management facilities for the Santa Clara-Calleguas Basin The issues of the study are related to groundwater use and oper-
ce are shown in Figure 10.1 L.8. ation of facilities under the control of the United Water Conserva-
nd The groundwater basin is approximately 300 mi2 (800 km2) in tion District (UWCD). During the 1984-1989 period, annual use of
ler area, with at least five different aquifers having been identified in groundwater in the basin plus water delivered by pipelines by
ity the Santa Clara-Calleguas Basin, grouped into an upper aquifer UWCD was approximately 360 ft3/s (10.3 m3/s), of which 80%
system (Oxnard and Mugu aquifers) and a lower system was for agriculture. The GMA has the responsibility for developing
aquifer system (Hueneme, Fox Canyon, and Grimes Canyon). The and implementing a plan to control seawater intrusion. They also
upper system is more transmissive and more easily recharged of have the authority to limit pumping and collect fees within sectors
nta the two aquifer systems. For purposes of modeling, two layers 1-9. GMA strategy has been to increase artificial recharge,
were used in the Oxnard Plain (sectors l-4 in Figure I0.11.6) and decrease pumpage from the upper aquifer, and increase pumpage
Jb- one layer elsewhere. Only the lower aquifer was assumed to be from the lower aquifer system.
ted
(continues)
10f
~or
tis
119"15" 119"00'
!eS)
10MilES
1---r----L-.....--___..J
10 KILOMffiRS
~M>e$1nleel
OJ Weter1iemand sector ~
l...ocatloo of Subareas
Subarea name Sector Subarm name
rrill1J Not ol model
part
1-4 Oxnard Plain 10 Plru
CAUFORNIA Pipeline 5 Pleasant Valley 11 All more
e Santa Rosa 12 Santa Paula
7-9 Las Posas 13 M<md
ft
Figure 10.11.6. Santa Clara-Calleguas Basin (Reichard68).
bl
Groundwater Management Modeling Applications (continued) ar
pi
10.1 1.3.2 Groundwater Model active cells in the lower aquifer. Model cells are 2,640 ft x 2,640 fll
ft (805 m x 805 m). The modeled groundwater system assumes fn
The groundwater study area model was developed by the Califor- heterogeneous, isotropic confmed flow in both layers. ca
nia Department of Water Resources21 · 22 and MODFLOW was Offshore boundaries represented outcrops of the two aquifers j(
used to perform the groundwater flow computations to develop and were modeled as specified-head boundaries. Drain boundaries
the response functions (see Reichard68) for the optimization were used to represent zones of rising water along the Santa Clara Tl
model. Two layers were used to represent the upper and lower River (see Figure 10.11.7). All other boundaries were no-flow fo
aquifers, with 1,044 active cells in the upper aquifer and 1,106
(continues) 01
fo
me
10Ma.ES Cl
10 ~llOt.IETH!S
Ow~ QfW$ '0 1r(l
m
VlllJ
Wato.-domond sector
Noc part of model (flg. Zb)
E3 Olloride :t500 mgll i

1..aym I and 2 Modti nodes
Oo Dr•ln
A C Artllklal rcchorg•
0 Production woU
Ji I'Toductlon well and •rtlfldal rodwgc
G111cllcnt constrolnts
II Chloric!~ ~
II Offshore bcundbry
Figure 10.11.7. Location of high-chlo-

ride areas and model nodes representing
drains, artificial recharge, central produc-
tion wells, and gradient-controllocations:
(a) upper aquifer and (b) lower aquifer
(Reichard68). Fi~
I0.1! Case SIUdies: Groundwater Management Modeling Applications 503
boundaries, representing either the contact between consolidated or

and unconsolidated materials, the presence of the faults, or the
Minimize the imposed water use reduction for a specified
pinching of the upper aquifer. 68 Natural recharge, from stream- amount of supplemental water.
.Q flow and precipitation, was detennined from modified estimates
or
from water-budgets. Many other details of the model development
can be found in Reichard.68 Minimize the expected sum of squared deviations from current
rs average pumping rates for a specified reduction in water use.
JO.ll.3.3 Model Fonnulation
es Subject to the constraints:
ra The. statement of the mathematical optimization model is as l. Availability of supplemental water
•W follows:
2. Water level constraints
·s) Objectives 3. Hydraulic gradient constraints
Minimize the relative amount of supplemental water obtained 4. Demand requirements
for a specified water use (.i.e., minimize the quantity of supple-
mental water as a fraction of surface water diverted from the Santa
Clara River).
7
/ - · - · - 1 SECTOR 2
n
<..., ;....._, SECTOR4u
· Oxnard -
I I
.....l.._.-·-·-..i ,r·---·-·-·7
L·- ·:Pt Mugu;_·/·
'-·-·-·-·-
lo-
cing
!due-
dons:
:er
Figure 10.11.8. Schematic of water management facilities in the Santa CJ.ara-Calleguas Basin (Reichard68).
Groundwater Management Modeling Applications (continued) no

10.11 eASE STUDIES
ot
ab
5. Continuity on water routing including operation of pipelines, The nonlinear optimization was solved using MINOS through
10
artificial recharge facilities, and the central production wells the GAMS modeling system (Brooke et al. 20).
011
for every year Application of the groundwater management model indicated
6. Artificial recharge and pipeline capacity constraints that a large quantity of supplemental water or significant pumpage
th<
reductions, particularly in the lower aquifer, may be required to
7. Groundwater hydraulics constraints pe
control seawater intrusion with current facilities. The supplemen-
be
The model planning period is T years. Constraints on the water tal water would be delivered directly to users through pipelines
101
levels and hydraulic gradients are imposed at the end of T years. and artificially recharged. The current artificial recharge facilities
str
Groundwater modeling simulations were used to develop the had been valuable and new artificial recharge facilities would be
su;
response functions used in the groundwater hydraulic constraints. beneftcial.
sy:
re~
PROBLEMS
Us
10.2.1 A pumping test is a frequently used field technique to aquifer in Figure 10.8.1 where Ill.= 80 m, T = 1000 m2fday, Wmin cal
detennine the properties of aquifers. It is done by pumping a well =500 m/day, h0 = 40 m, and h5 =35m. Solve this model for h1, SUI
at a constant rate and observing the drawdown of the piezometric ...• h4 and WI, ... ' w4. De
surface or water table in an observation well at some distance Vlll
10.8.2 Set up the LP model to determine the optimal pumpage for
from the pump well. Then an appropriate analytical solution for 10
the one-dimensional confined aquifer in Figure 10.8.1 where At=
groundwater flow is used to determine the transmissivity and stor-
10ft, T =10,000 ft2/day, ho =100ft, h4 =110ft, and Wmin = 100 ca1
age coefficient. Therefore, aquifer properties so detenn.ined will Th
ft/day.
be valid in the vicinity of the pumping well when flow satisfies the res
conditions used in developing the analytical solution. Utilize the 10.8.3 Solve the LP model set up in Problem 10.8.2.
un
Cooper-Jacob equation to determine the optimal transmissivity 10.8.4 Set up and solve the LP model to determine the optimal
Fi!
(T) and storage coefficient (S) that minimizes the sum of square of pumpage for the one-dimensional unconfined aquifer in Figure fiv
errors between the observed and calculated drawdowns (listed in 10.8.2 where ill:= 80 m, K =150m/day, Wmin = 500 m/day, ho = ste
the table below) at two observation wells over time. 40 m, and h5 = 35 m.
10.2.2 Resolve Problem 10.2.1 to minimize the sum of absolute 10.8.5 Set up the complete LP model for the two-dimensional
error between the observed and calculated drawdowns at two confined aquifer described in Example 10.8.3.
observation wells. 10.8.6 Set up and solve the LP model for the two-dimensional cru
10.5.1 Develop a report describing the groundwater laws in your confined aquifer shown in Figure 10.8.3 to determine the optimal fro
state or region of your country. How have these laws promoted or pumpage from cells (2, 2), (3, 2) and (3, 3); Ill.= 1,000 ft, T = en'
hindered the development of groundwater? 10,000 f~/day, Wmin = 2 ft/day. The boundary condition is a con-
10.8.1 Set up and solve the linear programming model to deter- stant head of 20 ft.
mine the optimal pumpage for the one-dimensional confined 10.8.7 Set up the complete LP model for the transient, one-
dimensional confined aquifer problem in Example 10.8.4. s
¢
Drawdown Data for Problem 10.2.1 10.8.8 Solve the model for the transient, one-dimensional con-
Drawdown (m) fined aquifer developed in Problem 10.8.7 for lh =100ft, !lt =I
day, T = 10,000 ft2/day, S = 0.001, h0 = 120ft, h5 = 105 ft, and
Time (days) Observation wells Wmin= 200ft/day. Att = 0, h 1 = 118.5 ft, h2 =116ft, h3 =113ft,
Well 1 (r = 100 m) We112 (r =200m) h4 :::: 103.5 ft, and Wl,O:::: W2,0 = W3,0 = W4,0:::: 0.
10.9.1 Develop a groundwater allocation model for an irrigation
0.001 0.087 0.015 district to determine the optimal cropping pattern and groundwa-
0.005 0.200 0.100 ter allocation that maximizes the net return from agricultural pro-
O.ot 0.252 0.147 duction. The irrigation district has the responsibility for
0.05 0.370 0.270 detennining the optimal cropping pattern (acreage devoted to each
0.1 0.435 0.320 crop) with the district's authority. Water is pumped from an exten-
0.5 0.555 0.450 sive confined aquifer system for which a steady-state model is
1 0.610 0.500 considered for the two-dimensional response of the aquifer. The
5 0.745 0.630 profit per acre for a crop is (the unit revenue of a crop times the
yield per acre) minus (the unit cost of groundwater times the
10 0.805 0.690
amount of groundwater applied per acre). The yield would be a Fig
Problems 505
nonlinear function defined by a production function. Define the point k and well location l; and T =aquifer transmissivity. Assume
objective function, constraints, parameters, and decision vari- that the radius of influence of all pump wells is 700 feet (213
ables. What would be the solution methodology? meters) and the aquifer transmissivity is 5,000 gallon/day/ft
10.9.2 Develop a conjunctive groundwater and suriace water plan- (0.0007187 m2/sec). Based on the information about the maxi-
ning model that can optimally distribute, over time, the water mum allowable drawdown and pump capacities given in Figure
~d IO.P.l. formulate an LP model to detennine the optimal pumpage
resources for various water demands. The decisions to be made are
ge for each well.
the groundwater and suriace water allocation in each planning
to period. The objective function is to maximize the net discounted 10.9.4 Refer to the hypothetical groundwater system as shown in
n- benefits from operating the system over the planning horizon, Figure IO.P.I. Consider now that the management is to be made
.es
including costs for capital, operation, and maintenance. Con- for three periods of 50 days each. This is a management problem
1es
straints would include (a) conservation or balance equations for the for transient groundwater systems which can be approximately
be
surface water system, (b) capacity limitations of the surface water described by the Theis equation or Cooper-Jacob equation.
system, (c) groundwater system flow equations to define aquifer (a) Using the Cooper-Jacob, equation define the unit response
response, and (d) bound constraints on groundwater levels. Others function.
may be required such as for artificial recharge, pumping schedules.
(b) Assume that the storage coefficient of the aquifer is 0.002 and
Use GWij and SWkJ to represent, respectively, groundwater allo-
transmissivity is 5,000 ft2/day (465 m2/day). Further, the maxi-
cated from groundwater source i to demand j in time period t and
mum allowable drawdowns (in ft) at each control point are given as
surface water allocated from source k to demand j in time period t.
Define the objective function, constraints, parameters, and decision Period Control point.
variables. What would be the solution methodology? 2 3 4
for 5
.x= 10.9.3 In the planning stage of groundwater management, appli-
cations of analytical solutions of groundwater flow can be made. 5 5 8 5 5
100
Those analytical equations can be used to calculate the unit 2 8 8 10 8 8
response matrix in groundwater management models. Consider an 3 10 10 15 10 10
undeveloped, homogenous, confined groundwater aquifer (see
mal Figure JO.P.l) in which there are three potential pump wells and Develop the LP model and solve it for maximizing total pumpage
:ure five control points at which the drawdowns are to be observed. For for all three 50-day periods.
1o= steady-state management, the Thiem equation 10.9.5 Assume that the transmissivity of the hypothetical aquifer
(Figure IO.P.l) is a lognormal random variable with mean 5,000
mal ski=
ln(r01 I'*') q, gallonlday/ft and coefficient of variation of 0.3. Convert the LP
2nT
model developed in Problem 10.9.1 into a chance-constrained
:mal can be used in which ski= drawdown at control point k resulting model such that the specified drawdown at the five control points
imal from a pumpage of q1 at the well location l; r01 =radius of influ- will not be exceeded with 95 percent reliability. Also, solve the
T= ence of potential production well; rkl =distance between control model.
con-
Distance (ft.) between potential pump wells and control points
one- Pump
Control points
Pumping capacity
con- well 2 3 4 5 (gpd)
t=1 158 381 158 255 430 200,000
. and 2 515 255 292 474 158 200,000
13 ft,
3 447 447 200 200 200 200,000
ation Maximum
allowable 7 7 15 7 7
dwa-
drawdown (ft)
l pro-
' for
.each 0 Pumping well
:xten-
0 Control point
del is
:. The r31 Distance between control
point i and well location j
~the
~s the
lbea Figure IO.P.l. Location of pumping wells and control points for a hypothetical groundwater basin (Problems 10.9.3-10.9.5).
REFERENCES 41
1. Aguado, E., and I. Remson, Ground-water management with fixed

charges, Jour. Water Resources Planning Mgmt. Div., Amer. Soc. Civil
ground water quality model, District report, Southern District, Los
Angeles, CA, 1974.
4"
m
Engrs., v. 106, pp. 375-382, 1980. 22. California Department of Water Resources, Compilation of technical
PI
2. Aguado, E., I. Remson, M. F. Pikul, and W. A. Thomas, Optimum information for the Ventura County cooperative investigation, Vol. II,
4:
pumping for aquifer dewatering, Jour. Hydraulics Div., Amer. Soc. Civil Sacramento, 1975.
R.
Engrs., v. 100, pp. 860-877, 1974. 23. Chang, L. C., C. A. Shoemaker, and P. L. F. Liu, Optimal time-varying
4~
3. Aguado, E., N. Sitar, and I. Remson, Sensitivity analysis in aquifer pumping rates for groundwater remediation: Application of a constrained
cc
studies, Water Resources Research, v. 13, pp. 733-737, 1977. optimal control algorithm, Water Resources Research, v. 28, no. 12,
Rt
4. Ahfeld, D. P., and M. Heidari, Applications of optimal hydraulic pp.3157-3173, 1992.
control to groundwater systems, Jour. Water Resources Planning and 24. Clendenen, F. B., Economic utilization of ground water and surface
Mgt., v. 120, no. 3, May/June 1994. water storage reservoirs, Paper presented before meeting of Amer. Soc.
5. Ahfeld, D. P., and A. Mulligan, Optimal Management of Flow in Civil Engrs., San Diego, Feb. 1955.
Wi
Groundwater Systems, Academic Press, 2000. 25. Culver, T. B., and C. A. Shoemaker, Dynamic optimal control for
6. Ahlfeld, D. P., and C. S. Sawyer, Well location in capture zone design groundwater remediation with flexible management periods, Water
using simulation and optimization techniques, Ground Water, v. 28, no. 4, Resources Research, v. 28, no. 3, pp. 629-641, March 1992.
pp. 507-512, 1990. 26. Culver, T. B., and C. A. Shoemaker, Optimal control for groundwater 47
remediation by differential dynamic programming with quasi-Newton gr;
7. Ahlfeld, D. P., R. H. Page, and G. F. Pinder, Optimal ground-water
V.,
remediation methods applied to a Superfund site: From formulation to approximations, Water Resources Research, v. 29, no. 4, pp. 823-831, 1993.
implementation, GroundWater, v. 33, no.l, pp. 58-70, Jan.-Feb. 1995. 27. Daubert, J. T., and R. A. Young, Groundwater development in western 48
river basins: Large economic gains with unseen costs, Ground Water, v. 20 Ge
8. A1ley, W. M., and J. E. Schefter, External effects of irrigator's pumping
decisions, High Plains Aquifer, Water Resources Research, v. 23, no. 7, pp. 80-85, 1982. 49
pp. 1123-1130,1987. 28. Domenico, P. A., Economic aspects of conjunctive use of water, Smith ing
Valley, Nevada, USA, Inti. Assoc. Sci. Hydrology Pub/ 72, pp. 474-482, As:
9. Alley, W. M., T. E., Reilly, 0. L. Franke, Sustainability of Ground-
water Resources, U.S. Geological Survey Circular 1186, Denver, CO, 1967. 50.
1999. 29. Domenico, P. A. et al., Optimal ground-water mining, Water ing
10. Amer. Soc. Civil Engrs., Ground Water Management, Manual Resources Research, v. 4, pp. 247-255, 1968. Re:
Engmg. Practice 40, 216 pp., 1972. 30. Dougherty, D. E., and R. A. Marryott, Optimal Groundwater 51.
11. Andricevic, R., Coupled withdrawal and sampling designs for Management, 1. Simulated Annealing, Water Resources Research, v. 27, reg
pp.2493-2508, 1991. Re~
groundwater supply models, Water Resources Research, v. 29, no. 1, pp.
5-16, 1993. lOt
31. Espeland, W. N., The Struggle for Water: Politics, Rationality, and
12. Arizona Department of Water Resources (ADWR), Overview of Ari- Identity in the American Southwest, University of Chicago Press, Chicago, 52.
zona's groundwater management code, http://www.adwr.state.az.us/ IL, 1998. ~ mOr
Azwaterinfo, 1998. 32. Fowler, L. C., Ground-water management for the nation's future- 53.
13. Baker, T. L., R. Rae, J. E. Minor, and S. V. Connor, Water for the ground-water basin operation, Jour. Hydraulics Div., Amer. Soc. Civil wh<
Southwest: Historical Survey and Guide to Historic Sites, ASCE Histori- Engrs., v. 90, no. HY4, pp. 51-57, 1964. v. 1
cal Publication No.3, Amer. Soc. Civil Engrs., New York, 1973. 33. Fradkin, P.L., A River No More: The Colorado River and the West, 54.
14. Barlow, P.M., B. J. Wagner, and K. Belitz, Pumping strategies for University of Arizona Press, Tucson, AZ, 1981. mar
management of a shallow water table: The value of the simulation- 34. Galloway, D. L., W. M. Alley, P.M. Barlow, T. E. Reilly, P. Tucci, 55.
optimization approach, Ground Water, v. 34, no. 2, March/April, 1996. Evolving Issues and Practices in Managing Ground-Water Resources:
15. Basagaoglu, H., and M.A. Marino, Joint management of surface and Case Studies on the Role of Science, U.S. Geological Survey Circular 56.
ground water supplies, Ground Water, v. 37, no. 2, pp. 214-222, 1999. 1247,2003. basi
16. Bear, J., and 0. Levin, The optimal yield of an aquifer, Inti. Assoc. Sci. 35. Georgakakos, A. P., and D. A. Vlatsa, Stochastic control of ground 57.
Hydrology Pub/. 72, pp. 401-412, 1967. water systems, Water Resources Research, v. 27, no. 8, pp. 2077-2090, ifon
17. Bisschop, J. W., J. H. Candler, and G. T. O'Mara, The Indian Basin 1991. 58.
model: A special application of two-level linear programming, Math 36. Glennon, R., Water Follies: Groundwater Pumping and the Fate of watf
Program, v. 20, pp. 30-38, 1982. America's Rivers, Island Press, Washington, D.C., 2002. cont
18. Bowden, C., Killing the Hidden Waters, University of Texas Press, 37. Gorelick, S.M., A review of distributed parameter groundwater man- 59.
Austin, TX, 1977. agement modeling methods, Water Resources Research, v. 19, Yod
19. Bredehoeft, J.D., and R. A. Young, Conjunctive use of groundwater pp.305-319, 1983. 60.
and surface water for irrigated agriculture: Risk aversion, Water Resources 38. Grubb, H. W., The Edwards Aquifer: Conflicts surrounding use of a agen
Research, AGU, v. 19, pp. 1111-1121, 1983. regional water resource, Water Resources Update, Universities Council on 61. .
20. Brooke, A., D. Kendrick, and A. Meerhaus, GAMS: A User's Guide, Water Resources, Issue No. 106, Carbondale, IL, Winter 1997. ment
Scientific Press, Redwood City, CA, 1988. 39. Haimes, Y. Y., andY. C. Dreizen, Management of groundwater and 62. J
21. California Department of Water Resources, Mathematical modeling surface water via decomposition, Water Resources Research, AGU, v. 13, Plai1
of water quality for water resources management, 1. Development of the pp. 69-77, 1977. Reso
References 507
40. Hartman, L. M., Economics and ground-water development, Ground 63. McKinney, D. C., and M. D. Lin, Genetic algorithm sol ution of
Water, v. 3, no. 2, pp. 4-8, 1965. groundwater management models, Water Resources Research, v. 30, no. 6,
41 . Heidari, M.. Application of linear systems theory and linear program· pp. 1897-1906. 1994.
ming to groundwater management in Kansas, Water Resources Bull, v. 18, 64. Morel-Seytoux, H. J., and C. J. Daly, A discrete kernel generator for
II pp. 1003- 1012, 1982. stream-aquifer studies, Water Resources Research, v. I I, pp. 253-260,
I, 42. Hillier, F. S., and G. J. Lieberman, Introduction to Operations 1975.
Research, 7th ed., McGraw-Hill, New York, 200 I . 65. Morei-Seytoux, H. J., G. Peters, R. Young, and T. Illangasekare,
tg 43. ltlangasekare, T., and H. J. Morei-Seytoux, Stream-aquifer influence Groundwater modeling for management, international Symposium on
~d Water Resource Systems, Water Resour. Dev. and Training Cent., Univ. of
coefficients as tools for simulation and management, Water Resources
2. Research, v. 18, pp. 168-176, 1982. Roorkee, Roorkee, India, 1980.
44. Ingram, H. et al., Water scarcity and the politics of plenty in the Four- 66. Peters, H. J., Groundwater management, Water Resources Bull. v. 8,
ce Comer states, Western Political Quarterly. September 1979. pp. 188-197, 1972.
JC.
45. Johnson, A. 1., et al., Symposium on planning and design of ground- 67. Pfannkuch, H. 0., and B. A. Labno, Design and optimization of
water data programs, Water Resources Research, v. 8, pp. 177-241 , 1972. ground-water monitoring networks for pollution studies, Ground Water;
for v. 14, pp. 455-462, 1976.
46. Killgore, M. K., and D. J. Eaton, NAFTA Handbook for Water Resource
ter 68. Reichard, E. G., Groundwater-surface water management with
Managers and Engineers, Amer. Soc. Civil Engrs., New York, 1995.
47. LaBolle, E. M ., A. A. Ahmed, and G. E. Fogg, Review of the inte- stochastic surface water supplies: A simulation optimization approach,
Iter Water Resources Research, v. 31, no. II, pp. 2845-2865, November 1995.
grated groundwater and surface-water model (JGSM), Ground Water,
ton
~ 41,no. 2,pp.238-246,2003. 69. Reisner, M., Cadillac Desert, Viking Penguin, New York, 1986.
93.
48. Lasdon, L. S., and A. D. Warren, GRG2 User 's Guide, Department of 70. Remson, !., G. M. Hornberger, and F. J. Molz, Numerical Methods in
em
General Business, University of Texas, Austin, 1989. Subsurface Hydrology, Wiley-lnterscience, New York, 1971.
. 20
49. Lasdon, L. S., A. D. Warren, A. Jain, and M. Ratner, Design and test- 71. Remson, I., and S. M. Gorelick, Management models incorporating
ing of a generalized reduced gradient code for nonlinear programming, groundwater variables, in Operation Research in Agriculture and Water
nith
Assoc. Compw. Mach. Trans. Math. Software, v. 4, pp. 34-50, 1978. Resources, D. Yaron and C. S. Tapiero, eds., North-Holland, Amsterdam,
~82,
1980.
50. Lee, S. L., and P K. Kitanidis, Optimization estimation and schedul-
ing in aquifer remediation with incomplete information, Water Resources 72. Renshaw, E. F., The management of ground water reservoirs, Jour.
ater
Research, v.27,no.9,pp.2203-2217, 1991. Farm Economics, v. 45, pp. 285-295, 1963.
51 . Longley, G., and W. R. Jordan, Management of the Edwards Aquifer 73. Ritzel, B. J., J. W. Eheart, and S. Ranjithan, Using genetic algorithms
·ater
region: How the Endangered Species Act influenced action, Water to solve a containment problem, Water Resortrces Research, v. 30, pp.
. 27,
Resources Update, Universities Council on Water Resources, Issue No. 1589-1603, 1994.
106, Carbondale, IL, Winter 1996. 74. Rogers, L. L., and F. U. Dowla, Optimization of groundwater reme-
and
52: Maddock, T., III, Algebraic technological function for a simulation diation using artificial neural networks with parallel solute transport mod-
:ago,
model, Water Resources Research, v. 8, pp. 129-134, 1972. eling, Water Resources Research, v. 30, pp. 457-481, 1994.
53. Maddock, T., III, Nonlinear technological functions for aquifers 75. Sasman, R. T., and R. J. Schicht, To mine or not to mine groundwater,
Ire--
whose transmissivities vary with drawdown, Water Resources Research, Jour. Amer. Water Works Assoc., v. 70, pp. 156-161, 1978.
Civil
v. 10, pp. 877-S81, 1974. 76. Sclunandt, J., E. T. Smerdon, and J. Clarkson, State Water Policies, A
54. Maddock, T., IIl, and Y. Y. Haimes, A tax system for groundwater Study of Six States, Praeger, New York, 1988.
West,
management, Water Resources Research. v. 11, pp. 7- 14, 1975. 77. Shupe, S. J., and J. Folk-Williams, The Upper Rio Grande: A Guide
55. Maknoon, R., and S. J. Burges, Conjunctive use of ground and surface to Decision-Making. Western Network, Santa Fe, NM. 1988.
rucci,
trees: water, Jour. Amer. Water Works Assoc., v. 70, pp. 419-424, 1978. 78. Skaggs, R. L., L. W. Mays, and L. W. Vail, Simulated annealing with
-cular 56. Mann, J. F., Jr., Factors affecting the safe yield of ground-water memory and directional search for ground water remediation design, Jour.
basins, Trans. Amer. Soc. Civil Engrs., v. 128, Pt. ID, pp. 18~190, 1963. Amer. Water Resources Assn. , v. 37, no. 4, pp. 853-866, August 2001.
round 57. Mann, J. F., Jr., Ground-water management in the Raymond Basin, Cal- 79. Skaggs, R. L., L. W. Mays, and L. W. Vail, Application of enhanced
-2090, ifornia, Eng mg. Geol. Case Histories 7, Geol. Soc. Amer., pp. 61-74, 1969. annealing to ground water remediation design, Jour. Amer. Water
58. Marryott, R. A., D. E. Dougherty, and R. L. Stollar, Optimal ground- Resources Assn., v. 37, no. 4, pp. 867-875, August 2001.
'ate of water management, 2. Application of simulated annealing to a field-scale 80. Smerdon, E. T., and W. R. Jordan, eds., Issues in groundwater man-
contamination site, Water Resources Research, v. 29, pp. 847-S60, 1993. agement, Center for Research in Water Resources, University ofTel\as at
59. Mays, L. W., Optimal Control of Hydrosystems, Marcel Dekker, New Austin, Austin,TX, 1985.
rman-
v. 19, York, 1997. 81. Smith, Z. A., ed., Water and the Future of the Southwest, University
60. Mays, L. W., andY. K. Tung, Hydrosystems Engineering and Man- of New Mexico Press, Albuquerque, NM, 1989.
se of a agement, McGraw-Hill, New York, 1992. 82. Taha, H. D., Operations Research: An Introduction, 7th ed., Macmil-
ncilon 61 . McCleskey, G. W., Problems and benefits in ground-water manage- lan Publishing, NewYorlc,l987.
ment, Ground Water, v. 10, no. 2, pp. 2-5, 1972. 83. Tarlock, D., Water law, in Water Resources Handbook, L. W. Mays,
ter and 62. McGuire, V. L., and J. B. Sharp, Water-Level Changes i11 the High ed., McGraw-Hill, New York, 1996.
J, v. l3. Plams Aquifer-Predevelopment to 1995, U.S. Geological Survey Water 84. Templer, 0 ., and L.V. Urban, Conjunctive use of water on the Tel\as
Resources Investigations Report 97-4081, 1997. High Plains, Water Resources Update, Universities Council on Water
Resources, Issue No. 116, Integrated Water Management, pp. 102-108, 96. Waterstone, M., Of dogs and tails: Water policy and social policy in
Carbondale, IL, Winter 1997. Arizona, Water Resources Bull, American Water Resources Association, v.
85. Texas Water Development Board, GWSIM-Groundwater simulation 28,no.3,pp.479-486, 1992.
program, Program Document and User's Manual, UM 57405, Austin, 1X, 97. Wehmhoefer, R. A., Chapter 2 in Water and the Future of the South-
1974. west, Z. A. Smith, ed., University of New Mexico Press, Albuquerque,
86. Texas Water Development Board (TWDB), Water for Texas, Austin, NM, 1989.
TX, August 1997. 98. Whiffen, G. J., and C. A. Shoemaker, Nonlinear weighted feedback
87. Texas Water Development Board, GWSIM IT-Groundwater simula- control of groundwater remediation under uncertainty, Water Resources
tion program, Program Documem and User's Manual, UM16, Austin, Research, v.29,no.9,pp.3277-3289, 1993.
TX, 1978. 99. Willis, R., A unified approach to regional groundwater management,
88. Thomas, H. E., The conservation ofground water, McGraw-Hill, New in Groundwater Hydraulics, Water Resour. Monogr. 9, J. S. Rosenshein
York, 327 pp., 1951. and G. D. Bennett, eds., AGU, Washington, DC. 1984.
89. U.S. Bureau of Reclamation, Ground water manual, U.S. Dept. Inte- 100. Willis, R., and P. Liu, Optimization model for ground-water plan-
rior, 480 pp., 1977. ning, Jour. Water Resour. Plan. Manag. Div., Amer. Soc. Civil Engrs., v.
110,pp.333-347, 1984.
90. U.S. Department of Energy, Modeling evaluation of N-Springs barrier
and pump-and-treat system, DOE/RL-94-132, Rev. 0, Richland, WA, 101. Willis, R., and B. A. Newman, Management model for groundwater
1995. development, Jour. Water Resour. Plan. Manag. Div.. Amer. Soc. Civil
Engrs., v. 13, pp. 159-171, 1977.
91. U.S. Geological Survey, National Water Surnmt~ry 1983-Hydrologic
Events arn:llssues, U.S. Geological Survey Water-Supply Paper 2250, 102. Willis, R., and W. W.-G. Yeh, Groundwater Systems Planni118 and
1984. Management, Prentice-Hall, Englewood Cliffs, NJ, 1987.
92. Wagner, B. J., Recent advances in simulation-optimization ground- 103. Winter, T. C., J. W. Harvey, 0. L. Franke, and W. M. Alley, Ground
water management modeling, U.S. National Report to IUGG, 1991-1994, Water and Surface Water: A Single Source, U.S. Geological Survey
Rev. Geophys., v. 33, Supp!., 1995. Circular ll39, Denver, CO, 1998.
93. Wanakule, N., Optimal groundwater management models for the 104. Yeh, W.-G., Groundwater systems, in Water Resources Handbook,
Barton Springs-Edwards Aquifer, Report EARDC Rl-89, &!wards L. W. Mays, ed., McGraw-Hill, New York, 1996.
Aquifer Research and Data Center (EARDC), Southwest Texas State Uni- 105. Young, R. A., Safe yield of aquifers: An economic reformulation,
versity, San Marcos, TX, March 1989. Jour. lrrig. Drain. Div., Amer. Soc. Civil Engrs., v. 96, no. IR4,
94. Wanakule, N., L. W. Mays, and L. S. Lasdon, Development and test- pp.377-385, 1970.
ing of a model for determining optimal plumping and recharge of large- 106. Young, R. A., and Bredehoeft, J.D., Digital computer simulation for
scale aquifers, Technical Report CRWR-217, Center for Research in Water solving management problems of conjunctive groundwater and surface
Resources, The University of Texas, Austin, TX, October 1985. water systems, Water Resources Research, v. 8, pp. 533-556, 1972.
95. Wanakule, N., L. W. Mays, and L. S. Lasdon, Optimal management of
large-scale aquifers: Methodology and application, Water Resources
Research, v. 22, no. 4 , pp. 447--465, Aprill986. ·
1
n
I.
Chapter 11
1-
-.
>
es
lt, Surface Investigations of

:in
lil-
Groundwater.
Y.
ter
.vii
Although groundwater cannot be seen on the earth's surface, a variety of techniques can pro-
vide information concerning its occurrence and-under certain conditions-even its quality
uul
vey
from surface or above-surface locations. Surface investigations of groundwater are seldom
more than partially successful in that results usually leave the hydrogeologic picture incom-
w/c, plete; however, such methods are normally less costly than subsurface investigations. Geologic
methods, involving interpretation of geologic data and field reconnaissance, represent an
ion, important first step in any groundwater investigation. Remote sensing from aircraft or satellite
:R-4, has become an increasingly valuable tool for understanding subsurface water conditions.
Finally, geophysical techniques, especially electrical resistivity and seismic refraction meth-
1 for ods, provide only indirect indications of groundwater so that underground hydrologic data
face
must be inferred from surface data. Techniques have been developed for sensing buried wastes
and waste m.igration. 3 Correct interpretation requires supplemental data from subsurface inves-
tigations (described in Chapter 12) to substantiate surface findings.
11.1 GEOLOGIC METHODS

Geologic studies enable large areas to be rapidly and economically appraised on a preliminary
basis as to their potential for groundwater development. A geologic investigation begins with
the collection, analysis, and hydrogeologic interpretation of existing topographic maps, aerial
photographs, geologic maps and logs, and otber pertinen~ records. This should be supple-
mented, when possible, by geologic field reconnaissance and by evaluation of available hydro-
logic data on streamflow and springs; well yields; groundwater recharge, discharge, and levels;
and water quality. Such an approach should be regarded as a first step in any investigation of
subsurface water because no expensive equipment is required; furthermore, information on
geologic composition and structure defines the need for field exploration by other methods.
Knowledge of the depositional and erosional events in an area may indicate the extent and
regularity of water-bearing formations. The type of rock formation will suggest the magnitude
of water yield to be expected; one formation may be adequate for a domestic supply but
entirely unsatisfactory for an industrial or municipal supply. Stratigraphy and geologic history
of an area may reveal aquifers beneath unsuitable upper strata, the continuity and intercon-
nection of aquifers, or important aquifer boundaries. The nature and thickness of overlying
beds, as well as the dip of water-bearing formations, will enable estimates of drilling depths to
509
510 Chapter 11 Surface Investigations of Groundwater
be made. Similarly, confined aquifers may be noted and the possibility of flowing wells or
low pumping lifts foretold. Landforms can often reveal near-surface unconsolidated forma-
tions serving as aquifers, such as glacial outwash, eskers, terraces, and sand dunes. Faults,
which may form impenneable barriers to subsurface flow, frequently can be mapped from sur-
face traces.
The fundamental relationships between geology and groundwater are presented else-
where: geologic formations as aquifers in Chapter 2, the quality of groundwater as affected by
geologic sources in Chapter 7, and geologic logs in Chapter 12.
11.2 REMOTE SENSING

Photographs of the earth taken from aircraft or satellite at various electromagnetic wa.velength
ranges can provide useful information regarding groundwater conditions. 5 The technology of
remote sensing has developed rapidly in recent years, while its application to water resources
is still being discovered. 20• 21 • 27 Furthermore, the ready availability of photographs from com-
mercial firms and government agencies has stimulated their use.
Stereoscopic examination of black-and-white aerial photographs has gained steadily in
importance. Observable patterns, colors, and relief make it possible to distinguish differences
in geology, soils, soil moisture, vegetation, and land use. 3 Thus, photogeology can differ-
entiate between rock and soil types and indicate their permeability and areal distribution-and
hence areas of groundwater recharge and discharge. 37 Maps classifying an area into good, fair,
and poor groundwater yields can be prepared. 20 Table 11.2.1 summarizes the role aerial pho-
tographs can play as interpretive aids in groundwater studies.
Aerial photographs also reveal fracture patterns in rocks, which can be related to poros-
ity, permeability, and ultimately well yield. 25 • 39 Springs and marshy areas indicate relatively
shallow depths to groundwater. Hydrobotanical studies of vegetation in photographs can be
productive.9 Phreatophytes, which transpire water from shallow water tables, define depths to
groundwater; Figure 11.2.1 illustrates vegetation on an alluvial fan. Halophytes, plants with a
high tolerance for soluble salts, and white efflorescences of salt at ground surface indicate the
presence of shallow brackish or saline groundwater. Xerophytes, desert plants subsisting on
minimal water, suggest a considerable depth to the water table.
Other nonvisible portions of the electromagnetic spectrum hold promise for a whole array
of imaging techniques that can contribute to hydrogeologic surveys. Infrared imagery, which
records differences in apparent surlace temperatures, enables information on soil moisture,
groundwater circulation, and faults functioning as aquicludes to be obtained. 36 Near-infrared
imaging has outlined seepage patterns from canals. 1 One of the most interesting results of
infrared aerial imaging has been mapping coastal submarine springs---'both hot and cold-
in regions of basalt or limestone. 14 Figure 11.2.2 shows swirls of groundwater, colder than
seawater, emerging around the island of Hawaii.* Radar imagery can provide information
on the presence of moisture on or at shallow depths below ground surface. Finally, low-
frequency electromagnetic aerial surveys have outlined buried channels and zones of seawater
intrusion. 32
*The detection of submarine springs should prove most beneficial in the eastern Mediterranean area where such
springs are common in karstic limestone. One such spring 2 km offshore of Chekka, Lebanon, discharges an esti-
mated 17m3/sec at a depth of 43 m and produces potable water at the sea surface.
.:·_.,~
11.2 Remote Sensing Sll
If
Table 11.2.1 Surficial Features Identified on Aerial Photographs that Aid in Evaluating
l-
Groundwater Conditions (after Heath and Trainer, Introduction to Ground-Water Hydrology,
John Wiley, New York, 1968, and Mollard27)
[-
Topography
Appraisal of regional relief setting
Appraisal of local relief setting
Phreatopbytes and aquatic plants
Geologic land forms likely to contain relatively permeable strata
Modem alluvial terraces and floodplains
Stratified valley-fill deposits in abandoned meltwater and spillway channels
th Glacial outwash and glacial deltas
of Kames and kame-moraine complexes
es Eskerine-kame complexes
n- Alluvial fans
Beach ridges
m Partly drift-filled valleys marked by a chain of elongate closed depressions
:es Largely masked bedrock valleys cutting across modem valleys, indicated by local nonslumping of
~r- weak shale strata in valley sides
nd Local drift-filled valleys in extensive bedrock-exposed terrains
ur, Sand dunes assumed to overlie sandy glacioflm•ial sediments
!o- Lakes and streams
Drainage density of stream network
os- Localized gain or loss of streamflow
ely Nearby small perennial and intennittent lakes (e.g., lakes in outwash, elongate saline lakes in inactive
be drainage systems)
; to Perennial rivers and larger creeks in valleys having inactive floodplains
ha Small intermittent drainages (including misfit creeks in abandoned glacial spillways and meltwater
channels)
the
No defined drainage channel in former glacial spillways and meltwater channels
on
Moist depressions and seepages
ray Moist depressions, marshy environments, and seepages (significance depends on interpretation of
associated phenomena)
tich
String of alkali flats or lakes (playas, salinas) along inactive drainage systems
ure,
Salt precipitates (e.g., salt crusts), localized anomalous-looking "burnout" patches in the soil, and
~red
vegetation associated with salt migration and accumulation
s of
Springs (types tentatively inferred from aerial photographs)
d-
Depression springs (where land surface locally cuts the water table or the upper surface of the zone
:han of saturation)
tion Contact springs (permeable water-bearing strata overlying relatively impermeable strata-usually
.ow- along the sides of valleys that cut across the interface between different strata)
•ater Artesian springs occurring on undulating upland till plains (permeable water-bearing bed between
relatively impermeable ronfining beds, with enough head to discharge water at the ground surface)
Artesian springs occurring on or near the base of hillsides, valley slopes. and local scarps
Springs where the type could not be reasonably inferred from aerial photographs
Artificial water features
Wells
h Developed springs
;ti- Reservoirs
Canals
512 Chapter 11 Sulface Investigations of Groundwater
Figure 11.2.1. Tracing of

an aerial photograph
showing a strip of phreato-
phytes along the toe of an
alluvial fan in a desert
area. There should be a
good supply of shallow
groundwater along the
upslope portion of the
strip (after Mann26):
Figure 11.2.2. Infrared aerial photograph

showing submarine springs emerging along
the coast of Hila Bay, Hawaii. Dark ocean
areas contain cooler fresh water (courtesy
U.S. Geological Survey).
11.3 GEOPHYSICAL EXPLORATION

Geophysical exploration is the scientific measurement of physical properties of the earth's
crust for investigation of mineral deposits or geologic structure. 11• 18 With the discovery of oil
by geophysical methods in 1926, economic pressures for locating petroleum and mineral
deposits stimulated the development and improvement of many geophysical methods and
equipment. Application to groundwater investigations was slow because the commercial value
of oil overshadows that of water. In recent years, however, refinement of geophysical tech-
niques-as well as an increasing recognition of the advantages of the methods for groundwa-
ter study-has changed the situation. Today, many organizations concerned with groundwater
employ geophysical methods. The methods are frequently inexact or difficult to interpret, and
they are most useful when supplemented by subsurface investigations.
Geophysical methods detect differences, or anomalies, of physical properties within the
eanh's crust. Density, magnetism, elasticity, and electrical resistivity are properties most com-
monly measured; these are described in the following sections. Experience and research have
enabled pronounced differences in these properties to be interpreted in terms of geologic struc-
ture, rock type and porosity, water content, and water quality.
11.4 ELECTRICAL RESISTIVITY METHOD

The electrical resistivity of a rock fonnation limits the amount of current passing through the
fonnation when an electrical potential is applied. It may be defined as the resistence in ohms
between opposite faces of a unit cube of the material. If a material of resistance R has a cross-
sectional area A and a length L, then its resistivity can be expressed as
RA
p=- (11.4.1)
L
>f
Units of resistivity are ohm-m2/m, or simply ohm-m.
:o- Resistivities of rock formations vary over a wide range, depending on material, density,
n porosity, pore size and shape, water content and quality, and temperature. 23 There are no fixed
limits for resistivities of various rocks; igneous and metamorphic rocks yield values in the
range 102 to 108 ohm-m; sedimentary and unconsolidated rocks, 10° to 104 ohni-m. Figure
11.4.1 provides a representative guide to electrical resistivity ranges of various sediments and
rocks. In relatively porous formations, the resistivity is controlled more by water content and
quality within the formation than by the rock resistivity. For aquifers composed of unconsoli-
dated materials, the resistivity decreases with the degree of saturation and the salinity of the
groundwater. Clay minerals conduct electrical current through their matrix; therefore, clayey
formations tend to display lower resistivities than do permeable alluvial aquifers.*
Actual resistivities are determined from apparent resistivities, which are computed from
measurements of current and potential differences between pairs of electrodes placed in the
ground surface. The procedure involves measuring a potential difference between two elec-
trodes (P in Figure 11.4.2) resulting from an applied current through two other electrodes (C
in Figure 11.4.2) outside but in line with the potential electrodes. If the resistivity is uniform
in the subsurface zone beneath the electrodes, an orthogonal network of circular arcs will be
formed by the current and equipotential lines, as shown in Figure 11.4.2. The measured
tg
,../ ........... r-...
Clay
Soft shale r--" ~""""""-
Hard shale ~ "'-,
th's
foil
Till
,... /~
-
eral Sand
/ ~-
and
alue
~ch
Sandstone
--V '
/""-.~
r---
------
lwa- Porous limestone Figure 11.4.1. Representative ranges of
·ater electrical resistivity for various sediments
and Dense linestone and rocks. Values assume presence of fresh
groundwater; saline water will shift values
10-1
at least an order of magnitude to the left
lthe Resistivity (ohm-meters) (after Amer. Soc. Civil Engrs. 2).
:om-
have
true- *Clay and till when wet typically have low resistivities of 5-30 ohm-m, whereas wet sand and gravel have resistivi-
ties five to ten times higher; therefore, relatively high resistivity zones are of interest as shallow aquifers.
Figure 11.4.2. Electrical circuit for resistivity

determination and electrical field for a homoge-
neous subsurface stratum.
potential difference is a weighted value over a subsurface region controlled by the shape of the
network. Thus, the measured current and potential differences yield an apparent resistivity over
an unspecified depth. If the spacing betw~en electrodes is increased, a deeper penetration of
the electric field occurs and a different apparent resistivity is obtained. In general, actual sub-
surface resistivities vary with depth; therefore, apparent resistivities will change as electrode
spacings are increased, but not in a like manner. Because changes of resistivity at great depths
have only a slight effect on the apparent resistivity compared to those at shallow depths, the
method is seldom effective for determining actual resistivities below a few hundred meters.
Electrodes consist of metal stakes driven into the ground.* In practice, various standard
electrode spacing arrangements have been adopted; most common are the Wenner and
Schlumberger arrangements.
The Wenner43 arrangement, shown in Figure 11.4.3a, has the potential electrodes located at
the one-third points between the current electrodes. The apparent resistivity is given by the ratio
of voltage to current times a spacing factor. For the Wenner arrangement, the apparent resistivity
v
Pa =2na-
I (11.4.2)
where a is the distance between adjacent electrodes, Vis the voltage difference between the
potential electrodes, and I is the applied current.
The Schlumberger arrangement, shown in Figure 11.4.3b, has the potential electrodes
close together. The apparent resistivity is given by
(11.4.3)
*For the potential electrodes, porous cups filled with a saturated solution of copper sulfate are sometimes employed
to inhibit electric fields from forming around them.
I
c
~ lc
t 1~ I
J
t.~-------------L------------~·1
(a) (b)
Figure 11.4.3. Common electrode arrangements for resistivity determination. (a) Wenner43 (b) Schlumberger10
where L and b are the current and potential electrode spacings, respectively. Theoretically, L
» b, but for practical application good results can be obtained if L 2: 5b. 10
When apparent resistivity is plotted against electrode spacing (a for Wenner, and U2 for
Schlumberger) for various spacings at one location, a smooth curve can be drawn through the
points. The interpretation of such a resistivity-spacing curve in terms of subsurface conditions
is a complex and frequently difficult problem. The solution can be obtained in two parts: (1)
interpretation in terms of various layers of actual (as distinguished from apparent) resistivities
and their depths, and (2) interpretation of the actual resistivities in tenns of subsurface geologic
and groundwater conditions. Part 1 can be accomplished with theoretically computed
ity resistivity-spacing curves of two-, three-, and four-layer cases for various ratios of resistivi-
oge- ties. 5 Curves and explanations of curve-matchlng techniques have been published for the
Wenner configuration31 • 42 and the Schlumberger configuration. 10• 35 Part 2 depends on sup-
plemental data. Comparing actual resistivity variations with depth to data from a nearby
Jfthe logged test hole enables a correlation to be established with subsurface geologic and ground-
·over water conditions. This information can then be applied for interpretation of resistivity meas-
on of urements in surrounding areas.
, sub- Figure 11.4.4 illustrates the interpretation of a two-layer situation from measurements
trade with the Schlumberger electrode spacings. The field curve, plotted on logarithmic transparent
epths paper to the same scale as published master curves, is superposed on the two-layer master set.
s, the By keeping the coordinate axes parallel, the sheet is moved until a best fit of the field and the-
ers. oretical curves is obtained. The abscissa of the cross, which is the origin of the theoretical
ndard curve, equals the thickness of the first layer, while the ordinate of the cross defines the actual
rand resistivity p1 of the first layer. The asymptote of the end of the curve with the largest spacing
defines the actual resistivity p2 of the second layer. Physically, such a curve might represent a
1ted at clay layer overlying a sandy aquifer at a depth of 14m.
~ratio
;tivity
1.4.2) Asymptote
p 2 = 100 Um
~---
en the
Theoretical
trodes curve
Observed
curve
p 1 =200m
ll.4.3)
Figure 11.4.4. Interpretation of a two-
layer electrical resistivity measurement
5 10 20 50 100 200 500 1000 from Schlumberger electrode spacings
1ployed
Electrode spacing, L/2 (meters) (after Zohdy et a1.44).
Gravelly clay I Gravel I Clay Gravel I Clay I Gravel

Figure 11.4.5. Horizontal profile by
surface resistivity measurements over a
100 200 300 400 500 shallow gravel deposit in California and
Horizontal distance (meters) its interpretation (after Zohdy et al. 44).
Resistivity surveys can cover vertical variations (soundings) at selected locations by vary-
ing electrode spacings." More generally, they are conducted to obtain horizontal profiles of
apparent resistivity or apparent-resistivity maps of an area by adopting a constant electrode
spacing. Figure 11.4.5 shows a horizontal resistivity profile across a shallow gravel deposit
together with its geologic interpretation. Areal resistivity changes can be interpreted in terms
of aquifer limits and changes in groundwater quality, whereas sounding surveys may indicate
aquifers, water tables, salinities, impermeable formations, and bedrock depths.
Any factors that disturb the electric field in the vicinity of electrodes may invalidate the
resistivity measurements. These may include lateral geologic inhomogeneities; in addition,
buried pipelines, cables, and wire fences are common hazards. Advances to these methods
have been reported by Radstake et al.36 and Goyal et al. 14
Of all surface geophysical methods, electric resistivity has been applied most widely for
groundwater investigations. Its portable equipment and ease of operation facilitate rapid meas-
urement. The method frequently aids in planning efficient and economic test drilling pro-
grams.33 It is especially well adapted for locating subsurface saltwater boundaries because the
decrease in resistance when salt water is encountered becomes apparent on a resistivity-spac-
ing curve. Where subsurface conditions are relatively homogeneous, the technique can be
employed to detect the water table as the top of a relatively conductive layer. ln California,
locations of highly permeable zones for groundwater recharge were aided by resistivity meas-
urements.33 The method has also been employed for delineating geothermal areas and esti-
mating aquifer permeability. 17• 23
An important new application of resistivity surveys involves defining areas and magni-
tudes of polluted groundwater. Results correlate best with groundwater samples where a highly
conductive pollutant, such as soluble salt, is moving in a relatively shallow zone with uniform
geologic conditions. 7 Studies of pollution from landfills, wastewater disposal, industrial
wastes, and acid mine drainage have demonstrated the feasibility of the technique.
"It is often assumed lhal a given electrode spacing represents the deplh of resistivity measurement. Although this
rule of lhumb is untrue, the greater the current electrode separation, the greater the amount of currentlhat penetrates
a given depth 46
11.5 SEISMIC REFRACTION METHOD

The seismic refraction method involves the creation of a small shock at the earth's surface,
either by the impact of a heavy instrument or by a small explosive charge, and measuring the
time required for the resulting sound, or shock, wave to travel known distances. Seismic waves
follow the same laws of propagation as light rays and may be reflected or refracted at any inter-
face where a velocity change occurs. Seismic reflection methods provide infonnation on geo-
logic structure thousands of meters below the surface, whereas seismic refraction methods--of
interest in groundwater studies-go only about 100 meters deep.34 The travel time of a seismic
wave depends on the media through which it is passing; velocities are greatest in solid igneous
rocks and least in unconsolidated materials.
Characteristic seismic velocities for a variety of geologic materials are shown in Figure
11.5.1; these can be employed to identify the nature of alluvium or bedrock. In coarse alluvial
materials, seismic velocity increases markedly from unsaturated to saturated zones; conse-
a quently, the depth to water table can be mapped, often to an accuracy of 10 percent, where
nd
I.
ary-
s of
·ode
tOSit
:nns
cate
~the
tion,
h.ods
y for
leaS-
pro- 50 100 200 500 1000 5000
e the Velocity (meters/sec)
;pac- (a}
mbe
mua,
neas- Topsoil, organic material
. esti-·
tagru-
tighly
iform Compacted clay,
tstrial Sandstone, graywacke, conglomerate
Shale, tillite, argillite
Limestone, dolom
Weathered, fractured rocks, all types
Figure 11.5.1. Seismic velocity of geologic

50 100 200 500 1000 2000 5000 materials. (a) Unsaturated materials
this Velocity (meters/sec)
netrates (b) Saturated materials (after Amer. Soc.
(b) Civil Engrs.2)
geologic conditions are relatively uniform. Changes in seismic velocities are governed by
changes in elastic properties of the formations. The greater the contrast of these properties, the
more clearly the formations and their boundaries can be identified. In sedimentary rocks, the
texture and geologic history are more important than the mineral composition. Porosity tends
to decrease wave velocity, but water content increases it.
For consolidated formations with a uniform distribution of small pores, such as a sand-
stone, velocity and porosity can be related bi 9 · 41
1 a 1-a
-=-+-- (11.5.1)
v VL Vs
where vis the measured velocity, vL is the velocity in the liquid saturating the rock," v 5 is the
velocity of the solid rock matrix, and o: is porosity of the rock.
A spherical wave expands outward from a shock point, as shown in Figure ll.5.2a. It
travels at a speed governed by the material through which it is passing. Assume, for example,
a homogeneous unconsolidated material with a water table; when the wave reaches the water
table it will travel along the interface. As it travels, a series of waves is propagated back into
the unsaturated layer. Positions of the wave front drawn at intervals of a few milliseconds in
Figure l1.5.2a illustrate this refraction. At any location on the surface, the first wave will arrive
either directly from the shot point or from a refracted path. By measuring the time interval of
the first arrival at varying distances from the shot point, a time-distance graph can be plotted.
For the horizontal two-layer case, the depth H to the water table can be computed from the
velocities v1 and v2 and the distance s to the intersection on the graph, as shown in Figure
11.5.2b. The equation is
H =!.._ v2- vi
(11.5.2)
2 v2 + v1
s is the distance from the shockpoint to the point at which the direct wave and the refracted
wave arrive simultaneously.
Alternatively, the depth to the lower layer, H, can be determined using
Sl
p
(11.5.3)
where Ti is the intercept time, determined by projecting the second-line segment (for v2 = 2,000
m) backward to the time axis.
Multilayered problems can be solved in a similar manner, often aided by nomographs.
Different surface elevations, sloping formations, faults, and changes in the interfacial config-
uration require special analysis. 34• 44 Computational procedures are described in textbooks of
geophysics. 11• 16• 18
For a three-layered seismic refraction case with v1 < v2 < v3, the thickness of the first layer,
H 1, is computed using Equations 11.5.2 or 11.5.3, and the thickness of the second layer, H2, is
computed using
(11.5.4)
The field procedure for seismic refraction investigations has been simplified with the help
of compact and efficient instruments. A small charge of dynamite is placed in a hand-augered
'Seismic velocity in water under typical groundwater conditions approximates I,460 m/s.
Ground surface Shock point { 10 milliseconds
\ 5 5 10m
Unsaturated
.d-
.1)
the Saturated
.It 25 milliseconds
)le,
1ter (a)
nto
sin
nve
1! of
.ted.
the
?;Ure
.5.2)
ICted Distance (meters)

Ground surface
Shock/·····l:ilill-llllll!iiili~lllli:ii!IDiiil••.-
point
.5.3)
Figure 11.5.2. Seismic refraction method
applied to detennine depth to water table.
2,000 (a) Wave front advance (b) Time-distance
(b) graph
·aphs.
lnfig- hole about one meter deep, and the hole is backfilled. Seismometers, also known as geophones,
)ks of detectors, or pickups, are spaced in a line from the shock point 3 to 15 m apart. They receive
the shock wave and convert the vibration into electric impulses. An electric circuit connects the
layer, seismometers to an amplifier and a recording oscillograph, which automatically records the
H2,is instant of firing and the various first arrivals of the shock wave. Depth detenninations to 60 to
100m are typical with this equipment, although satisfactory work to depths of 300 m has been
accomplished. For investigations of depths less than about 20m, a sledgehammer blow on the
ll.5.4) ground surface can produce a recordable shock wave.
Interpretation of seismic refraction data assumes homogeneous layers bounded by inter-
facial planes. Where no distinct boundary exists, but rather a gradual transition zone, a curve
1e help replaces the break in slope on the time-distance graph. Fortunately, water tables approximate
ugered planes, so that many of the problems imposed by irregular configurations of geologic structure
are avoided. 12 Efficient application of the method requires skill in proper interpretation in
tenus of rock materials, depths, and irregularities. Other knowledge of subsurface conditions
aids in proper analysis of field records. The actual presence of groundwater is difficult to deter- R
mine without supplemental information because velocities overlap in saturated and unsatu- I.
rated zones. Seismic velocities must increase with depth in order to obtain satisfactory results; fili
as a result, a dense layer overlying an unconsolidated aquifer can mask the presence of the no.
aquifer. 2.
The seismic refraction method when applied in applicable areas, can rapidly and eco- En;
nomically eliminate areas unfavorable for test drilling. It is not readily adapted to small areas. 3.
Mi1
Minimum distances of a few hundred meters are needed for seismic profiles in different direc-
4.
tions. Local noise or vibrations from sources such as highways, airports, and construction sites
for
interfere with seismic work. A~
The seismic method, because it requires special equipment and trained technicians for 5. l
operation and interpretation, has been applied to only a relatively limited extent for ground- ing·
water investigations. It is commonly employed to map cross sections of alluvial valleys so that 6. l
variations in thickness of unconfined aquifers can be determined. The successful work of ll. J
Linehan and Keith26 in locating groundwater supplies in New England provides notable exam- Chr
ples of the applicability of the seismic method. Lateral seismic velocity variations in unsatu- 7. I
on l
rated sediments can represent lithology differences that correlate with water well yields.
Geo
8. (
11.6 GRAVITY AND MAGNETIC METHODS cau:
dow
The gravity method measures differences in density on the earth's surface that may indicate 9. (
geologic structure. Because the method is expensive and because differences in water content War,
in subsurface strata seldom involve measurable differences in specific gravity at the surface, 10.
the method has little application to groundwater prospecting. Under special geologic condi- Sout
Hag1
tions, such as a large buried valley, the gross configuration of an aquifer can be detected from
1!.
gravity variations.
McC
The magnetic method enables magnetic fields of the earth to be mapped. Indirect infor-
12. I
mation pertinent to groundwater studies, such as dikes that form aquifer boundaries or limits VIUID
of a basaltic flow, has been obtained using the method.44 13. I
!mag
14. (
modi
PROBLEMS v.29
15. (
11.2.1 Perform an Internet search to determine the latest 11.5.3 Given the first arrival time results for a two-layer seismic
Geop
advances made in remote sensing. refraction problem, explain how to determine v1, v2, the angle of
refraction, and H. 16. c
11.4.1 Refer to the references by Radstake et a!. 36 and Goyal et and(
a1. 14 to determine what advances have been made in the area of 11.5.4 Refraction seismology has been very successful in the
17. r
electrical resistivity methods. delineation of bedrock beneath a landfill. Study "A Refraction inNe
11.4.2 Perform an Internet search to determine what advances Statics Method for Mapping Bedrock" and write a summary of the
18. f.
have been made in electrical resistivity methods. methodology.40 calP1
11.5.1 Perform an Internet search to determine what advances 11.6.1 Gravity surveys have been used to delineate faults in 19. H
have been made in seismic refraction methods. buried bedrock valleys. Study ''The Successful Use ofMicrograv- tion l
11.5.2 For a three-layer seismic problem with the first arrival ity Profiling to Delineate Faults in Buried Bedrock Valleys" and 1380-
times measured in both the forward and reverse directions, explain write a summary of the methodology. 1 20. IC
how to determine the seismic velocities for each layer to find the 11.6.2 Gravity measurements have been used to determine the Amer.
thickness of the layers and the critical angle between layer 1 and specific yield of an alluvial aquifer. Study "Measurements of 21. In
layer2. Aquifer-Storage Change and Specific Yield Using Gravity Sur- Proc.
veys" and write a surrunary of the methodology. 35 Resou
References 521
~ter REFERENCES
atu- L Allen, D. M., and F. A. Michel, The successful use of microgravity pro- 22. Keller, G. Y., and F. C. Frischknecht, Electrical Methods in Geo-
ults; ftling to delineate faults in buried bedrock valleys, Ground Water, v. 34. physical Prospecting, Pergamon, Oxford, 517 pp., 1966.
: the no. 6, pp. 1132-1140, 1996. 23. Kelly, W. E., Geoelectric sounding for estimating aquifer hydraulic
2. American Soc. Civil Engrs., Ground Water Management , Manual conductivity, Ground Water, v. 15, pp. 420-425, 1977.
Engmg. Practice 40, New York, 216 pp.• 1972.
eco- 24. Lattman, L. H., and Parizek. R. R., Relationships between fracture
3. Avery, T. E.• lnJerpretation ofAerial Photographs, Burgess Publishing, traces and the occurrence of groundwater in carbonate rocks, Jour. Hydrol-
reas.
Minneapolis, 392 pp.• 1977. og~ v. 2,pp. 73-91,1964.
tree-
4. Benson, R., R. A. Glaccum, and M. R. Noel, Geophysical Techniques 25. Linehan, D., and S. Keith, Seismic reconnaissance for ground-water
sites for Sensing Buried Wastes and Waste Migration, National Water Well development. Jour: New England Water Works Assoc., v. 63, pp. 76-95,
Association, Dublin, OH, 1988. 1949.
s for 5. Bhattacharya, P. K., and H. P. Patra, Direct Current Geoelectric Sound· 26. Mann, J. F., Jr., Estimating quantity and quality of ground water in dry
und- ing-Principles and Interpretation, Elsevier, Amsterdam, 135 pp., 1968. regions using airphotos,/nt/. Assoc. Sci. Hydrology Pub/. 44. pp. 125-134,
>that 6. Bowden, L. W., and E. L. Pruit. eds., Manual of Remote Sensing, Vol. 1958.
rk of II, Interpretation and Applications, Amer. Soc. Photogrammetry, Falls 27. Mollard, J. D., The role of photo-interpretation in finding groundwa-
xam- Church, VA, pp. 869-2144, 1975. ter sources in Western Canada, Proc. 2nd Seminar on Air Photo loterpre-
7. Bugg, S. F., and J. W. Lloyd, A study of fresh water lens configuration tation in the Development of Canada, The 'Queen's Printer, Onawa, pp.
satu-
on the Cayman Islands using resistivity methods, Quanerly Jour. Engmg. 57-75, 1968.
Gtol., v. 9, pp. 291-302, 1976. 28. Mooney, H. M., and W. W. Wetzel, The Potentials About a Point
8. Chadwick, D. G., and L. Jensen, The detection of magnetic fields Electrode and Apparent Resistivity Curves for a Two-, Three-, and Four-
caused by groundwater and the correlation of such fields with water lAyered Eanh, Univ. Minnesota Press, Minneapolis, 146 pp. + set of
dowsing, Utah Water Research Lab., Logan, 57 pp., Jan. 1971. curves, 1956.
licate 9. Chikishev, A. G., ed., Plant Indicators ofSoils, Rocks, and Subsurface 29. Morley, L. W., ed., Mining and groundwater geophysics/1967, Ecoo.
mtent Waters, Consultants Bureau, New York, 210 pp., 1965. Geol. Rept. 26, Geological Survey of Canada, Onawa, 722 pp., 1970.
rface, 10. Compagnie Generale de Geophysique, Master Curves for Electrical 30. Morris, D. B., The application of resistivity methods to ground-water
:ondi- Sounding, 2nd rev. ed., European Assoc. Exploration Geophysicists, The exploration of alluvial basins in semi-arid areas, Jour. lnstn. Water Engrs.,
Hague, Netherlands, 49 pp., 1963. v. 18, pp. 59-65, 1964.
.from
11 . Dobrin. M. B., Introduction to Geophysical Prospecting, 3rd ed., 31. Musgrave, A. W., ed., Seismic Refraction Prospecting, Soc. Explor.
McGraw-Hill, New York. 630 pp., 1976. Geophysists, Tulsa, 604 pp.• l967. ·
infor- 12. Duguid, J. 0., Refraction determination of water table depth and allu- 32. Orellana, E., and H. M. Mooney, Master Tables and Curves for Veni-
limits vium thickness, Geophysics, v. 33, pp. 481-488, 1968. cal Electrical Sounding over Layered Structures, Interciencia, Madrid,
13. Fischer. W. A., et al., Fresh-Water Spri11gs of Hawaii from Infrared 150 pp., 66 tables, 1966.
Images, U.S. Geological Survey Hydrologic Atlas 218, 1966. 33. Page, L. M., Use of the electrical resistivity method for investigating
14. Goyal, V. C., Sri Niwas, and P. K. Gupta, Theoretical evaluation of geologic and hydrologic conditions in Santa Clara County, California,
modified Weiner array for shallow resistivity exploration, Ground Water, Ground Water, v. 6, no. 5, pp. 31-40, !968.
~ 29. no.4, pp. 582-586, 1991. 34. Pluhowski, E. J., Hydrologic Interpretations Based on Infrared
15. Grant, F. S., and G. F. West, Interpretation Theory in Applied Imagery oflAng Island, New York, U.S. Geological Survey Water-Supply
;eismic Paper 2009-B, 20 pp., 1972.
Geophysics, McGraw-Hill, New York. 583 pp., 1965.
ngle of
16. Griffiths, D. H., and R. F. King, Applied Geophysics for Engineers 35. Pool, D. R., and I. H. Eychaner, Measurements of aquifer-storage
and Geologists, Pergamon, Oxford, 223 pp., 1965. change and specific yield using gravity surveys, Ground Water. v. 33,
. in the l7. Hatherton, T., et al., Geophysical methods in geothennal prospecting no. 3,pp.425-432, 1995.
]-action in New Zealand, Bull. Volcanol., v. 29, pp. 485-498, 1966. 36. Radstake, R., et al., Applications of forward modeling resistivity pro-
y of the files, Ground Water, v. 20, no. I, pp. 13-17, 1991.
18. Hearst, J. R , P. H. Nelson, and F. L. PaiUet, Well Logging for Physi·
cal Properties, 2nd ed., John Wiley & Sons, New York, 2000. 37. Ray, R. G., Aerial Photographs in Geologic lnterpretation and Map·
lulls in 19. Howe, R. H., et al., Application of air photo interpretation in the loca- ping, U.S. Geological Survey Professional Paper 373, 227 pp., 1960.
:rograv- tion of ground water, Jour. Amer. Water Works Assoc., vol. 48, pp. 38. Setzer, J.. Hydrologic significance of tectonic fractures detectable on
ys" and 1380-1390, 1956. airphotos, Ground Water. v. 4, no. 4, pp. 23-27, 1966.
20. 1dso, S. B., et al., Detection of soil moisture by remote surveillance, 39. Stollar, R., and P. Roux, Earth resistivity surveys-A method for
tine the Amer. Sci., vol. 63, pp. 549-557, 1975. defining ground-water contamination. Ground Water, v. 13. no. 2, pp.
tents of 21. Inland Waters Branch, Instrumentation and observation techniques, 145-150, 1975.
ity Sur- Proc. Hydrology Symposium no. 7, Dept. of Energy, Mines and 40. Taucher, P., and B. N. Fuller, A refraction statistics method for map-
Resources, Ottawa, 343 pp.. 1969. ping bedrock, Ground Water. v. 32, no. 6, pp. 895-904, 1994.
41. Wyllie, M. R. J., The Fundamentals of Well Log Interpretation, 3rd 44. Zohdy, A. A., G. P. Eaton, and D. R. Mabey, Application of Surface
ed., Academic Press, New York, 238 pp., 1963. Geophysics to Ground- Water Investigations, U.S. Geological Survey
42. Van Nostrand, R. G., and K. L. Cook, Interpretation of Resistivity Techniques ofWater-Resources Investigations, Chap. Dl, Bk. 2, 116 pp.,
Data, U.S. Geological Survey Professional Paper 499, 310 pp., 1966. 1974.
43. Wenner, F., A method of measuring earth-resistivity, Bull. Bureau
Standards, v. 12, Washington, DC, pp. 469-478, 1916.
12.:
ace
vey
Chapter 12
pp.,
Subsurface Investigations of
Grotindwater
Detailed and comprehensive study of groundwater and conditions under which it occurs can
be made only by subsurface investigations. Whether the information needed concerns an
aquifer (its location, thickness, composition, permeability, and yield) or groundwater (its loca·
tion, movement, and quality), quantitative data can be obtained from subsurface examinations.
It should be emphasized that all work classed as subsurface investigations is conducted entirely
by personnel on the surface who operate equipment extending underground. Test drilling fur·
nishes information on substrata in a vertical line from the surface. Geophysical logging tech-
niques provide information on physical properties of geologic formations, water quality, and
well construction. Books dealing with the topic of subsurface investigations include those by
Bassiouni 1, Heath et al., 14 Kearey and Brooks, 18 Keys, 19· 20 Vogelsang,44 and Ward.46
12.1 TEST DRILLING

Test drilling of small-diameter holes to ascertain geologic and groundwater conditions is use-
ful in verifying other means of investigation and to obtain assurance of underground conditions
prior to well drilling. Many times, if a test hole proves fruitful, it is redrilled or reamed to a
larger diameter to form a pumping well. Test holes also serve as observation wells for meas-
uring water levels or for conducting pumping tests.
Almost any well-drilling method can be employed for test drilling; however, in unconsol-
idated formations, cable tool and hydraulic rotary methods are most common (see Chapter 5).
The former is slower but provides more accurate samples from the bailer; the latter is faster,
but it is sometimes difficult to determine the exact character of the formations. This fact is par-
ticularly true where fine-grained materials are encountered because these mix with the drilling
fluid. For great depths and fairly uniform sands, the hydraulic rotary method is quicker and
cheaper. Accurate samples can be obtained by pulling the drill stem and using a sampler at the
bottom of the hole, or if intact cores are desired in exploratory drilling, a hollow-stem cutter
head can be affixed to cut cylindrical cores. For test holes in soft ground and shallow depths,
drilling with an auger is quick and economic. Jetting has proved to be an economic method of
of
drilling shallow, small-diameter holes for investigational purposes. The rapidity the jetting
operation combined with its lightweight portable equipment gives it important advantages, but
the lack of good samples is a disadvantage. The choice of method for test drilling depends on
what information is necessary, the type of material encountered, drilling depth, and location.
523
524 Chapter 12 Subsurface Investigations of Groundwater
12.1.1 Geologic Log

A geologic log is constructed from sampling and examination of well cuttings collected at fre-
quent intervals during the drilling of a well or test hole. Such logs furnish a description of the
geologic character and thickness of each stratum encountered as a function of depth, thereby
enabling aquifers to be delineated. Figure 12.1.1 shows a geologic log for a well in unconsol-
idated alluvium.
Considering all types of logs, the geologic log is probably the most important, but prepa-
ration of a good geologic log can be difficult. One problem is that well cuttings are small and
mixed with mud. Particularly in rotary drilling, the drilling mud masks the presence of mate-
rial in the silt and clay ranges. Another problem stems from the fact that most geologic logs are
prepared by well drillers, who are busy with many activities during drilling operations and
sometimes lack formal geologic training; therefore, logs often tend to be prepared in a per-
functory manner. But experienced drillers, who recognize the value of geologic logs, make
diligent efforts to prepare them carefully and completely.
It is good practice to store samples of well cuttings systematically. These not only permit
detailed geologic logs to be prepared but also enable grain size analyses and correlations with
other nearby wells to be made after drilling is finished. Figure 12.1.2 shows well cuttings col-
lected by a bucket sampler during drilling; they have been arranged in order on the ground, 12.:
bagged, and labeled for storage.
Depth
(meters) Material
o. . . . . .-_,
Top soli and slit
Fine sand
Figure 12.1.1. Geologic

log of a well prepared
from well cuttings.
12.1 Test Drilling 525
e-
Je
JY
>1-
la-
nd
te-
are Figure 12.1.2. Well curtings,
lild taken at regular intervals from a
rotary-drilled well in Joufrah,
Libya, being tagged and labeled
for subsequent storage in com-
partmented boxes (photo by
mit David K. Todd).
vith
:ol-
md, 12.1.2 Drilling-Time Log
A drilling-time log is a useful supplement to test drilling. It consists of an accurate record of
the time, in minutes and seconds, required to drill each unit depth of the hole.23 The technique
is most practical with hydraulic rotary drilling although it is applicable to other methods as
well. Because the texture of a stratum being penetrated largely governs the drilling rate, a
drilling-time log may be readily interpreted in terms of fonnation types and depths. A portion
of a drilling time Jog obtained by the hydraulic rotary method is shown in Figure 12.1.3
together with the Jog of the test hole based on cuttings.
0 1.5
68
Pebbly clay
Muddy sand
I
Coarse clean sand
Figure 12.1.3 Drilling-

eologic time log and strata pene-
~ed trated (after Kirby23;
s. courtesy Johnson Div.,
UOP Inc.).
12.2 WATER LEVEL MEASUREMENT

One of the most common measurements in groundwater investigations is the determination of
the depth to groundwater. In both existing and new wells, such data are needed to define
groundwater flow directions, changes in water levels over time, and effects of pumping tests.
A simple and accurate method for obtaining water depth is lowering a steel tape into a
well. By adding chalk to the end of the tape, the length of submersion becomes apparent, thus
giving the distance from the top of the well to the water surface.
For repeated measurements and for depths exceeding 50 m, an electric water-level
sounder is preferable. A sounder consists of a battery, a voltmeter, a calibrated two-wire cable
on a reel, and an electrode. When the electrode contacts water, the circuit is completed and the
voltmeter shows a deflection. The depth is read directly from graduations along the cable.
Another widely employed technique is the air-line method. 11 A small-diameter tube is 1
placed in the annular space between the pump column and the casing. Commonly, the tube
is fastened to the pump column so that the two are installed simultaneously. The tube extends
below the water surface and is connected to a tire pump or small air compressor with a pres-
sure gauge. Air is pumped into the tube until a maximum pressure is observed; this pressure,
converted to depth of water, indicates the distance from the lower end of the tube to the
water surface. Although the air-line method is less accurate than the above methods, it is espe-
cially applicable in pumped wells where water splash and turbulence may invalidate other
techniques. ·
A unique and convenient method for measuring water levels in deep wells is the rock tech-
nique developed by Stewart. 41 He determined empirically the time required for a common Li
1.55-cm glass marble or a standard BB (air rifle shot) to fall to the water surface plus the time
for the sound of the splash to return to ground surface. Measuring the elapsed time by stop- Tc
watch, the depth to water can be read directly from Table 12.2.1. For water depths exceeding
57 m, the sphere reaches a constant terminal velocity; therefore, for depths greater than those Ef
listed in Table 12.2.1, the equation Cl
Pe
d=27.3t-47.6 (12.2.1)
Se
Sp
Gr
Table 12.2.1 Depth to Water Surface in a Well as a Function of the Time Interval of a Falling Lo
Sphere (after Stewart41 )
Me
Time Distance Time Distance Time Distance
lnf
(s) (m) (s) (m) (s) (m)
Di~
0.0 0.0 1.4 9.6 2.8 33.3
0.1 0.0 1.5 11.0 2.9 35.4 Sot
0.2 0.2 1.6 12.4 3.0 37.6 Cht
0.3 0.4 1.7 13.9 3.1 39.8 sa
0.4 0.8 1.8 15.3 3.2 42.1 Cor
........................................................ .............. " ................................................ .........................."'''''''''''''''''''''''''''''-''''''''''-'''''''''''''''''''"'''''''''''''''''''''''' .................................................................
of
0.5 1.2 1.9 16.8 3.3 44.4
Gui
0.6 1.8 2.0 18.4 3.4 46.9
0.7 2.4 2.1 20.0 3.5 49.2 Cen
0.8 3.1 2.2 21.7 3.6 51.6
0.9 4.0 2.3 23.5 3.7 54.1 Casi
1.0 4.9 2.4 25.4 3.8 56.7 Casi
1.1 5.9 2.5 27.3 3.9 59.1
1.2 7.1 2.6 29.3 4.0 61.6
1.3 8.3 2.7 31.3 4.1 64.3
12.3 Geophysical Logging 527
can be employed where dis the depth to water in meters and tis the time interval in seconds.
The method is accurate to within 1.5 m.*
of Automatic water level recorders or pressure transducers are installed in observation welts
ne where short-term fluctuations are of interest, such as near intennittently operating wells or for
ls.
pumping tests. A typical recorder consists of a float and counterweight, a gear linkage that
,a
rotates a chart drum, and a recording pen driven across the chart by a clock mechanism.
iUS
Where multiple aquifers exist with differing water levels, individual observation wells
screened in only one aquifer are often drilled. Alternatively, individual small piezometer tubes
vel extending down to the levels of the various aquifers are placed inside a large single perforated
ble casing. The casing is backfilled with sand and sealed by grout between adjoining aquifers.
the
12.3 GEOPHYSICAL LOGGING
~ is
Jbe Geophysical logging involves lowering sensing devices in a borehole and recording a physi-
nds cal parameter that may be interpreted in terms of formation characteristics; groundwater
:es- quantity, quality, and movement; or physical structure of the borehole.21 · 31 A wide variety of
Jre, logging techniques are available; Table 12.3.1 lists the types of information that can be
the obtained from various logging techniques described in this chapter.
;pe-
ther Table 12.3.1 Summary of Logging Applications to Groundwater Hydrology (after Keys and MacCrar/ 1)
?ch- Required information Possible logging techniques

non Lithology and stratigraphic correlation of aquifers Resistivity, sonic, or caliper logs made in open holes; radiation
ime and associated rocks logs made in open or cased holes
top- Total porosity or bulk density Calibrated sonic logs in open holes; calibrated neutron or
ding gamma-gamma logs in open or cased holes
hose Effective porosity or true resistivity Calibrated long-normal resistivity logs
Clay or shale content Natural gamma logs
Penneability Under some conditions, long-normal resistivity logs
.2.1) Secondary permeability-fractures, solution openings Caliper, sonic, or television Jogs
Spedic yield of unconfined aquifers Calibrated neutron logs
Grain size Possible relation to formation factor derived from resistivity logs
Location of water level or saturated zones Resistivity, temperature, or fluid conductivity logs; neutron or
gamma-gamma logs in open or cased holes
Moisture content Calibrated neutron logs
Inflltration Time-interval neutron logs
Dispersion, dilution, and movement of waste Fluid conductivity or temperature logs; natural gamma logs
for some radioactive wastes
Source and movement of water in a well Fluid velocity or temperature logs
Chemical and physical characteristics of water, including Calibrated fluid conductivity or temperature logs;
salinity, temperature, density and viscosity resistivity logs
Construction of existing wells, diameter and position Gamma-gamma, caliper, casing, or television logs
of casing, perforations, screens
Guide to screen setting All logs providing data on the lithology, water-bearing
characteristics, and correlation and thickness of aquifers
Cementing Caliper, temperature, or gamma-ganuna logs; acoustic logs
for cement bond
Casing corrosion Under some conditions, caliper, casing, or television logs
Casing leaks and/or plugged screen Fluid velocity logs
'Table 12.2.1 applies only to the two specified spheres; ordinary pebbles give erratic results because of their irregu-
lar shapes. Stewart41 also pointed out that BBs make a sharp "ping" sound, while marbles cause a short "blurred"
- ··.:
sound. Furthennore, he noted, BBs are deflected by spiderwebs, bul marbles are not.
Geophysical logs furnish continuous records of subsurface conditions that can be corre-
lated from one well to another. 16 They serve as valuable supplements to geologic logs. Data
from geophysical logs can be digitized, stored on magnetic tape, or transmitted by radio or
telephone for interpretation. Graphic displays of log data permit rapid visual interpretations
and comparisons in the field so decisions regarding completion and testing of wells can be
made immediate! y.
The application of geophysical logging to groundwater hydrology lags far behind its com-
parable use in petroleum exploration. It is doubtful if more than a few percent of the new water
wells drilled each year are logged by geophysical equipment. The primary reason for this is
cost. Most water wells are shallow, small-diameter holes for domestic water supply; logging
costs would be relatively large and usually unnecessary. But for deeper and more expensive
wells, such as for municipal, irrigation, or injection purposes, logging can be economically jus-
tified in terms of improved well construction and performance.
Another deterrent to geophysical logging is the lack of experience among drillers, engi-
neers, and geologists in the interpretation of logs. As logging techniques become more sophis-
ticated, the data they produce become more complex. The interpretation of many logs is more
of an art than a science; log responses are governed by numerous environmental factors, mak-
ing quantitative analysis difficult. In general, best results are obtained with experience and with
supplemental hydrogeologic information.
In the following sections, the geophysical logging techniques that are most important in
groundwater hydrology are described. Emphasis is on concepts and applications. Figure 12.3 .1
is a schematic diagram showing several of the logs and their typical relative responses in var-
ious unconsolidated and consolidated geologic formations.
Spontaneous
potential Caliper Resistivity Gamma
SP K RES GR
- + ____. ~
Figure 12.3.1. Schematic A
~
____. ~
diagram of various geophys-

~ Values increase ~ Values increase
icallogs showing their rela-
1 Medium-grained sand 5 Silty fine sand 1 Claystone 5 Sandstone tive responses in (a) uncon-
2 Boulder clay (till} 6 Brown coal 2 Marly limestone 6 Sandstone solidated rocks and (b) con-
3 Coarse sand 7 Clay 3 Limestone 7 Sandstone
4 Fine sand 8 Clayey silt 4 Marlstone 6 but with solidated rocks (after Brown
salt water et al.,6 reproduced by per- Fi
(a) (b) mission of UNESCO). (c
12.4 RESISTIVITY LOGGING

ta
Within an uncased well, current and potential electrodes can be lowered to measure electric
or
resistivities of the surrounding media and to obtain a trace of their variation with depth. The
ns
result is a resistivity (or electric) log. Such a log is affected by fluid within a well, by well
be
diameter, by the character of surrounding strata, and by groundwater.
Of several possible methods for measuring underground resistivities, the multielectrode
n-
method is most commonly employed, because it minimizes effects of the drilling fluid and
ter
well diameter and also makes possible a direct comparison of several recorded resistivity
is
curves. 25 Four electrodes, two for emitting current and two for potential measurement, consti-
ng
tute the system. Recorded curves are termed normal or lateral, depending on the electrode
lVe
arrangement, as shown in Figure 12.4.1. In the normal arrangement, the effective spacing is
JS-
considered to be the distance AM (Figure 12.4.1) and the recorded curve is designated AM.
Sometimes a long normal curve (AM') is recorded based on the same electrode arrangement
gi-
as the normal but with a larger AM distance (Figure 12.4.1). The spacing for lateral (AO)
liS-
curves is taken as the distance AO, measured between A and a point midway between the elec-
ore
trodes M and N (Figure 12.4.1c). Boundaries of formations having different resistivities are
ale-
located most readily with a short electrode spacing, whereas information on fluids in thick per-
lith
meable formations can be best obtained with long spacings.
An electric log of a well usually consists of vertical traverses that record the short and long
tin
normals, the lateral, and the spontaneous potential curves (see following section). An illustra-
.3.1
tion of an electric log is given in Figure 12.4.2. Accurate interpretation of resistivity logs is dif-
ficult, requires careful analysis, and is best done by specialists. 18 • 26 • 47
Resistivity curves indicate the lithology of rock strata penetrated by the well and enable
fresh and salt waters to be distinguished in the surrounding material. 22 • 24 In old wells, exact
AC gen. AC gen. AC gen.
q
l
N
B
15m
15m
15m
M
A
Reference point
--- AM'= 1.6 m
M 1.6 m Normal AO= 5.7 m
Reference point
AM= 0.4 m j
ltiC
phys-
rela-
0.4 m Normal
A A Reference point M _L
5.7 m Lateral
lCOll- N
. con- (a) (b) (c)
~rown
Jer- Figure 12.4.1 Typical electrode arrangements and standardized distances for resistivity logs. (a) Short normal (b) Long normal
(c) Lateral (after Keys and MacCrary21 )
Spontaneous potential Resistivity (ohm-meters)

(millivolts) AM 0 500 1000
AO 5000 10,000
I~A
r
meters
Depth
~
_L I~
Figure 12.4.2. Spontaneous potential

and resistivity logs of a well (courtesy
Schlumberger Well Surveying Corp.).
locat~~ms of casings can be detennined. Resistivity logs may be used to detennine specific
resistivities of strata, or they may indicate qualitatively changes of importance. As mentioned
in the previous chapter, resistivity of an unconsolidated aquifer is controlled primarily by
porosity, packing, water resistivity, degree of saturation, and temperature. Although specific
resistivity values cannot be stated for different aquifers, on a relative basis, shale, clay, and salt-
water sand give low values; freshwater sand moderate to high values; and cemented sandstone
and nonporous limestone high values (see Figure 11.4.1). Casings and metallic objects will
indicate very low resistivities. Correlation of rock samples, taken from wells during drilling,
with resistivity curves furnishes a sound basis for interpretation of curves measured in nearby
wells without available samples.
Resistivity of groundwater depends on ionic concentration and ionic mobility of the salt
solution. This mobility is related to the molecular weight and electrical charge, so that differ-
ences exist for various compounds. For example, the ion mobility of a sodium chloride solu-
tion is several times that of a comparable calcium carbonate solution. Relationships between
resistivity and total dissolved solids for several salt solutions and natural groundwaters are
shown in Figure 12.4.3.
As the temperature of groundwater increases, it has a greater ionic mobility, associated
with a decrease in viscosity. Hence, an inverse relation exists between resistivity and temper-
ature. The relation, expressed as a correction factor, is shown in Figure 12.4.4. Resistivity at
the measurement temperature when multiplied by the correction factor for that temperature
yields the resistivity at the standard temperature of 25°C.
One of the most common uses of an electric log is to detennine the proper place to set well
screens. A log provides a basis for selecting proper lengths of screens and for setting them
• •• • • • • NaHC03 solution
----- MgS04 solution
-- NaCI solution
- ·- MgCI2 solution
-- Average of natural
groundwaters
Figure 12.4.3. Resistivity-concentration

curves for various salt solutions and natural
groundwaters (after Agric. Handbook 60, U.S.
Resistivity (ohm-meters, at 25"C) Dept. of Agriculture).
ial
~sy
l.).
c:
~ 1.5
al
t 1.0
cific ()
~ned 0.5 Figure 12.4.4. Correction factor to con-

y by vert resistivities at other temperatures to
00 10 20 30 40 50 60 70 80 90 100 resistivities at 25°C (after Jones and
dfic
Temperature (°C) Buford 1\
salt-
;tone
; will
opposite the best fonnations. Because of this application, many well drillers have their own
lling, "loggers" for this purpose.
!arby Investigations of Louisiana aquifers by Jones and Buford 17 and later by Turcan42 have
extended the applicability of resistivity logs to the estimation of groundwater quality. First, a
e salt
field-formation factor F for an aquifer is detennined from previous data by
liffer-
solu- F= .£Q_ (12.4.1)
tween Pw
rs are
where Po is the resistivity of the saturated aquifer and Pw is the resistivity of the groundwater
ciated in the aquifer: Second, specific conductance is related to chloride content or total dissolved
mper-
vity at
~rature
•It should be noted that the resistivity of groundwater is the reciprocal of its specific conductance (see Chapter 7).
The relation has the fonn
Pw = I <fiE,
et well where p,. is in ohm-m and Ec is specific conductance in !iS/em.
; them
:i:
Hydrologic Conditions in Pumping Wells

Hydrologic Conditions in Nonpumping Wells Pumping drawdown overcomes effects of
Depend upon relative positions of salt and fresh-water aquifers and upon their respective heads ordinary head differences
(a) (b) (e)
Salt-water aquifer is below Salt-water aquifer is below Salt-water aquifer is below
fresh-water aquifer and has higher head fresh-water aquifer and has lower head fresh-water aquifer
L_
Head In sak-water a~er
l
AdiUSied (actual) water level
water level Head i'l lresh-water aquite<
~ust_ed level
Water level
------ Static level
Head In lresh-waler aquller Head on satt-waler aquiter Increasing Pumping level Water level
~· v
Fresh-water '----- sa5nity
~·
Fresh-water
' -+-- fresh-waler
(I)
+-- ~
r,..- i!
-- aqu~er aqu~er
~ Decreasing
aqu~er a
I ~ resisti>ity :f
I. I~ :;
~ 01)
"'
1
IJ\J Suction
1
-=-=- }\t_
:~
a;
I! Increasing <ll
~\t Salt-water a; salirity
~ 5.Jl-
<I)
.n--.
I 'I
?"' :~er .· -----+
Decreasing
I ------11 resistivity
(c) ,I (d) (I)

Salt-water aquifer is above Salt-water aquifer is above Salt-water aquifer is above
fresh-water aquifer and has higher head fresh-wa1er aquifer and has lower head fresh-water aquifer
L----__.__
Head in salt-water aqlifer

~ -..aterlevl!l
Water level
Actual watet level
I/
Head In tresh-wate< aql.ifer Water level
I ,___
'- .._,•
'"-111 n sh-water aquiler Head in san-water aquifer 1
~ ~----~~--, ~ Sa.,_ ..
- -..,' 'r< - - ~ 'v+-- 3!JIIe<
i!
i3
~
::<
- -:-</'\.
r ~ - --
Fresh-water
~~ rresh~wate(
:
<I)
Fresh-wale<
I
-+-- aquifer
aquiter aquiler +--
II
,..:;_
Figure 12.4.5. Hydrologic conditions and resistivity curves for wells penetrating two aquifers of different salinities (after Poland
and Morrison36).
12
solids for the aquifer, as in Figure 12.4.3. Finally, with these relationships known, p0 is read
from the long-normal resistivity curve in an aquifer; this enables Pw and then the salinity of the
groundwater to be calculated. The method yields best results in uniform clastic aquifers such
as sand and sandstone, consisting mainly of intergranular pores saturated with water.
Another application of long-normal curves has been suggested by Croft8 to estimate
permeability. A value ofF is determined as above; then from a previously established rela-
tionship, permeability is calculated directly from F. Again, the method should be limited to
clastic rock formations. 12.
Resistivity logs can also aid in identifying wells that intersect both fresh and saline zones.
Circulation within such a well under non-pumping conditions depends on the relative hydro-
static heads, water densities, aquifer locations and thicknesses, and the physical structure and
condition of the welL Various hydrologic conditions for pumping and nonpumping wells are
shown diagrammatically in Figure 12.4.5 together with corresponding resistivity curves.
Resistivity logs are also employed for locating aquifers, detennining bed sequences, correlat-
ing aquifers, and estimating changes in groundwater quality.
12.6 Radiation Logging 533
12.5 SPONTANEOUS POTENTIAL LOGGING

The spontaneous potential method measures natural electrical potentials found within the
earth.* Measurements, usually in millivolts, are obtained from a recording potentiometer con-
nected to two like electrodes. One electrode is lowered in an uncased well and the other is con-
nected to the ground surface, as illustrated by electrodes M and N in Figure 12.4.1. The
potentials are primarily produced by electrochemical cells formed by the electrical conductiv-
ity differences of drilling mud and groundwater where boundaries of permeable zones inter-
sect a borehole. In some instances, electrokinetic effects of fluids moving through permeable
formations are also responsible for spontaneous potentials. Therefore, potential logs indicate
permeable zones but not in absolute terms; they can also aid in determining casing lengths10
and in estimating total dissolved solids in groundwater.45 Where no sharp contrasts occur in
permeable zones, as often happens in shallow alluvial formations, potential logs lack relief and
contribute little. In urban and industrial areas, spurious earth currents may occur, such as from
electric railroads, which interfere with potential logging.
Potential values range from zero to several hundred millivolts. By convention, potential
logs are read in terms of positive and negative deflections from an arbitrary baseline, usually
associated with an impermeable formation of considerable thickness. The sign of the potential
depends on the ratio of the salinity (or resistivity) of the drilling mud to the formation water. 9
Spontaneous potentials resulting from electrochemical potentials can be expressed by
SP = -(64.3 + 0.239T) log pf (12.5.1)

Pw
) where p1 is the drilling fluid resistivity in ohm-m, Pwis the groundwater resistivity in ohm-m,
and Tis the borehole temperature in °C. Therefore, for measured SP, Pp and T values, the resis-
tivity and hence salinity of groundwater can be determined. 21• 45 It should be noted, however,
that the formula applies only where the groundwater is very saline, NaCl is the predominant
salt, and the drilling mud contains no unusual additives.
In practice, potential and resistivity logs are usually recorded together as shown in Figure
12.4.2. The two logs often indicate the same subsurface conditions and thereby supplement
i each other; however, occasionally the two types of logs will furnish information not available
directly from either alone.
12.6 RADIATION LOGGING

read Radiation logging, also known as nuclear or radioactive logging, involves the measurement of
fthe fundamental particles emitted from unstable radioactive isotopes. Logs having application to
;uch groundwater are natural gamma, gamma-gamma, and neutron. These are promising but not
widely used hydrogeologic tools. An important advantage of these logs over most others is that
nate they may be recorded in either cased or open holes that are filled with any fluid.
rela-
:d to
12.6.1 Natural-Gamma Logging
mes. Because all rocks emit natural-gamma radiation, a record of this constitutes a natural-gamma
1dr0-
log. The radiation originates from unstable isotopes of potassium, uranium, and thorium. In
~ and
general, the natural-gamma activity of clayey formations is significantly higher than that of
s are
rves.
relat-
'Potentials are also referred to as self·potentials or simply SP.
quartz sands and carbonate rocks. The most important application to groundwater hydrology
is identification of lithology, particularly clayey or shale-bearing sediments, which possess the
highest gamma intensity. 28 • 34 Because most of the gamma rays detected originate within
15-30 em of the borehole wall, logs run before and after well development can reveal zones
where clay and fine-grained material were removed. Borehole dimensions and fluid, casing,
and gravel pack all exert minor influences on gamma probe measurements.
Figure 12.6.1 shows the natural-gamma log of a test hole in unconsolidated sediments
together with its geologic interpretation.
12.6.2 Gamma-Gamma Logging

Gamma radiation originating from a source probe and recorded after it is backscattered and 1:
attenuated within the borehole and surrounding formation constitutes a gamma-gamma log.
The source probe generally contains cobalt-60 or cesium-137, which is shielded from a sodium
iodide detector built into the probe. Primary applications of gamma-gamma logs are for iden-
tification of lithology and measurement of bulk density and porosity of rocks. The porosity a
can be determined by
GeoloQic interpretation
0
, Sand and clay
3
Sand and gravel
10
11
Till
20 20
Sand and gravel
21 Till
22
Sand and gravel
24
26
m
Sand and gravel 12
34
Till, very sandy, or silt
37
Till, dark
39
40
Sand and gravel
46
Till
50
51 Sand and gravel
52
Till Figure 12.6.1. Natural-ganuna
54
Sand and gravel log of a test hole in Moraine City,
56 nu
57 Sand and gravel Ohio, together with its geologic
Increasing 59 .
interpretation (after USGS Water-
-----
60 radioactivity Bedrock, shale
61 Supply Paper 1808).
12.7 Temperature Logging 535
y
1e
(12.6. 1)
n
!S where PG is grain density (obtained from cuttings or cores), p8 is bulk density (measured from
g, a calibrated log), and PF is the fluid density. Also, within the same geologic formation it
should be possible to derive specific yield from the difference in bulk density measured above
Its and below a water table. Finally, gamma-gamma logs can assist in locating casing, collars,
grout, and zones of hole enlargement. As with natural-gamma logs, borehole and fluid condi-
tions affect readings.
12.6.3 Neutron Logging

nd
)g. Neutron logging is accomplished by a neutron source and detector arranged in a single probe,
iliD which produces a record related to the hydrogen content of the borehole environment. In most
fonnations the hydrogen content is directly proportional to the interstitial water; therefore,
neutron Jogs can measure moisture content above the water table and porosity below the water
table. 33 Neutrons have a relative mass of 1 and no electric charge; therefore, the loss of energy
when passing through matter is by elastic collisions. Neutrons are slowed most effectively by
collisions with hydrogen because the nucleus of a hydrogen atom has approximately the same
mass as a neutron. • Several designs of neutron probes are currently available, utilizing sources
of beryllium combined with radium-226, plutonium-239, or americium-241. Probes for meas-
urement of soil moisture (moisture meters) are compact and designed to fit snugly in a small-
diameter access tube for accurate quantitative results. For porosity determination in
large-diameter holes, larger probes are employed. By measuring moisture contents above and
below the water table, the specific yield of unconfined aquifers can be determined. 27 The lat-
eral penetration of neutron logs is in the range 0.2-0.6 m. Neutron log results are influenced
by hole size; therefore, in large uncased holes information on hole diameter is required for
proper interpretation.
Figure 12.6.2 shows a neutron log of a shallow well in unconsolidated alluvium together
with the geologic log. This log is calibrated in moisture content as a percentage of bulk vol-
ume. Note that the capillary fringe is apparent above the water table as well as high porosity
clay layers in the saturated zone.
12.7 TEMPERATURE LOGGING

A vertical traverse measurement of groundwater temperature in a well can be readily obtained
with a recording resistance thermometer. Such data can be of value in analyzing subsurface
conditions.3· 38 Ordinarily, temperatures will increase with depth in accordance with the geo-
thermal gradient, amounting to roughly 3°C for each 100m in depth. Departures from this nor-
mal gradient may provide information on circulation or geologic conditions in the well.4· 40
Abnormally cold temperatures may indicate the presence of gas or, in deep wells, may suggest
recharge from ground surface. Likewise, abnormally warm water may occur from water of
deep-seated origin. Temperatures may indicate waters from different aquifers intersected by a
well. In a few instances, temperature logs have aided the location of the approximate top of
new concrete behind a casing,2 because the heat generated during setting produces a marked
temperature increase of the water within the casing.
runa
1eCity,
logic
Water· 'Dynamically, the slowing of a neutron by a hydrogen atom is analogous to a golf ball losing energy upon collision
with another golf ball but rebounding elastically with little energy loss from a large mass such as a concrete wall.
Hole caliper l
Recording
in operation
panel
Unsaturated Caliper log, average
zone hole diameter, em
40 30 20
! Water
l-4-------'- table
0 r - -·-n
t Casing
Clay bed
~
Soft
!!?
(!)
water 12
"ai sand
s 20 Sand and
.I::.
a. sandy silt 4 cav1ng
.
Q
(!) ~
!!? shales
2(!)
25 s.r::.
a.
Q)
0
6 +- Liner
35
8
0 10 20 30 40 50
Moisture content, percent
Figure 12.6.2. Neutron log of a shallow well in

unconsolidated alluvium near Garden City,
Kansas, together with its geologic interpre-
tation (courtesy U.S. Geological Survey). Figure 12.8.1. Hole caliper and corresponding caliper log (after
Bays and Folk2).
12.8 CALIPER LOGGING

2
A caliper log provides a record of the average diameter of a borehole. Caliper tools are
designed either with arms hinged at the upper end and pressed against the hole wall by springs
or with bow springs fastened at both ends. These logs aid in the identification of lithology and 3
stratigraphic correlation, in the location of fractures and other rock openings, 30 and in cor-
recting other logs for hole-diameter effects. During well construction, caliper logs indicate the
size of casing that can be fined into the hole and enable the annular volume for gravel packing Figu:
to be calculated. Other applications include measuring casing diameters in old wells and locat- show
ing swelling and caving zones. A hole caliper and the resulting log are shown in Figure 12.8.1. sition
12.10 Fluid-Velocity Logging 537
12.9 FLUID-CONDUCTIVITY LOGGING

A continuous record of the conductivity of fluid in a borehole is a fluid-conductivity log. The
probe measures the AC-voltage drop across two closely spaced electrodes and is governed by
the resistivity of the fluid between the electrodes. Fluid resistivity is generally measured in
ohm-m; its reciprocal, conductivity, is measured in J.!Sicm (see Equation 12.4.2). Use of the
termfluid-conductivity log avoids confusion with a resistivity log, which measures rock and
fluid conditions outside a borehole. Temperature logs should be made in conjunction with
fluid-conductivity logs so that values can be corrected to a standard temperature.
Fluid-conductivity logs enable saline water zones to be located, furnish information on
fluid flow within a well, and provide a means to extrapolate water-sample data from a well.
Figure 12.9.1 illustrates how a fluid-conductivity log can define the location and transition
zone of saline water underlying fresh water within a well.
12.10 FLUID-VELOCITY LOGGING

Measurement of fluid movement within a borehole constitutes a fluid-velocity log. Such data
reveal strata contributing water to a well, flow from one stratum to another within a well,
hydraulic differences between aquifers intersected by a well, and casing leaks. Several
flowmeter designs have been developed for boreholes; it is important that they be compact and
sensitive to small water movements and directions. 32 Figure 12.10.1 illustrates a fluid-velocity
log for a pumping well.
/
I
I
50
Well
casing
100
Ul
Q;
Q)
~r
_§, 150
.c
a.
Q)
0 ~--Well Upper
screen aquifer
200
250
s are ------(
rings
yand Ascending
1---+-• velocity
t cor- 10,000 20,000 30,000 40,000 50,000
Specific conductance (microsiemens/cm) (a) (b)
te the
eking Figure 12.9.1. Fluid conductivity log in basalt at Honolulu, Hawaii, Figure 12.10.1. Example of a fluid-velocity log for a
locat- showing fresh water overlying saline water and separated by a tran- well tapping two confined aquifers. (a) Well
2.8.1. sition zone (courtesy Honolulu Board of Water Supply). (b) Fluid-velocity log
12.11 MISCELLANEOUS LOGGING TECHNIQUES

12.11.1 Television Logging
A convenient tool with increasing use is a television camera lowered in a well. Specially
designed wide-angle cameras, typically less than 7 em in diameter, are equipped with lights
and provide continuous visual inspection of a borehole; with videotape a record of the interior
can be preserved. Among the variety of applications are locating changes in geologic strata,
pinpointing large pore spaces, inspecting the condition of the well casing and screen, check- 1~
ing for debris in wells, locating zones of sand entrance, and searching for lost drilling toots?·
12
Photographs taken within a well at close intervals, termed a photolog, 24 can be employed for
the same purposes. As an example, photographs of cased and uncased portions of a well in
Honolulu appear in Figure 12.11.1.
12.11.2 Acoustic Logging

Acoustic, or sonic, logging measures the velocity of sound through the rock surrounding an
uncased, fluid-filled hole. Sound velocity in rock is governed by the velocity of the rock matrix
and the fluid filling the pore space (see Chapter 11); therefore, the greater the porosity, the
closer the measured sound velocity approaches that of the fluid. Chief applications of the
acoustic log include determining the depth and thickness of por:ous zones, 35 estimating poros-
ity, identifying fracture zones, and determining the bonding of cement between the casing and
the formation.
1
Th
Ge
ne<
wa
is c
gra
um
of
ch<
ter.
Oa
we
ran
cal
Oa:
og)
dUI
inc:
Figure 12.11.1. Photographs tivi
inside a water supply well in the
Honolulu, Hawaii. (a) !alE
Tuberculated condition within a res;
wrought iron casing 83 years old vie·
(b) Fractured and cavernous lava
formation below the casing 12.
(courtesy Honolulu Board of trar
Water Supply) pro
12.13 Case Study: Oasis Valley, Nevada 539
12.11.3 Casing Logging

A casing-collar locator is a useful device for recording locations of casing collars, perforations,
and screensY The instrument consists of a magnet wrapped with a coil of wire; voltage fluc-
ly tuations caused by changes in the mass of metal cutting the lines of flux from the magnet are
ts recorded to form the log.
or
:a,
k- 12.12 OTHER SUBSURFACE METHODS
7,
'or Besides the array of logging techniques described in this chapter, it should be recognized that
m
other subsurface methods can yield important information about hydrogeologic conditions.
These have been described in earlier chapters and by way of summary are listed here:
1. Tracer tests for groundwater flow (Chapter 3)
2. Groundwater level measurements for flow directions and aquifer conditions (Chapter 3)
an 3. Pumping tests of wells for aquifer characteristics (Chapter 4)
rix 4. Groundwater level fluctuation measurements for aquifer characteristics (Chapter 6)
the 5. Groundwater samples for water quality determination (Chapter 7)
the
OS-
md
12.13 CASE STUDY Oasis Valley, Nevada
This case study is reported by Robledo et al.37 In 1997, the U.S. that were identified in this borehole. Figure 12.13.3 shows the
Geological Survey installed a network of wells in the Oasis Valley geologic and geophysical logs for the borehole. The six. water-
near Beatty, Nye County, Nevada, to develop a long-term ground- bearing zones in Figure 12.13.2 were identified with the aid of
water-monitoring network in the Oasis Valley. The Oasis Valley acoustic televiewer, temperature, caliper, and heat-pulse flowme-
is of concern to the U.S. Department of Energy because it is down ter logs (Hess 19 ). The large scatter of the flowmeter log data in the
gradient from the Pahute Mesa, one of the sites of intensive interval where the flow was the greatest made it difficult to inter-
underground nuclear testing at the Nevada Test Site. The objective pret the flowmeter log. Most of the data scatter of the flowmeter
of the monitoring network is to define spatial and temporal log was caused by the rugged borehole, as shown by the caliper
changes in the chemical and isotopic character of the groundwa- log. This did not allow a good seal between aquifer zones. The
ter. Seven sites up gradient from the spring-discharge area in temperature log is nearly isothermal in the interval between 210
Oasis Valley were selected for installation of single or multiple and 340 ft, supporting the interpretation that the greatest ambient
wells. Twelve monitoring wells (in 11 boreholes) were installed, flow (both inflow and outflow) is in this interval. The flow is fast
ranging in depth from 65 to 642ft. Figure 12.13.1 show the typi- enough that it does not cool much as it travels up the borehole.
cal construction of single- and multiple-well boreholes in the The differences in inflow rates of the water production zones
Oasis Valley. are proportional to transmissivity and may therefore be used to
The boreholes were geophysically logged to define the lithol- determine relative transmissivities of the zones (Paillet29).
ogy, stratigraphy, and geohydrologic character of the rock units Amount of flow entering from a bed or zone is proportional to the
during drilling. The typical suite of logs for each borehole product of the transmissivity of the zone and the change in
included a caliper log, natural-gamma log, conductivity log, resis- hydraulic head across the borehole wall. The relative transmissiv-
~s tivity log, resistance log, and spontaneous-potential log. Some of ity for a given zone can be determined using the following equa-
n the boreholes also have short- and long-normal resistivity logs, tion that describes the change in flow, 6Q, for a given fracture
lateral logs, acoustic velocity logs, fluid temperature logs, fluid zone,
tlin a resistivity logs, fluid specific conductance logs, acoustic tele· 6Q=6h XT
rs old viewer logs, and heat-pulse flowmeter logs.
s Java A borehole video log was run in well ER-OV-06a. Table where M is the change in gradient from the fracture zone to the
12.13.1 is a summary of the inflow and outflow zones and relative borehole between ambient and injection conditions and Tis the
of transmissivities in the borehole. Figure 12.13.2 shows fl ow-
producing fractures or bedding planes in six water-bearing zones (continues)
Single-Well Borehole Multiple-Well Borehole
Measuring-point altitude
Capped and locked
well protector
Land-surface altitude
Surface casing
Neat cement
Depth of surface casing
Bentonite
Borehole {below
surface casing)
'Iller pack Interval

Interval t
e
Figure 12.13.1. Typical
construction of single- 1
Blank casing (sump) and multiple-well bore- lc
holes in Oasis Valley, l
Well depth Borehole depth Nevada (Robledo et al.37). ic
Problems 541
Table 12.13.1 Summary of Inflow and Outflow Zones and Relative Transrnissivities in Borehole ER-OV-06a, Oasis Valley, Nevada
Depth of inflow Probable inflow Inflow or outflow Percent difference

or outflow or outflow (gallons per minute) between ambient
zone (feet below point (feet below and injection Relative
Zone land surface) land surface) Ambient 1 Injection 1 Difference inflow or outflow2 transmissivity
A 43.0-48.5 45 -0.10 -0.10 0.00 <1 Low
B 62.5-99.0 75 -.30 -.32 .02 5.1 Low
c 195.0-213.0 205 -.20 -.37 .17 43.6 High
D 325.5-346.0 340 .50 .34 .16 4LO High
E 397.0-417.0 400 .06 .03 .03 7.7 Low
F 483.0-536.0 515 .04 .03 .01 2.6 Low
Total 0.00 -.39 .39 100
1
By convention, inflow (water flowing inco well) is shown as positive number and oucflow is shown as negative number.
2Percentdifference for zone is calculated using the following equation: I00 x [(ambient inflow)- (injection inflow)]/(sum of differences
between ambient and injection inflow for all zones). Precision implied by percent differences for zone exceeds intended accuracy.
Source: Robledo et al. 37
12.13 CASE STUDY Oasis Valley, Nevada (continued)
transmissivity of the fracture zone. Assuming that the change in This difference between the inflow under two head conditions can
head difference or gradient between the borehole and the aquifer be used to determine the relative transmissivity. Another good
at each zone is the same, then the change in flow is directly pro- example case study is presented by Brendle5 for the San Luis Val-
portional to the difference in transmissivity of the fracture zones. ley, Colorado.
PROBLEMS
12.3.1 Refer to ASTM Reference D5753-95-Standard Guide 12.13.3 Perfonn a search of the ASTM standards and develop a
for Planning and Conducting Borehole Geophysical Logging and list of the various standards that apply to subsurface investiga-
write a summary of the guidelines. tions of groundwater. Briefly define how the standard applies.
12.13.1 Develop a detailed summary of the Oasis Valley, Nevada, 12.13.4 Perform an [ntemet search to determine the most recent
case study. Refer to the report by Robledo et a!Y for addi- advances that have been made in logging techniques.
tional details. Your instructor may identify additional points to 12.13.5 Develop a categorized list of the.required information on
;a}
emphasize. the properties of roclcs, fluid, wells, and groundwater systems.
12.13.2 Discuss the methodologies and results of the geophysical Then identify the available logging techniques that might be
logging procedures used in the U. S. Geological Water Resources utilized for each. A starting point might be to refer to Keys and
Investigation Report 02-4058 by Brendle.5 Your instructor may MacCary.21
identify additional points to emphasize.
Caliper Fluid temperature Flow in borehole
I Ambient
8I ·/
dala
<0(1 l
o I. •
/ ~r-
lnjedion ·I
Zone of
data ~ •
outllow 'I
from
botellole 8'I . •
150
q•
jl
I
R•
.e _,
I o 1•
I
I
• I 0
/:VI_,c •
ln41otp"etalk>n ) 0I •
ornow •
dls1ributlon) Q>•
:. - - ZqneD··. under slress .........,_.___
ccnlflllons , ~\
350
Zone of
Inflow to lnt8lprelation
borehole olftow
dlslribulion
und..-amblent
400 oonditions
-- - Zone E
. '
500
Figure 12.13.2. Caliper and fluid-
. ·- -- Zone F temperature logs and heat-pulse
flowmeter interpretation from well
10 16 75 80 85 90 ·1 0
Degrees
ER-OV-06a. Oasis Valley, Nevada
Borehole diameter (inches) Gallons per minute
J<ohrenhei1 (Robledo et al. 37).
Problems 543
Stratigraphic Natural Induction

Caliper Conductivity
unit resistivity
0 ,........,::-----,:---.
: .....
50
100
600 0
Borehole American petroleum Millimhos per Ohmmeters Figure 12.13.3. Geologic and geo-
diameter, (inches) institute units meter physical logs from well ER-OV-06a,
Oasis Valley, Nevada (Robledo et alY).
Fluid Fluid Fluid specific

Resistance temperature Resistivity conductance
1.
S<
2.
1-- - pt
u
3.
AI
1-- -
4.
m
Rt
- 5.
I tn
.fiJ:
G,
- htl
g
Q) 6.
0
~
:::!
H:
Cl)
-g 250 7.
~ ge
3:
0
Qj 8.
.0
:5
Lc
g.soo - 19
0
9.
pn
19
350 -
10
Wt
11
- In
Im
12
Wc
450 -
13
of
14
500 - Pr,
536 L--.."--...l...--'----_j
0 200
Ohms
400 70
Degrees
95 10
I
Ohm
25 400
I 900
Microsiemens per
Wi
15.
vel
pp.
fahrenheit meters centimeter at 25
degrees celsius 16.
Figure 12.13.3 (continued). (Robledo et al. 37 ) 19:
17.
wa
18.
rat
19.
gal
20.
tig;
Wa
twr
21.
wa1
References 545
REFERENCES
I . Bassiouni, Z., Theory: Measurement, and lnterprelation of Well Legs, tions Reports, U.S. Geological Survey Techniques of Water-Resources
Society of Petroleum Engineers. Houston, 1994. lnvs.. Bk. 2. Chap_El, hup://water.usgs.gov/pubs/twri/twri2-el. 126 pp ,
2. Bays, C. A.• and S. H. Folk, Developments in the application of geo- 1971.
physics to ground-water problems, Illinois Geological Survey Circ. 108, 22. Keys, W. S., and L. M. MacCrary, /..()cation and Characteristics ofthe
Urbana, 25 pp., 1944. lnlerface between Brine and Fresh Water from Geophysical lAgs of Bore-
3. Birman, J. H.. Geothermal exploration for ground water, Geol. Soc. holes in the Upper Bra~os River Basin, Texas, U.S. Geological Survey
Amer. Bull.. v. 80, pp. 617-630, 1969. Professional Paper 809-8, 23 pp., 1973.
4. Bredehoeft, J.D., and I. S. Papadopulos, Rates oi vertical ground water 23. Kirby, M. E., Improve your work with dtilling-time logs, Johnson
movement estimated from earth's thermal profile, Water Resources Nalional Drillers Jour., v. 26, no. 6, pp. 6-7, 14, 1954.
Research, v. 1, pp. 325-328. 1965. 24. Lao, C., et al., Application of electric well logging and other wei! log-
5. Brendle, D. L., Geophysical Logging to Detennine Construction, Con- ging methods in Hawaii, Tech. Rept. 21, Water Resources Research Cen-
tributing Zones, and Appropriate Use of Water Le~eLs Measured in Con- ter, Univ. of Hawaii, Honolulu, 108 pp., 1969.
fined-Aquifer Network Wells, San Luis Valley, Colorado, 1998-2000, U.S. 25. LeRoy, L. W., Subsurface Geologic Methods, 2nd ed., Colorado
Geological Survey Water Resources Investigations Reporl 02-4058, School of Mines, Golden, CO, 1156 pp., I 951.
http://water.usgs.gov/pubs/wri/wrir02-4058. 2002. 26. Lynch, E. J.. Fonnation £val11ation, Harper and Row. New York, 422
6. Brown. R. M., et at.. Ground-water Studies, Studies and Repts. in pp.. 1962.
Hydrology 7, UNESCO, Paris. vars. pp., 1972. 27. Meyer, W. R., Use of a Neulron Probe 10 Derermine the Storage Coef
7. Callahan, J. T.. et al., Television-A new tool for the ground-water ficienr of an Unconfined Aquifer, U.S. Geological Survey Professional
geologist, Ground Water. v. I, no. 4, pp. 4-6, 1963. Paper450E, pp. EI74-El 76, 1963.
8. Croft, M. G., A Method for Calculating Permeability from Electric 28. Norris, S. E., The use of gamma logs in determining the character of
lAgs, U.S. Geological Survey Professional Paper 150-B, pp. B265-B269, unconsolidated sediments and well construction features, Ground Water,
1971. v. 10, no. 6. pp. 14-21, 1972.
9. Doll, H. G.. The S. P. log: Theoretical analysis and principles of inter- 29. Paillet, F. L., Flow modeling and permeability interpretation using
pretation, Trans. Amer. lnst. Min. and Mer. Engrs., v. 179, pp. 146-185, borehole flow logs in heterogeneous fractured formations, Water
1949. Resources Research, v. 34, no. 5, pp. 997-10 I0, 1998.
10. Frimpter, M. H., Casing detector and self-potential logger, Ground 30. Parizek, R. P., and S. H. Siddiqui, Determining the sustained yield of
Water, v. 7, no. 6, pp. 24-27, 1969. wells in carbonate and fractured aquifers, Ground Water, v. 8, no. 5, pp.
II . Garber, M. S., and F. C. Koopman, Methods of measuring water levels 12-20, !970.
in deep wells, U.S. Geological Survey Techniq«es of Water-Resources 31. Patten, E. P., Jr., and G. D. Bennett, Application of Electrical and
Investigations, Bk. 8, Chap. AI , 23 pp., 1968. Radioactive Well Logging to Ground-Water Hydrology, U.S. Geological
12. Gorder, Z. A., Television inspection of a gravel pack well, Jour. Amer. Survey Wmer-Supply Paper 1544-0, 60 pp., 1963.
Water Works Assoc., v. 55, pp. 31-34, 1963. 32. Patten, E. P., Jr., and G. D. Bennett, Melhods of Flow Measurement in
13. Guyod, H., and 1. A. Pranglin, Analysis Charts for the Determination Well Bores, U.S. Geological Survey Water-Supply Paper 1544-C, 28 pp.,
of True Resislivity from Electric U:Jgs, Houston, TX. 202 pp., 1959. 1962.
14. Heath, J. R., P. H. Nelson, and F. L. Paillet, Well lAgging for Physical 33. Peterson, F. L .. Neutron Well lAgging in Hawaii, Tech. Rept. 75,
Properties: A Handbook for Geologists, and Engineers, 2nd ed., John Water Resources Research Center, Univ. of Hawaii, Honolulu, 42 pp.,
Wiley & Sons, New York, 2000. 1974.
15. Hess, A. E., Identifying hydraulically conductive features with a slow- 34. Pickel!, 1. J., and J. G. Heacock, Density logging, Geophysics, v. 25,
velocity borehole flowmeter, Canadian Geotechnical Journal, v. 23, no. I, pp. 891-904, 1960.
pp. 69- 78, 1986. 35. Pickett, G. R., The use of acoustic logs in the evaluation of sandstone
16. Jakosky, J.J., Exploration Geophysics, Trija, Los Angeles, 1195 pp., reservoirs, Geophysics, v. 25, pp. 250-274, 1960.
!950. 36. Poland, J. F., and R. B. Morrison, An electrical resistivity-apparatus
17. Jones, P. H., and T. B. Buford, Electric logging applied to ground for testing well-waters, Trans. Amer. Geophysical Union, v. 21, pp. 35-46,
water exploration, Geophysics, v. 16, pp. 115-139,1951. 1940.
18. Kearey, P., and M. Brooks, An Introduction lo Geophysical Explo- 37. Robledo, A. R., P. L. Ryder, J. M. Fenelon, and F L. Paillet, Geo-
ration, 2nd ed., Blackwell Scientific Publications, Oxford, 1991. hydrology of Monitoring Wells Drilled ill Oasis Valley near Beatty, Nye
I9. Keys, W. S., Borehole Geophysics Applied to Ground-Water Investi- County, Nevada, 1997, U.S. Geological Waler-Resources Investigations
gations, National Water Well Association, Dublin, OH, 1989. Report 98-4184, httpJ/water.usgs.gov/pubs/wri/wri984184, 1998.
20. Keys, W. S., Borehole geophysics applied to Ground-water inves- 38. Schneider, R., An Application ofThemwmerry to the Study of Ground
tigations, Book 2, Chapter E2, U.S. Geological Survey Techniques of Water, U.S. Geological Survey Water-Supply Paper 1544-B, 16 pp.. 1962.
Water-Resources lnvesligations Reporls, http://water.usgs.gov/pubs/ 39. Shuter, E., and W. E. Teasdale, Application of drilling, coring, and
twri/twri2-e2, 1990. sampling techniques to test holes and wells, Bk. 2, Chapt. Fl , U.S. Geo-
21. Keys, W. S., and L. M. MacCrary, Application of borehole geophysics logical Survey Techniques of Water-Resources Investigations Repons,
water-resources investigations, Techniques of Water-Resources [nvestiga- http://water.usgs.gov/pubs/twri/twri2-fl, 1989.
40. Sorey, M. L., Measurement of vertical groundwater velocity from 44. Vogelsang, D., Environmental Geophysics: A Practical Guide,
temperature profiles in wells, Water Resources Research, v. 7, pp. Springer Verlag, Berlin, 1995.
963-970, 1971. 45. Vonhof, J. A., Water quality determination from spontaneous-poten-
41. Stewart, D. M., The rock and bong techniques of measuring water tial electric log curves, lour. Hydrology, v. 4, pp. 341-347, 1966.
levels in wells, Ground Water, v. 8, no. 6, pp. 14-18, 1970. 46. Ward, S., ed., Geotechnical and environmental geophysics, vols. I, II,
42. Turcan, A. N., Jr., Calculation of water quality from electrical logs- and III, Investigations in Geophysics, SEG Series, SEG, Tulsa, 1990.
theory and practice, Water Resources Pamphlet 19, Louisiana Geological 47. Wyllie, M. R. J., The Fundamentals of Well Log Interpretation, 3rd
Survey, Baton Rouge, 23 pp., 1966. ed., Academic Press, New York, 238 pp., 1963.
43. U. S. Geological Survey, A Guide to Safe Field Operations, U.S. Geo-
logical Survey Open-File Report 95-777, http://water.usgs.gov/pubs/
of/of95-777, 1995.
13
Chapter 13
1-
1,
rd
Artificial Recharge of Groundwater
In order to increase the natural supply of groundwater, people artificially recharge ground-
water basins. Artificial recharge may be defined as augmenting the natural movement of sur-
face water into underground formations by some method of construction, by spreading of
water, or by artificially changing natural conditions. A variety of methods have been devel-
oped, including water spreading, recharging through pits and wells, and pumping to induce
recharge from surface water bodies. 5• 18· 32 · 34• 66 The choice of a particular method is governed
by local topographic, geologic, and soil conditions; the quantity of water to be recharged; and
the ultimate water use. In special circumstances, land value, water quality, or even climate may
be an important factor.
13.1 CONCEPT OF ARTIFICIAL RECHARGE

Artificial recharge projects are designed to serve one or more of the following purposes:
1. Maintain or augment the natural groundwater as an economic resource.
2. Coordinate operation of surface and groundwater reservoirs.
3. Combat adverse conditions such as progressive lowering of groundwater levels,
unfavorable salt balance, or saline water intrusion.
4. Provide subsurface storage for local or imported surface waters.
5. Reduce or stop significant land subsidence.
6. Provide a localized subsurface distribution system for established wells.
7. Provide treatment and storage for reclaimed wastewater for subsequent reuse.
8. Conserve or extract energy in the form of hot or cold water.
Thus, in most situations, artificial recharge projects not only serve as water-conservation
mechanisms but also assist in overcoming problems associated with overdrafts. 16 The role of
artificial recharge in groundwater management is described in Chapter 9; control of seawater
intrusion is described in Chapter 14.
To place water underground for future use requires that adequate amounts of water be
obtained for this purpose. In some localities, storm runoff is collected in ditches, basins, or
reservoirs for subsequent recharge. Elsewhere, recharge water is imported into a region by
pipeline or aqueduct from a distant surface water source. A third possibility involves utiliza-
tion of treated wastewater.
547
548 Chapter 13 Artificial Recharge of Groundwater
Recharging began in Europe early in the nineteenth century and in the United States near
the end of the century; since then recharge installations have steadily increased throughout the
world. Recharge basins form integral parts of many Swedish municipal water supply systems.
Artificial recharge is widely practiced in Germany to meet industrial and municipal water
demands. In the Netherlands, water supply systems for Amsterdam, Leiden, and The Hague
include basins for recharging surface water into coastal sand dunes. 9
Artificial recharge is accomplished by applying water to the soil for infiltration and down-
ward movement through the unsaturated or "vadose" zone to the groundwater. Bouwer 14
defined artificial recharge systems according to permeable materials in which they can be
placed, as illustrated in Figure 13.1.1. The surface system (a) requires soils that are sufficiently
permeable, vadose zones that have no clay or other restricting layers, and aquifers that are
unconfined. Artificial recharge can be accomplished with excavated basins (b) that are suffi-
ciently deep to reach permeable matetial where permeable soils are not available at relatively
small depths (e.g., 1m). Seepage trenches (c) can be used if the permeable material is too deep
for removal of overlying material, but is within trenchable depth (e.g., less than about 7 m).
Trenches are also suitable in soils that are highly stratified with alternating layers of fine and
coarse materials. 14 Large-diameter wells, pits, or shafts (d) in the vadose zone can, used when
permeable material is too deep for trenches. These shafts can be drilled with bucket augers to a
depth of about 50 m with a diameter of about 1 m. 14 Recharge wells, penetrating the aquifer (e),
can be used in situations where permeable surface soils are not available, vadose zones are not
sufficiently permeable to transmit water, or aquifers are confined.
13.2 RECHARGE METHODS

13.2.1 Methods
A variety of methods have been developed to recharge groundwater artificially. The most
widely practiced methods can be described as types of water spreading-releasing water over
the ground surface in order to increase the quantity of water infiltrating into the ground and
then percolating to the water table. Although field studies of spreading have shown that many
factors govern the rate at which water will enter the soil, from a quantitative standpoint, area
of recharge and length of time that water is in contact with soil are most important. Spreading
(a) (b) (c) (d) (e)
:._:;· ·. ·.. ·..

: ... :·.·
:; ·.·... :. . ·.-.->=·~>:::··.. : .·
Figure 13.1.1. Recharge sys-

tems for increasingly deep
al:
permeable materials: surface
basin (a), excavated basin (b),
trench (c), shaft or vadose
zone well (d), and aquifer
well (e) (Bouwer 14).
- -~
:f efficiency is measured in terms of the recharge rate, expressed as the velocity of downward
e water movement over the wetted area.
Spreading methods may be classified as basin, stream channel, ditch and furrow, flooding,
!f and irrigation. These, together with techniques employing pits and recharge wells, are
e described in the following sections.
t-
14
13.2.1.1 Basi11 Method
•e Water may be recharged by releasing it into basins formed by construction of dikes or levees
ly or by excavation. Generally, basin sizes and shapes are adapted to Land surface slope. Silt-free
re water aids in preventing sealing of basins during submergence. Most basins require periodic
.1- maintenance to improve infiltration rates by scarifying, disking, or scraping the bottom sur-
ly faces when dry. Where local storm runoff is being recharged, a single basin will normally suf-
!p fice, but where streamflow is being diverted for recharge, a series of basins, often parallel to
l). the natural stream channel, becomes advantageous.8 Water from the stream is led by a ditch
1d into the uppennost basin. As the first fills, it spills into the second, and the process is repeated
!0 through the entire chain of basins. From the lowest basin, any excess water is returned to the
•a stream channel. Figure 13.2.1 illustrates a typical plan of a multiple-basin recharge project.
?), This method permits water contact over 75 to 90 percent of the gross area.
10t Multiple basins provide for continuity of operation when certain basins are removed from
service for drying and maintenance. Furthermore, where streamflow from storm runoff is being
spread, upper basins can be reserved for settling silt. Figure 13.2.2 shows an aerial photograph
of an extensive series of recharge basins in Los Angeles, California. Figure 13.2.3 on page 551
shows the recharge operations of Orange County Water District.
Basins, because of their general feasibility, efficient use of space, and ease of mainte-
nance, are the most favored method of recharge. Long-time recharg_e rates vary widely.
JSt Infiltration basins (SAT systems, see Section 13.4) are by far the most widely used
{er method for groundwater recharge and municipal waste removal (U.S. Environmental Protec-
nd tion Agency66 ). Because of the economical attractiveness and low-cost maintenance, many
my SAT systems have been installed or are planned to be installed in the United States and other
rea countries throughout the world. Los Angeles County implemented the use of SAT systems for
ing groundwater recharge in the Montbello Forebay in the early 1960s. After over 20 years of
operation and extensive study, no measurable impact on groundwater quality or human health
could be established (Nellor et al.47). Bouwer 12• 13 studied SAT systems in the Phoenix metro-
politan area with effluents from two different wastewater treatment plants. SAT systems have
also been studied and used in Israel (ldelovitch et a1. 33 ) and in Australia (Ho et al. 31). Wilson
et al. 69 has studied water quality changes and the fate of disinfection byproducts at the Tucson
Sediment
retention
:ys- Figure 13.2.1. Typical plan of a
basin
multiple-basin recharge project
ce diverting water from a stream
(b), (after Amer. Soc. Civil Engrs.,
Ground Water Management, Man.
and RepiS. on Engmg. Practice 40,
1972).
'\
Figure 13.2.2. Aerial view of spreading

basins adjoining the San Gabriel River, Los
Angeles, California, and temporary finger
dikes within the river channel (courtesy Los
Angeles County Flood Control District).
Sweetwater Recharge Site. Arizona State University et al. 4 have performed extensive studies
on the feasibility of a large-scale SAT project in Phoenix, Arizona.
13.2.1.2 Stream-Channel Method

Water spreading in a natural stream channel involves operations that will increase the time and
area over which water is recharged from a naturally losing channel. This involves both
upstream management of streamflow and channel modifications to enhance infiltration.
Upstream reservoirs enable erratic runoff to be regulated and ideally limit streamflows to rates
that do not exceed the absorptive capacity of downstream channels.
Improvements of stream channels may include widening, leveling, scarifying, or ditching
to increase infiltration. In addition, low check dams and dikes can be constructed across a
stream where a wide bottom occurs; these act as weirs and distribute the water into shallow 8t
ponds occupying the entire streambed (see Figure 13.2.4). These structures are normally PI
BE
8<
Figure 13.2.4. Channel spreading with rock-and-

wire check dams in Cucamonga Creek near Upland, Fit
California (courtesy D. C. Muckel). ye;
ba:
Miller
Basin
Kraemer
Basin
Carbon Creek
:s Diversion Channel
Glassell
Basin -
Warner
Basin e lnllatable Dam
td Bypass
th Five Coves
n. Basin ---
es
1W
Burris
ly Pit --
Figure 13.2.3. The efficiency of Orange County Water District's recharge operations has been greatly increased over the past seven
years by a major capital improvement program. As a result, up to 400,000 acre-feet of water can percolate into the groundwater
basin in a year's time (Orange County Water District49).
temporary, consisting of river-bottom material, and sometimes protected by vegetation, wire,

or rocks. Such works quickly collapse when high streamflows occur; if permanent structures
are placed in a channel, it is important that they do not create a flood hazard.
L-shaped finger levees can be seen in the stream channel in Figure 13.2.2, each of which
impounds water. These are simply constructed annually by bulldozer at the end of the high-
streamflow season.
Channel spreading can also be conducted without a specific spreading works. In streams
with storage reservoirs primarily for flood control, releases of clear water may be entirely
recharged into downstream reaches. A majority of the spreading works in and near Los Ange-
les County are part of an integrated water conservation and flood protection plan.
13.2.1.3 Ditch-and-Furrow Method

In the ditch-and-furrow method, water is distributed to a series of ditches, or furrows, that are
shallow, flat-bottomed, and closely spaced to obtain maximum water-contact area. One of
three basic layouts is generally employed: (I) contour, where the ditch follows the ground con-
tour and by means of sharp switchbacks meanders back and forth across the land; (2) tree-
shaped, where the main canal successively branches into smaller canals and ditches; and (3)
lateral, where a series of small ditches extend laterally from the main canal.
Ditch widths range from 0.3 to 1.8 m. On very steep slopes, checks are sometimes placed
in ditches to minimize erosion and to increase the wetted area.
Gradients of major feeder ditches should be sufficient to carry suspended material through
the system. Deposition of fine-grained material clogs soil surface openings. Although a vari-
ety of ditch plans have been devised, a particular plan should be tailored to the configuration
of the local area. A collecting ditch is needed at the lower end of the site to convey excess water
back into the main stream channel.
The method is adaptable to irregular terrain but seldom provides water contact to more
than about 10 percent of the gross area. Figure 13.2.5 shows typical spreading ditches on an
alluvial plain.
13.2.1.4 Flooding Method

In relatively flat topography, water may be diverted to spread evenly over a large area. In
practice, cartals and earth-distributing gullies are usually needed to release the water at intervals
over the upper end of the flooding area. It is desirable to form a thin sheet of water over the land,
which moves at a minimum velocity to avoid disturbing the soil. Tests indicate that highest infil-
tration rates occur on areas with undisturbed vegetation and soil covering. Compared with other
spreading methods, flood spreading costs least for land preparation. In order to control the water
Figure 13.2.5. Spreading ditches in Tujunga

Wash, Los Angeles, California (courtesy
City of Los Angeles Department of Water
and Power).
Control tower
Sand and
gravel
pipe
Bedrod<.
e Figure 13.2.6. Cross section of a recharge pit

,f 0 10 20 30 at Peoria, Illinois (after Suter and
Horizontal scale (meters) Vertical scale (meters) Harrneson61 ).
I)
at all times, embankments or ditches should surround the entire flooding area. To obtain maxi-
mum efficiency, a person should be on the grounds during flooding operations-frequently,
movement of a few shovelsful of dirt will effectively increase the wetted area.
;h
1- 13.2.1.5 lrrigatUm Method
m
In irrigated areas, water is sometimes deliberately spread by irrigating cropland with excess
er
water during dormant, winter, or nonirrigating seasons. The method requires no additional cost
for land preparation because the distribution system is already installed. Even keeping irriga-
re
tion canals full will contribute to recharge by seepage from the canals. Where a large portion
an
of the water supply is pumped, the method has the advantage of raising the water table and
consequently reducing power costs. Consideration needs to be given to the effects of the leach-
ing action of the percolating water both in carrying salts from the root zone to groundwater and
in removing soil nutrients, iliereby possibly reducing crop yields.
In
als 13.2.1.6 Pit Method
1d,
A pit (see Figure 13.2.6) excavated into a permeable formation serves as an ideal facility for
fil-
groundwater recharge. Because the cost of excavation and removal of material is high, use of
her
abandoned excavations, such as gravel pits, is most economic. In areas where shallow subsur-
tter
face strata, such as hardpans and clay layers, restrict the downward passage of water, pits can
effectively reach materials wiili higher infiltration rates.35· 41
Besides the small capital cost of pits constructed for a purpose other than recharge, the steep
sides provide a high silt tolerance. Silt usually settles to the bottom of the pit, leaving the walls
relatively unclogged for continued infiltration of water.8 Nonsilty water should be recharged
whenever possible so as to minimize silt accumulation and periodic removal costs.23 Attention to
the geometry of a recharge pit is important in order to obtain the maximum infiltration rate. 20•58
Studies by the illinois State Water Survey29 of the filtration efficiency of coarse media
resulted in the equation
(13.2.2)
mga
where SS0 is the suspended solids concentration (mg/1) transmitted through the filter layer, His the
~r
filter layer thickness (em), dis the mean diameter (mm) of particles forming the filter layer, Q0 is
the rate of recharge (m/day), and SSi is suspended solids concentration of the recharged water.
13.2.1.7 Recharge Well Method

A recharge well may be defined as a well that admits water from the surface to freshwater
aquifers. Recharge wells are also known as disposal wells and drain wells. They should be dis-
tinguished from injection wells, described in Chapter 8, which recharge brines and toxic indus-
trial wastes to deep, saline-water aquifers. A recharge well's flow is the reverse of a pumping
well, but its construction may or may not be the same. Well recharging is practical where deep,
confined aquifers must be recharged, or where economy of space, such as in urban areas, is an
important consideration.
If water is admitted into a well, a cone of recharge will be formed that is similar in shape
but is the reverse of a cone of depression surrounding a pumping well. The equation for the
curve can be derived in a similar manner to that for a pumping well (see Chapter 4). For a
confined aquifer with water being recharged into a completely penetrating well at a rate Qf' the
approximate steady-state expression
(13.2.3)
is applicable. (Symbols are identified in Figure 13.2.7a.) For a recharge well penetrating an
unconfined aquifer (see Figure 13.2.7b),
(13.2.4)
I
By comparing the discharge equations for pumping and recharge wells, it might be antic- 1
ipated that the recharge capacity would equal the pumping capacity of a well if the recharge a
cone has dimensions equivalent to the cone of depression. Field measurements, however,
rarely support this reasoning; recharge rates seldom equal pumping rates. The difficulty lies in
the fact that pumping and recharging differ by more than a simple change of flow direction.
As water is pumped from a well, fine material present in the aquifer is carried through the
coarser particles surrounding the well and into the well. On the other hand, any silt carried by
water into a recharge well is filtered out and tends to clog the aquifer surrounding the well. 51· 54
Similarly, recharge water may carry large amounts of dissolved air, tending to reduce the per-
meability of the aquifer by air binding. Recharge water may also contain bacteria, which can
form growths on the well screen and the surrounding formation, thereby reducing the effective
flow area. Chemical constituents of the recharge water may differ sufficiently from the normal
o,
o,
Piezometric surface ~
Unconfined
Confined aquifer
b
aquifer
~
(a) (b)
Figure 13.2.7. Radial flow from recharge wells penetrating (a) confined and (b) unconfined aquifers.
Irrigation tail
water pond
!f \
s- Lid
s- Soil
tg Surface casing
p,
ill
~e Ogallala
he formation
·a
Water Sludge
he
table
c Porous lava
formation --z._
.3)
an
Clay
.4) (a) (b)
Figure 13.2.8. Examples of recharge well designs. (a) Combined irrigation and recharge well in alluvium in the High Plains of
:ic- Texas (b) Recharge well for disposal of septic tank effluent into a lava fonnation in Central Oregon (after Hauser and Lotspeich30
rge and Sceva.56)
rer,
)Ill
groundwater to cause undesired chemical reactions-for example, defloculation caused by
n.
reaction of high-sodium water with fine soil particles. These factors all act to reduce recharge
the
rates; as a result, well recharging has been limited to a few areas where experience, particularly
Jby
1,54 with water treatment and redevelopment of wells, has shown the practicality of the method. 67
It should be noted that supply wells can alternate as recharge wells, as shown in Figure
>er-
13.2.8a Recharge wells serve as convenient means for disposal of septic tank effluent, excess
can
irrigation water, and surface runoff into the permeable volcanic terrains of the northwestern
tive
United States.55•64 A typical domestic wastewater disposal system including a recharge well is
mal
shown in Figure 13.2.8b.
An extensive series of recharge wells has been successfully operated since 1953 along the
- coasts of Los Angeles and Orange Counties, California. These wells create and maintain a pres-
sure ridge of fresh water to control seawater intrusion (see Chapter 14). Experience gained on the
project demonstrated that gravel-packed wells operate most efficiently; a typical dual-aquifer
well is shown in Figure 13.2.9. Favorable recharge rates have been maintained by chlorination
and deaeration of the water supply and by a comprehensive well-maintenance program involv-
ing periodic pumping of the wells. It was also found that a concrete seal should be provided on
the outside of the casing where it passes through the impermeable zone above the confined allu-
vial aquifer to prevent upward movement of water. Figure 13.2.10 illustrates the Scottsdale Water .
Campus located in Scottsdale, Arizona, which has been in operation since 1999.
Finally, it should be mentioned that field and laboratory studies have demonstrated the feasi-
bility of temporary storage of fresh water in saline water aquifers through wells first recharged and
later pumped.21 • 37 The efficiency of the procedure increases with each recharge-storage-
withdrawal cycle. The technique has application in flat coastal areas underlain by saline water
aquifers where no surface reservoir sites are available to provide freshwater supplies on a year-
round basis.
.-,., ';
Vault bottom-s
91.4 em steel-surface
protective casing
7.6 em plastic ~~++-'-'-t----1-'-;t---+t
conductor pipe .
30.5 em AC pipe- ~,;............~~
blank casing
10.2 em plastic tremie pipe ---'--c-'~.-"--:+!~o..o

30.5 em perforated AC casing --'-'-~-'-'-l:~IO'l
Gravel packing -----'c-'-~+f'Ci'!ilhl-.~~
30.5 em AC blank casing ---'-'-______,_~~;;~-.,
Lower-zone measuring pipe ~~,---,--.~:~~F..~
Packer pressure-line ~~--:-ft'll

Figure 13.2.9. Design of a
Steel landing-plate ..:..:,...;___,;-.....,...;:B!Hitij~ dual-aquifer recharge well for
with drill-collar box
control of sea water intrusion,
Los Angeles County, California
(after Department of Economic
and Social Affairs 19 ).
13.2.1.8 Incidental Recharge

Incidental, or unplanned, recharge occurs where water enters the ground as a result of a human
activity whose primary objective is unrelated to artificial recharge of groundwater. 19 Included
in this category is water from irrigation, cesspools, septic tanks, water mains, sewers, landfills,
waste;.disposal facilities, canals, and reservoirs. The quantity of incidental recharge normally far
exceeds that deliberately accomplished by artificial recharge projects. Because several of these
sources introduce polluted waters into the underground, degradation of the quality of ground- (I
water can occur (see Chapter 8).
13.2.2 Recharge Rates

The economy of water spreading hinges on maintenence of a high infiltration rate. Typical rate
curves, however, show a pronounced tendency to decrease with time. Determining the cause of
this decrease and how to counteract it has led to extensive research programs. A variety of soil
and water treatments and operational methods have been undertaken to study the problem,
from the 1960s7• 11 · 55 to more recent studies (Arizona State University et al. 4).
Figure 13.2.11 shows a typical curve of recharge rate versus time. The initial decrease is
attributed to dispersion and swelling of soil particles after wetting; the subsequent increase
accompanies elimination of entrapped air by solution in passing water; the final gradual
decrease results from microbial growths clogging the soil pores. Laboratory tests with sterile
soil and water give nearly constant maximum recharge rates, thereby substantiating the effect
of microbial growths.
(a) Outside view
Figure 13.2.10. Scottsdale Water Campus (courtesy of

Black & Veacb 10 and the City of Scottsdale, AZ).
This facility includes a 43-mgd conventional water
treatment plant, a 10-mgd wastewater treatment plant,
a 10-mgd dual membrane water reclamation plant, and
groundwater recharge facilities.
The dual membrane system, consisting of micro fil-
tration (MF) and reverse osmosis (RO), is being used
to treat both surface water from the Colorado River
and tertiary effluent prior to groundwater recharge. The
water then percolates through the soil and eventually
mixes with local groundwater. The city receives
groundwater withdrawal credits based on the volume
recharged from the Arizona Department of Water
Resources within its service area.
The MF system treats both surface water and sec-
.s, ondary treated sewage effluent.
After MF, the filtrate is further treated using RO for
desalting before groundwater recharge. This is to pre-
(b) Filtration system vent salt buildup in the groundwater.
tte
(I)
of ~
)i\ (I)
0'1
m, (ij 0.5
J::.
u
(I)
a:
is Figure 13.2.11. Typical time variation of
tse 20 40 60 80 recharge rate for water spreading on undis-
1al Time (days) turbed soil (after Mucke\45).
ile
!Ct Recharge rates generally decrease as the mean particle size of soil on a spreading area
decreases. Efforts to maintain soil pores free for water passage have led to additions of organic
matter and chemicals to the soil as well as to growing vegetation on the spreading area. Chem- ]
ical soil conditioners, which tend to aggregate the soil, show promise in soils of certain
textures. Alternating wet and dry periods on a basin generally furnishes a greater total recharge
than does continuous spreading, in spite of the fact that water is in contact with the soil for as
little as one-half of the total time. Drying kills microbial growths, and this, combined with
scarification of the soil surface, reopens the soil pores. 47
Other factors contribute to recharge rates. Studies in small ponds have confirmed that
infiltration rates are directly proportional to head of water. Where less pervious strata lie below
the surface stratum, the recharge rate depends on the rate of subsurface lateral flow. Hence,
spreading only in narrow, widely spaced strips recharges nearly as much water as spreading
over an entire area. 20• 58 Water containing silt or clay is known to clog soil pores, leading to
rapid reductions in recharge rates. Wave action in large, shallow ponds can stir bottom sedi-
ments and seal pores that would otherwise remain open. Water quality can be an important fac-
tor; thus, recharging water of high sodium content tends to de flocculate colloidal soil particles
and thereby hinder water passage. Because a high water table limits the downward flow of
recharged water, this surface should be at least 3 to 6 m below spreading surfaces.
Clogging effects can be modeled considering the hydraulic effects. The hydraulic charac-
teristics on hydraulic conductivity K, water content ro, and pressure head h. have been described
by Kool and van Genuchten: 36
(13.2.5)
and
(13.2.6)
where 005 and ror are the saturated and residual water content; a, ~, and 'Yare curve-fitting
parameters; Se is the degree of saturation, Se = (ro - ror)l(ros - ror), and Kc is the saturated
hydraulic conductivity affected by the clogging effect, defined as
Kc =Ksfclog (13.2.7)
where K 5 is the saturated hydraulic conductivity without clogging. The factorfclog expresses the
effect of surface clogging (z =0) and biological clogging (z ~ 0).
General expressions for the clogging effect can be described using the affected hydraulic
conductivity Kc, based on Ahuja2 and Moore: 43
(13.2.8) E:
where K1is the lowest hydraulic conductivity value, ac is the coefficient related to character- Ph
istics of soil and clogging substance, and 1 is the cumulative influx of clogging substances as J
a function of time: e
t
1 =Jqs(aic1 + B2c2 )d't (13.2.9)

0
where ai is the percent of each clogging substance being intercepted by the soil surface, ci
Tw
denotes concentration of clogging substances i in ponding water, and i = 1 for suspended solids
R
and i =2 for algae. eJ
The effect of biological clogging is described according to Taylor and Jaffe 65 by
2
-r
Jclog
(
z, t ) ={exp( aB + bB ) B 5: B0 (13.2.10)
fo B > B0
where a, b, fo and B0 are empirical values and B is the biomass.

13.3 Wastewater Recharge for Reuse 559
1- 13.3 WASTEWATER RECHARGE FOR REUSE

m In recent years, increasing attention has been focused on reuse of municipal wastewater.
~e
Almost all uses are nonpotable, such as for irrigation or industrial purposes, because of ques-
as tionable health effects, 59 and one of the mostly widely favored is recharge to supplement
th groundwater resources.
The major concerns associated with recharge to provide high-quality potable or non potable
water from wastewater effluent include organic carbon, nitrogen species, and pathogens. As an
example, priority pollutants in biologically treated wastewater from the 9lstAvenue Wastewater
:e, Treatment Plant in Phoenix, Arizona, have satisfied primary drinking water standards since strict
rrg industrial pretreatment requirements were enforced (Arizona State University et al. 4). The levels
to of organic carbon, nitrogen species, and pathogens found in biologically treated wastewater tend
ii- to be at levels greater than those found in surface waters and provide serious concerns for reuse.
lC-
The organics in biologically treated wastewaters are analogous to the organics present in sur-
les face waters and are of concern just as with organics occurring naturally in both surface waters and
of groundwaters. This concern is the reaction of the organics with the commonly used disinfectant
chlorine to produce disinfection byproducts. Both the removal of organic carbon during soil
lC- aquifer treatment and the effect on the formation of disinfection byproducts have been studied.
>ed Nitrogen species, in particular nitrate-nitrogen, are one of the most common reasons that
groundwaters do not meet drinking water standards. Nitrogen can be removed effectively
before soil aquifer treatment to ensure compliance with nitrogen standards. Significant trans-
:.5) formations of nitrogen occur during recharge to provide further removal of nitrogen species.
A wide range of nitrogen characteristics in wastewater eft1uents has been studied to determine
impacts on nitrogen transfonnations. 4 Table 13.3.1 presents the effluents studied and their
wide range of characteristics.
~.6)
The removal of all three major classes of pathogens (parasites, bacteria, and viruses)
:ing needs to be considered when wastewater is reclaimed. Pathogenic bacteria and viruses can be
tted present in large numbers in biologically treated wastewater; however, disinfection prior to
recharge is effective in eliminating pathogens. During previous studies on recharge, no enteric
viruses or enteric bacteria have been found in aquifers that were recharged with wastewater.
2.7) The study by Arizona State University et al. 4 particularly focused on examining the fate of
the Cryptosporidium during soil aquifer treatment and Hepatitis A, since these two pathogens are
of concern and very little data exists on their fate during recharge.
ulic
Table 13.3.1 Typical Range of Water Quality Parameters (mg!L)
2.8) Effluent Type BOD ss Total N TOC
~ter- Phoenix 91 st Dechlorinated, 3-7 3-7 1-3 1-6 4-10 8-10
!S as Ave. WWTP denitrified
effluent Chlorinated, 3-7 3-7 1-3 1-6 4-10 8-10
denitrified
.2.9) High rate activated 10-15 10-25 10-30 0-8 10-40 13-15
sludge
(chlorinated)
:e, ci
Tucson Roger Secondary 20-25 20-30 15-25 0-2 15-30 15-20
olids
RoadWWTP trickling
effluent filter*
Tertiary-pressure 10-15 5-10 15-25 0-2 15-30 10-15
filter
2.10) Primary 50-80 60-100 20-35 0-1 25-40 40-50
*03 secondary effluent was also studied
Source: Arizona State University et al. 4
Recharge of wastewater (usually after secondary treatment) improves its quality by

removal of physical, biological, and some chemical constituents; provides storage until sub-
sequent reuse; reduces seasonal temperature variations; and dilutes the recharged water with
native groundwater. Land application practices involve irrigation, spreading, overland flow,
and recharge wells. Selection of a given system is governed by soil and subsurlace conditions,
climate, availability of land, and intended reuse of the wastewater.
Irrigation Method
Eft1uent can be applied by sprinklers or surlace irrigation techniques to irrigate cropland.
Application rates are low, ranging from 0.05 to 0.2 m/week; also, only the portion not con-
sumed by plants percolates downward to the water table. In humid regions, low evapotranspi-
ration rates plus dilution with rainwater contributes good-quality water to the groundwater; in
arid regions, brackish water that degrades the groundwater can result.
Spreading Method
Recharging of effluent onto bare ground or native vegetation for infiltration and percolation
functions as a tertiary treatment plant, producing reclaimed water for reuse. Application rates
are high, ranging from 0.5 to 10 m/week, depending on local conditions. High-rate systems
require deep permeable soils (sandy loams to loamy sands) and a water table that does not rise
to ground surlace. Flooding is conducted intermittently-for example, 2 to 14 days wet alter-
nating with 5 to 20 days dry. Movement of effluent through the soil removes bacteria and
viruses, almost all biochemical oxygen demand and suspended solids, up to 50 percent of
nitrogen, and 60 to 95 percent of phosphorus. Because municipal use adds up to 300 mg/1 of
dissolved solids to water and this cannot be removed by recharging, wastewater can deterio-
rate the quality of groundwater unless adequate subsurlace dilution is available.
Overland Flow Method

Where soils have low infiltration rates, such as clays and clay loams, wastewater is applied by
irrigation or spray techniques to the upper end of sloping vegetated plots and allowed to flow
in a shallow sheet to runoff collection ditches. Only a minor fraction of the applied water infil-
trates; hence, this method contributes little to groundwater recharge.
Recharge Well Method

High-quality, tertiary-treated effluent can be placed underground through recharge wells. The
high cost of recharging effluent into wells can be justified economically only where some spe-
cial purpose, such as control of land subsidence, control of seawater intrusion, or delayed
development of a costly alternative water supply source, can be served. 6• 57
The removal of pollutants from secondary effluent by recharging depends on the time and
distance of travel underground and on the type and properties of the soils and subsurface for-
mations. In general, for percolation through fine-textured alluvial deposits, bacteria and
viruses are removed, nitrogen is reduced, soluble salts are not removed, and trace elements and
heavy metals may be reduced.
13.4 SOIL AQUIFER TREATMENT (SAT) SYSTEMS

13.4.1 What Are SAT Systems?
Artificial recharge of sewage effluent has an important role to play in water reuse. Treated
effluent can be placed in recharge basins, allowing for infiltration into the ground for the
recharge of aquifers. As the effluent moves through the soil and the aquifer, it undergoes sig-
JY Greater
b- To direct metropolitan
nonpotable area
.th
reuse
w,
lS,
!d.
m-
pi-
m
lOll
tes Figure 13.4.1a. An
ms SAT system and its
components (Puizona
ise
State University et al. 4)
:er-
md
of
l of
10-
I by
low
tfil-
The
;pe-
yed
and Figure 13.4.1b. Avra Valley recharge facility near

fer- Tucson, Arizona (courtesy of Central Arizona Project,
and photo by M. Early).
and
nificant quality improvements through physical, chemical, and biological processes. Collec-
tively, these processes and the water quality improvement obtained are called soil aquifer
treatment (SA1). SAT is premised on the infiltration of treated wastewater into the soil and per-
colation through the vadose zone. Improvements in water quality can occur due to many dif-
ferent mechanisms, including infiltration, biological degradation, physical adsorption. ion
ated exchange, and precipitation.
· the An SAT system consists of five major components (see Figures 13.4.1a and b): (1) the
sig- pipeline that carries the treated effluent from the wastewater treatment plant, (2) percolation
....0
...::l
Q,
.5
(a) (b)
Pumped well Conservation well ~
....0
...ca
Q)
Ci)
(c) (d)
Figure 13.4.2. Schematic of recharge and recovery SAT systems: (a) with natural drainage of renovated water into stream, lake, or
low area, (b) collection of renovated water by subsurface drain (c) infiltration areas in two parallel rows and line of wells midway
between, and (d) infiltration areas in center surrounded by a circle of wells (Bouwert 4).
..
0
..
0
::l
0.
::l
(infiltration) basins where the treated effluent infiltrates into the ground, (3) the soil immedi-
ately below the infiltration basins (vadose zone), (4) the aquifer where water is stored for a
long duration, and (5) the recovery well where water is pumped from the aquifer for potable
or nonpotable reuse. Recharge systems for SAT can be designed as infiltration-recovery sys-
tems in which basically all ef11uent water is recovered, as shown by systems A, B, and C in
Figure 13.4.2, or after blending with the native groundwater as shown by system D.
Figure 13.4.3 illustrates the SAT system dynamics including inputs (design inputs), the
state of the system, and the output of the system. The inputs are the soil type, water quality,
operation schedule, and environment. The state of the system, which controls the residence
time of water in the vadose zone and the level of microbial activity, includes the soil moisture
profile, level of oxygen in the system, algal growth, and the soil hydraulic conductivity. Of par-
ticular importance is the level of oxygen, which is related to the microorganism distribution
and oxygen demanding substrates. Algal growth is related to clogging layer fonnation on the
soil surface and effective soil hydraulic conductivity. Soil moisture is directly affected by soil
type and the environment, and indirectly by the water quality, which affects algal growth and
the soil hydraulic conductivity. Oxygen in the vadose zone is affected by the soil moisture pro-
file, the water quality of treated effluent, the operation schedule, and the algal growth. Algal 13.·
growth is affected by the water quality of the treated effluent, the operation schedule and the
environment. Soil hydraulic conductivity is affected by the algal growth, the soil type, and the
environment.
The critical outputs of the system are the total infiltration, the total organic carbon
removal efficiency, the nitrogen removal efficiency, and the pathogen removal efficiency. The
total organic carbon removal efficiency is primarily affected by the amount of oxygen in the
system through the organic oxidation process and the presence of acclimated microorganisms.
Nitrogen removal efficiency is dependent upon oxygen levels and the availability of biodegrad-
able organic carbon through the nitrification-denitrification process. Pathogen removal effi-
ciency is dependent upon the soil moisture profile and the soil hydraulic conductivity through
the adsorption-inactivation process.
EC1)
.......... Water quality Environment
Soil type Operation schedule
0 U) Primary effuent, Temperature,
... > Sand, silt, Wetting & drying time,
:I U) secondary effluent, solar radiation,
c. ... clay, etc ... loading rate, pond depth
etc ... evaporation, etc ...
c ca
- U)
r--
J
- - - -
- - - - -
- -
- - ,_
,, . +
r ~ Algal growth ~ - ,,
0 tn
cu>
... U)
E
........
C1)
ca ...
...
cnU) ca
Soil moisture
profile
Residence time
' ,, ,,
Oxygen in soil
system
Organic carbon release,
clogging layer formation
r-
"""
I
~
......
Soil hydraulic
conductivity
Clogging effect
Microorganism
~
0 distribution
;:;:: t: t:
'0
.l!l -c:: ~~
e.::.
~ u t: t: .Q t: t: o<::
.2 ·c:~
~~ :2~ {g~
Jr ca
~~ ~~ .s
ii
(/) <:(
:§ 0 ~ {g~
r r 2:'0
<:( .s ,, r
.....0 E
... U)
...
C1)
Total
Total organic Nitrogen Pathogen
Figure 13.4.3. Soil
aquifer treatment system
:I> carbon removal removal removal
Q.U) infiltration
efficiency efficiency efficiency dynamics (Arizona State
'S;
QU) University et al. 4).
:di-
1r a
ble The major purification processes in the soil aquifer system are infiltration, chemical
ys- precipitation/dissolution, organic biodegradation, nitrification, denitrification, disinfection,
: in ion exchange, and adsorption/desorption. SAT is a key component of overall water reuse
strategies, providing for (1) mechanical filtration of suspended particles, (2) biologically medi-
the ated transformation of organics and nitrogen, and (3) physical-chemical retention of inorganic
lity, and organic dissolved constituents (e.g., phosphorus, potassium, trace elements) from biolog;.
nee ically treated wastewater.
ture The removal of nitrogen, organic compounds, and biochemical oxygen demand (BOD)
?af- can be a continuous biological process that is sustainable. Removal of trace metals, phospho-
tion rus, and refractory organics by abiotic mechanisms can result in accumulation in the soil and
the eventual breakthrough. Bouwer13 has estimated that SAT-related accumulation of trace metals
soil and phosphorus could become a problem in future decades.
and
pro-
.lgal 13.4.2 Operation of SAT Systems
l the Cyclic flooding and drying of the SAT infiltration basins is necessary both to improve infil-
l the tration rates and to control aerobic/anoxic conditions in the soil. Basins are flooded until infil-
tration rates decrease due to development of a surface-clogging layer. Most SAT-related
·bon treatment occurs in the upper strata of soil, near the clogging layer (see Figure 13.4.4). Dur-
The ing normal conditions (native conditions), unsaturated porous soils (see Figure 13.4.5) contain
l the voids consisting of water and air. Accumulation of biofilm, algae, and suspended solids reduce
sms. the void space as illustrated in Figure 13.4.6. The clogging layer development and reduced
;rad- infiltration rates can dominate SAT system performance, adversely affecting the operation.
effi- Cyclic flooding/drying of the basins is necessary to renew hydraulic capacity. Cycle times also
)Ugh intluence the transport of oxygen with depth in the vadose zone, causing cycle times to be
critical for the control and efficiency of biological processes.
Operational controls:
Pumping rate
Pumping schedule
(wetting-drying) Sewage eHiuent
Surface clogging layer

Infiltration
Physical clogging
Biological clogging
Unsaturated zone
Unsaturated flow
Solute transport
Biochemical reaction (nitrogen and organics removal)
Biological clogging
Figure 13.4.4. Processes related to operation of SAT

system. The saturated zone represents the point at
which groundwater enters the water table and a
mound develops (Arizona State University et al. 4).
Soild Water
D Air
• • • •
Solid
partlcle
Algae Water Blofllm
D
Air Suspended
solid
Figure 13.4.5. A cross section of soil in the vadose zone con- Figure 13.4.6. After SAT, the air spaces become filled with
sisting of a matrix of soil, water, and air. Prior to SAT, the air biofilm, algae, and SS. In addition, a mat of algae can cover
spaces are void and allow for infiltration of water.4 the surface, greatly reducing infiltration rate.4
Figure 13.4.7. Flow dia-

gram of the nitrogen cycle
NITROGEN REMOVAL ORGANIC REMOVAL in an SAT system.
Removal of organic carbon and nitrogen in SAT systems are related (see Figure 13.4.7) and
the removal mechanisms include adsorption-desorption and biodegradation. During the flood-
ing period, dissolved oxygen in the vadose zone often becomes limited and the primary removal
mechanism for organic carbon is adsorption, while ammonia is removed by ion exchange.
Larger gravel and sands may hold larger air pockets, which increases aerobic biodegradation of
carbon and nitrification of ammonia during flooding. Soils that contain more clays and silts
have large adsorptive and ion-exchange capacities to remove organic carbon and ammonia. Dur-
ing drying, the soil in the vadose zone becomes aerobic; however, desiccation of the upper soil
layer can decrease biological activity and hinder removal since most removal occurs in the
upper soil layer. As the flooding cycle begins again, the nitrate that was produced from nitrifi-
cation becomes soluble. Denitrification will occur if sufficient organic carbon and low dissolved
oxygen are simultaneously present. Longer flooding cycles and/or the use of primary effluent
tend to increase available organic carbon and increase total nitrogen removal.
13.4.3 Modeling SAT Systems

Abdulrazzak et al. 1 performed a combined field experiment and analytical approach to con-
clude that the Green-Ampt infiltration model provides adequate representation of the actual
infiltration process for artificial recharge systems. Zomorodi70 developed an infiltration equa-
tion that accounted for the change in infiltration rate due to clogging from turbid water and the
concentration of suspended sediments. Using several simplified assumptions, the analysis con-
cluded that a system of intermittent multibasin recharge is more efficient than either continu-
ous or intermittent recharge on a single basin when clogging occurs. This work did not account
for redistribution of water within the soil, and required many parameters to be defined.
Mushtaq et al. 46 developed optimization models for determining the operation cycles of
recharge basins with the objective of maximizing the infiltration over many cycle times. They
defined each cycle time as consisting of an application period (X), a draining period (Y), and
a drying period (Z). During the application period, water is discharged into the basin. During
the draining period, water is drained from the basin to begin the drying period. The
Green-Ampt infiltration method was used to define the infiltration process. A kinematic wave
model (see Section 3.11) was used to define the redistribution of the soil-moisture profile as a
function of time as the soil moisture moves downward through the soil. The kinematic-wave-
redistribution model developed by Charbeneau 17 was used to define the wetting front profile
and water propagation (refer to Figure 3.11.1). Beginning with a zone of saturated soil, the
~nded ·
soil-moisture wave propagates downward through the soil. The optimization model has the fol-
lowing assumptions: (1) infiltration model is one dimensional; (2) redistribution model is one
dimensional; (3) interference between adjacent basins is neglected; (4) soil-moisture-
h
redistribution model applies at the end of the drain time; (5) hydraulic conductivity remains
constant; (6) clogging effects are ignored; and (7) water quality effects are ignored.
Tang et al. 63 · 64 and Arizona State University et al.4 developed a simulation model
(NITRINFIL) to quantitatively describe the chemical, physical, and biological interactions that
take place between wastewater constituents and the soil system. NITRINFJL (a modified ver-
sion of the HYDRUS model by Kool and van Genuchten36) is a one-dimensional variably sat-
urated flow and interactive multicomponent solute transport model for cyclic operation of soil
aquifer systems. NITRINFIL consists of three major submodels, describing unsaturated flow
(hydrological), solute transport, and solute biotransformation. The functions incorporated into
the unsaturated flow hydraulics simulator are (1) pending on the surface; (2) clogging effects
(surface clogging and biological clogging); (3) infiltration; (4) solute transport process (dis-
solved oxygen, ammonium nitrogen, nitrite/nitrate nitrogen, biodegradable organic carbon,
nonbiodegradable organic carbon, bacteria); (5) biochemical reaction process including nitro-
gen removal (nitrification and denitrification) and organics removal (oxidation); (6) tempera-
ture effects on unsaturated flow and biochemical reaction; and (7) system performance
evaluation (removal efficiency of nitrogen and organic carbon and infiltration volume and
infiltration volume and infiltration rate). Figures 13.4.8 and 13.4.9 are operation curves gen-
erated through a series of computer simulations of the hydrological submodel of NITRINFJL.
The NITRINFIL model was interfaced with an optimization procedure to develop an opti- ·
mization model referred to as SATOM (SAT Operation Model) for use in determining optimal
cycle times for SAT operation (Arizona State University et al.4).
Climate conditions: summer time

Soli type: silt
Degree of clogging: moderate
X= 10
X= 12
>:
«<
"0
X= 14
E X= 16 13
!::.. X= 18
Q)
~ X= 20
c
0
'P
~
!;:::
. .!;
Q)
Ol
e
Q)
13.
>
<%;
*9
X 6
X 5
X=4
Figure 13.4.8. SAT opera-
tion curve showing relation-
X=3
ships between average infil-
X=2 tration rate and operation
------x=1 schedule (applying time and
drying time) during summer
time (Arizona State
Y + Z: Draining time and drying time (day) University et al. 4).
Climate conditions: winter time

Soil type: silt
Degree of clogging: moderate
'• >: X= 10
m
u
...__ X::. 12
- E
.2- X=14
e X= 16
d ~ X= 18
c X= 20
0
l-
~
'E
i- Cl)
tl en
~
Cl)
>
<(
~~~ Figure 13.4.9. SAT opera-

X=6
X=5 tion curve showing relation-
X=4
X=3 ships between average infil-
X=2 tration rate and operation
X=1 schedule (applying time and
drying time) during winter
10 time (Arizona State
Y + Z: Draining time and drying time (day) University, et al.4).
13.5 RECHARGE MOUNDS

Groundwater mounds resulting from artificial recharge and stonnwater infiltration basins have
been studied by several investigators (Bouwer et al., 14 Glover,22 Griffen and Warrington,24
Guo,25•26• 27 Hantush,28 Marino,39· 40 Molden et al.,42 Morel-Seytoux et al.,44 Ortiz et al., 50 Rao
and Sarma,52 Rastogi and Pandy,53 Sumner et al.,60 Swamee and Ojha,62 and Warner et al. 68).
13.5.1 Perched Groundwater Mounds

A perched mound occurs when a mound is created above a restricting layer (see Figure 13.5.1 ).
In the case of a zero pressure head for the water at the bottom of the restricting layer, the height
of the perched mound above the restricting layer, LP, can be calculated using (Bouwer et al. 14)
1-
·n- (13.5.1)
il-
nd
ter where Lr is the thickness of the restricting layer; i is the infiltration rate, Kr is the hydraulic
conductivity of the restricting layer, and Ks is the hydraulic conductivity of soil above the
restricting the layer.
1
Figure 13.5.1. Geometry and symbols for perched
mound above restricting layer in vadose zones
(Bouwer 14 ).
I
Often, the infiltration rate is much smaller thanKs because surface soils are finer textured
than deeper soils. Also, there may be a clogging layer on the surface soil that reduces infiltra-
tion. The infiltration rate is often much larger than K,. For these conditions, Equation 13.5.1
can be simplified to
L =il2_ (13.5.2)
P Kr
LP should be small enough so that the top of the perched mound is deep enough to avoid reduc-
tions in infiltration rates. The above equations will overestimate the height of perched mounds
above noncontinuous perching layers (lenses or strips), because they are not as high as above
continuous perching layers with the same L, and K, because there is no lateral flow in the
perched mound.
The average annual recharge over an area of 9.57 square miles is 0.52 ft/year. A restricting layer of 10ft
thickness, having K, = 0.026 ft/day, is detected in the subsurface. If the hydraulic conductivity of the soil
above the restricting layer is 5.6 ft/day, estimate the height of the perched mound.
SOLUTION The given information in the problem statement is i =0.52 ftlyear =1.425 x 10-3 ft/day; Ks =5.6 ft/day;
K, =0.026 ft/day; and L, =10ft.
Substituting this information into Equation 13.5.1 yields a negative height because the hydraulic
conductivity of the so-called restricting layer is higher than the infiltration rate. In fact, the infiltration rate SO,
must be at least 0.026 ft/day or 9.49 ft/year in order for a perched mound above the restricting layer to
fu~ •
If the same area in Example 13.5.1 is artificially recharged at a rate of 0.09 ft/day, estimate the height of
EXAMPLE 13.5.2
the perched mound above the restricting layer.
SOLUTION Again, substituting the given information into Equation 13.5.1 yields
[
_i -l] 0.09 ft/day
1
LP = L, K, =(lOft) 0.026 ft/day (10ft) (3. 46 - ) =25ft
i ] 1- 0.09 ft/day (1- 0.016)
[l- Ks 5.6 ft/day
•
13.5.2 Steady-State Equations for Groundwater Mounds
For long strip basins (length of at least five times the width) groundwater flow away from the
strip can be approximated as linear horizontal flow (Dupuit-Forchheimer flow). Below an
Figure 13.5.2. Geometry and symbols

for groundwater mound below long
infiltration area (strip) (Bouwer 14).
d infiltration area, lateral flow can be assumed to increase linearly with distance from the center
l-
(see Figure 13.5.2). Lateral flow can be assumed to be constant between· the edge of the
1 recharge system at a distance W/2 from the center and the constant control water table at a dis-
tance L,1 from the edge (Figure 13.5.2). With these assumptions, Bouwer et aL 15 developed an
equation for the ultimate rise of a groundwater mound below the center of the recharge strip
~) in the case of equilibrium between recharge and pumping from the aquifer:
H -H =iW(W
- -+L ) (13.5.3)
c n 2T 4 n
is
te where He is the height of the groundwater mound in the center of the recharge area, Hn is the
1e
height of the groundwater table at the control area, i is the average infiltration rate in the
recharge area (total recharge divided by total area), W is the width of the recharge area, Ln is
the distance between the edge of the recharge area and the control area, and Tis the transmis-
sivity of the aquifer.
EXAMPLE 13.S.J A recharge basin 300m x 2,500 m is proposed for a new area. The recharge rate is 63,000,000 m 3 per
year. The depth to natural groundwater in the area is 46 m and the aquifer has a saturated thickness of 20
m. The transmissivity of the aquifer is 2,200 m2/day. If the maximum allowable rise of the recharge
mound is 38 m, how far should the control area be established from the-centerline of the recharge basin?
lie
1te SOLUTION SinceUW = 2500 m/300 m = 8.3 > 5, we can use Equation 13.5.3 developed for long strip basins. Given:
to CHc)max = 20 m + 38 m =58 m; Hn = 20 m; W = 300m; T = 2,200 m2/day; and
• 3
. 63,000,000 m /year = 84 rnlyear= 0.23 m/day,
t=
(300 mx2,500 m)
of
iW(W
- -+L )
H c -Hn =2T 4 n
(0.23rn/day)(300m)(300m )
58m-20m= 2
--+Ln
(2)(2200m /day) 4
L. =2350 m
Thus, the control area should be established 2,350 m + 150m= 2,500 m = 2.5 km or closer from
the centerline of the recharge basin. •
•
For a round or square recharge area, the groundwater flow is in a radial direction from the
recharge area. Bouwer et al. 15 used radial flow theory to develop an equation for the equilibrium
the height of the mound below the center of the recharge system, above the constant groundwater
an table at a distance Rn from the center of the recharge system.
(j)
~~~~;~~~
~~h-A-l
1'~ ® hi-- Hn Figure 13.5.3. Geometry and sym-
bols for groundwater mounds
below round infiltration area
(Bouwer 14).
2
H -H =iR- ( 1+2ln-
Rn) (13.5.4)
c n 4T R
where R is the radius or equivalent radius of the recharge area and Rn is the distance from the
center of the recharge area to the water table control area (see Figure 13.5.3).
Equations 13.5.3 and 13.5.4 can be used to determine where groundwater should be
recovered and to what depth groundwater levels should be pumped to keep the mound from
rising too high. These equations can also be used to determine the dimensions of the recharge
basins and to determine allowable recharge (infiltration) rates.
A circular recharge area with a radius of 500 m will be established over an unconfined aquifer where
depth to groundwater is 73 m. If the estimated recharge rate is 0.14 m/day and the groundwater mound
should be kept at least 10 m below the bottom of the recharge basin, detennine the distance from the cen-
ter of the recharge area to the control area where the original groundwater table is maintained. The trans-
missivity of the aquifer is 440 m2/day.
SOLUTION The following information is given: (Hc)max- Hn = 73 m- 10 m = 63 m; i = 0.14 m/day; R =500 m; T =

440m2/day. Equation 13.5.4 is used to compute Rn:
2
H -H =iR- ( 1+2ln-
Rn)
c n 4T R
(0.14 m/ day)(500 mf ( Rn )
63m= 1+2ln-
(4)(440m2/day) 500
Rn =1478m
Sl
Thus, the control area should be established approximately 1.5 km from the center of the recharge area .
•
13.5.3 Hantush Equation
Hantush28 developed the following equation to determine the height of the water table (mound)
as a function of location (x, y) and timet (see Figure 13.5.4):
hx,y,r- H = :~ {F( (W/2+ x)n,(U2 + y)n]+ F[(WI2+x)n,(U2- y)nj

(13.5.5)
+ F[ (W/2 - x)n, (U2 + y )n] + F[ (W/2 - x)n, (U2 - y )n]}
X
[l-
.4)
:he Figure 13.5.4. Geometry and symbols for

recharge system and groundwater mound
be (Bouwer 14).
)m
rge
where hx,y,t is the height of the water table above the impermeable layer at x, y, and time t; His
the original height of the water table above the impermeable layer; va is the arrival rate at the
water table of water from the infiltration basin; tis the time since the start of the recharge;fis
1ere the tillable porosity (1 > f> 0); Lis the length of the recharge basin (in y direction); W is the
und width of the recharge basin (in x direction); n = (4tT/f)- 112 ; and F(a,p) =J~ erf(a't- 112 ).
;en-
ms-
erf(p-r-112 )d't The values of F(a, P) are tabulated in Table 13.5.1, where a= (W/2 + x)n or
(W/2- x)n and p= (L/2 + y)n or (L/2- y)n.
T= The average leakage from the artificially constructed Tempe Town Lake in Tempe, Arizona, is 1.5 ft/day.
Depth to bedrock in the area is about 150 ft and the regional aquifer is 100 ft thick. The average trans-
missivity in the area is 10,000 ft 2/day and the tillable porosity of the aquifer is 0.2. The geometry of the
lake can be approximated by a rectangle that is 1,000 ft wide and 2 miles long. Assuming no natural gra-
dient of groundwater and no mound control by groundwater pumping, how far away from the center of
the lake must one move before the groundwater system doesn't feel the effect of the lake? How long does
it take before the area becomes waterlogged?
SOLUTION The answer to this problem is obtained by utilizing the Mathcad software and the exact solution proposed
by Hantush. The most important assumption/limitation in the present solution is that there is no natural
area.
• gradient of the regional groundwater. The application of the Hantush equation yielded the contour plots
of groundwater levels (shown in Figure 13.5.5) at t = 4, 6, 8, 12 days. The maximum rise of the ground-
water mound occurs directly below the center point of the lake and reaches 150 ft above the bedrock at
about 12 days (i.e., reaches the ground surface or the bottom of the lake).
Figure 13.5.6 shows the propagation of the groundwater mound along they-axis passing through the
center point of the lake (i.e., x = 0) as well as the propagation along the x-axis (i.e., y =0). It is difficult
1und)
to determine the distance from the center of the lake where the mound rise is negligible since the solu-
tion is still transient when the area starts to become waterlogged. On the other hand, a 0.5-ft rise. of the
mound is observed at about 2,680 ft from the center of the lake along the x-direction at t =12 days. Sim-
ilarly, a 0.5-ft increase is observed 7,150 ft from the center point along they-axis.
Groundwater mounding beneath the Tempe Town Lake has been controlled to minimize the water
~.5.5)
loss and the impact on regional groundwater levels by means of recovery wells located around the
Table 13.5.1 Values of the Function F (a.,~) in Equation 13.5.5 for Different Values of a. and~ 0.
-......)
N
13 0.02 0.04 0.06 0.08 0.10 0.14 0.18 0.22 0.26 0.30 0.34 0.38 0.42 0.46 0.50 0.54 0.58 0.62
(l
n
::::r
0.02 0.0041 0.0073 0.0101 0.0125 0.0146 0.0184 0.0216 0.0243 0.0267 0.0288 0.0306 0.0322 0.0337 0.0349 0.0361 0.0371 0.0380 0.0387 $:>:>
'"0
0.0073 0.0135 0.0188 0.0236 0.0278 0.0353 0.0416 0.0470 0.0518 0.0559 0.0596 0.0628 0.0657 0.0683 0.0705 0.0725 0.0743 0.0759 (t
0.04 ......
..........
0.06 0.0101 0.0188 0.0266 0.0335 0.0398 0.0509 0.0602 0.0684 0.0754 0.0817 0.0871 0.0920 0.0963 0.1001 0.1035 0.1065 0.1091 0.1115 V.l
0.08 0.0125 0.0236 0.0335 0.0425 0.0508 0.0652 0.0776 0.0884 0.0978 0.1060 0.1133 0.1197 0.1254 0.1305 0.1350 0.1389 0.1425 0.1456
0.10 0.0146 0.0278 0.0398 0.0508 0.0608 0.0786 0.0939 0.1072 0.1188 0.1290 0.1381 0.1461 0.1532 0.1595 0.1650 0.1700 0.1744 0.1783 ~
51
(")
0.14 0.0184 0.0353 0.0509 0.0652 0.0786 0.1025 0.1232 0.1414 0.1573 0.1714 0.1839 0.1941 0.2048 0.2135 0.2212 0.2281 0.2343 0.2397 §.:
0.18 0.0216 0.0416 0.0602 0.0776 0.0939 0.1232 0.1490 0.1716 0.1916 0.2094 0.2251 0.2391 0.2515 0.2626 0.2724 0.2812 0.2890 0.2959 ::=t'
(I>
0.22 0.0243 0.0470 0.0684 0.0884 0.1072 0.1414 0.1716 0.1984 0.2222 0.2433 0.2621 0.2789 0.2938 0.3071 0.3189 0.3295 0.3389 0.3472 (")
::::r
0.26 0.0267 0.0518 0.0754 0.0978 0.1188 0.1573 0.1916 0.2222 0.2494 0.2737 0.2954 0.3147 0.3320 0.3474 0.3612 0.3735 0.3844 0.3941 ~
(I>
0.30 0.0288 0.0559 0.0817 0.1060 0.1290 0.1714 0.1094 0.2433 0.2737 0.3009 0.3252 0.3470 0.3665 0.3839 0.3995 0.4134 0.4257 0.4368 0.......,
0.34 0.0306 0.0596 0.0871 0.1133 0.1391 0.1839 0.2251 0.2621 0.2954 0.3252 0.3520 0.3761 0.3976 0.4169 0.4341 0.4495 0.4633 0.4756
0.38 0.0322 0.0628 0.0920 0.1197 0.1461 0.1949 0.2391 0.2789 0.3147 0.3470 0.3761 0.4022 0.4256 0.4466 0.4654 0.4823 0.4973 0.5108 ~
~
:::s
0.42 0.0377 0.0657 0.0963 0.1254 0.1532 0.2048 0.2515 0.2938 0.3320 0.3665 0.3976 0.4256 0.4508 0.4734 0.4937 0.5119 0.5281 0.5427 0...
::8
0.46 0.0349 0.0683 0.1001 0.1305 0.1595 0.2135 0.2626 0.3071 0.3474 0.3839 0.4169 0.4466 0.4734 0.4975 0.5161 0.5385 0.5559 0.5715
~
......
0.50 0.0361 0.0705 0.1035 0.1350 0.1650 0.2212 0.2724 0.3189 0.3612 0.3995 0.4341 0.4654 0.4937 0.5191 0.5420 0.5626 0.5810 0.5975
0.54 0.0371 0.0725 0.1065 0.1389 0.1700 0.2281 0.2812 0.3295 0.3735 0.4134 0.4495 0.4823 0.5119 0.5385 0.5626 0.5842 0.6036 0.6209
0.58 0.0380 0.0743 0.1091 0.1425 0.1744 0.2343 0.2890 0.3389 0.3844 0.4257 0.4633 0.4973 0.5281 0.5559 0.5810 0.6036 0.6238 0.6420
0.62 0.0387 0.0759 0.1115 0.1456 0.1783 0.2397 0.2959 0.3472 0.3941 0.4368 0.4756 0.5108 0.5427 0.5715 0.5975 0.6209 0.6420 0.6609
0.66 0.0394 0.0773 0.1136 0.1484 0.1718 0.2445 0.3020 0.3547 0.4027 0.4466 0.4865 0.5227 0.5556 0.5854 0.6122 0.6364 0.6582 0.6778
0.70 0.0401 0.0785 0.1154 0.1509 0.1849 0.2488 0.3075 0.3612 0.4104 0.4553 0.4962 0.5334 0.5672 0.5977 0.6254 0.6503 0.6728 0.6929
0.74 0.0406 0.0796 0.1117 0.1531 0.1876 0.2526 0.3123 0.3671 0.4172 0.4630 0.5048 0.5429 0.5774 0.6087 0.6371 0.6627 0.6857 0.7064
0.78 0.0411 0.0806 0.1185 0.1550 0.1900 0.2559 0.3166 0.3722 0.4232 0.4699 0.5125 0.5513 0.5865 0.6185 0.6475 0.6736 0.6972 0.7184
0.82 0.0415 0.0814 0.1198 0.1567 0.1921 0.2589 0.3203 0.3768 0.4286 0.4760 0.5192 0.5587 0.5946 0.6272 0.6567 0.6834 0.7074 0.7291
0.86 0.0419 0.0822 0.1209 0.1582 0.1940 0.2615 0.3237 0.3808 0.4333 0.4813 0.5252 0.5653 0.6017 0.6348 0.6648 0.6920 0.7165 0.7386
0.90 0.0422 0.0828 0.1219 0.1595 0.1957 0.2638 0.3266 0.3844 0.4374 0.4860 0.5305 0.5711 0.6080 0.6416 0.6721 0.6996 0.7245 0.7469
0.94 0.0425 0.0834 0.1228 0.1607 0.1971 0.2658 0.3292 0.3875 0.4411 0.4902 0.5351 0.5762 0.6136 0.6476 0.6784 0.7063 0.7316 0.7543
0.98 0.0428 0.0839 0.1236 0.1617 0.1984 0.2676 0.3314 0.3902 0.4442 0.4938 0.5392 0.5807 0.6184 0.6528 0.6840 0.7123 0.7378 0.7608
1.00 0.0429 0.0842 0.1239 0.1622 0.1990 0.2684 0.3324 0.3914 0.4457 0.4955 0.5410 0.5827 0.6206 0.6552 0.6865 0.7150 0.7406 0.7638
1.20 0.0437 0.0858 0.1263 0.1654 0.2030 0.2740 0.3396 0.4001 0.4558 0.5070 0.5540 0.5969 0.6362 0.6719 0.7044 0.7339 0.7605 0.7846
1.40 0.0441 0.0866 0.1275 0.1669 0.2049 0.2767 0.3431 0.4043 0.4608 0.5127 0.5603 0.6039 0.6438 0.6801 0.7132 0.7432 0.7704 0.7949
1.80 0.0444 0.0871 0.1283 0.1680 0.2062 0.2785 0.3454 0.4071 0.4641 0.5165 0.5645 0.6086 0.6489 0.6856 0.7190 0.7494 0.7769 0.8018
2.00 0.0444 0.0871 0.1284 0.1681 0.2064 0.2787 0.3457 0.4075 0.4645 0.5169 0.5651 0.6092 0.6495 0.6863 0.7198 0.7502 0.7778 0.8027
2.20 0.0444 0.0872 0.1284 0.1682 0.2065 0.2788 0.3458 0.4076 0.4646 0.5171 0.5653 0.6094 0.6497 0.6865 0.7200 0.7505 0.7781 0.8030
2.50 0.0444 0.0872 0.1284 0.1682 0.2065 0.2788 0.3458 0.4077 0.4647 0.5172 0.5653 0.6095 0.6498 0.6867 0.7202 0.7506 0.7782 0.8032
3.00 0.0444 0.0872 0.1284 0.1682 0.2065 0.2789 0.3458 0.4077 0.4647 0.5172 0.5654 0.6095 0.6499 0.6867 0.7202 0.7506 0.7782 0.8032
,.._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _""'""'=-<'' .... .,. __ ........... ~"""'''..;~~
le 13.5.1 (continued) Values of the Function F ( ('l_ f\) in F.mmt1on 1 ~. c; c; for n1ffP.rP.nt V::~lnP.<:: of N ::~n£1 R
Table 13.5.1 (continued) Values of the Function F (a, j3) in Equation 13.5.5 for Different Values of a and j3
13 0.62 0.66 0.70 0.74 0.78 0.82 0.86 0.90 0.94 0.98 1.00 1.20 1.40 1.80 2.00 2.20 2.50 3.00
a
0.02 0.0387 0.0394 0.0401 0.0406 0.0411 0.0415 0.0419 0.0422 0.0425 0.0428 0.0429 0.0437 0.0441 0.0444 0.0444 0.0444 0.0444 0.0444
0.04 0.0759 0.0773 0.0785 0.0796 0.0806 0.0814 0.0822 0.0828 0.0834 0.0839 0.0842 0.0858 0.0866 0.0871 0.0871 0.0872 0.0882 0.0882
0.06 0.1115 0.1136 0.1154 0.1171 0.1185 0.1198 0.1209 0.1219 0.1228 0.1236 0.1239 0.1263 0.1275 0.1283 0.1284 0.1284 0.1284 0.1284
0.08 0.1456 0.1484 0.1509 0.1531 0.1550 0.1567 0.1582 0.1595 0.1606 0.1617 0.1622 0.1654 0.1669 0.1680 0.1681 0.1682 0.1682 0.1682
0.10 0.1783 0.1818 0.1849 0.1876 0.1900 0.1921 0.1940 0.1957 0.1971 0.1984 0.1990 0.2030 0.2049 0.2062 0.2064 0.2065 0.2065 0.2065
0.14 0.2397 0.2445 0.2488 0.2526 0.2559 0.2589 0.2615 0.2638 0.2658 0.2676 0.2684 0.2740 0.2777 0.2785 0.2787 0.2788 0.2788 0.2788
0.18 0.2959 0.3020 0.3075 0.3123 0.3166 0.3203 0.3237 0.3266 0.3292 0.3314 0.3224 0.3396 0.3431 0.3454 0.3457 0.3454 0.3454 0.3454
0.22 0.3472 0.3547 0.3612 0.3671 0.3722 0.3768 0.3808 0.3844 0.3875 0.3902 0.3914 0.4001 0.4043 0.4071 0.4075 0.4076 0.4077 0.4077
0.26 0.3941 0.4027 0.4104 0.4172 0.4232 0.4286 0.4333 0.4374 0.4411 0.4442 0.4457 0.4558 0.4608 0.4641 0.4645 0.4646 0.4647 0.4647
0.30 0.4368 0.4466 0.4553 0.4630 0.5699 0.5760 0.4813 0.4860 0.4902 0.4938 0.4955 0.5070 0.5127 0.5165 0.5159 0.5171 0.5172 0.5172
0.34 0.4756 0.4865 0.4962 0.5048 0.5125 0.5192 0.5252 0.5305 0.5351 0.5392 0.5410 0.5540 0.5603 0.5645 0.5651 0.5653 0.5653 0.5654
0.38 0.5108 0.5227 0.5334 0.5429 0.5213 0.5587 0.5653 0.5711 0.5762 0.5807 0.5827 0.5969 0.6039 0.6086 0.6092 0.6094 0.6095 0.6095
0.42 0.5427 0.5556 0.5672 0.5774 0.5865 0.5946 0.6017 0.6080 0.6136 0.6184 0.6206 0.6362 0.6438 0.6489 0.6495 0.6497 0.6498 0.6499
0.46 0.5715 0.5854 0.5977 0.6087 0.6185 0.6272 0.6348 0.6416 0.6476 0.6528 0.6552 0.6719 0.6801 0.6856 0.6863 0.6865 0.6867 0.6867
0.50 0.5975 0.6122 0.6254 0.6371 0.6475 0.6567 0.6648 0.7721 0.6784 0.6840 0.6865 0.7044 0.7132 0.7190 0.7198 0.7200 0.7202 0.7202
0.54 0.6209 0.6364 0.6503 0.6627 0.6736 0.6834 0.6920 0.6996 0.7063 0.7123 0.7150 0.7379 0.7432 0.7494 0.7502 0.7505 0.7506 0.7506
0.58 0.6420 0.6482 0.6728 0.6857 0.6972 0.7074 0.7165 0.7245 0.7316 0.7378 0.7406 0.7605 0.7704 0.7799 0.7778 0.7781 0.7782 0.7782
0.62 0.6609 0.6778 0.6929 0.7064 0.7784 0.7291 0.7386 0.7469 0.7543 0.7608 0.7638 0.7846 0.7949 0.8018 0.8027 0.8030 0.8032 0.8032
0.66 0.6778 0.7953 0.7110 0.7250 0.7375 0.7486 0.7584 0.7671 0.7748 0.7816 0.7846 0.8064 0.8171 0.8243 0.8252 0.8255 0.8257 0.8257
0.70 0.6929 0.7110 0.7272 0.7417 0.7546 0.7660 0.7762 0.7852 0.7932 0.8002 0.8034 0.8259 0.8370 0.8445 0.8454 0.8458 0.8460 0.8460
0.74 0.7064 0.7250 0.7414 0.7566 0.7698 0.7816 0.7921 0.8014 0.8096 0.8168 0.8201 0.8434 0.8549 0.8627 0.8636 0.8640 0.8642 0.8642
0.78 0.7184 0.7375 0.7546 0.7698 0.7834 0.7956 0.8063 0.8159 0.8243 0.8317 0.8351 0.8591 0.8710 0.8789 0.8799 0.8803 0.8805 0.8805
0.82 0.7291 0.7486 0.7660 0.7816 0.7956 0.8080 0.8190 0.8288 0.8374 0.8450 0.8485 0.8731 0.8853 0.8935 0.8945 0.8949 0.8951 0.8951
0.86 0.7386 0.7584 0.7762 0.7921 0.8063 0.8190 0.8302 0.8402 0.8491 0.8569 0.8604 0.8855 0.8980 0.9065 0.9075 0.9079 0.9081 0.9081
0.90 0.7469 0.7671 0.7852 0.8014 0.8159 0.8288 0.8402 0.8504 0.8594 0.8674 0.8710 0.8966 0.9094 0.9180 0.9191 0.9195 0.9187 0.9187
0.94 0.7543 0.7748 0.7932 0.8096 0.8243 0.8374 0.8491 0.8594 0.8686 0.8767 0.8803 0.9064 0.9195 0.9282 0.9294 0.9298 0.9300 0.9300
0.98 0.7608 0.7816 0.8002 0.8168 0.8317 0.8450 0.8569 0.8674 0.8767 0.8849 0.8886 0.9151 0.9284 0.9373 0.9384 0.9389 0.9391 0.9391
1.00 0.7638 0.7846 0.8034 0.8201 0.8351 0.8485 0.8604 0.8710 0.8803 0.8886 0.8924 0.9191 0.9324 0.9414 0.9426 0.9430 0.9432 0.9433 .........
w
1.20 0.7846 0.8064 0.8259 0.8434 0.8591 0.8731 0.8855 0.8966 0.9064 0.9151 0.9191 0.9472 0.9614 0.9709 0.9722 0.9726 0.9728 0.9729 i..Jl
1.40 0.7949 0.8171 0.8370 0.8549 0.8710 0.8853 0.8980 0.9094 0.9195 0.9284 0.9324 0.9614 0.9759 0.9858 0.9871 0.9875 0.9878 0.9878
:;o
(1)
1.80 0.8018 0.8243 0.8445 0.8627 0.8789 0.8935 0.9065 0.9180 0.9282 0.9373 0.9414 0.9709 0.9858 0.9959 0.9972 0.9972 0.9879 0.9980 (")
:::r
2.00 0.8027 0.8252 0.8454 0.8636 0.8799 0.8945 0.9075 0.9191 0.9594 0.9384 0.9426 0.9722 0.9871 0.9972 0.9985 0.9990 0.9992 0.9934 ~
(1)
2.20 0.8030 0.8255 0.8458 0.8640 0.8803 0.8949 0.9079 0.9195 0.9298 0.9389 0.9430 0.9726 0.9875 0.9977 0.9990 0.9995 0.9997 0.9998
2.50 0.8032 0.8257 0.8460 0.8642 0.8805 0.8951 0.9081 0.9197 0.9300 0.9391 0.9432 0.9728 0.9878 0.9979 0.9992 0.9997 1.0000 1.0000
3::
0
~
3.00 0.8032 0.8257 0.8460 0.8642 0.8805 0.8951 0.9081 0.9197 0.9300 0.9391 0.9433 0.9729 0.9878 0.9980 0.9993 0.9998 1.0000 1.0000 ::::
0..
[/)
Source: From Hantush28

Ul
.....:a
VJ
15 115 105
110
110
105 r 130
125
110
105
110 120 120
50 50
~JH20)
10~~
Jrr
20 40 20 40
t= 4 days f= 6 days
Figure 13.5.5. Growth of groundwater mound beneath Tempe Town Lake using Hantush Equation (Scale: 1 unit= 100ft). Fi
periphery of the lake since its construction. It should be noted that the solution presented here assumes
no artificial mound control. •
BXAMPLE 13. 6 A square recharge basin 800 m x 800 m is constructed over an area where an unconfined aquifer is found
in the subsurface. The thickness of the aquifer is 50 m and the average transmissivity is 1,140 m2/day. The
tillable porosity of the vadose zone is 0.25. If the aquifer is recharged at a rate of25,000,000 m3 per year
and the maximum permissible rise of the groundwater mound is 25 m above the original water table, how
long does it take for the mound to rise to its maximum level? How much water can be stored underground
until that time?
SOLUTION We are interested only in the maximum rise of the mound. The given information is W = 800 m;
L = 800 m; x = 0 m; y = 0 m;J= 0.25; H =50 m; T= 1,140 m2/day and
25 106 3
va = x m /year = 39.0625 m/year = 0.107 m/day
800mx800 m
ho,o,r =50m+25m=75m
and
12 2
4tT]- ' [4(t)(1,140 m /day)]-!IZ _112 -1
n = [- = =0.0074(t)
f 0.25
100 100
105
140
I
105
110
115 135
115
110
140
r 130
125
120
115
i1 135 130 110105

50 50
1
110 il
130
12
\o1
I II
110 130
115
1~:5
20 40 20 40
t= 8 days t= 12 days
Figure 13.5.5 (continued). Growth of groundwater mound beneath Tempe Town Lake using Hantush Equation (Scale: 1 unit = 100 ft).
~--------------------150~----------------------~
nes
•
Jnd
fhe
rear
lOW
1qQ
und ::
e ji
125
~ ;:
lm;
Figure 13.5.6. Propagation of

groundwater mound along x- and
y-axes
Solid line: Propagation along y-
. /~).n \\~;;~\-
'1 00
i,___...J-~~'------L--..:.:.;···..:..:··.;L,.:··;:....:·
i
·. •..........
axis for t =1 day, 2 days, 4 days,
8 days, and 12 days
Dashed line: Propagation along
-1 0,000 -8000 -6000 -4000 -2000 0 2000 4000 6000 8000 10,000
x-axis for t = 1 day, 2 days, 4 days,
(x, 0) or (0, y) coordinate
8 days, and 12 days
Table 13.5.2 Results for Example 13.5.6
t (years) n a=~ F(a, ~) ho,o,t (m)

0.50 5.481E-04 0.22 0.1984 65.50
0.74 4.505E-04 0.18 0.1490 67.22
1.23 3.495E-04 0.14 0.1025 69.70
2.40 2.502E-04 0.10 0.0608 72.80
3.75 2.001E-04 0.08 0.0425 74.90
6.70 1.497E-04 0.06 0.0266 77.84
15.00 l.OOlE-04 0.04 0.0135 81.63
Substituting the above information into Equation 13.5.5 yields
ho,o,t -H= :~ {F[(WI2+x)n,(U2+y)n]+F[(WI2+x)n,(U2-y)n]

+ F[(WI2-x)n,(U2+ y)n]+ F[(W/2- x)n,(U2- y)n]}
(0.107 m/day)(t) { }
h0,0,, -50 m = (0. S) 4 F( 400n, 400n)
4 2
The values of ho,O,t can be found for a range oft values using Table 13.5.1 as shown in the following Table
13.5.2. Note that t values in the table are chosen such that no interpolation was necessary while using
Table 13.5.1. The results show that the height of the mound would reach 75 m above the impermeable
bed in about 3.75 years. A more accurate answer could be obtained using the integral expression for F(a,
~) instead of Table 13.5.1, while the answer above is sufficient for practical purposes. Finally, the vol-
ume of water that can be stored underground in 3.75 years is given by
V = (25 x 106 m3/year)(3.75 year)= 93.75 million m3
•
13.6 STORMWATER INFILTRATION BASIN MOUND DEVELOPMENT
Groundwater mounds resulting from stormwater infiltration basins have been studied by sev-
eral investigators (Griffen and Warrington, 24 Guo, 25 · 26· 27 Marino, 39• 40 Molden et al., 42 Morel-
Seytoux et al., 44 Ortiz et al., 50 Sumner et al., 60 and Swamee and Ojha62 ).
13.6.1 Potential Flow Model for a Trench

Guo 25 • 26• 27 utilized potential flow theory to depict the seepage flow through soils, considering
both vertical and lateral movements. Guo used a stream function for infiltrating flow under a
trench, defined to satisfy two-dimensional continuity and irrotationality (Guo25 ) as 'P =if!D)xy.
Here,f is the infiltration rate, D is the saturated depth, xis the horizontal distance from the cen-
tral axis of the basin, and y is the vertical distance below the basin. Variables and parameters
are illustrated in Figure 13.6.1. At a point C(B, D) the stream function is 'Pc =q =fB, where q
is the infiltration volume rate and B is the half-width of the trench. In order to maintain conti-
nuity along the streamline, 'Pc• between the basin and any cross section at a vertical distance
y, the width of the wetting front, w, can be determined using 'P =if/D)xy, so that w =BD!y.
The velocity vectors of seepage flow are determined by taking the derivative of 'P =
(j!D)xy with respect to x andy to obtain u =(j!D)x and v =-(j/D)y. For steady-state conditions,
the flow rate is constant at sections AC, PR, and RF, shown in Figure 13.6.1. When the soil
suction is ignored, the hydraulic gradient at PR can be approximated as unity (i = -1, down-
ward), so that from Darcy's law, the flow of recharge at PR is v =q!W =KJWIW =-KY, where
KY is the hydraulic conductivity in the vertical direction. Also at y = H, v =-(j!D)H and also v
=-KY, so that (j!D)H =Ky- Solving forD= if/Ky)H, then we find D ='Ayfl, where Ay =f/Ky.
13.6 Stormwater Infiltration Basin Mound Development 577
U=O
8----J, - - - - - - - - - - - - - - -
----------- -----------
Z D
Groundwater
H
Figure 13.6.1. Potential
j 0
flow pattern for infiltrating
water under a trench
V= 0 E V= 0 F (Guo 27 ).
Streamlines of recharge flow near the mound are concentrated in the upper or active thick-
ness of the aquifer, with much less flow and almost stagnant water in the deeper portion of the
a, aquifer (Bouwer 14). The active flow depth, H, below the initial groundwater table can be esti-
Jl- mated by the Dupuit-Forchheimer equation,
(D2 - H2)
• q = \f
c
= Kx
2 W-B
(13.6.1)
in which Kx is the hydraulic conductivity in the later~J direction.

Guo 25 reasoned that storm water infiltration basins are small in size and are operated using
~v- short loading periods, so that the height of the water mound decays toward the initial ground-
d- water table. Under this condition, Equation 13.6.1 can be approximated to the edge of the area
of recharge, that is, point R(W, H) in Figure 13.6.1. Equation 13.6.1 divided by the effective
flow depth, H, yields a sectional velocity which must be equal to u =(j/D)x at x = W; resulting
in an effective thickness of the aquifer expressed as
ing (13.6.2)
~r a
~xy.
where
:en-
ters
re q
'nti- Using D = 'Afi, we can compute the saturated depth:
mce
ly.
¥=
D =1..
B Y
J 21..x
'Ay +1
(13.6.3)
ons,
soil
lWn- Consider an infiltration basin 20ft by 180ft with an average soil infiltration rate of 1.8 in/hr (3.6 ft/day)
'EX:AMPLE 13.6.1
here and a hydraulic conductivity of 1.8 ft/day. The subsurface medium is isotropic: Kx = Ky = K. Determine
lso v the saturated depth, D, and the effective flow depth, H, for a half-width of the trench of B = 10ft. Also
ry. determine Z, the required distance to the groundwater table.
SOLUTION The ratio of the average infiltratration rate to the hydraulic conductivity is jj K = 2.0. The subsurface
medium is isotropic so that A.x = A.Y = f)K< :::;f)Ky = 2.0. From Equation 13.6.3, we have
2 22
D =A,) Ax =2 ( .0) =2.309 andD= B(D B)=(!Oft)(2.309)=23.09ft.
B · Ay+1 2.0+1
The effective flow depth, H, is computed using Equation 13.6.2, so that
22
H
B
=J 2/..x = { ( .of=l.l55andH=B(H B)=(l0ft)(1.155)=11.55ft.
1.),+1 ~2.0+1
The required distance to the groundwater table is Z = D- H = 23.09- 11.55 = 1l.54 ft.
•
13.6.2 Potential Flow Model for Circular Basin
The flow pattern below a circular basin (see Figure 13.6.2), described by the stream function,
'¥ = rt(j!D)?y, satisfies the continuity and irrotationality equations for three-dimensional
axisymmetric flow (Guo 25). Streamlines below the basin are distributed as concentric circles
with 'I' =0 along the y-ax.is and 'I'= infiltration volume rate at the circumference of the basin
bottom (i.e., C(r, y) = (R0 , D) in Figure 13.6.2). R0 is the radius of the circular basin.
The infiltration volume rate released from the circular basin is q =frtRl To maintain con-
tinuity of flow, then q =frtR02 ='¥ = 1t(f!D)?y, so solving for r gives r = ~ Dly~. Then at sec-
tion PR in Figure 13.6.2, the radius of the area of recharge must pass pointR(R, H), expressed as
R =~ D/ H~- Using the same approach as for the rectangular basin, the following relation-
ships can be developed for the subsurface geometry,
H I/..,ln/..Y
~ = ~ 2(t..; -1) (13.6.4)
Ground surface
Unsaturated
zone
uand q
0 F
Saturated zone Figure 13.6.2. Flow pattern under circular
Impervious layer basin (Guo27 ).
13.6 Storrnwater Infiltration Basin Mound Development 579
where
and Kr is the hydraulic radius in the radial direction. Substituting
into Equation 13.6.4 gives the following expression for computing the saturation depth:
I
(13.6.5)
1,
EXAMPLE 13.6.2 A stormwater infiltration basin has a radius of 72ft. The basin is lined with loamy sand and the result-
al
ing average infiltration rate is 1.60 in/hr. Determine the required saturation depth, Y0, between the bot-
~s
tom of the basin and the groundwater table if the hydraulic conductivity of the vadose zone is 1.6 ft/day.
lll
Assume an isotropic subsurface.
n-
c- SOLUTION The following information is given:f= 1.60 inlhr; K, = KY = 1.6 ft/day = 0.8 inlhr; and R0 =72ft. The
as required saturation depth, Y0, is equal to (D- H). Thus, combining Equations 13.6.4 and 13.6.5 yields
n-
Since the subsurface is isotropic,

4)
Then
2
Yo =(2.0-1) (~ 2) =0.481
R0 2(2 -1)
Y0 =0.481(72 ft) = 34.6 ft
Thus, the minimum required distance between the bottom of the basin and the groundwater table is about
35 ft, so that there will be a sufficient hydraulic gradient in the subsurface. •
EXAMPLE 13.6.3 The saturation depth, the distance between the bottom of the basin and the groundwater table, beneath a
storm water infiltration basin is 5.5 m. If the design infiltration rate from the basin is 5.13 cmlhr, deter-
mine and plot the required basin radius as a function of the hydraulic conductivity. Assume the hydraulic
conductivity beneath the basin is isotropic.
SOLUTION The required saturation depth Y0 is given by (D- H). As in the previous example, combining Equations
13.6.4 and 13.6.5 yields
1200,-----------------,450
-o- Maximum permissible
basin radius 400
1000 ~ Maximum infiltration
volume rate 350
800 300
£
M- 600
25o I
,S
()
200 tr:.o
400 150
100
200
50
5 4 3 2 Figure 13.6.3. Results for

K (cm/hr) Example 13.6.3
However, since the saturation depth is specified and fixed in this problem, the basin radius is the design
parameter as a function of the unknown hydraulic conductivity. Thus, we must solve this relationship
for R0:
Assuming an isotropic formation
and substituting into above the expression, we have

_ 5.5 m
Ro-
5.13ln 5.13
(
5.13 lJ K K
K- z(e;3r -IJ
where K is in cmlhr and the basin radius, R0 , is in meters. A plot of R0 versus K is shown in Figure 13.6.3.
If K ~ 5.13 cm/hr, this would be the case of a direct downward recharge to the groundwater. In fact,
(R0)max -7 + oo asK -7 5.13 cm/hr, as shown in Figure 13.6.3. On the other hand, when K~ 5.13 cm/hr,
there is a maximum permissible basin radius depending on the K value while the saturation depth is a
fixed value. This constraint is to assure there will be sufficient hydraulic gradients beneath the basin to
prevent the infiltration system from being backed up (Guo26 •27 ). Note that as the permissible basin radius
decreases with decreasing K values, the corresponding infiltration volume rates decrease even more sig-
nificantly. This shows that a system with a highftK ratio, while the saturation depth is fixed, may prove
to be very inefficient as a stormwater infiltration system. •
13.6.3 Mound Growth

The growth (before steady state is reached) of a mound for flow under an infiltration trench is
depicted in Figure 13 .6.4. To define the storage effect in the soil as a function of time, the con-
tinuity equation between the intlow at section AC and the outflow at section PR is
13.6 Stormwater Infiltration Basin Mound Development 581
dV
- = q-qo
dt
where Vis the soil water storage volume, tis time, q is the infiltrating volumetric rate, and q0
is the outflow rate to recharge. The incremental volume, dV, at depth y is
D
dV == Syw dy =Sy - Bdy
y
in which SY is the soil yield. The outflow rate, q0 , at Section PR is defined using Darcy's law as
D
qo =SyKu W =SyKu -H B
in which Ku = ~ K,KY is the representative or equivalent hydraulic conductivity for the
growth of the mound and
BD
w=-
y
The growth function of an infiltration trench can be developed by substituting the expres-
~n
sion dV = SY(Diy)B dy into the continuity equation,
ip
dV
-=q - qo
dt
and integrating from the thickness of the aquifer, H, to the depth y to obtain
y =He-laiD (13.6.6)
in which k is a pseudo-hydraulic conductivity for the growth of a water mound, expressed as
k=[L-K
S Dl
"H
J
(13.6.7)
Ground surface
Unsaturated
.3. zone
ct,
hr.
;a z Direction of
to
.US D
tgrowth of mound
ig-
.Ye
is 0 F
Saturated zone
ifl- Figure 13.6.4. Growth of water mound under-
Impervious layer neath a trench (Guo27).
Equation 13.6.6 is an increasing function with depth y, which cannot exceed Din Equation
13.6.3 for an infiltration trench or Equation 13.6.5 for a circular infiltration basin. Equation ·
13.6.6 can be used to estimate the growth of a water mound for an elapsed timet. For circular
infiltration basins, the same derivation procedure can be used to show that Equation 13.6.6 also
applies. Generally, the exponent (-kt!D) is small enough so that Equation 13.6.6 reduces to
(Guo26. n)
(13.6.8)
EXAMPLE 13.6.4 A circular stonnwater infiltration basin has a radius of 29 m. The saturation depth (distance from the
basin bottom to the impermeable bedrock) is 18.5 m and the saturated thickness of the unconfined
aquifer beneath the basin is 13.5 m. The saturated hydraulic conductivity of the aquifer is 2.4 rnfday.
The infiltration rate from the basin is 1.25 cmlhr and the soil yield, Sy, is estimated as 0.25. Assuming a
representative unsaturated hydraulic conductivity of 0.23 m/day, compute the pseudo-hydraulic
conductivity.
SOLUTION To determine the pseudo-hydraulic conductivity, we solve Equation 13.6.7 withf = 1.25 cmlhr = 0.3
m/day to obtain
k= [L-
Sy
Ku Dl =[OJ0.25
H
rnfday
18 5
(0.23 m/day) ( · m)l =0.8848 m/day
(13.5 m)
To estimate the growth of the mound, we would solve Equation 13.6.6.

•
13.6.4 Mound Recession
A mound begins to recede during the time of no infiltration from a trench or basin by gradu-
ally recharging the aquifer (see Figure 13.6.5). The volume release from the soil medium is
equal to the volume of recharge, at Section PR to the groundwater table. The recession of a
mound starts at the bottom. If the receding water front maintains the shape as the saturated, the
volume released from the soil is estimated as
Vr :::: Sy y]: dy = Sy yt BD dy = SyDB In[ Dl (13.6.9)

y=y y=y y y
Considering vertical flow through section PR, the volume of recharge to the groundwater
table is
(13.6.10)
Equating the above two expressions for the volume gives the following recession function for
a mound:
y =De-{K,t/HJ (13.6.11)
For a small exponent, the recession function reduces to
(13.6.12)
13.7 Innovative Approaches 583
n Ground surface
ll t-:-:-:-:-: Water:-:-:-:-:-:1
----- --r ~-- ,,
r
0
~ t and q Unsaturated
0
zone
t----1C(8, D)
:)
! Direction of
growth of mound
te y
:d
y. Groundwater table I
a
ic
0 F
Saturated zone
Figure 13.6.5. Recession of water mound
.3 Impervious layer (Guo 2\
The unsaturated hydraulic conductivity, Ku, is the most sensitive parameter in using the
above models to predict the gmwth and recession of a water mound. Ku varies with respect to
the soil moisture content as discussed in Chapter 3. Guo26 compared the results of using the
• recession function described above with results from using the unsaturated flow simulation
model VS2D developed by Lappala et al. 38
u- 13.7 INNOVATIVE APPROACHES

Is Several innovative approaches employing artificial recharge are described by Alley et al3 as
'a follows:
he
13.7.1 Wildwood, New Jersey
.9) Wildwood, New Jersey, is a resort town on a barrier island along the Atlantic Coast. As a resort
corrununity, it has a large influx of tourists in the summertime. The population can increase from
about 5,000 during the winter to 30,000 at the height of the summer tourist season. The Wild-
ter wood Water Utility withdraws water from wells located about 5 miles inland from the barrier
island. To supply water from these wells to meet the island's needs during the summer would
require a large pumping facility and transmission lines that would be little used the rest of the
.0) year. To avoid these excessive costs, the utility injects groundwater into a shallow aquifer on the
island during periods of low demand and withdraws the water in the summer by using dual injec-
for tion and recovery wells. This system, operated since 1967, represents perhaps the oldest opera-
tional aquifer storage recovery (ASR) project in the United States. In ASR, water is injected
underground, commonly into nonpotable or saline aquifers, where it forms a lens of good qual-
ll) ity water for later recovery from the sanne well(s). Operation of a typical ASR installation is
designed to smooth out annual variability in water demand by recharging aquifers during peri-
ods of low demand and recovering the water during periods of high demand. Advantages of ASR
12) over other artificial recharge schemes are that it uses very little land (especially compared to sur-
face spreading) and reduces the cost and maintenance of separate injection and recovery wells.
13.7 .2 Orange County, California

1
Orange County, near Los Angeles, California, receives an average of only 13 to 15 inches of
rainfall annually, yet sustains a population of approximately 2.5 million people. A groundwater
basin that underlies the northwestern half of the county supplies about 75 percent of the total
water demand. As the area developed from a thriving agricultural center into a highly urban-
ized area, increased demands for groundwater resulted in a gradual lowering of the water table
below sea level and encroachment of saltwater from the Pacific Ocean. To prevent further salt-
water intrusion and to replenish the groundwater supply, the Orange County Water District
operates a hydraulic barrier system composed of a series of 23 multipoint wells that inject
fresh water into the aquifer, blocking further passage of seawater. The source of injection water
is a blended combination of deep well water and recycled secondary effluent. The recycled
product meets drinking-water standards through advanced treatment processes of reverse
osmosis and activated carbon adsorption. F
p
13.7.3 Long Island, New York
Groundwater is the sole source of fresh water for the more than 3 million people who live on
Long Island, outside the metropolitan New York City boundary. To help replenish the aquifer,
as well as reduce urban flooding and control saltwater intrusion, more than 3,000 recharge
basins dispose of storm runoff at an average rate of about 150 million gallons per day. Initially,
many of these basins were abandoned gravel pits, but since 1936 urban developers are required
to provide recharge basins with new developments. Practically all basins are unlined excava-
tions in the upper glacial deposits and have areas from less than 0.1 to more than 30 acres.
13.7.4 Orlando, Florida

Large volumes of reclaimed water, which has undergone advanced secondary treatment, are
reused through land-based applications in a 40-square-mile area near Orlando, Florida. These
applications include citrus crop irrigation and artificial recharge to the surficial aquifer through
rapid infiltration basins.
13.7.5 Dayton, Ohio

Dayton, Ohio, is heavily dependent upon groundwater to meet municipal and industrial water-
supply needs. Nearly one-fourth of all groundwater used in Ohio is withdrawn from wells
completed in a sole-source sand and gravel aquifer that underlies the Dayton metropolitan
area. Much of the water is pumped from a 30- to 75-foot thick shallow aquifer that underlies
the Mad River Valley. To ensure that groundwater levels are maintained high enough to allow p
for large drawdowns by high-capacity wells, an artificial recharge system has been in place
since the 1930s. The source of recharge is streamt1ow diverted from the Mad River into a series 1~
of interconnected infiltration ditches and lagoons that occupy about 20 acres on Rohrers ta
Island. To meet increasing demand, a new municipal well field was developed in the 1960s in be
re
a section of the aquifer north of Dayton. Here, water is pumped from the Great Miami River
of
into a series of ponds and lagoons, some of which also serve as water hazards on a city-owned be
golf course. Recharge lagoons at both well fields are periodically drained, and accumulated th
muck and silt are excavated to maintain a high rate of infiltration into the underlying aquifer. re
12
13.8 INDUCED RECHARGE A
er
Direct methods of artificial recharge described above involve the conveyance of surface water lal
to some point where it enters the ground. Distinguished from these is the method of induced 1(
Problems 585
Ground surface
/ Water table
--:-----
;( i
~ ~ .... \ \ \
/
I tI /
\
' -- --- ------
/
\ \
-- I
I
I
\
\
' , ----------------- -- \ \
/
/ I
I
\
'
\ \
-- ''~------------------ ' '"-
/
---.... ~-~--------------
Impermeable Impermeable
(a) {b)
Figure 13.8.1. Induced recharge resulting from a well pumping near a river. (a) Natural flow pattern (b) Flow pattern with
pumping well
recharge, accomplished by withdrawing groundwater at a location adjacent to a river or lake

so that lowering of the groundwater level will induce water to enter the ground from the sur-
face source. The schematic cross sections of a river valley in Figure 13.8.1 show flow patterns
with and without induced infiltration from a stream. On the basis of thjs definition, wells
located directly adjacent to and fed largely by surface water provide artificial recharge. The
hydraulics of wells located near streams is described in Chapter 4; the construction and oper-
ation of wells, co1lectors, and galleries is described in Chapter 5.
Induced infiltration where supplied by a perennial stream ensures a continuing water sup-
ply even though overdraft conditions may exist in nearby areas supplied only by natural
recharge. The method has proved effective in unconsolidated formations of permeable sand
and gravel hydraulically connected between stream and aquifer. The amount of water induced
into the aquifer depends on the rate of pumping, permeability, type of well, distance from sur-
face stream, and natural groundwater movement. It is important that the velocity of the surface
stream be sufficient to prevent silt deposition from sealing the streambed.
Studies of water quality have shown that induced recharge can furnish water free of
organic matter and pathogenic bacteria. Because surface water commonly is less mineralized
than groundwater, water obtained by induced infiltration, being a mixture of two water
sources, possesses a higher quality than natural groundwater.
PROBLEMS
13.5.1 What should be the maximum recharge rate if the water fe/day. If the leakage from the lake is about 1.5 ft/day and the
table should be at least 1 m below the soil surface (bottom of maximum pennissible rise of groundwater mound is 140ft above
basin) in order to maintain unimpeded infiltration rates? A the bedrock, how far should the control area be established from
restricting layer 3 m thick is detected in the vadose zone. The top the centerline of the lake? Assume steady-state conditions and no
of the restricting layer is 8.6 m below the bottom of the recharge natural groundwater gradient.
basin. The hydraulic conductivities of the restricting layer and 13.5.3 A square recharge basin 1 mile x 1 mile is proposed over a
the soil above the restricting layer are 0.022 m/day and 7.8 m/day, new area. The average transmissivity of the regional aquifer is
respectively. about 8,500 ft 2/day and the unconfined aquifer is 150ft thick. If the
13.5.2 The artificially constructed Tempe Town Lake in Tempe, maximum allowable rise of the groundwater mound is 70 ft above
AZ, is approximately two miles long and 800-1,200 ft wide, cov- the original water table, plot the permissible recharge (infiltration)
ering 222 acres of surface area. The depth to bedrock under the rate versus Rn, the distance from the center of recharge area to the
lake is about 150 ft and the saturated thickness of the aquifer is water table control area. Assume that the original groundwater
100 ft. The average transmissivity in the area is about 10,000 table will be maintained along the perimeter of the control area.
J
13.5.4 How far should a water table control area be established 2.4 ft/day and the depth from the basin bottom to the groundwa-
I
from the recharge area in Example 13.5.6 in order to prevent the ter table is 26 ft. Determine the maximum permissible infiltration
groundwater mound from rising more than 25m above the origi- rate through the basin bottom. Assume an istotropic subsurface ~
nal water table?

13.5.5 The rise of a groundwater mound directly below the cen-
ter of gravity of a rectangular recharge area 0.5 km x 1.5 km is 41
min three years. If the recharge rate is approximately 37 million
medium.
13.6.3 A new stormwater infiltration basin will be constructed
over an area where the depth to the groundwater table is about 6.8
m. Determine the required radius of a circular infiltration basin if
I g
n
1
m3· per year and the fillable porosity of the vadose zone is 0.22, the design infiltration rate is 4.82 cm/h and the hydraulic conduc- c
estimate the aquifer transmissivity beneath the recharge area. tivity of the formation beneath the basin is 0.58 m/day. Assume
the hydraulic conductivity of the formation is isotropic. tr
13.5.6 A rectangular recharge area, 0.3 mi x 1.6 mi, recharges an
unconfined aquifer at a rate of 2.25 x 109 ft 3 per year. The initial 13.6.4 A new stormwater infiltration basin will be constructed at
n
water table of the unconfined aquifer is 97 ft above the imperme- a site where the design volumetric infiltration rate based on the
1'
able bedrock and the depth to the horizontal water table is 220 ft. design rainfall event selected is 0.00625 m3/s. The hydraulic con-
rt
The tillable porosity of the vadose zone is 0.25. The average ductivity of the medium beneath the basin is about 0.64 m/day. If
Pl
transmissivity in the area is 14,500 ft 2/day. If the maximum per- the distance from the basin bottom to the groundwater level is
21
missible rise of the groundwater mound is 290 ft above the 10.2 m, determine the required basin radius and the corresponding in
bedrock to prevent waterlogging, when and where should a water design infiltration rate that will prevent the infiltration system
PI
table control area be established? How much water can be stored from being backed up. 2:
underground during that time? 13.6.5 Show that growth function for a mound below a circular bi
13.6.1" A new stormwater in.(iJtration basin will be designed at a infiltration can be described by Equation 13.6.6. To begin the v.
site where the average soil infiltration rate at the land surface is derivation, the incremental water storage volume in the soil 2~
1.27 em/hr. The distance between the bottom of the basin and the medium is EJ
groundwater table will be approximately 13 m. The hydraulic 2 D 2~
conductivity of the subsurface material is isotropic and estimated dV =S,rtr dy =Syrt- RJ dy w.
y
to be in the range of 0.80-1.00 em/hr. Design a square infiltration 24
basin that can sustain the design infiltration rate for the given and the rate of recharge at Section PR in Figure 13.6.2 is Wi
range of hydraulic conductivity values. D 2 69
13.6.2 Rework Example 13.6.2 if the hydraulic conductivity in
qo =SyKurtR 2 =SyKu h RQ 25
the vadose zone were anisotropic. Assume Ky = 0.5 ft/day and Jo
13.6.6 Show that the recession function for a mound from a cir-
K, =1.6 ft/day. cular infiltration basin is identical to the recession function for a
26
eft
13.6.2 A circular stormwater infiltration system has a diameter of mound from an infiltration trench.
27
70 ft. The hydraulic conductivity of the subsurface soil is
St1
M(
28
REFERENCES res
1. Abdulrazzak, M. J., A. Sorman, and A. Alhames, Techniques of artifi- 7. Behnke, J. J., Clogging in surface spreading operations for artificial pp.
cial recharge from an ephemeral wadi channel under extreme arid condi- ground-water recharge, Water Resources Research, v. 5, pp. 870-876, 1969. 29.
tions, Proceedings, International Symposium on Artificial Recharge of 8. Bianchi, W. C., and D. C. Muckel, Ground-Water Recharge Hydrology, Jot
Groundwater, A. Z. Johnson and D. J. Finlayson, eds., Am. Soc. of Civil ARS 41-161, Agric. Research Service, U.S. Dept. Agric., 62 pp., 1970. 30.
Engrs .• New York, pp. 602-611, 1988. 9. Biemond, C., Dune water flow and replenishment in the catchment thr•
2. Ahuja, L. R., J. W. Naney, and R. D. Williams, Estimating soil water area of the Amsterdam water supply, Jour. Instn. Water Engrgs., v. 11, pp. 31.
characteristics from simpler properties or limited data, Soil Sci. Soc. Am. 196-213, 1957. me
Proc., v. 45, pp. 1100-1105, 1985. 10. Black and Veach, www.bv.com/services/projects/water/scottsdale.asp. Wa
3. Alley, W. M., T. E. Reilly, and 0. L. Franke, Sustainability of Ground- 11. Bouwer, H., Renovating municipal wastewater by high rate infiltra- FO\
water Resources, U.S. Geological Survey Circular 1186, 1999. tion of groundwater recharge, Jour. Amer. Water Works Assoc., v. 66, pp. 32.
4. Arizona State University, University of Arizona, and University of Col- 159-162, 1974. Pit1
orado, Soil Treatability Pilot Studies to Design and Model Soil Aquifer 12. Bouwer, H., R. C. Rice, J. C. Lance, and R. G. Gilbert, Rapid infil- 33.
Treatment Systems, AWWA Research Foundation and American Water tration research: The Flushing Meadows project, Arizona, Jour. Water Poll. wat
Works Association, Denver, CO, 1998. Control Fed., v. 42, no. 10, pp. 2457-2470, 1980. Am.
5. Asano, T., ed., Artificial Recharge of Groundwater, Butterworth Pub- 13. Bouwer, E. J., P. L. McCarty, H. R. Bouwer, and R. C. Rice, Organic 34.
lishers, Stoneham, MA, 1985. contaminant behavior during rapid infiltration of secondary wastewater at Grc
6. Baier, D. C., and G. W. Wesner, Reclaimed waste for groundwater the Phoenix 23rd Avenue project, Water Research, v. 18, pp. 463-472, 35.
recharge, Water Resources Bull., v. 7, pp. 991-1001, 1971. 1984. 196
References 587
t- 14. Bouwer, H., Artificial recharge of groundwater: Systems, design, and 36. Kool, J. B., and M. Th. Van Genuchten, HYDRUS: One dimensional
·n management, Ch. 24, Hydraulic Design Handbook, L. W. Mays, ed., variably saturated flow and transport model, including hystersis and root
:e McGraw-Hill, New York, 1999. uptake, Version 3.31, U.S. Salinity Laboratory, U.S. Department of Agri-
15. Bouwer, H., J. T. Back, and J. M. Oliver, Predicting infiltration and culture, Pineside, CA, 1991.
ground-water mounds for artificial recharge, Jour. of Hydro!. Engrg, v. 4, 37. Kumar, A., and 0. K. Kimbler, Effect of dispersion, gravitational seg-
:d
no. 4 pp. 350-357, Oct. 1999 . regation, and formation stratification on the recovery of freshwater stored
.8 in saline aquifers, Water Resources Research, v. 6, pp. 1689-1700, 1970.
16. Brown, R. F., and D. C. Signor, Artifical recharge-State of the art,
if
Ground Water, v. 12, pp. 152-160, 1974. 38. Lappala, E. G., R. W. Healy, and E. P. Weeks, Documentation of Com-
c- puter Program VS2D to Solve the Equations of Fluid Flow in Variably Sat-
17. Charbeneau, R. J., Kinematic models for soil moisture and solute
le urated Porous Media, USGS Water Resources Investigations Report, No.
transport, Water Resources Research, v. 17, no. 3, pp. 600-706, 1984.
83-409, U.S. Geological Survey, Lakewood, CO, 1987.
18. Crook, J., Water reclamation and reuse, Ch. 21, Water Resources
at Handbook, L. W. Mays, ed., McGraw-Hill, New York, 1996. 39. Marino, M. A., Artificial ground water recharge, I. Circular recharg-
he ing area, Jour. Hydrology, v. 25, pp. 201-208, 1975a.
19. Dept. Econ. and Social Affairs, Ground-water storage and artificial
n- recharge, Natural Resources/Water Ser. 2, United Nations, New York, 270 40. Marino, M. A., Artificial ground water recharge, II. Rectangular
lf pp., 1975. recharging area, Jour. Hydrology, v. 26, pp. 29-37, 1975b.
is 20. Dvoracek, M. J., and V. H. Scott, Ground-water flow characteristics 41. McWhorter, D. B., and J. A. Brookman, Pit recharge influenced by
ng sub-surface spreading, Ground Water, v. 10, no. 5, pp. 6-11, 1972.
influenced by recharge pit geometry, Trans. Amer. Soc. Agric. Engrs., v. 6,
:m pp. 262-265, 267, 1963. 42. Molden, D. J., D. K. Sunada, and J. W. Warner, Microcomputer model
of artificial recharge using Glover's solution, Ground Water, v. 22, no. 1,
21. Esmail, 0. J., and 0. K. Kimbler, Investigation of the technical feasi-
pp. 73-79, 1984.
lar bility of storing fresh water in saline aquifers, Water Resources Research,
v.3,pp.683-695, 1967. 43. Moore, I. D., Infiltration equations modified for surface effects, Jour.
he
lrrig. Div., Trans. ASAE, v. 107, no. IR 1, pp. 71-79, 1981.
oil 22. Glover, R. E., Groundwater movement. U.S. Bureau of Reclamation,
Engineering Monograph, 31, 1964. 44. Morel-Seytoux, H. J., C. Miracapillo, and M. J. Abdulrazzak, A
reductionist physical approach to unsaturated aquifer recharge from a
23. Goss, D. W., et al., Fate of suspended sediment during basin recharge,
circular spreading basin, Water Resources Research, v. 26, pp. 771-777,
Water Resources Research, v. 9, pp. 668-675, 1973.
1990.
24. Griffin, D. M., Jr., and R. 0. Warrington, Examination of2-D ground-
45. Muckel, D. C., Replenishment of Ground Water Supplies by Artificial
water recharge solution, Jour. of Irrig. Drain. Engrg., v. 114, no. 4, pp.
Means, Tech. Bull. 1195, Agric. Research Service, U.S. Dept. Agric., 51
691-704, Nov. 1988.
pp., 1959.
25. Guo, J. C. Y., Surface-subsurface model for trench infiltration basins,
46. Mushtaq, H., L. W. Mays, and K. E. Lansey, Optimum operation of
Jour. ofWater Res. Plan. Mgt. v. 124, no. 5, pp. 280-284, Sept./Oct. 1998.
recharge basins, Jour. Water Resources Plan. Mgt., v. 129, no. 6,
::ir- 26. Guo, J. C. Y., Design of circular infiltration basin under water mound pp. 927-943, 1994.
)f a
effects, Jour. ofWater Res. Plan. Mgt., Jan 2001.
47. Nellor, M. H., R. B. Baird, and J. R. Smyth, Health Effects Study:
27. Guo, J. C. Y., Design of infiltration basins for stormwater, Ch. 9, Final Report. County Sanitation District of Los Angeles County, Whittier,
Stormwater Collection Systems Design Handbook, L. W. Mays, ed., CA, 1984.
McGraw-Hill, New York, 2001.
48. Nightingale, H. I., and W. C. Bianchi, Ground-water recharge for urban
28. Hantush, M. S., Growth and decay of groundwater-mounds in use: Leaky Acres project, Ground Water; v. 11, no. 6, pp. 36-43, 1973.
response to uniform percolation, Water Resources Research, v. 3, 49. Orange County Water District, The Groundwater Management Plan,
Icia! pp. 227-234, 1967. Fountain Valley, CA, 1994.
969. 29. Harmeson, R. H., eta!., Coarse media filtration for artificial recharge, 50. Ortiz, N. V., D. Q. Zachmann, D. B. McWhorter, and D. K. Sunada,
ogy, Jour. Amer. Water Works Assoc., v. 60, pp. 1396-1403, 1968. Effects of in-transit water on groundwater mounds beneath circular and
'0. 30. Hauser, V. L., and F. B. Lotspeich, Artificial groundwater recharge rectangular recharge areas, Water Resources Research, v. 15, no. 3,
nent through wells, Jour. Soil and Water Conservation, v. 22, pp. 11-15, 1967. pp. 577-582, 1979.
,pp. 31. Ho, G. E., K. Mathew, and P. W. G. Newman, Water quality improve- 51. Rahman, M. A., et al., Effect of sediment concentration on well
ment of treated wastewater by soil percolation, in Proceedings of seminar, recharge in a fine sand aquifer, Water Resources Research, v. 5, pp.
.asp. Water Quality, Its Significance in Western Australia, Water Research 641-646, 1969.
Foundation of Australia, Perth, 1983. 52. Rao, N.H., and P. B.S. Sarma, Ground water recharge from rectan-
lltra-
., pp. 32. Huisman, L., and T. N. Olsthoom, Artificial Groundwater Recharge, gular areas, Ground Water, v. 19, no. 3, pp. 270-274, 1981.
Pitman Publishing, Mansfield, MA, 1983. 53. Rastogi, A. K., and S. N. Pandy, Modeling of artificial recharge basins
infil- 33. Idelovitch, E., R. Terkeltoub, and M. Michall, The role of ground- of different shapes and effect on underlying aquifer system, Jour. Hydro/.
Poll. water recharges in wastewater reuse: Israel's Dan Region Project, Jour. Engrg., v. 123, no. 3, pp. 62-68, Jan. 1998.
Am. Water Works Assn., v. 72, no. 7, pp. 391-400, 1980. 54. Rebhun, M., and J. Schwartz, Clogging and contamination processes
~anic 34. Johnson, A. I., and D. J. Finlayson, eds., Artificial Recharge of in recharge wells, Water Resources Research, v. 4, pp. 1207-1217, 1968.
ter at Groundwater; Amer. Soc. Civil Engrs., 1988. 55. Ripley, D.P., and Z. A. Saleem, Clogging in simulated glacial aquifers
-472, 35. Kelly, T. E., Artificial recharge at Valley City, North Dakota, 1932 to due to artificial recharge, Water Resources Research, v. 9., pp. 1047-1057,
1965, Ground Water, v. 5, no. 2, pp. 20-25, 1967. 1973.
56. Sceva, J. E., Liquid waste disposal in the lava terranes of Central
Oregon, U.S. Federal Water Pollution Control Admin., Corvallis, OR, 2
ings, Second International Symposium on Wastewater Reclamation and
Reuse, Iraklio, Greece, pp. 925-935, 1995.
I
vols., 162 pp. 1968. 64. Tang, Z., Li, G., L. W. Mays, and P. Fox, New methodology for opti-
57. Schicht, R. J., Feasibility of recharging treated sewage effluent into a mal operation of soil aquifer treatment systems, Water Resources Man-
deep sandstone aquifer, Ground Water, v. 9, no. 6, pp. 29-35, 1971. agement, Kluwer Acadmic Publishers, European Water Resources
58. Scott, V. H., and G. Aron, Aquifer recharge efficiency of wells and Association, The Netherlands, Vol. 14, no. 1, pp. 13-33, Feb. 2000.
trenches, GroundWater, v. 5, no. 3, pp. 6-14, 1967. 65. Taylor. S. W., and P. R. Jaffe, Subsurface and biomass transfer in a
59. State Water Resources Control Board, Dept. Water Resources, and porous medium, Water Resources Research, v. 26, no. 2, pp. 2181-2194,
Dept. Health, A "State of the An" Review of Health Aspects of Wastewater 1990.
Reclamation for Groundwater Recharge, Water Information Center, Hunt- 66. U.S. Environmental Protection Agency (EPA), Arid Water Reclama-
ington, NY, 240 pp. 1978. tion Reuse Guidelines, Cincinnati, OH, 1991.
60. Sumner, D. M., D. E. Rolston, and M. A. Marino, Effects of unsatu- 67. Valliant, J., Artificial recharge of surface water to the Ogallala For-
rated zone on groundwater mounding, Jour. of Hydro/. Engrg., v. 4, no. 1, mation in the High Plains of Texas, Ground Water, v. 2, no. 2, pp. 42-45,
pp. 65-69, 1999. 1964.
61. Suter, M., and R. H. Harrneson, Artificial ground-water recharge at 68. Warner, J. W., D. Molden, M. Chehata, and D. K. Sunada, Mathemat-
Peoria, Illinois, Bull. 48, Illinois State Water Survey, Urbana, 48 pp., ical analysis of artificial recharge from basins, Water Resources Bull.,
1960. v.25,no.2,pp.401-41l,April 1989.
62. Swamee, P. K., and C. S. P. Ojha, Ground-water mound equation for 69. Wilson, L. G., Water quality changes during soil aquifer treatment of
rectangular recharge area, Jour. Irrig. and Drain. Engrg., v. 123, no. 3, pp. tertiary effluents, Water Environment Research, v. 67, pp. 371-376, 1995.
215-217, May/June 1997. 70. Zomorodi, K., Optimal artificial recharge in intermittent multibasin
63. Tang, Z., G. Li, L. W. Mays, and P. Fox, Development of methodol- system, Jour. Water Resources Plan. Mgt. v. 116, no. 5, pp. 639-651,
ogy for the optimal operation of soil aquifer treatment systems, Proceed- Sept./Oct. 1990.
Chapter 14
Saline Water Intrusion in Aquifers
Saline water is the most common pollutant in fresh groundwater. Intrusion of saline water
occurs where saline water displaces or mixes with freshwater in an aquifer. The phenomenon
can occur in deep aquifers with the upward advance of saline waters of geologic origin, in
shallow aquifers from surface waste discharges, and in coastal aquifers from an invasion of sea-
water. The interrelations of two miscible fluids in porous media have been studied extensively
both theoretically and under field conditions. Management techniques that enable development
of fresh water and at the same time control of saline intrusion are discussed in this chapter.
14.1 OCCURRENCE OF SALINE WATER INTRUSION

Saltwater intrusion into fresh groundwater formations generally results inadvertently from
human activities.* Most large sources of fresh groundwater are in close proximity to the sea,
to natural bodies of saline groundwater, or to salts from effluent wastes released by human
activities. 44 Typically, shallow freshwater overlies saline water because the flushing action dur-
ing recent times removes salts from ancient marine deposits. But at greater depths groundwa-
ter movement is much less, so that displacement of saline water is slower. Furthermore, at
depths of several thousand meters, brines are normally encountered.
Saline water in aquifers may be derived from any of several sources: 59
1. Encroachment of seawater in coastal areas
2. Seawater that entered aquifers during past geologic time
3. Salt in salt domes, thin beds, or disseminated in geologic formations
4. Water concentrated by evaporation in tidal lagoons, playas, or other enclosed areas
5. Return flows to streams from irrigated land
6. Human saline wastes
The mechanisms responsible for saline water intrusion fall into three categories. 19
(1) Reduction or reversal of groundwater gradients, which permits denser saline water to dis-
place fresh water. This situation commonly occurs in coastal aquifers in hydraulic continuity
with the sea when pumping of wells disturbs the natural hydrodynamic balance. (2) Destruction
of natural barriers that separate fresh and saline waters. An example would be the construction
*It should be noted that saline groundwater can represent a valuable resource, particularly in arid inland regions.
Potential uses include industrial processes such as cooling; irrigation, where the mineral content is moderate; and
desalination for local domestic purposes.
589
590 Chapter 14 Saline Water Intrusion in Aquifers
of a coastal drainage canal that enables tidal water to advance inland and to percolate into a
freshwater aquifer. (3) Subsurface disposal of waste saline water, such as into disposal wells,
landfills, or other waste repositories (see Chapter 8).
The occurrence of saline water intrusion is extensive and represents a special category of
groundwater pollution. A comprehensive listing of examples of intrusion has been pre-
pared;3· 4• 59 this reveals thallhe problem exists in localities of most parts of the United States.
Seawater intrusion along coasts has received the most attention. 32. 35· 39· 67 Internationally the
problem has received attention in populated coastal areas. 7· 31 · 63 · 72• 73· 74 Many small oceanic
islands are completely underlain with aquifers containing sea water; these pose special prob-
lems in meeting water supply demands. Reilly and Goodman50 presented a historical perspec-
tive on the quantitative analysis of saltwater-freshwater relationships.
14.2 GHYBEN-HERZBERG RELATION BETWEEN

FRESH AND SALINE WATERS
More than 100 years ago two groups of investigators, 18• 27 working independently along the
European coast, found that salt water occurred underground, not at sea level but at a depth
below sea level of about 40 times the height of the fresh water above sea leveL This distribu-
tion was attributed to a hydrostatic equilibrium existing between the two fluids of different
densities. The equation derived to explain the phenomenon is generally referred to as the
Ghyben-Herzberg relation, after its originators."
The hydrostatic balance between fresh and saline water can be illustrated by the U-tube
shown in Fig. 14.2.1 Pressures on each side of the tube must be equal; therefore,
(14.2.1)
where p5 is the density of the saline water, p1is the density of the fresh water, g is the acceler-
ation of gravity, and z and h1 are as shown in Figure 14.2.1. Solving for z yields
Fresh water
Saline water
Figure 14.2.1. Hydrostatic balance between

fresh water and saline water illustrated by a
U-tube.
•Almost unnoticed in the hydrologic literature is the much earlier contribution of Joseph DuCommun, a French
teacher at West Point Military Academy. He first stated clearly and correctly the fresh-salt water balance existing in
coastal aquifers (Du-Commun. J., On the cause of fresh water springs, fountains, etc., Amer. Jour. Science and Arts,
v. 14, pp. 174-176, 1828).
14.2 Ghyben-Herzberg Relation Between Fresh and Saline Waters 591
h,
--~-~A-F- ----- ------ T -~ MSL
--
Fresh water,
density Pr
--
Ocean
Interface z
Saline water, Figure 14.2.2. Idealized sketch of

density p5 occurrence of fresh and saline ground-
water in an unconfined coastal aquifer.
z= Pt hr (14.2.2)
Ps -pf
which is the Ghyben-Herzberg relation. For typical seawater conditions, let Ps = 1.025 g/cm3
and Pt= 1.000 g/cm3, so that
(14.2.3)
Translating the U-tube to a coastal situation, as shown in Figure 14.2.2, h1 becomes the
elevation of the water table above sea level and z is the depth to the fresh-saline interface
below sea level. This is a hydrodynamic rather than a hydrostatic balance because fresh water
is flowing toward the sea. From density considerations alone, without flow, a horizontal inter-
face would develop with fresh water everywhere floating above saline water. Where the flow
is nearly horizontal, the Ghyben-Herzberg relation28 gives satisfactory results. Only near the
shoreline, where vertical flow components become pronounced (see Figure 14.2.2), do signif-
icant errors in the position of the interface occur.
For confined aquifers, the above derivation can also be applied by replacing the water
table by the piezometric surface. It is important to note from the Ghyben-Herzberg relation
that fresh-salt water equilibrium requires that the water table, or piezometric surface (1) lie
above sea level and (2) slope downward toward the ocean. Without these conditions, seawater
will advance directly inland.
Starting from the work ofHubbert,28 the Ghyben-Herzberg relation has been generalized
by Lusczynski and others40• 41 for situations where the underlying saline water is in motion
with heads above or below sea level. The result for nonequilibrium conditions has the form
(14.2.4)
where h1 is the altitude of the water level in a well filled with fresh water of density p1 and ter-
minated at depth z, while h5 is the altitude of the water level in a well filled with saline water
of density Ps and also terminated at depth z (see Figure 14.2.3). When h5 = 0, the saline water
is in equilibrium with the sea, and Equation 14.2.4 reduces to Equation 14.2.2.
EXAMPLE 14.2.1 The steady fresh-saline interface is located 43 m below sea level 500 m inland from the shoreline in an
unconfined aquifer. Determine the elevation of the water table above this point.
Water table
1-,------
----------------MSL
Ocean Pressure surface

z in the saline water
Saline water
J Fresh water
Figure 14.2.3. Diagram illustrating different heads

for nonequilibrium conditions in fresh and saline
water in an unconfmed coastal aquifer.
SOLUTION Using the fresh and saline water densities for typical seawater conditions, we have Ps = 1.025 g/cm3 and
Pr= 1.000 g/cm) Substituting into Equation 14.2.2 yields Equation 14.2.3: z = 40 hr
43m
Given z = 43 m, hr =--= 1.075 m
40
Thus, the groundwater table is located 1.075 m above sea level at this location.
•
14.3 SHAPE OF THE FRESH-SALT WATER INTERFACE
Recognizing the approximations inherent in the Ghyben-Herzberg relation, more exact solu-
tions for the shape of the interface have been developed from potential flow theory. 14· 16 The
result by Glover (see Cooper et al. 16) has the form
Z2 = 2pqx
~pK
+ (..2i_l
~pK
2
(14.3.1)
where z and x are as shown in Figure 14.3.1, ~p = Ps- Pp pis the density of freshwater, K is
the hydraulic conductivity of the aquifer, and q is the freshwater flow per unit length of shore-
line. The corresponding shape for the water table is given by
[ l
l/2
h _ 2~p qx
(14.3.2)
f - (p+~p) K
The widthx0 of the submarine zone through which fresh water discharges into the sea can
be obtained for z = 0, yielding
X = pq (14.3.3)
2~pK
0
The depth of the interface beneath the shoreline Z0 occurs where .x =0 (see Figure 14.3 .1) so that
z = ..2.i_ (14.3.4)
o ~pK
The fresh-water flow from a coastal aquifer into the sea per unit length of shoreline is estimated as 1.5
(m3/day)/m. The hydraulic conductivity of the unconfined aquifer is 13.6 rnlday. Compute the water table
14.3 Shape of the Fresh-Salt Water Intetface 593
~ Ground surface
/
X Ocean-=='"""
-=-Water table}
Fresh water
Saline water Figure 14.3.1. Flow pattern of fresh water in

an unconfined coastal aquifer.
profile and shape of the fresh-saline interface between the shoreline and 1 km inland. Use both the
Ghyben-Herzberg relation and the solution by Glover. Assume steady conditions.
SOLUTION The solution by Glover has the following form (Equation 14.3.1):
z = 2pqx + [___E.LJ2
~pK ~pK
Note that the constant second term inside the square bracket has been added to the Ghyben-Herzberg
model to account for the missing seepage face that allows for the vertical components of flow and dis-
charge of the fresh water onto the sea floor. 22 The fresh-saline interface intercepts the water table at the
coastline in the Ghyben-Herzberg model.
Thus, the Ghyben-Herzberg model has the following form of Equation 14.3.1:
z=~2pqx
~pK
The resulting water table profile and the fresh-saline interface estimated by the two models are summa-
rized in Table 14.3.1. Note that the water table profile for Ghyben-Herzberg model is estimated using
Thble 14.3.1 Example 14.3.1 Results
X z (Ghyben-Herzberg) z (Glover) h1(Ghyben-Herzberg) hr(Glover)

20 13.28 14.00 0.33 0.33
40 18.79 19.30 0.47 0.46
60 23.01 23.43 0.58 0.57
80 26.57 26.93 0.66 0.66
100 29.70 30.03 0.74 0.73
200 42.01 42.24 1.05 1.04
300 51.45 51.64 1.29 1.27
400 59.41 59.57 1.49 1.47
500 66.42 66.57 1.66 1.64
600 72.76 72.89 1.82 1.80
700 78.59 78.71 1.96 1.94
800 84.02 84.13 2.10 2.07
900 89.11 89.22 2.23 2.20
1000 93.93 94.04 2.35 2.32
Il
Distance from shoreline (m)
Ia; 1o
>
~
(lj
Q)
(/)
?::
0
a;
..c
..c
0..
Q)
25
0
--Glover Figure 14.3.2. Example

------ Ghyben-Herzberg 14.3.1 results showing
.- ·- Water table interface.
z
hr=-
40
The Ghyben-Herzberg model, which does not allow for an outflow face, underestimates the depth to salt-
water interface compared to the more exact solution by Glover. The difference becomes more significant
closer to the shoreline, especially less than 20 m from the shoreline. On the other hand, the water table
profile does not change significantly between the two models. The shape of the interface and the water
table profile between the shoreline and 100m inland are shown in Figure 14.3.2.
The width of the submarine zone through which fresh water discharges into the sea can be computed
using Equation 14.3.3:
•
EXA'MfLE 14.3.2 The salt-water interface 80 m from the shoreline is located 27.5 m below sea level in an unconfined
coastal aquifer. If the hydraulic conductivity of the aquifer is 11.8 m/day, determine the fresh water dis-
charge from the aquifer into the sea using Ghyben-Herzberg model and Glover's model. Assume steady-
state conditions.
SOLUTION Ghyben-Herzberg model:
z= ppqx
~pK
2(1.000 g/cm 3 )(q)(80 m)

27.5 m= ( ) .
0.025 g/cm 3 {11.8 m/day)
3
q = 1.394 (m /day ym
Glover's relation:
z = 2pqx +(__2i_J2
~pK ~pK
14.3 Shape of the Fresh-Salt Water Interface 595
2
3 3
2(1.000 glcm )(q)(;:) [ (1.000 glcm ){q) ]
27.5m =
(0.025 glcm 3){1L8 m/day) + (0.025 glcm 3 )(lt.8 m/day)
(ll.491)q 2 +(542.373)q -756.25 =0

q =l.35s(m 3 /dayym
Glover's relation provides a more exact answer, although the result from the Ghyben-Herzberg model
does not differ significantly from that of Glover's in this problem. It should be noted that the difference
between the two models becomes more significant as one gets closer to the shoreline. •
EXAMPLE 14.3.3 The steady fresh-water discharge from an unconfined coastal aquifer into the sea is about 1.6 (m3/day)/m
along the shoreline. The hydraulic conductivity of the aquifer is 8.6 m/day and the horizontal impervi-
ous stratum is located 57 m below sea level. Determine the shape of the fresh-saline interface and the
location of the toe of the interface from the shoreline.
SOLUTION First, the exact location of the toe of the interface (where the interface meets the bedrock), z =57 m, must
be substituted into Equation 14.3.1, yielding
(·
z= 2pq.x + (__£1_)2
ll llp K llp K
.e
2
:r 3
2(40)(1.6 m /day!m)(.x) [( 40)(1.6 m 3/day/m)]
57 m = + -~----:-----.!.
(8.6 m/day) (8.6 m/day)
where p/llp =40 assuming typical fresh and salt water densities. Equation 14.3 .1 is used to solve for z
for x ranging from the shoreline (x = 0) to the toe at x = 215 m. Glover's method can be solved through
• using a spreadsheet solution. The resulting shape of the interface is shown in Figure 14.3.3.
•
w
is-
y-
Distance from coastline (m)
--Glover
- ·- Water table
·- • • • • Bedrock
Figure 14.3.3. Example

14.3.3 results showing
interface.
14.4 STRUCTURE OF THE FRESH-5ALT WATER INTERFACE

The sharp interfacial boundary between fresh and saline water as described above does not
occur under field conditions. Instead. a brackish transition zone of finite thickness separates the
two t1uids. This zone develops from dispersion by flow of the fresh water plus unsteady dis-
placement of the interface by external influences such as tides, recharge, and pumping of wells.
[n general, the greatest thicknesses of transition zones are found in highly permeable coastal
aquifers subject to heavy pumping. Observed thicknesses vary from less than 1m to more than
100m.· Figure 14.4.1 shows a transition zone in a highly penneable limestone aquifer in
Miami, Florida. Note that the isochlors approach the base of the aquifer perpendicularly; this
results because the flow parallels the base of the aquifer, thereby restricting vertical mixing. 16
An important consequence of the transition zone and its seaward flow is the transport of
saline water to the sea. This water originates from the underlying saline water; hence, from
continuity considerations, there must exist a small landward flow in the saline water region.60
Figure 14.4.2 schematically illustrates the flow patterns in the three subsurface zones. Field
measurements at Miami 16 and experimental studies 11 have confmned the landward move-
ment of the saline water body. Where tidal action is the predominant mixing mechanism,
fluctuations of groundwater-and hence the thickness of the transition zone-become great-
est near the shoreline.
Figure 14.4.1. Cross

section through the tran-
sition zone of the
Biscayne aquifer near
Miami, Florida.
Numbered lines are
1000 500 0
isochlors in mgll (after
Distants from slloreline, meton Cooper et al. 16).
-
-+--
unconfined coastal
aquifer (after Todd61 ;
reprinted from the
Journal American Water
Works Association, vol.
66, by pennission of the
Association, copyright ©
~ 1974 by American Water
........... ..............
Saline water --.
-- Works Association, 6666
West Quincy Avenue,
Denver, CO. 80235).
·As an exlreme case, concentrated pumping in the Honolulu-Pearl Harbor area of Hawaii has created localized lran-
sition zones more than 300 m thick; these occupy essentially the entire vertical extent of the aquifer. 67
0 ,.--------,
)t
1e
s-
s.
al ..
m
water
in
is Salinity--+ 0.1 10 50 90 99 99.9
6 Figure 14.4.3. Increase in salinity with
Relative salinity SR, percent
Df depth through the transition zone.
(a) (b) (a) Linear scale (b) Probability scale
m
60
ld Within the transition zone, the salinity of the groundwater increases progressively with
e- depth from that of the fresh water to that of the saline water. Typically, the distribution of salin-
n, ity with depth varies as an error function, as shown in Figure 14.4.3a. 16 It then becomes advan-
lt- tageous to calculate the relative salinity SR as a percentage by
SR =100 __c -c!_ J (14.4.1)

( cs -cf
where cis the salinity* at a particular depth within the transition zone, and c1 and c5 are the
salinities of the fresh and saline waters, respectively.
Plotting values of SR versus depth on probability paper will generally yield a straight line,
as demonstrated in Figure 14.4.3b. Because detailed data on the structure of a transition zone
are difficult to obtain in the field, this graphic technique enables the zone to be estimated from
)- any two point measurements of salinity The 50 percent value of SR• representing the midline
of the transition zone, defines the position of the interface without mixing.
Where inhomogeneities occur in coastal aquifers, stratifications and irregularities in the dis-
tribution of fresh and saline waters occur. 25 · 47 Considerable research has been done on seawater
intrusion in layered aquifers 18• 20• 21 and on unsteady movement of the transition zone. 33 • 34
Anderson et al} Bear et al., 6 Bear and Verruijt, 8 Guvanasen et al., 24 Huyakorn et al, 29 • 30
Ledoux et al., 37 McElwee,43 Padilla and Cruz-Sanjulian,46 Segal and Pinder, 55 Person et al., 48
Reichard,49 Verruijt, 66 and Wilson and Sada Costa70 are just a few of the references relating to
the simulation of saltwater intrusion in groundwater.
14.5 EFFECT OF WELLS ON SEAWATER INTRUSION

Strack56 developed a technique for evaluating the effect of wells on seawater intrusion. 5 As a
demonstration of the application of the technique, Strack considered a case of a single well
er located at a distance xw from the shoreline as illustrated in Figure 14.5.1. For an isotropic,
l.
homogeneous aquifer with a horizontal impervious base and with the assumptions of essen-
he
tially horizontal fresh water flow and no flow in the salt water zone, the following relationship
t©
1ter
is used for determining the position of the toe of an interface under steady conditions: 5
i66
(14.5.1)
ran-
*salinity can be measured as total dissolved solids, chloride, or electrical conductivity.
Unconfined --+-~--of------ Unconfined flow

interface flow ------
~ow
~-------1- Xw -----~ r
:v----
II
II
II
II
Zone 1 Zone 2 II
II
II
II
II Impervious stratum
Tip of saltwater tongue

(a)
Figure 14.5.1. Shallow coastal

aquifer with a well (after
(b) Strack56 ).
where
(14.5.2)
B: depth to bedrock below mean sea level

Q~x: freshwater flow per unit length of shoreline (assumed uniform from infinity to the
coast)
Qw: constant pumping rate of the well superimposed on Q~x
K: hydraulic conductivity of the aquifer
xw: distance between the well and the shoreline (see Figure 14.5.1); and
(x ,y): x-y coordinates of the toe of the interface (see Figure 14.5.1).
For the stagnation point (xs, Ys), Strack56 obtained
(14.5.3)
Assuming that an unstable critical situation occurs when the toe of the interface passes through
the stagnation point, he obtained the following expression for computing the critical well dis-
charge, Qw, for unconfined aquifers:
A. = z(1-.!::)I/2 +1:: In[ I-(1- wnt2]

1t 1t I+ (1- Wrtf 2 (14.5.4)
where
(14.5.5)
and
(14.5.6)
The corresponding expressions for confined conditions.were presented by Strack. 56
EXAMPLE 14.5.1 If a production well pumping at a constant rate of 5 Us is constructed 800 m from the shoreline of the
coastal aquifer as described in Example 14.3.3, detennine the new location of the toe of the fresh-saline
interface under steady conditions. If the well is fully penetrating the aquifer, does brackish water occur?
SOLUTION Strack's solution estimates the location of the toe of the interface using Equation 14.5.1 given the fol-
lowing: B =57 m, Q'Ox =1.6 (m3/day)/m, Q,.. =5 Us= 432m3/day, K =8.6 m/day, x.. =800 m.
Substituting 0= 40 for typical sea water conditions andy= 0 to find the location ofthe tip only yields
.!(Ito)~= Qox x+
2 8
2
K
Q,.. ln[(x-x,..)
4nK (x+x,..) +i
2
2
+/]
2 3 3 2 2
1 (57m) (L6(m /dayym) (432m /day) [(x-800m) +(0m) ]
-(1 + 40) - - = X t ln ..:.....___----'-=---'-.:..,..
2 40 2 8.6rnlday 4n(8.6m/day) (xt800m) +(0m) 2 2
x=252m
Note that the toe of the interface was located 215m from the shoreline (see Example 14.3.3). Thus, the
toe of the interface moves 252m- 215 m"' 37m inland because of the well pumping at a constant rate
of5 Us.
To determine if brackish water occur in the well or not, we must determine the critical well discharge
beyond which the interface would advance rapidly until a new equilibrium is reached, with the interface
toe landward of the well. Using· Strack's solution for the critical situation (Equation 14.5.5 is used to com-
pute)
"J..
"'
( J(I+S) =[(1.6(8.6mmldayX57
KB
QQxxw 7
2
m) ](1+40) 559
/day!m)(&OO m) 4()2 =0.3
2
Subs~tuting A= 0.559 into Equation 14.5.4 and solving for J.l yields
0.559 = 2(1-t.J'/2 +.!: ln l-(1- J.l/7t)~

1t 1t l+(l-J.l/1t)'
J.l;;;; 1.524
Finally, the critical well discharge is computed using Equation 14.5.6:

3)
1.524= Qw
sh (1.6 m 3/day/m)(800 m)
is- 3
Qw "' 195 I m /day
1
1
j
Since the proposed pumping rate, 5 Us =432 m3/day, is below the critical pumping rate, the toe of the inter- ~
face will stabilize some distance away from the stagnation point or the water divide defining the region of 1
water supply to the well. Thus, there won't be brackish water in the pumping well of this problem. • ~
l
14.6 UPCONING OF SALINE WATER I
When an aquifer contains an underlying layer of saline water and is pumped by a well pene- I
trating only the upper freshwater portion of the aquifer, a local rise of the interface below the
well occurs. This phenomenon, known as upconing, is illustrated by Figure 14.6.1. Here the I
interface is horizontal at the start of pumping when t = t 0. With continued pumping the inter-
face rises to successively higher levels until eventually it can reach the well. This generally
necessitates the well having to be shut down because of the degrading intluence of the saline
water. When pumping is stopped, the denser saline water tends to settle downward and return
to its fanner position.
Upconing is a complex phenomenon. Only in recent years has significant headway been
made in research studies to enable criteria to be fonnulated for the design and operation of
wells for skimming fresh water from above saline water. 1· 33 • 51 · 52• 69 • 71 From a water-supply
standpoint it is important to detennine the optimum location, depth, spacing, pumping rate,
and pumping sequence that will ensure production of the largest quantity of fresh groundwater,
while at the same time striving to minimize any underground mixing of the fresh saline water.
Most investigations of upconing have assumed an abrupt interface between the two flu-
ids, 17• 26 as shown in Figure 14.6.1. This situation would pertain between immiscible fluids, but
for miscible fluids such as fresh and saline groundwater, a mixing, or transition, zone having
a finite thickness occurs. Although an abrupt interface neglects the physical reality of a transi-
tion zone found in groundwater, the assumption has the advantage of simplicity. Furthennore,
an interface can be considered as an approximation to the position of the 50-percent relative
salinity in a transition zone.
Q
Ground surface
Impermeable
Fresh water,
density Pt
Interface reaching well
___+________ _
Figure 14.6.1. Diagram
Critical
of upconing of underly-
ing saline water to a
pumping well (after
Saline water, Schmorak and
density p5 Mercado54 ).
14.6 Upconing of Saline Water 601
An approximate analytic solution for the upconing directly beneath a well, based on the
Dupuit assumptions and the Ghyben-Herzberg relation, is given bi 4
z= Q (14.6.1)
27tdK(L1plp f)
where Llp = Ps- Pp K is the hydraulic conductivity, and all other quantities are defined by Fig-
ure 14.6.1. This equation indicates an ultimate rise of the interface to a new equilibrium posi-
tion that is directly proportional to the pu~ping rate Q.
Hydraulic model experiments have revealed that the relation in Equation 14.6.1 holds only
if the rise is limited. If the upconing exceeds a certain c1itical rise (see Figure 14.6.1), it accel-
erates upward to the well. The critical rise has been estimated to approximate z/d =0.3 to 0.5.
Thus, adopting an upper limit of z}d =0.5, it follows that the maximum pennissible pumping
rate without salt entering the well is
Qmax ~ miK(t1plp1) (14.6.2)
For anisotropic aquifers where the vertical permeability is less than the horizontal, a maximum
f
well discharge larger than that for the isotropic case is possible. 13
In the real-world situation, a transition zone with a finite thickness of brackish water
occurs above the body of undiluted saline water. The water at the upper edge of the zone is
essehtially fresh water and moves accordingly. Upward movement of the almost-fresh water
occurs readily along with the adjoining fresh water; consequently, even with a relatively low
pumping rate, no limiting critical rise exists above which saline water will not rise. It follows
that with any rate of continuous pumping, some saline water must sooner or later reach a well.
Examination of Equation 14.6.1 supports the previous statements. In a transition zone, the
,, salinity changes gradually; hence, dp in an incremental width at the top of the zone approaches
zero. As a result, z must tend toward infinity, indicating that there can be no finite limit to z.
e
Similarly, in Equation 14.6.2, as Llp approaches zero, so also must Qmax approach zero.
A comparison of the arrivals of salinity at a pumping well for an abrupt interface and for a
transition zone is shown qualitatively in Figure 14.6.2. With an abrupt interface, if Q > Qmax' the
salinity appears later and increases more rapidly than with a transition zone. For Q < Qmax' there
will be no salinity reaching the well in the abrupt case; however, a gradual invasion of saline
water will occur from a rising transition zone. The ultimate well-water salinity with upconing
approaches an intermediate value between the extremes of the fresh and saline waters; empirical
data indicate this lies in the range of 5 to 8 percent of the salt concentration in the saline water. 54
Reilly and Goodman51 presented an analysis of saltwater upconing beneath a pumping well.
It follows from the above analysis that upconing can be minimized by the proper design and
operation of wells and galleries. For given aquifer conditions, wells should be separated as far as
Figure 14.6.2. Well water

-~F h--t- --------0~ Omaxwith__7 __ _ salinity curves for upconing
res wa er abrupt interface of an abrupt interface and a
transition zone (after
Duration of pumping - + Schmorak and Mercado54).
>,
ell
1J
........
(")
40
~ 440
Q)
"ID
E "§,
(")
0 30 E
..... o) 400
o) "0
"§ "§
0')
:c
c 20 ()
360
·a.
E
::l
a..
10 320
1964 1965 1966 1967 1968 1969 1964 1965 1966 1967 1968 1969
Figure 14.6.3. Effect of annual variation in pumping rate on upconing of underlying saline water in Honolulu, Hawaii. Chloride con-
tents were measured in an observation well at a depth of 128m and located 430 m from the pumping well (after Todd and Meyer63 ).
possible vertically from the saline zone and pumped at a low uniform rate. Well and gallery
designs for regions where a thin layer of fresh water overlies saline water are discussed in the
section in this chapter on oceanic islands. Field tests have shown that scavenger wells, which
pump brackish or saline water from below the fresh water, can also successfully counteract
upconing. 38 5
The correspondence of pumping rate and salinity in Figure 14.6.3 dramatically illustrates
the effects of upconing caused by increased pumping to meet annual midsummer demands in
Honolulu.
E»A'MP.LE 14.6.1 A pumping well partially penetrates a confined aquifer with the bottom of the well screen at an elevation
of +25m. The impervious top and bottom elevations of the aquifer are +64 m and +8 m, respectively. If
a horizontal fresh-salt water interface exists at an elevation of +12 m prior to pumping, determine the
maximum allowable pumping rate in order to prevent salt water from entering the well. The aquifer's
hydraulic conductivity is K = 25 m/day.
SOLUTION Given is d =25m-12m= 13m. The critical rise is approximately z/d = 0.3 to 0.5 and pjl1p = 40 for
typical fresh-salt water densities. Substituting into Equation 14.6.2 yields
2 2
nd K n(13 m) (25 m/day) _z __ 0.
Q =--= =332m 3/day for 5
max 40 40 d
or
2 2
= 0.61td K 0.6n(13 m) (25 m/day)
Qmax -----'---~------'-=199m 3/day for ~ = 0.3
40 40 d
•
EXAMPLE 14.6.2 If the desired pumping rate is 450m3/day in Example 14.6.1 and the interruptions of the well pumping
are to be minimized, determine the maximum permissible length of the well screen and compare it to the
given well screen length.
SOLUTION Since the desired pumping rate exceeds the critical pumping rates found in the solution to Example
14.6.1, the critical rise of the interface will often be experienced and the pumping will have to be stopped
for the interface to return to its former position. However, the interruptions can be minimized if the well
screen length is adjusted such that
z/d = 0.3 (assuming this is the limit for critical rise)
Substituting into Equation 14.6.1 yields
14.7 Fresh-Salt Water Relations on Oceanic Islands 603
0.3d = Q
27tdK( ~p/p f)
d= Q
0.6nK(.1p/p 1 )
3
d = 450 m /day x 40
0.6n(25 m/day)
d = 19.5 m
Since the interface is initially located at +12 m, the bottom of the screen must be at +12 m + 19.5 m =
+31.5 m or above.
This results in a maximum permissible screen length of L5 = +64 m- 31.5 m = 32.5 m
•
E-XAMPLE 14.6.3 A partially penetrating pumping well is installed in an unconfined coastal aquifer for which the horizontal
saltwater interface lies 90 ft below the bottom of the well screen. The hydraulic conductivity of the
aquifer is 40ft/day. If the well pumps from the aquifer at a constant rate of 1,200 ft 3/hr, how close can
the saltwater edge approach the bottom of the screen before the water quality in the pumping well is
affected?
SOLUTION
First we must determine the ultimate rise of the interface by substituting ~p/p1 = 1/40, Q = 1200 ft 3/hr =
28800 ft 3/day, K =40ft/day, and d =90ft into Equation 14.6.1:
z= Q
27tdK( ~p/p f)
[1
28800 ft 3/day
f
2n(90 ft)(40 ft/day)(l/40)
e
=51ft
and
_£ = 51 ft = 0.57
>r d 90ft
Since the ultimate rise of the interface exceeds the critical rise estimated to approximate zld = 0.3 to 0.5,
the pumping should be stopped when z = 0.3 x 90ft= 30ft (assuming an upper limit of zld = 0.3 for the
critical rise). In other words, the pumping should be stopped when the interface is located 60ft below the
bottom of the well screen. •
I 14.7 FRESH-SALT WATER RELATIONS ON OCEANIC ISLANDS

Most small oceanic islands are relatively permeable, consisting of sand, lava, coral, or lime-
stone, so that seawater is in contact with groundwater on all sides. Because fresh groundwater
originates entirely from rainfall, only a limited quantity is available. A fresh-water lens, shown
schematically in Figure 14.7.1, is formed by the radial movement of the fresh water toward the
coast. This lens tloats on the underlying salt water; its thickness decreases from the center
1le toward the coast. 15• 23
ed From the Dupuit assumptions and the Ghyben-Herzberg relation, an approximate fresh-
ell water boundary can be determined. Assume a circular island of radius R, as shown in Figure
14.7 .1, receiving an effective recharge from rainfall at a rate W, the outward flow Q at radius r is
dh
Q = 2nrK(z +h)- (14.7.1)
dr
Recharge rate W
Figure 14.7.1. Freshwater lens in an oceanic

island under natural conditions.
where K is the hydraulic conductivity and hand z are defined in Figure 14.7.1. Noting that h
=(ilp/p)z and that from continuity Q =1t?W, we have
6:l
d Wrdr
, , =2K[l +~I
(14.7.2)
Integrating and applying the boundary condition that h =0 when r =R yields
z2 = w( R2 - r2)
2K[l+ il:][ ~ l (14.7.3)
Thus, the depth to salt water at any location is a function of the rainfall recharge, the size of the
island, and the hydraulic conductivity. For almost all island conditions, it can be shown that this
approximate solution is indistinguishable from more exact solutions by potential theory. 16
Tidal, atmospheric, and rainfall fluctuations (see Chapter 6), together with dispersiOJ!, cre-
ate a transition zone along the interface in an oceanic island. The close proximity of this -~\-
boundary zone to the water table can introduce saline water into a well by upconing. Therefore,
care must be exercised in development of underground water supplies so that pumping causes
a minimal disturbance to the fresh-salt water equilibrium. To avoid the danger of entrainment
of saline water, island wells should be designed for minimum drawdown, just skimming fresh
water from the top of the lens. If small diameter wells are employed, they should be shallow,
dispersed, and pumped at low uniform rates.
In areas where water tables are shallow, an infiltration gallery (see Chapter 5), consisting
of a horizontal collecting tunnel at the water table, is advantageous.53 Drawdowns of a few
centimeters can in many instances furnish plentiful water supplies. Installations of infiltration
galleries for local water supplies were developed on Be.nnuda and the Bahamas in the Atlantic
Ocean and on the Gilbert and Marianas islands in the Pacific Ocean.42• 45 Where water tables
are deep, dug wells or shafts are sunk to the water table with horizontal tunnels (adits) extend-
ing outward to intercept the uppermost layer of fresh water.
Illustrative of the occurrence of fresh water on an oceanic island is Barbados (see Figure
14.7.2). Here a thin highly permeable coral limestone layer serves as the aquifer. Recharge
from rainfall percolates through solution channels along the bottom of the limestone until it
reaches sea level. From there to the coast the fresh water floats in a layer (known locally as
sheet water) above the underlying saline water. Water is extracted from large-diameter dug
wells connected to horizontal adits at the water table.
14.8 Seawater Intrusion in Karst Terrains 605
so.----.----.-----,----.-----,----.----,.---~
60 /Pumping well --~)

Ground surtace ~ .. -·
:::::J
(/)
~
5
40
20 Shoreline____ ~
___..,./ .
)~ -\
Water
table
c:: Fresll water
.Q -20
iii
5i -40
w
-60 Figure 14.7.2. Geologic profile showing
Impermeable
-BO clay and shale interrelations of fresh and saline water in
-1Q0L---~ ____ _ L_ _ _ __ L_ _ _ _~----L----J----~----~
an unconfined aquifer at Bridgetown,
1 0 2 3 4 5 6 7 Barbados (courtesy Waterworks
Distance from shoreline (km) Department, Government of Barbados).
1000
800
lnflowing
600 groundwater
400 Ground
:::::J surface
(/) I
~ 200
E Water table ~
c::
_______ j_~
0
0
'; I
~ -200 I
w I
-400
s Figure 14.7.3. Geologic profile showing
-600
Permeable interrelations of fresh and saline water at
basalt Honolulu, Hawaii. The thickening of the
;-
-BOO
6 4 2 0 2 4 6 8 transition zone resulted from pumping
s
.,, Distance from shoreline (km) begun in 1880 (after Todd and Meyer63) .
$
It Another example of the balance between fresh and saline groundwater on an oceanic island
h is Honolulu (see Figure 14.7.3). Permeable basalt forms the aquifer, but impermeable
v, caprock (erosional material and marine deposits) acts as a groundwater dam. Before wells were
first drilled in 1880, groundwater discharged as springs either at the inland or at submarine
.g boundaries of the caprock. At that time the transition zone was narrow and nearly horizontal.
w Subsequent development of water by wells has lowered the water table and expanded the transi-
lfl tion zone so that the volume of fresh water has been substantially reduced. Depths of production
ic wells have decreased gradually from 450 m to 85 mas upconing has progressively increased.63
~s
:1-
14.8 SEAWATER INTRUSION IN KARST TERRAINS
re Coastal aquifers consisting of karstic limestone pose special problems of seawater intrusion.
se Irregular fissures and solution openings enable seawater to enter the aquifer in configurations
it that may differ appreciably from those for more homogeneous aquifers.57• 58 Unique features
as sometimes found in karst are intermittent brackish springs, which can result from channels
Jg connecting with the sea or where saline heads under high tides exceed inland freshwater heads,
causing seawater to be discharged inland.
A common feature in the karst regions surrounding the Mediterranean Sea are solutions
channels, which discharge fresh water as submarine springs. But pumping of these channels to
prevent wastage of fresh water often yields saline water within hours because seawater can
freely enter the channel if the freshwater flow is reduced.
14.9 CONTROL OF SALINE WATER INTRUSION

Methods for controlling intrusion vary widely depending on the source of the saline water, the
extent of intrusion, local geology, water use, and economic factors. Table 14.9.1 summarizes the
l
~
generally recognized methods for controlling intrusion from various sources. Because as little ~
as two percent of seawater in fresh water can render water unpotable, considerable attention has
been focused on methods to control seawater intrusion. Alternative methods are discussed in the l
following paragraphs. 9· 61 Measures for coping with upconing were described in earlier sections
of this chapter. l
I
l
Modification of Pumping Pattern
Changing the locations of pumping wells, typically by dispersing them in inland areas, can aid
in reestablishing a stronger seaward hydraulic gradient. Also, reduction in pumping of exist-
ing wells can produce the same beneficial effect.
Artificial Recharge
Groundwater levels can be raised and maintained by artificial recharge, using surface spread-
ing for unconfined aquifers and recharge wells for confined aquifers. This necessitates devel-
opment of a supplemental water source.
Extraction Barrier
An extraction barrier is created by maintaining a continuous pumping trough with a line of
wells adjacent to the sea. Seawater flows inland from the ocean to the trough, while fresh water
within the basin flows seaward toward the trough, as shown in Figure 14.9.1. The water
pumped is brackish and normally is discharged into the sea.
Table 14.9.1 Methods for Controlling Saline Water Intrusion
Source or cause of intrusion Control methods

Seawater in coastal aquifer Modification of pumping pattern
Artificial recharge
Extraction barrier
Injection barrier
Subsurface barrier
Upconing Modification of pumping pattern
Saline scavenger wells
Oil field brine Elimination of surface disposal
Injection wells
Plugging of abandoned wells
Defective well casings Plugging of faulty wells
Surface infiltration Elimination of source
Saline water zones in freshwater aquifers Relocation and redesign of wells
14.9 Control of Saline Water Intrusion 607
Figure 14.9.1. Control of seawater

e intrusion by an extraction barrier
fanning a pumping trough parallel to
e seawater
wedge the coast.
e
s
e
s Injection Barrier
This method maintains a pressure ridge along the coast by a line of recharge wells. 10· 62
Injected fresh water flows both seaward and landward, as indicated in Figure 14.9.2. High-
quality imported water is required for recharge into wells. A combination of injection and
extraction barriers is feasible; this reduces both recharge and extraction rates but requires a
d larger number of wells.
t-
Subsurface Barrier
Construction of an impermeable subsurface barrier parallel to the coast and through the verti-
cal extent of the aquifer can effectively prevent the inflow of seawater into the basin (see
l- Figure 14.9.3). Materials to construct a barrier might include sheet piling, puddled clay, emul-
1- sified asphalt, cement grout, bentonite, silica gel, calcium acrylate, or plastics. Chief problems
are construction cost and resistance to earthquakes and chemical erosion.
lf
er Ground surface
er
-- \
Aquifer
-
Figure 14.9.2. Control of sea-
water intrusion by an injection
barrier forming a pressure ridge
parallel to the coast.
Ground surface
Piezometr1c surface
Impermeable barrier
Aquifer
Figure 14.9.3. Control of seawater
intrusion by an impenneable sub-
Fresh water
surface barrier parallel to the coast.
14.10 EXAMPLES OF SEAWATER INTRUSION

Barlow and Wild4 developed a bibliography on the occurrence and intrusion of saltwater in
aquifers along the U.S. Atlantic coast. Of the many localities along the coast of the United
States experiencing seawater intrusion, those on Long Island and Southern California are per-
haps most significant. Brief descriptions of each of these situations follow.
14.10.1 Long Island, New York

Long Island is underlain by a wedge-shaped mass of unconsolidated sediments that extends
more than 600 m in depth. Major intrusion has occurred in western Long Island as a result of
development of groundwater and of a decrease in recharge by improved drainage and sewer
systems. By the mid-1930s, pumping had lowered water levels at the western end to as much
as 10 m below sea level and caused extensive intrusion. Along the southern coast, similar
development and reduction of recharge caused active intrusion at depth. 41
Intrusion has been largeIy controlled by abandoning pumping and using an imported
water supply and by requiring that water pumped from industrial wells be recharged back into
the ground after use. Farther east in Nassau County, intrusion is not yet a serious problem;
however, to prevent this tendency an extensive program to salvage stonn runoff and recharge
it through infiltration basins has been initiated (see Chapter 13). In addition, experimental
work is underway to reclaim wastewater for recharge to fonn an injection barrier. Figure
14.10.1 shows a cross section of the distribution of saline water in southwestern Nassau
County; the advance of the saline wedge inland varies from 3 to 60 m per year depending on
local pumping conditions.
14.10.2 Southern California

Los Angeles, California
Intrusion appeared in the early 1930s along the west coast of Los Angeles County and the rate
accelerated rapidly with development of the area in the 1940s. The first and largest injection
barrier project has been constructed along a 11-km portion of this shoreline. Some 94 recharge
wells fonn a pressure ridge in the confined aquifers so that seawater is effectively separated
from the overpumped basin inland. 1o, 36
Ground surface Gardiners clay

Upper glacial aquifer
Figure 14.10.1. Geologic profile showing interrela-

tions of fresh and saline water at Far Rockaway,
Nassau County, Long Island (after Perlmutter and
Distance from shoreline, kilometers Geraghty47 ).
14.10 Examples of Seawater Intrusion 609
4
n
d 2
Q)~
r- ()II)
~·~
:::l"' 0
~$
;e
·- Cll
-2
E ..:
0~
N•-
Q) :::l
·- 0'
~t( -4
Is c:,
0 0
I
·-0 all
~ -6
I Figure 14.10.2. Piezometric surface profiles per-
Jf ~~
_!!II)
w.E -8 pendicular to the seawater intrusion injection barri-
er I er, Los Angeles County, after various time inter-
;h -10
lnjectio'n barrier
vals. Note establishment of the pressure ridge fol-
ar Seaward...,__ I ~Landward
lowing initiation of the barrier at the end of 1963
(courtesy Los Angeles County Aood Control
~d Distance from injection barrier, meters District).
1t0
m;
ge
tal Figure 14.10.2 shows a profile of the piezometric surface perpendicular to the barrier dur-
tre ing the years that this portion of the barrier was established. It can be seen how the landward
.au
gradient in the injected Silverado Aquifer was transformed into a pressure ridge extending
on above sea level.
A total of 267 observation wells are monitored to study effects of the injection barrier.
Oxnard Pwin, Ventura, California

In Section 10.11, a case study is presented of using simulation-optimization modeling to look
ate the Santa Clara-Calleguas Basin in the Oxnard Plain in southern California. Reichard49 pre-
ion sented an application of a response matrix model for groundwater-surface water management
rge to identify efficient strategies for meeting water demand (with uncertain surface water
.ted supplies) and controlling regional water quality (seawater intrusion) in the Santa
Clara-Calleguas Basin in southern California, shown in Figures 10.11. 6-10.11.8.
Figure 14.10.3 illustrates the chloride concentrations in water from wells in the upper
aquifer system in the Oxnard Plain from 1955 to 1989. Seawater entered the aquifers through
outcrop areas in the Hueneme and Mugu submarine canyons (see Figure 14.10.4) in the mid-
1950s and advanced inland in response to changes in the amount and distribution of pumping,
as illustrated in Figure 14.10.3. By 1989, approximately 23 square miles of the upper aquifer
system was believed to be intruded by seawater. 31 Because of the increasing chloride concen-
trations, pumping was shifted to the lower system from the upper system. As a result the water
levels in the lower system declined to below sea level. A combination of groundwater man-
agement strategies and increased availability of water from the Santa Clara River for ground-
water recharge, by 1993, caused water levels in wells near the coast to rise above sea level and
above water levels in the perched aquifer. 31
ela- The studies by Reichard49 showed that a significant reduction in pumping, particularly in
the lower aquifer system, combined with a large quantity of additional water (either recharged
d to the aquifers or used in place of groundwater pumping), would be required to control the sea-
water intrusion.
A21W
'
J
I
1
'•.
Figure 14.10.3.
Chloride concen-
trations in water
from wells in the
upper aquifer sys- v
tern in the Oxnard D
r:··:::;j Unconsolidated deposits Chloride concentrations (mg/L) Plain, 1955-89. f

~:i:Z:l Consolidated rocks D 1oo-25o • 5oo-1.ooo Data from
~\:(' Bathymetry: Depth of • 250-500 • > 1,000 California c
/ water, in feet below sea Department of b
level. Interval 50 feet Water Resources 0
0 Hueneme Canyon
I I
and County of 0
® Mugu Canyon 5 Kilomelers Ventura Public (;
Works Agency 0
(lzbick:i31 ).
610
Problems 611
sw Nl 400
400
Sea
1
1 Sea PaCJ1ic Ocean Level
Level
g 400
Q)
400 lJ
::J
g = 800
Q)
~
lJ 800
~
-~
(( 1,200
1,200
1,600
1,600
0 SMiles
I
0
1 1
1 1 1 I I
I
5 Ki\otmMers
2,000 Vertical scale is greafty exaggerale<
Figure 14.10.4. Generalized geologic sections of the two submarine canyons in Figure 14.10.3. Hueneme Canyon is shown as A
and Mugu Canyon is shown as Bin Figure 14.10.3 (Izbicki31 ).
PROBLEMS
14.3.1 The water table 30m from the shoreline is located 0.65 m (c) the new location of the toe after the well installation;
I
above sea level in an unconfined coastal aquifer. Determine the (d) the location of the water divide.
I
j
depth to the saltwater interface at this location using the Ghyben-
Herzberg relation.
14.3.2 Rework Problem 14.3.1 using Glover's relation. The
hydraulic conductivity of the aquifer is 8.1 rn/day. Which method
provides a more exact solution?
14.5.3 Freshwater flows from an unconfined coastal aquifer into
the sea at a steady rate of 30 (ff/day)/ft. The impenneable bedrock
underlying the aquifer is horizontal and found at a depth of 140ft.
The hydraulic conductivity of the aquifer is 45 ft/day. If a new
I
fully penetrating pumping well were constructed 500 ft from the
14.3.3 The saltwater interface 40 ft from the shoreline is found shoreline, what is the maximum allowable pumping rate in order to
48 .5 ft below sea level in a coastal aquifer. The hydraulic conduc- prevent bracldsh water from occurring in the pumping well?
tivity of the subsurface medium is 68 ft/day. Determine the rate of
14.5.4 Determine the location of the toe of fresh-saline interface
fresh water input from the aquifer into the sea and the width of the
in Problem 14.5.1 without the well pumping using
submarine zone through which fresh water discharges.
(a) the Ghyben- Henberg relation;
14.3.4 Determine the depth of the interface beneath the shore-
line in Example 14.3.2 using the Ghyben-Herzberg model and (b) Glover's relation;
Glover's model. (c) Strack's solution.
14.5.1 The seaward specific discharge in an unconfined coastal 14.6.1 A well pumps at a constant rate of 15 Us in an unconfined
aquifer is 2.2 (m3/day)/m. The aquifer is underlain by a horizon- coastal aquifer. The aquifer is 40 rn thick and has a hydraulic con-
tal impervious layer 38 m below sea level. The hydraulic conduc- ductivity of 68 m/day. An initial horizontal saltwater interface lies
tivity of the aquifer is estimated as 20m/day. If a fully penetrating 14m below the top of well screen. If the local rise of the interface
well with a constant pumping rate of 850m3/day is installed 350 below the well is monitored continuously, when should the pump-
m from the shoreline, determine if brackish water occurs in the ing be stopped before the quality of the pumped water is affected?
pumping well
14.6.2 A new pumping well for potable water extraction will be
14.5.2 The freshwater discharge into the sea in an unconfined constructed in a 35-m thick confined aquifer that has a hydraulic
coastal aquifer is 50 (ft3/day)/ft. The horizontal impermeable conductivity of 50 rnlday. A 6-m thick horizontal layer of saline
bottom lies 100ft below sea level and the hydraulic conductivity water is detected at the bottom of the aquifer before the well
of the aquifer is 80 ft/day. If a new well pumping at a constant rate installation. If the desired pumping rate is 800m3/day, determine
of 750 fr/hr is installed 1,000 ft from the shoreline, determine the maximum permissible length of the well screen in order to
(a) the location of the toe of the interface before the well installation; minimize the interruption of pumping. Assume that the well
(b) if brackish water occurs in the pumping well; screen starts from the upper confining layer.
REFERENCES
1. Ackermann, N. L., and Y. Y. Chang, Salt water interface during 21. Essaid, H. 1., The Computer Model Sharp, A Quasi-Three-Dimen-
groundwater pumping, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. sional Finite Difference Model to Simulate Freshwater and Saltwater Flow
97, no. HY2, pp. 223-232, 1971. in Layered Coastal Aquifer Systems, U.S. Geological Survey, Water
2. Anderson, P. F., J. W. Mercer, and H. 0. White, Jr., Numerical model- Resources Investigations Report 90-4130, 1990b.
ing of salt-water intrusion at Hallandale, Florida, Ground Water, v. 26, 22. Fetter, C. W., Applied Hydrogeology, 4th ed., Prentice Hall, Upper
no.5,pp.619-630, 1988. Saddle River, NJ, 2001.
4
3. Atkinson, S. F., G. D. Miller, D. S. Curry, and S. B. Lee, Salt Water 23. Fetter, C. W., Jr., Position of the saline water interface beneath oceanic t!
Intrusion-Status and Potential in the Contiguous United States, Lewis islands, Water Resources Research, v. 8, pp. 1307-1315, 1972. p
Publishers, Chelsea, MI, 1986. 24. Guvanasen, V., S.C. Wade, and M. D. Barcelo, Simulation of regional 4
4. Barlow, P. M., and E. C. Wild, Bibliography on the Occurrence and groundwater flow and salt water intrusion in Hernando County, Florida, 6
Intrusion of Saltwater in Aquifers along the Atlantic Coast of the lJnited Ground Water, v. 39, no. 5, 2000.
~tates, U.S. Geological Survey Open-File Report 02-235, Northborough, 25. Harris, W. H., stratification of fresh and salt water on barrier islands 4
MA, 2002. as a result of differences in sediment permeability. Water Resources I:
5. Bear, J., Hydraulics of Groundwater, McGraw-Hill, New York, 1979. Research, v. 3, pp. 89-97, 1967.
4
6. Bear, J., A. H. D. Cheng, S. Sorek, D. Ouazar, and Ismael, eds., Sea- 26. Haubold, R. G. Approximation for steady interface beneath a wel.l 0
water Intrusion in Coastal Aquifers-Concepts, Methods, and Practices, pumping fresh water overlying salt water, Ground Water, v. 13,
4
Kluwer Academic, Dordrecht, The Netherlands, 1999. pp.254-259, 1975.
d
7. Bear, J., and G. Dagan, Intercepting freshwater above the interface in a 27. Herzberg, B., Die Wasserversorgung einiger Nordseebader, Jour. }\
coastal aquifer, Inti. Assoc. Sci. Hydrology Pub!. 64, pp. 154-181, 1964. Gasbeleuchtung und Wasserversorgung, v. 44, pp. 815-819, 842-844,
4
8. Bear, J., and A.Verruijt, Modeling Groundwater Flow and Pollution, Munich, 1901.
0
Reidel, Dordrecht, The Netherlands, 1987. 28. Hubbert, M. K., The theory of ground-water motion, Jour. Geol., v. 1
9. Bruington, A. E., Control of salt-water intrusion in a ground-water 48,pp. 785-944,1940.
4
aquifer, Ground Water, v. 7, no. 3, pp. 9-14, 1969. 29. Huyakorn, P. S., P. F. Anderson, J. W. Mercer, and H. 0. White, Jr., c:
10. Bruington, A. E., and F. D. Seares, Operating a sea water barrier proj- Saltwater intrusion in aquifers: Development and testing of a three- R
ect, Jour. Irrig. Drain. Div., Amer. Soc. Civil Engrs., v. 91, no. IR1, pp. dimensional finite element model, Water Resources Research, v. 23, no. 2,
5
117-140, 1965. pp.293-312, 1987.
tf
11. Cahill, J. M., Hydraulic Sand-Model Study of the Cyclic Flow of Salt 30. Huyakorn, P. S., Y. S.Wu, and N. S. Park, Multiphase approach to the SJ
Water in a Coastal Aquifer, U.S. Geological Survey Professional Paper numerical solution of a sharp interface saltwater intrusion problem, Water
5
575-B, pp. 240-244, 1967. Resources Research, v. 32, no. 1, pp. 93-102, 1996.
b
12. Calvache, M. L., and A. Pulido-Bosch, Modeling the effects of salt- 31. lzbicki, J. A., Seawater Intrusion in a Coastal California Aquifer, U.S.
5
water intrusion dynamics for a coastal karstified block connected to a Geological Survey Fact Sheet 125-96, http://ca. water.usgs.gov/archive/
si
detrital aquifer, Ground Water, v. 32, no. 5, pp. 767-777, 1994. fact_sheetslb07 /ref.html, 1996.
c
13. Chandler, R. A., and D. B. McWhorter, Upconing of the salt- 32. Kashef, A. 1., On the management of ground water in coastal aquifers,
5
water-fresh-water interface beneath a pumping well, Ground Water, v. 13, Ground Water, v. 9, no. 2, pp. 12-20, 1971.
lc
pp. 354-359, 1975. 33. Kemblowski, M., Saltwater-freshwater transient upconing-An 91
14. Charmonman, S., A solution of the pattern of fresh-water flow in an implicit boundary-element solution, Jour. Hydrology, v. 78, pp. 35-47,
5·
unconfined coastal aquifer, Jour. Geophysical Research, v. 70, 1985. ir
pp.2813-2819, 1965. 34. Kemblowski, M., The impact of the Dupuit-Forcheimer approxima- 5.
15. Chidley, T. R. E., and J. W. Lloyd, A mathematical model study of tion on salt-water intrusion simulation, Ground Water, v. 25, no. 3, 5:
freshwater lenses, Ground Water, v. 15, pp. 215-222, 1977. pp.331-336, 1987. ir
16. Cooper, H. H., Jr. et al., Sea Water in Coastal Aquifers, U.S. Geo- 35. Lau, L. S., and J. F. Mink, A step in optimizing the development of the }I
logical Survey Water-Supply Paper 1613-C, 84 pp., 1964. basal water lens of Southern Oahu, Hawaii, Inti. Assoc. Sci. Hydrology 51
17. Dagan, G., and J. Bear, Solving the problem oflocal interface upcon- Pub!. 72, pp. 500-508, 1967. le
ing in a coastal aquifer by the method of small perturbations, Jour. Hydr. 36. Laverty, F. B., and H. A. van der Goot, Development of a fresh-water 5'
Research, v. 6, pp. 15-44, 1968. barrier in southern California for the prevention of sea-water intrustion, w
18. Drabbe, J., and W. Badon Ghyben, Nota in verband met de Jour. Amer. Water Works Assoc., v. 47, pp. 886-908, 1955. u
voorgenomen putboring nabij Amsterdam, Tijdschrift van het Koninklijk 37. Ledoux, E., S. Sauvagnac, and A. Rivera, A compatible single-
Instituut van Ingenieurs, The Hague, Netherlands, pp. 8-22, 1888-1889. phase/two-phase numerical model: Modeling the transient salt-
19. Ernest, L. F., Groundwater flow in the Netherlands delta area and its water/fresh-water interface motion, Ground Water, v. 28, no. 1, pp. 79-87,
influence on the salt balance of the future Lake Zeeland, Jour. Hydrology, 1990.
v. 8, pp. 137-172, 1969. 38. Long, R. A., Feasibility of a Scavenger-Well System as a Solution to
20. Essaid, H. 1., A multilayered sharp interface model of coupled fresh- the Problem ofVertical Salt-Water Encroachment, Water Resources Pam-
water and saltwater flow in coastal systems: Model development and phlet 15, Louisiana Geological Survey, Baton Rouge, 27 pp., 1965.
application, Water Resources Research, v. 26, no. 7, pp. 1431-1454, 39. Louisiana Water Resources Research Inst., Salt-water encroachment
1990a. into aquifers, Bull. 3, Louisiana State Univ., Baton Rouge, 192 pp., 1968.
References 613
40. Lusczynski, N. J., Head and flow of ground water of variable density, 58. Stringfield, V. T., and H. E. LeGrand, Effects of karst features on
Jour. Geophysical Research, v. 66, pp. 4247-4256, 1961. circulation of water in carbonate rocks in coastal areas, Jour. Hydrology,
41. Lusczynski, N.J., and W. V. Swarzenski, Salt-Water Encroachment in v. 14, pp. 139-157, 1971.
Southern Nassau and Southeastern Queens Counties, Long Island, New 59. Task Committee on Salt Water Intrusion, Saltwater intrusion in the
York, U.S. Geological Survey Water-Supply Paper 1613-F, 76 pp., 1966. United States, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 95,
42. Mather, J. D., Development of the groundwater resources of small no. HY5, pp. 1651-1669, 1969.
limestone islands, Quarterly Jour. Engmg. Geol., v. 8, pp. 141-150, 1975. 60. Todd, D. K., Salt water intrusion of coastal aquifers in the United
43. McElwee, C. D., A model study of salt-water intrusion to a river using States, Inti. Assoc. Sci. Hydrology Publ. 52, pp. 452-461, 1960.
the sharp interface approximation, Ground Water, v. 23, no. 4, 61. Todd, D. K., Salt water intrusion and its control, Jour. Amer. Water
pp.465-475, 1985. Works Assoc., v. 66, pp. 180-187, 1974.
44. Newport, B. D., Salt Water Intrusion in the United States, Rept. EPA- 62. Todd, D. K., and L. Huisman, Ground water flow in the Netherlands
600/8-77-011, U.S. Environmental Protection Agency, Ada, OK, 30 pp., coastal dunes, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 85,
1977. no. HY7, pp. 63-81, 1959.
45. Ohrt, F., Water development and salt-water intrusion on Pacific 63. Todd, D. K., and C. F. Meyer, Hydrology and geology of the Honolulu
Islands, Jour. Amer. Works Assoc., v. 39, pp. 979-988, 1947. aquifer, Jour. Hydraulics Div., Amer. Soc. Civil Engrs., v. 97, no. HY2, pp.
46. Padilla, F., and J. Cruz-Sanjulian, Modeling sea-water intrusion with 233-256, 1971.
open boundary conditions, Ground Water, v. 35, no. 4, pp. 704-712, 1997.
64. Vacher, H. L., Dupuit-Ghyben-Herzberg analysis of strip-island
47. Perlmutter, N. M., and J. J. Geraghty, Geology and Ground-Water Con- lenses, Geol. Soc. Amer. Bull., v. 100, pp. 580-591, 1988.
ditions in Southern Nassau and Southeastern Queens Counties, Long Island,
65. Vacher, H. L., and T. N. Wallis, Comparative hydrogeology of fresh-
N.Y., U.S. Geological Survey Water-Supply Paper 1613-A, 205 pp., 1963.
water lenses of Bermuda and Great Exuma Island, Bahamas, Ground
48. Person, M., J. Z. Taylor, and S. L. Dingman, Sharp interface models Water, v. 30, no. I, pp. 15-20, 1992.
of salt water intrusion and wellhead delineation on Nantucket Island,
66. Verruijt, A., A finite element model for interface problems in ground-
Massachusetts, Ground Water; v. 36, no. 4, pp. 731-742, 1998.
water flow, Microcomputers in Engineering Applications, B. Schrefler and
49. Reichard, E. G., Groundwater-surface water management with sto- R. W. Lewis, eds., John Wiley, New York, pp. 251-271, 1987.
chastic surface water supplies: A simulation optimization approach, Water
Resources Research, v. 31, no. 11, pp. 2845-2865, Nov. 1995. 67. Visher, F. N., and J. F. Mink, Ground-Water Resources in Southern
Oahu, Hawaii, U.S. Geological Survey Water-Supply Paper 177B, 133 pp.
50. Reilly, T. E., and A. S. Goodman, Quantitative analysis of saltwa-
1964.
ter-freshwater relationships in groundwater systems-A historical per-
spective,Jour. Hydrology, v. 80, pp. 125-160, 1985. 68. Wallis, T. N., H. L. Vacher; and M. T. Stewart, Hydrogeology of fresh-
water lens beneath a Halocene strandplain, Great Exuma, Bahamas, Jour.
51. Reilly, T. E., and A. S. Goodman, Analysis of saltwater upconing
Hydrology, v. 125, pp. 93-109, 1991.
beneath a pumping well, Jour. Hydrology, v. 89, no. 3-4, pp. 169-204, 1987.
l
69. Wang, F. C., Approximate theory for skimming well formulation in
( .$~ 52. Reilly,T. E., M. H. Frimpter, D. R. LeBlanc, and A. S. Goodman, Analy-
the Indus Plain of West Pakistan, Jour. Geophysical Research, v. 70,
sis of steadystate salt-water upconing with application at Truro well field,
a Cape Cod, Massachusetts, Ground Water, v. 25, no. 2, pp. 194-206, 1987. pp. 5055-5063, 1965.
;,
I
-j
53. Rochester, E. W., Jr., and G. J. Kriz, Potable water availability on
long oceanic islands, Jour. San. Engng. Div., Amer. Soc. Civil Engrs., v.
70. Wilson, J. L., and A. Sada Costa, Finite element simulation of a salt-
water-freshwater interface with interface toe tracking, Water Resources
n 96, no. SA5, pp. 1235-1248, 1970. Research, vol. 18, pp. 1078-1080, 1982.
' 54. Schmorak, S., and A. Mercado, Upconing of fresh water-sea water 71. Wiroganagud, P., and R. J. Charbeneau, Saltwater upconing in uncon-
interface below pumping wells, field study, Water Resources Research, v. fined aquifers, Jour. Hydr. Engrg., v. 111, no. 3, pp. 417-434, 1985.
l- 5,pp. 1290-1311,1969. 72. Wu, P., Y. Xue, J. Liu, J. Wang, Q. Jiang, and H. Shi, Sea-water intru-
\, 55. Segol, G., and G. F. Pinder, Transient simulation of saltwater intrusion sion in the coastal area of Laizhou Bay, China: 2. Sea-water intrusion
in southeastern Florida, Water Resources Research, v. 12, no. 1, pp. 65-70, monitoring, GroundWater, v. 31, no. 5, pp. 740-745, 1993.
1e 1976. 73. Xue, Y., J. Wu, P. Liu, J. Wang, Q. Jiang, and H. Shi, Sea-water intru-
~y sion in the coastal area of Laizhou Bay, China: 1. Distribution of sea-
56. Strack, 0. D. L., A single-potential solution for regional interface prob-
lems in coastal aquifers, Water Resources Research, v. 12, no. 6, Dec. 1976. water intrusion and its hydrochemical characteristics, Ground Water, v. 31,
!r
57. Stringfield, V. T., and H. E. LeGrand, Relation of sea water to fresh no.4,pp.532-537, 1993.
n, water in carbonate rocks in coastal areas, with special reference to Florida, 74. Yechiele, Y., Fresh-saline ground water interface in the western Dead
USA., and Cepha1onia, Greece, Jour. Hydrology, v. 9, pp. 387-404, 1969. Sea area, Ground Water, v. 38, no. 3, pp. 615-623, 2000.
~-
t-
7,
to
n-
nt
8.
1
J AppendixA
1
~
~
~
METRIC UNITS AND ENGLISH EQUIVALENTS
Metric Units
Quantity Unit Symbol Formula
Base units
Length meter m
Mass kilogram kg
Time second s
Electric current ampere A
Thermodynamic
temperature kelvin K
Plane angle radian rad
Solid angle steradian sr
Derived units
Acceleration meter per second m/s2
squared
Activity (of a
radioactive source) disintegration per (disintegration)/s
second
Angular acceleration radian per second rad/s 2
squared
Angular velocity radian per second rad/s
Area square meter m2
Density kilogram per cubic kg/m3
meter
Electric capacitance farad F A·s/V
Electric conductance siemens s A/V
Electric field strength volt per meter V/m
Electric inductance henry H WIA
Electric potential
difference volt v WIA
Electric resistance ohm ~ VIA
Electromotive force volt v WIA
Energy joule I N·m
Force newton N kg·m/s 2
Power watt w }Is
Pressure pascal Pa N/m 2
Quantity of electricity coulomb c A·s
Quantity of heat joule I N·m
Radiant intensity watt per steradian W/sr
Specific heat joule per
kilogram-kelvin J!kg·K
Stress pascal Pa N/m2
Thermal conductivity watt per
meter-kelvin W/m·K
615
616 Appendix A
Metric Units
Quantity Unit Symbol Formula
Velocity meter per second m/s
Viscosity, dynamic pascal-second Pa·s
Viscosity, kinematic square meter per m2/s
second
Voltage volt v WIA
Volume cubic meter m3
Work joule J N·m
Common Metric Equivalents

Length _Area Volume
1m= 103 mm =
1 hectare (ha) 104 m2 1 liter (1) = to-3 rna
= 102 em = t0-2 km 2
= to-3 km Mass
1 kg= 103 g
Metric Prefixes
Prefix Symbol Multiplication Factor
tera T 1012
giga G 109
mega M 106
kilo k 103
hecto h 102
deka da 101
deci d to- 1
centi c to- 2
milli m to-3
micro to- 6
nano
pico
"'pn to- 9
t0-12
fern to f to-15
atto a to-ts
Metric-English Equivalents
Length Flow Rate
1 em = 0.3937 in. 1 IIs = 15.85 gpm
1m= 3.281 ft = 0.02282 mgd
1 km = 0.6214 mi = 0.03531 cfs
1 m3/ s 1.585 X 104 gpm
Area = 22.82 mgd.
1 cm 2 = 0.1550 in. 2 = 35.31 cfs
1 m2 = 10.76 ft 2 1m3/day= 0.1834 gpm
1 ha = 2.471 acre = 2.642 X 10-4 mgd
=
1 km2 0.3861 mi 2 = 4.087 X 10-4 cfs
Appendix A 617
Metric-English Equivalents
Volume Velocity
t cm 3 = 0.06t02 in. 3 1 m/s = 3.28t ftls
t I = 0.2642 gal = 2.237 mi/hr
= 0.0353t ft 3 1 km/hr = 0.9113 ft/s
t m3 = 264.2 gal = 0.62t4 mi/hr
= 35.31 ft 3
= 8.106 X to- 4 acre-ft Temperature
degree Celsius
Mass = kelvin - 273.t5
t g = 2.205 x to- 3 lb (mass) =(degree
1 kg= 2.205 lb (mass) Fahrenheit - 32}/1.8
=9.842 X to-4 long ton
Pressure Viscosity
t Pa = 9.872 X to-6 atmosphere t Pa • s = 1.000 X 103 centistoke
= t.OOO X to- 5 bar = 10.00 poise
= 0.01000 millibar =0.02089 lb (force)· s/ft 2
= 10.00 dyne/ em 2 2
1 m Is = 1.000 X 106 centistoke
= 3.346 X to-4 ft H20 (4 °C) = 10.76 ft 2/s
= 2.953 X t0-4 in. Hg (0°C)
= O.t020 kg (force)/m 2 Force
= 0.02089 lb (force)/ft 2 1N = 1.000 X t05 dyne
= 0.1020 kg (force)
Heat = 0.2248 lb {force)
1 J/m 2 = 8.806 >4 10-5 BTU /ft 2
= 2.390 X 10-5 calorie/cm 2 Power
t Jlkg = 4.299 X to- 4.BTU/lb (mass) 1 W == 9.478 X 10-4 BTU/s
= 2.388 X to-• calorie/g = 0.2388 calorie/s
== 0.7376 ft-lb (force)/s
Hydraulic Conductivity
1 m/day = 24.54 gpd/ft 2 Energy
1 cm/s = 2.121 X t0 4 gpd/ft2 t J= 9.478 X to-• BTU
t m/day = t.198 darcy {for water = 0.2388 calorie
at 20°C) = 0.7376 ft-lb (force)
1 cm/s = 1035 darcy (for water at = 2.778 x to- 7 kw-hr
20°C)
Water Quality
1 mg/l = 1 ppm
= 0.0584 grain/gal
Equivalent weight of ion = atomic weight of ion/valence of ion
meq/1 of ion = mg/1 of ion/equivalent wt of ion
1 meq/1 = 1 me/1
= 1 epm
li'S/cm = lJ.Lmho/cm
li'S/cm = 0.65 mg/1 } Approximations for most natural
waters in the range of 100 to
= O.tO meq/1 of cations 5000 I'Sicm at 25°C
618 Appendix A
Numerical Values for Physical Properties

Quantity Metric English
Gravitationa~ acceleration,
g (std., frel fall) 9.807 m/s 2 32.2 ft/s 2
Density of w ter, p
@50° F/10~ C 1000 kg/m 3 1.94 slugs/ft3
Specific weight of water, y
@50° F/10° C 9.807 X 103 Nfm3 62.4lb/ft 3
Dynamic viscosity of water, p.
@50° F/10° C 1.30 X 10-3Pa • s 2.73 X 10-5 lb • s/ft2
@68° F/20° C 1.00 x 1o-3Pa • s 2.05 x to- 5lb • s/ft2
Kinematic viscosity of water, ,
@50° F/10° C 1.30 x 10- 6 m~/s 1.41 x to- 5 ft 2 /s
@68° F/20° C 1.00 X 10-6 m2 /s 1.06 x 10-5 ft 2 /s
Atmospheric pressure,
p (std.) 1.013 X lOsPa 14.70 psia
Appendix B
FOLLOWING IS A LIST OF U.S. GOVERNMENT

WEB SITES THAT ARE AVAILABLE:
U.S. Geological Survey www.usgs.gov
Environmental Protection Agency (EPA) www.epa.gov
Office of Groundwater and Drinking Water www.epa.gov/ogwdw
U.S. Department of Agriculture www.usda.gov
Natural Resources Conservation Service www.nrcs.usda.gov
U.S. Department of Interior www.doi.gov
Bureau of Reclamation www.usbr.gov
U.S. Army Corps of Engineers www. usace.army.mil
Fish and Wildlife Service www.fws.gov
Federal Legislation http://thomas.loc.gov
Library of Congress http://lcweb.loc.gov
National Environmental Methods Index www.nemi.gov
National Oceanic and Atmospheric Administration www.noaa.gov
FOLLOWING IS A LIST OF NONGOVERNMENTAL

ORGANIZATIONS:
American Geophysical Union www.agu.org
American Institute of Hydrology www.aihydro.org
American Society of Civil Engineers www.asce.org
American Water Resources Association www.awra.org
American Water Works Association www.awwa.org
Geological Society of America www.geosociety.org
Groundwater Management Districts Association www.gmdausa.org
Groundwater Foundation www. groundwater.org
International Association of Hydrogeologists www.iah.org
International Association of Hydrological Sciences www.cig.ensmp.fr/-iahs
International Water Association www.iwahq.org.uk
International Water Resources Association www.iwra.siu.edu
National Ground Water Association www.ngwa.org
National Institutes for Water Resources http://wrri.nmsu.edu/ni wr
National Water Resources Association www.nwra.org
Universities Council on Water Resources www.uwin.siu.edu/ucowr
Water Education Foundation www. water-ed.org
Water Environment Federation www. wef.org
619
Appendix C
TabJe 9.9.1 Input Data for Example 9.9.1 (part of Example 9.9.1)
INPUT FILE
Basic package
steady state problem
7 7 4
FREE
0 1
1 1(713) -1 6. DATA IBOUND Array for Layer 1
-1 -1 -1 -1 -1 -1 -1
1 1 -1
1 1 -1
1 1 -1
1 1 1 1 1 1 -1
1 1 1 -1
1 1 1 -1
0
1(7G14.0) -1 8. DATA Starting Heads in Layer 1
10 9 8 6 4 2 0
10 10 10 10 10 10 3
10 10 10 10 10 10 6
10 10 10 10 10 10 8
10 10 10 10 10 10 12
10 10 10 10 10 10 15
10 10 10 10 10 10 20
365 1 1
BLOCK CENTERED FLOW PACKAGE

50 0 0 0 0 0
0
11 1(1G14.0) -1 3. DATA TRPY(): Anisotropy factor o'f layers
11 1(7G14.0) -1 4.DATA: DELR(NCOL)

500 500 500 500 500 500 500
11 1(7G14.0) -1 5.DATA: DELC(NROW)
500 500 500 500 500 500 500
11 1(7G14.0) -1 7. DATA Tran() = user specified Transmissivity of layer 1
500 500 500 500 500 500 500
500 500 500 500 500 500 500
500 500 500 500 500 500 500
500 500 500 500 500 500 500
500 500 500 500 500 500 500
500 500 500 500 500 500 500
500 500 500 500 500 500 500
WELL PACKAGE
50
5 3 -8000
SIP PACKAGE
50 5
1 .01 0
620
Appendix C 621
RECHARGE PACKAGE
50 0
0
18 1(7G14.0) -1 Recharge
.001 .001 .001 .001 .001 .001 .001
.001 .001 .001 .001 .001 .001 .001
.001 .001 .001 .001 .001 .001 .001
.001 .001 .001 .001 .001 .001 .001
.001 .001 .001 .001 .001 .001 .001
.001 .001 .001 .001 .001 .001 .001
.001 .001 .001 .001 .001 .001 .001
OUTPUT CONTROL PACKAGE
0 0 51 52
1 1 1 1 PERIOD: 1 TIME STEP: 1
0 0 LAYER: 1
Table 9.9.2 Output Data for Example 9.9.1

OUTPUT FILE
11 ITERATIONS FOR TIME STEP 1 IN STRESS PERIOD 1
VOLUMETRIC BUDGET FOR ENTIRE MODEL AT END OF TIME STEP 1 IN STRESS PERIOD 1
CUMULATIVE VOLUMES L**3 RATES FOR THIS TIME STEP L**3fr
IN: IN:
CONSTANT HEAD = 1269498.8800 CONSTANT HEAD = 3478.0791
WELLS= 0.0000 WELLS= 0.0000
RECHARGE= 3285000.2500 RECHARGE= 9000.0010
TOTAL IN= 4554499.0000 TOTAL IN= 12478.0801

OUT: OUT:
CONSTANT HEAD= 1635180.3800 CONSTANT HEAD = 4479.9463
WELLS= 2920000.0000 WELLS= 8000.0000
RECHARGE=· 0.0000 RECHARGE= 0.0000
TOTAL OUT= 4555180.5000 TOTAL OUT= 12479.9463
IN- OUT= -681.5000 IN- OUT= -1.8662
PERCENT DISCREPANCY= -0.01 PERCENT DISCREPANCY = -0.01
TIME SUMMARY AT END OF TIME STEP 1 IN STRESS PERIOD 1

SECONDS MINUTES HOURS DAYS YEARS
TIME STEP LENGTH 3.15360E+07 5.25600E+05 8760.0 365.00 0.99932
STRESS PERIOD TIME 3.15360E+07 5.25600E+05 8760.0 365.00 0.99932
TOTAL TIME 3.15360E+07 5.25600E+05 8760.0 365.00 0.99932
HEAD IN LAYER 1 AT END OF TIME STEP 1 IN STRESS PERIOD 1

7 7
10 9 8 6 4 2 0
9.582833 9.00734 8.179432 7.097559 5.900103 4.558105 3
9.241158 8.766907 8.112735 7.810666 7.444777 6.832418 6
8.873644 8.206147 7.193868 8.087639 8.736045 8.826828 8
8.673501 7.489902 3.868852 8.109984 10.085 11.2388 12
9.156809 8.710836 8.181492 9.898345 11.7551 13.54326 15
9.585927 9.514864 9.747688 11.04657 12.99357 15.67894 20
622 Appendix C
Table 9.9.3 Input Data for Example 9.9.2

INPUT FILE
Basic Package
19 19
FREE
0
1 1(1913) -1 6. DATA IBOUND Array for Layer 1
-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
~1 1 1 1 1 1 1 1 1 1 1 1 .1 1 1 1 1 1 -1
~1 1 1 1 1 1 1 1 1 -1
~1 1 1 1 1 -1
~1 1 1 -1
~1 1 ~1
-1 ~1
~1 ~1
~1 -1
~1 1 ~1
~1 1 ~1
~1 1 -1
~1 1 1 ~1
~1 1 1 -1
-1 1 1 1 1 1 1 -1
~1 1 1 1 1 1 1 1 -1
-1 1 1 1 1 1 1 1 1 1 -1
-1 1 1 1 1 1 1 1 1 1 1 1 1 1 ~1
~1 ~1 -1 -1 ~1 ~ 1 ~ 1 ~ 1 -1 ~1 ~ 1 ~ 1 ~ 1 -1 -1 ~ 1 -1 -1 -1
~999.99
1 1(19G14.0) -1 8. DATA Starting Heads in Layer 1
0 .OOOE+OO
BLOCK CENTERED FLOW PACKAGE

50 -1E+30 0 .1 1 0
0
11 1(1G14.0) -1 3. DATA TRPY(): Anisotropy factor of layers
1
11 1(19G14.0) -1 4.DATA: DELR(NCOL)
63.75 40 40 40 40 40 20 10 5 2.5 5 10 20 40 40 40 40 40 63.75
11 1(19G14.0) -1 5.DATA: DELC(NROW)
63.75 40 40 40 40 40 20 10 5 2.5 5 10 20 40 40 40 40 40 63.75
11 1(19G14.0) -1 7. DATA Tran() = user specified Transmissivity of layer 1
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
164.3 164.3
I
I
Appendix C 623
WELL PACKAGE
50
10 10 -425
SIP PACKAGE
50 5
1 .01 0
OUTPUT CONTROL PACKAGE

0 0 51 52
1 1 PERIOD: 1 TIME STEP: 1
0 0 LAYER: 1
Table 9.9.4 Grid Scheme for Example 9.9.2

Row number, i =Colum n number, j DELC (i) =DELR U) (m)
1 63.75
2 40
3 40
4 40
5 40
6 40
7 20
8 10
9 5
10 2.5
11 5
12 10
13 20
14 40
15 40
16 40
17 40
18 40
19 63.75
Table 9.9.5 Output Data for Example 9.9.2

OUTPUT FILE
11 ITERATIONS FOR TIME STEP 1 IN STRESS PERIOD 1
VOLUMETRIC BUDGET FOR ENTIRE MODEL AT END OF TIME STEP 1 IN STRESS PERIOD 1
CUMULATIVE VOLUMES L**3 RATES FOR THIS TIME STEP L**3/T
IN: IN:
CONSTANT HEAD = 424.5620 CONSTANT HEAD = 424.5620
WELLS= 0.0000 WELLS= 0.0000
TOTAL IN= 424.5620 TOTAL IN= 424.5620
OUT: OUT:
CONSTANT HEAD= 0.0000 CONSTANT HEAD = 0.0000
WELLS= 425.0000 WELLS= 425.0000
TOTAL OUT= 425.0000 TOTAL OUT= 425.0000
IN- OUT= -0.4380 IN- OUT= -0.4380
PERCENT DISCREPANCY = -0.10 PERCENT DISCREPANCY= -0.10

~
624 Appendix C
Table 9.9.6 Results for Example 9.9.2

r (m) Drawdown (Thiem) Drawdown (MODFLOW) r (m) Drawdown (Thiem) Drawdown (MODFLOW)
0.5 2.634 2.737 70 0.599 0.596
2 2.063 2.349 80 0.544 0.541
4 1.777 1.879 100 0.452 0.438
8 1.492 1.605 120 0.377 0.365
12 1.325 1.363 140 0.314 0.296
16 1.207 1.263 160 0.259 0.242
20 1.115 1.163 180 0.210 0.189
24 1.040 1.063 200 0.167 0.144
30 0.948 0.965 220 0.128 0.099
35 0.884 0.909 240 0.092 0.058
40 0.830 0.853 260 0.059 0.017
45 0.781 0.797 280 0.028 0.000
50 0.738 0.741 300 0.000 0.000
60 0.663 0.651
Table 9.9.7 Two Grid Schemes for Example 9.9.3.

DELC (i) = DELR (j) (m) DELC (i) =DELR (j) (m)
Row number, Row number,
i =Column number, j Grid-1 Grid-2 i =Column number, j Grid-1 Grid-2
300 300 16 100 30
2 200 300 17 150 30
3 150 300 18 200 40
4 100 300 19 300 60
5 80 300 20 80
6 60 300 21 100
7 40 200 22 150
8 30 150 23 200
9 30 100 24 300
10 20 80 25 300
11 30 60 26 300
12 30 40 27 300
13 40 30 28 300
14 60 30 29 300
15 80 20
Index
Index note: The f or t following a page number refers to a figure or table, respectively, on that page.
Acoustic logging, 538 Area of influence, of a well, 152 Irrigation method, 560
Adsorption-desorption, 563, 565 Arizona's groundwater management code, Overland flow method, 560
Advection-dispersion equation, 382-384 477-478 Recharge well method, 560
Analytical procedure, 384 Goals of, 478 Spreading method, 560
Peclet number, 384 Arsenic, Assessment monitoring, 386
Aeration zone, 1, 45-50 In groundwater, 329-330 Atmospheric moisture flow,
Capillary fringe, 47-48 In U.S., 33lf In hydrologic cycle, 14
Capillary zone, 46f, 47-48 Artesian aquifers, 56 Auger, 210f
Divisions of subsurface water, 47f Artesian wells, 56 Auger hole tests,
Intermediate zone, 46-47, 46f Artesian springs, 67 For hydraulic conductivity determination,
Soil water zone, 46f Artificial recharge, 547-585, 606 99-100
Capillary water, 46 Concept of, 547-548 Diagram of, lOOf
Gravitational water, 46 Purposes of, 547 Available water, 50
Hygroscopic water, 46 Induced recharge, 584-585 Average interstitial velocity, 89
Age, Recharge methods, 548-556
Of groundwater, 16, 35-36 Basin method, 549, 549f, 550f Balcones fault zone aquifer, 419f, 420f
Residence time, 35 Disposal wells, 554 Cell map for simulation model, 420f
Algal growth, 562f, 563f Ditch-and-furrow method, 552, 552f San Antonio region map, 419f
Alluvial deposits, 52-53 Drain wells, 554 Bank storage, 285-287
Alluvial valleys, in U.S., 79f Flooding method, 552 Barometric efficiency of, 295
Anisotrophy, 101 Incidental recharge, 556 Barrier boundary, 186
Appropriation law, 477 Irrigation method, 553 Barriers for saline water intrusion, 606-607
Aquiclude, 37 Pit method, 553 Extraction barrier, 606, 607f
Aquifer restoration techniques, 398 Recharge well method, 554-556, 555f, Injection barrier, 607, 607f
Aquifers, 36, 52-58 556f Subsurface barrier, 607
Barometric efficiency of, 295 Stream-channel method, 550 Barton Springs-Edwards Aquifer management
Compaction of, 312f Water spreading, 548 model, 496, 496f, 497f
Compression of, 31 Of Recharge mounds, 567-576 Finite-difference grid, 497f
Definition of, 36, 52 Hantush equation, 570-576 Water level contours, 496f
Geologic origin of, 53t Perched mounds, 567-568 Barton Springs-Edwards Aquifer model, 421f
Tidal efficiency of, 301-302 Steady-state equations for, 568-570 Basin,
Types of, 56-58 Recharge rates, 556-558 Data collection, 466-469
Artesian, 56 Clogging effects, 558 Geologic data, 468
Confined, 56 Saline water intrusion, 606 Hydrologic data, 468
Homogeneous, 57 Soil aquifer treatment (SAT) systems, Topographic data, 467
Idealized, 57 560-566 Deferred perennial yield of, 470
Isotropic, 57 Components of, 561 f Investigations, 466
Leaky, 57 Definition of, 561 Management, 464-469
Pressure, 56 Dynamics of, 563f Concepts of, 464-469
Unconfined, 56 Modeling of, 565-566 Investigations, 466f
Aquifer storage recovery (ASR), 583 NITRINFIL, 566 Maximum perennial yield of, 471
Aquifer systems, Nitrogen cycle, 565f Mining yield of, 469
Compaction of, 312f Operation conditions, 566f, 567f Mound development, 567-583
Conceptual model of, 16 Operation of, 563-565 Perennial yield of, 469
Example of, 19f Optimization of, 565 Overdraft of, 470
Major aquifer systems in U.S., 17f Processes, 564f Safe yield, 469
Midwestern Basin and Arches Aquifer Purification processes, 563 Yield of, 469-473
System, 20f Recovery and recharge systems, 562f Deferred perennial yield, 470
Aquifuge, 37 Systems for increasing depth, 548f Maximum perennial yield, 471
Aquitard, 37, 57f Wastewater recharge, 559-560 Mining yield, 469
625
626 Index
Perennial yield, 469 Cone of depression, 152 WATSTORE, 30

Base flow, 287 Confined aquifer, 56 Delayed yield, 172, 173f
Bernoulli equation, 87 Steady radial flow in, 152-156 Effect on pump test, 173f
Biochemical oxygen demand (BOD), 563 Steady unidirectional flow in, 146-147 Delayed index, 173, 174f
Biodegradation, 565 Unsteady radial flow in, 163-172 Depression springs, 67
Bioremediation, 396 Confining bed, 36-37 Detection monitoring, 386
Boreholes, 540f Aquiclude, 37 Dewatering,
Multiple-well borehole, 540f Aquifuge, 37 Organic soils, 317
Single well, 540f Aquitard, 37 Diffusion, 380, 382
Bouwer and Rice method, 262-268 Conjunctive use, 473-476 Apparent (effective) coefficient 382
Brooks-Corey equation, 137-138 Advantages of, 476t Molecular diffusion, 380, 382
Bubbling pressure, 130 Disadvantages of, 476t Dentrification, 351f
Bulk density, 39 Schematic diagram, 475f Desorption, 385
Connate water, 35, 36f Dispersion, 119-122, 380, 382-383
Cable tool method, 218-219 Construction of wells, see well construction Concept of, 119-120
Caliper logging, 536, 536f Consumptive use, 293, 468 Definition of, 119
Hole caliper, 536f Contact springs, 67 Equation for, 120
Capillary fringe, 47-48 Contaminant migration prevention, 398 Hydrodynamic dispersion, 380, 382
Capillary rise, 47, 48f Contaminant release scenarios, 360f Longitudinal coefficient of, 382
In samples of unconsolidated materials, 48t Dissolved contaminant release, 360f Mechanical dispersion, 382-383
Capillary tube, 48f DNAPL release, 360f Of two dye streams, 122f
Capillary water, 46 Inorganic releases, 360f Tracer, 119-120
Capillary zone, 46f, 47-48 LNAPL release, 360f Distribution of, 121 f
Carbonate rocks, Control volume approach, 122-124 Lateral dispersion of, 12lf
Cavity distribution in, 54f Control surface, 122-123 Longitudinal dispersion of, 120f
Characteristic well loss, see well loss Control volume, 122-123 Dispersivity, 380, 383
Chemical analysis of groundwater, 337-340 For derivation of conservation of Dynamic dispersivity, 382
Chemical equivalence, 338-339 mass, 380f Dissolved contaminant release, 360f
Concentrations by weight, 337 Control volume equation for continuity, DNAPL release, 360f
Hardness, 339 126,380 Drawdown, 152
Hardness classification of, 340t Cooper, Bredehoeft, Papadopoulos method, Drawdown curve, 152
Total dissolved solids, 339 255-259 Drilling methods, 214-222
Chemical characteristics of groundwater, 336f Cooper-Jacob method, 167-169 Drilling mud circulation system, 22lf
Chemical composition changes in Corrective action monitoring, 386 Dupuit assumption, 147
groundwater, 348-349 Crustal uplift, 319 Dupuit equation, 148
Atrazine, 352f, 353f Cryptosporidium, 559 Dupuit-Forchheimer equation, 577
Basic exchange, 348 Cumulative infiltration, 134, 135f Dupuit-Ghygen-Herzberg model, 592-593
Calcium to lakes, 350f Green-Ampt method for, 137 Dynamic dispersivity, 382
Chemical precipitation, 348
Chemical reduction, 348 Dagan method, 268-270 Edwards Aquifer, 54-56, 70-73, 70f, 71t, 72f,
Dentrification, 351 f Darcy flux, 126-127, 136,415 73,496-497
Dissolved organic carbon from seepage, 351 f Darcy's law, 86-91, 104-105, 124, 127, 146, Annual recharge to, 71t
Equilibrium, 348 380,415-416,422-425 Balcones fault zone aquifer, 419f, 420f
Forests, 350f Advection, 380 Cell map for simulation model, 420f
Ion exchange, 348 Application to vertically downward flow, San Antonio region map, 419f
Quality variation, 349f 104f Barton Springs-Edwards Aquifer
Wetlands, 350f
Chemical constituents in groundwater, 333t,
334t, 335t
Chemical equivalence, 338-339
Average interstitial velocity, 89
Darcy velocity, 88-89
Experimental verification of, 86-88
For finite difference models, 415-416,
management model, 496, 496f, 497f
Finite-difference grid, 497f
Water level contours, 496f
Barton Springs-Edwards Aquifer model,
I
·.~
Chemical precipitation, 348 423-425 42lf

Chemical reduction, 348
Chow method, 169-170
Clay, 55
Climate change, 321
For infiltration, 136
For kinematic wave model, 133
For unsaturated flow, 127
Validity of, 90
Discharge from, 73
Map of, 70f, 72
Effective hydraulic conductivity equation, 422
Effective porosity, 37, 137, 138t
I
i
~
Collector wells, 240, 241 f
Carnal Springs, 70, 72f
Darcy velocity, 88-89
Data collection platform (DCP), 31f, 32f
Effective saturation, 137
Electrical resistivity method, 513-516
1;
Comprehensive Emergency Response and Data sources, 29-32 Electrical circuit for, 514f
Compensation Act (CERCLA), 387 Data collection platform (DCP), 31f, 32f Electrode arrangements for, 515f 1f
I
Compression, NWISWeb, 30-31 Horizontal profile, 516f
Due to subsidence, 31 Of Real-time data, 31-32 Interpretation of, 515f
I
Index 627
Representative ranges of, 513f Flow lines, 106f, 162f Gravimetric water content, 39
Resistivity equations, 513-514 For anisotropic conditions, 107f, 162f Gravitational water, 46
For Schlumberger arrangement, 514, 515f For a dam, 107f Gravity methods, for surface investigations, 522
For Wenner arrangement, 514, 515f For discharging well in unconfined aquifer, Gravity springs, 67
Schlumberger arrangement, 514, 515f 158f Types of, 67
Wenner arrangement, 514, 515f For discharging real well, 181 f Artesian springs, 67
Equilibrium equation, 153 For partially penetrating well in confined Contact springs, 67
Equipotential lines, 106£, 162f aquifer, 155f Depression springs, 67
Equivalent hydraulic conductance, 430 For recharging image well, 181 f Fracture springs, 67
Equivalent hydraulic conductivity, 101-103 For seepage from channel, 107f Impervious springs, 67
Evaporation, 13-14 For well penetrating confined aquifer, 162f Tubular springs, 67
In global hydrologic system, 15f Fluid-conductivity logging, 537 Green-Ampt,
In hydrologic cycle, 14 Fluid conductivity log, 537f Infiltration equation, 137, l39f
Evapotranspiration, 13-14 Fluid-velocity logging, 537 Parameter values for soil classes, l38t
In global hydrologic system, l5f Example of, 537f Effective porosity, 137, 138t
In hydrologic cycle, 14 Fracture springs, 67 Hydraulic conductivity, 138t
Ex-situ remediation techniques, 396, 398t, 399t Fresh-salt water interface, 592-597 Porosity, 138t
Extensometers for monitoring land subsidence, Biscayne aquifer, Florida, 596f Wetting front soil suction, 138t
322-323 Flow pattern, 593f, 596f Green-Ampt method, see infiltration
Extensometer construction, 322f Ghyben-Herzberg model, 593 Groundwater,
Extensometer instrumentation, 323f Relative salinity, 597, 597f Arsenic in, 331
Extraction barrier, 606, 607f Shape of, 593-595 Budget, 21, 22f
Extraction systems, 396, 407 Structure of, 596-597 Chemical analysis of, 337-340
DNAPL dual-extraction system, 397f Freundlich isotherm, 385 Chemical characteristics of, 336f
LNAPL dual-extraction system, 397f .Fumerole, 74-75 Chemical constituents in, 333t, 334t, 335t
Extraction/injection scheme design, 407 Definition of, 1
Down-gradient pumping, 407 Gaining stream, 284 Hardness of, 340f
Source control with down gradient Genetic types of water, see water, genetic types Phreatic water, 45
pumping, 407 Geochemical cycle, 332f Pollution of, 359-408
Source control with mid-plume and Geohydrology, definition of, l Quality of, 329-358
down-gradient pumping, 407 Geologic formations, Utilization of, 9-13
Extraction trench systems, 397f As aquifers, 52-55 Withdrawals 17f
Alluvial deposits, 52-53 Groundwater age, see age
Fick's law, 380 Clay, 55 Groundwater basin, see basin
Fick's second law, 382 Igneous rock, 55 Groundwater budget, see hydrologic budget
Field capacity, 50 Limestone, 53-54 Groundwater chemistry, 330t,
Finite difference equations, 417-419 Metamorphic rock, 55 Principal chemical constituents, 333t,
Derivation of for three dimensions, 420-427 Sandstone, 55 334t, 335t
Two-dimensional model, 417-419 Volcanic rock, 55 Groundwater contours, 108-112
Finite difference grids, 416f, 420f, 421f, 422f, Geologic methods, for surface investigations, Contour map, 109f
423f, 424f, 427f, 428f, 433f, 441f, 442f, 509-510 Contour map of piezometric surface, 11 Of,
443~445~451~452~454f Geophysical exploration, for surface 112f
Barton Springs-Edwards aquifer, 421f investigations, 512 Groundwater flow, 14-19,114-119
Block-centered grid system, 423f Geophysical logging, 527-528 Across hydraulic conductivity boundary,
Boundary and cell designations, 427f Definition of, 527 114-115
Cell flow, 424f Schematic diagram of, 528f Refraction across layers of sand, 115f
Edwards (Balcones fault zone) aquifer, 420f Summary of applications, 527t Refraction of flow Jines across
Lake-Five 0, 445f Geothermal energy resources, 75-76 boundary, 114f
Modeling low conductivity unit, 429f Dry rock, 76 Across water table, 112-113
Point-centered grid system, 423f Dry steam field, 75 Case study, 116-119
Telescopic mesh refinement, 440-443 Low-temperature field, 75 In global hydrologic system, 15f
Application of, 442f, 443f Wet steam field, 75 In hydrologic cycle, 14
Definition of, 440 Geyser, 73 Near streambeds and banks, 18f, 19f
Illustration of, 441f Ghyben-Herzberg relation, 590-592 Regional flow patterns, 115-116
MODTMR,440 Heads for nonequilibrium conditions, 592f Steady flow to a well, 152-161
Three dimensional, 422f Global hydrologic system, 14-15 Confined aquifer, 152-156
Vertical discretization, 428f Atmospheric water system, 14 Equilibrium equation, 153
Fissure springs, 67 Block-diagram of, 15f Thiem equation, 153
Flow lines, 106f, 162f Subsurface water system, 14 Unconfined aquifer, 156-159
Flow nets, 105-106, 107f, 158f,l62f, 181f Surface water system, 14 Unconfined flow with uniform recharge,
Equipotential lines, 106f, 162f Gravimetric method, 49 159-161
628 Index
Groundwater flow (continued) External loads, 308 Salinity, sources of, 330-335
Steady unidirectional flow, 146-152 Land subsidence, see land subsidence Salt distribution, 348f
Base flow to a stream, 149-152 Meteorological phenomena, 295-299 Temperature, 351, 353f, 354f
Confined aquifer flow, 146-147 Atmospheric pressure effect, 295 Water characteristics that affect, 337t
Dupuit equation, 148 Barametric efficiency, 295 Water quality criteria, 346-348
Unconfined aquifer flow, 147-149 Earthquakes, 306-308 Drinking water standards, 3465
Uniform flow, 147-149 Frost, 298, 299f Industrial water criteria, 346
Unsteady radial flow, Rainfall, 297 Irrigation water criteria, 347
Confined aquifer, 163-172 Tides, 299-303 Laws regulating, 346t
Chow method, 169-170 Earthtides, 302f, 303 Groundwater quality monitoring, see
Cooper-Jacob method, 167-169 Oceantides, 299-300f monitoring
Nonequilibrium equation, 163-164 Time variation of, 279-280 Groundwater recharge, see recharge, artificial
Theis equation, 163 Seasonal variation, 280 recharge
Theis method, 164-167 Secular variation, 279-280 In global hydrologic system, 15f
Recovery test, 170-171 Short-term variation, 280 Groundwater regions, in U.S., 77-82
Unconfined aquifer, 172-177 Groundwater management models, see Alluvial valleys, 79
Delayed yield, 172 management models Hydrologic characteristics of, 80t
Delay index, 173 Groundwater modeling software support, see Map of, 78f
Groundwater flow directions, I 05-119 Modeling support Physical characteristics of, SOt
Flow nets, 105-106 Groundwater models, see Models for Groundwater remediation, see remediation
Equipotential lines, 106f simulation design,
For anisotropic conditions, 107f Groundwater mounds, see recharge mounds Groundwater reservoir, 36
For a dam, 107f Groundwater quality, 329-358 Groundwater withdrawals,
For seepage from channel, 107f Biological analysis, 345 Major aquifer systems, 17f
Flow lines, 106f Chemical analysis of, 337-340 Groundwater yield, see yield
Groundwater flow equations, 122-124 Chemical equivalence, 338-339 Gulf Coastal Plain Aquifer System, 61, 66f
Control volume approach, 123-124 Concentrations by weight, 337
Diffusion equation, 124 Hardness, 339 Hanford Site, Richland, Washington, 377-389
General control volume equation for Hardness classification of, 340t Location of, 377f
continuity, 122 Total dissolved solids, 339 Distribution of carbon tetrachloride, 379f
Groundwater flow paths, 19f Chemical characteristics of, 336f Spread of tritium contamination, 378f
Intennediate, 19f Chemical composition changes, 348-349 Hardness, 339-340
Local, 19f Atrizine, 352f, 353f Classification of, 340t
Regional, 19f Basic exchange, 348 Of groundwater in U.S., 340f
Groundwater flow rates, 103-105 Calcium to lakes, 350f High Plains Aquifer, 59-61, 62f, 63-64
Groundwater flow systems, 15-20,59-66 Chemical precipitation, 348 Homogeneous aquifer, 57
Basins, 59-66 Chemical reduction, 348 Horizontal pipes, 240
Definition of, 16 Dentrification, 351 f Horizontal wells, 239-242
Flow paths of, 19f Dissolved organic carbon from seepage, Collector wells, 240, 241f
Intermediate, 19f 351f Horizontal pipes, 240
Local, 19f Equilibrium, 348 Infiltration gallery, 239-240, 240f
Regional, 19f Forests, 350f Horton's infiltration equation, 139t
In hydrologic cycle, 15-20 Ion exchange, 348 Hot springs, 73
lntennediate system of, 16 Quality variation, 349f Hydraulic conductance, 425
Local system of, 16 Wetlands, 350f Definition of, 425
Regional flow system, 60f Chemical constituents in, 333t, 334t, 335t Effective hydraulic conductivity
Regional systems, 16,59-66 Dissolved gases, 349 equation, 422
Gulf Coastal Plain Aquifer System, 61, 66f Geochemical cycle, 332f Equivalent hydraulic conductance, 430
High Plains Aquifer, 59-61, 62f, 63-64 Graphical representations of, 340-344 Hydraulic conductance equations,
Groundwater hydrology, definition of, 1 Circular diagrams, 343f 430-431
Groundwater levels, 279-308 Pattern diagrams, 343f In MODFLOW·2000 (Layer property
And streamflow, 284-290 Schoeller semilogarithmic diagram, 344f flow package), 436-440
Bank storage, 285 Trilinear diagrams, 344f Arithmetic-mean interblock
Baseflow, 287 Vector diagrams, 343f transmissivity method, 436
Recession curve, 289 Vertical bar graphs, 341f Arithmetic-mean thickness and
Evapotranspiration, 291-295 Groundwater samples, 345 logarithmic-mean hydraulic
Evaporation effects, 291 Measures of, 337 conductivitymethod, 436
Recharge, 294f Natural groundwater quality, 329-330 Dewatered conditions, 438-439
Relationship with water table, 294f Physical analysis, 345 Horizontal hydraulic conductance, 436-437
Transpiration effects, 291, 292f, 293f Saline groundwater, 354-355 Logarithmic-mean interblock
Variation of, 294f Depth to in U.S., 356f transmissivity method, 436
Index 629
Hygroscopic water, 46 Case study, MODFLOW application,

Vertical hydraulic conductance, 437-438
Hyporheic zone, 19f, 286f 444-449
Hydraulic conductivity, 88,91-100, 128-133,
Hvorslev method, 259-261 Boundary conditions, 444--446
138,306
Hydrothermal system, 73-76 Calibrated hydraulic conductivity, 446f
Determination of, 94-100
Schematic diagram of, 74f Finite difference grid, 445f
Auger hole tests, 99-100
Model calibration, 444-447
Formulas, 94-95
Idealized aquifer, 57 Model results, 449f
Laboratory methods, 95-96
Igneous rock, 55 Recharge function, 448
Pump tests, l 00
Linear isotherm, 385 Sensitivity analysis, 447f, 448f
Tracer tests, 98
Sensitivity to specific storage, 447
Equivalent hydraulic conductivity, 10 1-l 03 Image well, 180
Systems, 182f, 189f, 190f, 19lf Sensitivity to specific yield, 447
For unsaturated flow, 128-133
Vertical discretization, 446f
See unsaturated hydraulic conductivity Impervious springs, 67
Particle tracking analysis, 449-450, 450f
Intrinsic permeability, 91 Infiltration,
Cumulative, 134, 135f Flow paths, 450f, 451 f
Of anisotropic aquifers, 100-103
Basins, 551 MODPATH, 449
Of geologic materials, 92-94
Equations, 139t Land subsidence, 308-321
Of proportions of sands, 93f
Green-Ampt, 137, 139f Aquifer system compaction, 312f
Range of values, 94f
Horton's, 139t Compression of aquifers, 31 Of
Representative values of, 93t
Philip's, 139t Crustal uplift, 319
Transmissivity, 92
SCS method, 139t Development effects, 313
Unit hydraulic conductivity, 91
Gallery, 239-240, 240f Dewatering organic soils, 317
Urbanization effects, 306f
Green-Ampt method, 134-137 Hydrocompaction, 315-317
Values for soil classes, 138t
Green-Ampt parameters, 138t Inelastic compression, 311 f
Hydraulic containment, 394, 403f
In global hydrologic system, 15f In Arizona, 316f
Examples of, 403f In Sacramento-San Joaquin Delta, 318f
Hydraulic barriers, 395 In hydrologic cycle, 14
Moisture profile, 134f In U.S., 310f
Nonselective physical barriers, 395
Moisture zones during, 134f Monitoring of, 319
Selective physical barriers, 395
Piezometric surface lowering, 311-315
Types of physical barriers, 396t Saturated zone, 134f
Transmission zone, 134f Sinkhole formation, 317
Hydraulic fracturing, 231
Wetting front, 134f Trends in Houston, Texas, 314f, 315f
Hydraulic jetting, 230, 231 f
Laurel Bay, Florida application of MOC3D, 458f
Hydraulic management models, 484-491 Wetting zone, 134f
Simulation of benzene results, 458f
Confined aquifers, 484-486, 487-490 Ponding time, 140f
Rate, 134, 135f Simulation of MTBE results, 458f
Steady-state, one-dimensional, 484-486
Injection barrier, 607, 607f Leaky aquifer, 57, 58f
Steady-state, two-dimensional, 487-489
Inorganic releases, 360f Unsteady radial flow in, 177-180
Transient, one-dimensional, 489-490
In-situ air sparging, 399f Limestone, 53-55
Unconfined aquifers, 486-487,490-491
In-situ remediation techniques, 396, 398t, 399t LNAPL release, 360f
Steady-state, one-dimensional, 486-487
Logging, see subsurface investigations,
Steady-state, two-dimensional, 490-491 Interception,
In hydrologic cycle, 15f Losing streams, 284
Hydrocampaction, 315-317
International Groundwater Modeling Center Lost river, 53
Hydrodynamic dispersion, 380, 382
Longitudinal coefficient of, 382 (IGWMC), 460
Internet locations, 27 Magmatic water, 35, 36f
Hydrogeology, definition of, 1 Magnetic methods, for surface investigations,
Hydrologic budget, 20-27, 306f Interstices, 37, 38f
Intrinsic permeability, 91 522
Diagram of, 22f
Ion exchange, 348 Management, 464-469
Groundwater system budget, 21
Isotherms, 385-386 Concepts of, 464-469
For Lake Five-0, Florida, 24-27
Freundlich isotherm, 385 Examples, 479-481
For Nassau and Suffolk Counties, 23f
Linear isotherm, 385 Edwards aquifer management, 479
Surface water system budget, 21
Sorption isotherm, 385 High Plains aquifer, 479
System budget, 21
Isotropic aquifer, 57 Investigations, 466-469
Hydrologic cycle, 13-20 Sequence of activities, 466f, 467f
Definition of, 13
Juvenile water, 35, 36f Studies, 466-467
Global hydrologic system, 14-15
Define project, 467
Atmospheric water system, 14
Karst terrain, 53, 605-606 Feasibility, 467
Block-diagram of, 15f
Sea water intrusion in, 605--606 Preliminary examination, 466
Subsurface water system, 14
KGS model, 271-272 Reconnaissance, 467
Surface water system, 14
Kinematic-wave model, 133-134 Management models, 481-504
Groundwater system in, 15-20
Classification of, 483f
Hydrologic system, 14
Lake-Five-0, Florida Definition of, 481
Hydrology, definition of, 1
Inverse solution procedures, 481
Hydroseims, 306 Case study of flow, 116-119
Case study, hydrologic budget, 24-27 Simulation-management models, 481
HYDRUS,568
630 Index
Management models (continued) SUTRA,459 Wells, 436

Flowchart of, 484f ZONEBUDGET, 459 Hydraulic conductance (Layer property flow
Hydraulic management models, 484-491 Models for simulation, 415-431 package), 436-440
See hydrau1ic management models Calibration, 415 Arithmetic-mean interblock
Optimal control approach, 483, 495-497 Calibration sensitivity analysis, 415 transmissivity method, 436
For Barton Springs-Edwards Aquifer, Definition of, 413 Arithmetic-mean thickness and logarithmic-
Texas,496,496~497f Finite difference method, 414 mean hydraulic conductivity method, 436
For groundwater remediation design, Finite element method, 414 Dewatered conditions, 438-439
497-499 Galerk.in's method, 415 Horizontal hydraulic conductance,
For N-Springs Site, Hanford, Edwards (Balcones fault zone) aquifer, 436-437
Washington, 497-499 419f, 420f Logarithmic-mean interblock
Optimization methods, 482 Barton Springs-Edwards aquifer model, transmissivity method, 436
Constraints, 482 421f Vertical hydraulic conductance, 437-438
Feasible solution, 482 Cell map for simulation model, 420f Introduction to, 431
Linear programming, 482 San Antonio region map, 419f LPF package, 436-443
Nonlinear programming, 482 GWSIM,419 Packages, 432t
Objective function, 482 Node, 421 Particle tracking analysis, 449-451, 450f, 45lf
Optimization problem, 482 Steps in development of, 415 Definition of, 449
Simplex method, 482 MODFLOW, see MODFLOW-2000 Flowpaths, 450f, 451f
Policy evaluation and allocation models, Three-dimensional model, 420 MODPATH, 449
492-495 Block-centered grid system, 423f Processes, 432t
Response matrix approach, 492, 499-504 Boundaries, simulation of, 427 Processors for, 459-460
Algebraic technological function, 492 Conductance, 425 RIVGRD, 440
Santa Clara-Calleguas Basin, California, Constant~head cells, 427 Solver packages, 440
499-504 Darcy's law, 422-425 Space discretization, 432
Types of defined, 482 Derivation of finite difference equations, Telescopic mesh refinement, 440-443
Embedding approach, 482-483 420-427 Application of, 442f, 443f
Heuristic methods, 483 Discretized hypothetical aquifer system, Definition of, 440
Hydraulic management models, 482, 484 422 Illustration of, 441 f
Optimal control approach, 483 Effective hydraulic conductivity MODTMR,440
Policy evaluation models, 482 equation, 422 Time discretization, 432
Policy evaluation and allocation Equivalent hydraulic conductance, 430 TMRDIFF, 440
models, 483, 492 Governing equation, 420 MODFLOW-GUI PIE, 460
Response matrix approach, 483 Hydraulic conductance, 425 MODFLOW-2000 (Layer property flow
Maramac Spring, 68, 69f Hydraulic conductance equations, package),436-440
Mass transport processes defined, 378, 430-431 Arithmetic-mean interblock transmissivity
380-382 Inactive cells, 427 method, 436
Advection, 380, 38lf No~flow cells, 427 Arithmetic-mean thickness and logarithmic-
Diffusion, 380 Point-centered grid system, 423 mean hydraulic conductivity method, 436
Dispersivity, 380 Quasi-three dimensional approach, 429, Dewatered conditions, 438-439
Lateral (transverse) dispersion, 380, 381f 432 Horizontal hydraulic conductance, 436-437
Longitudinal dispersion, 380 Three-dimensional cell, 424f Logarithmic-mean interblock transmissivity
Hydrodynamic dispersion, 380 Vertical discretion, 427f, 428f method, 436
Mechanical mixing, 380 Two-dimensional model, 415-420 Vertical hydraulic conductance, 437-438
Molecular diffusion, 380 Case study, 419-420 MODPATH, 449-450
Matric potential, 127 Darcy's law, 415-416 MODTMR,440
Metamorphic rock, 55 Finite difference equations, 417-419 Moisture profile, 134f
Metamorphic water, 35, 36f Governing equations, 415-417 Moisture zones during, 134f
Meteoric water, 35, 36f Iterative alternating direction implicit Saturated zone, 134f
Modeling support, 459-460 procedure, 419 Transmission zone, 134f
International Groundwater Modeling Center Verification of, 415 Wetting front, 134f
(IGWMC), 459 MODFLOW-2000, 431-443 Wetting zone, 134f
U.S. EPA Center for Exposure Assessment Additional capabilities, 432f Monitoring of,
Modeling (CEAM), 459 Example applications of, 450-455 Groundwater quality, 386-392
Center for Subsurface Modeling Support, External sources and stresses, 433-436 Assessment monitoring, 386
(CSMoS), 459 Areal recharge, 434 Corrective action monitoring, 386
U.S. Geological Survey, 459-460 Drains, 434-435 Detection monitoring, 386
MODFLOW, 459 Evapotranspiration, 435 Equipment setup, 392f
MODPATH, 459 General head boundary, 436 Monitoring plan outline, 388t
MOC3D,459 Rivers and streams, 433-434 Monitoring systems, 388f
I
t
Index 631
Sequence of activities, 387f Overdr;.tft, 470 Pick's law, 380

Performance monitoring, 386 Overland flow, Pick's second law, 382
Well cluster design, 391f In global hydrologic system, 15f Hydrodynamic dispersion, 380, 382
Well design for NWQA Program, 390f, Longitudinal coefficient of, 382
391£ Partially-penetrating wells, 195-197 Transport ofreactive pollutants, 385-386
Land subsidence, 319-323 Particle-size distribution graph, 41 Desorption, 385
Extensometer construction, 322f Particle tracldng analysis, 449-451, 450f, 45lf Distribution coefficient, 385
Extensometer instrumentation, 323f Definition of, 450 Freundlich isotherm, 385
Nested piezometers, 320f Flowpaths, 450f, 45lf Linear isotherm, 385
Mounds, see recharge mounds MODPATH, 449 Relative velocity, 386
Mudpot, 74 Peclet number, 384 Retardation, 385
Perched mound, 567, 568f Retardation equation, 385-386
National Water-Quality Assessment (NAWQA) Geometry of, 568f Sorption, 385
Program, 389 Height of, 568 Sorption isotherm, 385
Flowpath studies, 389 Perched water table, 57f Transport processes defined, 378, 380-382
Land-use studies, 389 Perennial yield, 469-473 Advection, 380, 381 f
Study-unit surveys, 389 Calculation of, 472 Diffusion, 380
Water quality constituents measured, 389t Evaluation of, 471--473 Dispersivity, 380
Natural groundwater chemistry, 330t, 333t, Factors governing, 471-472 Lateral (transverse) dispersion, 380, 381f
334t, 335t Maximum, 471 Longitudinal dispersion, 380
Neutron probe, 49 Variability of, 472--473 Hydrodynamic dispersion, 380
NITRINFIL, 568 Performance monitoring, 386 Mechanical mixing, 380
Nitrogen cycle, in SAT, 567f Permafrost, 76-77 Molecular diffusion, 380
Nonequilibrium equation, 163-164 Location below ground, 76f Streamlines, 381 f
Assumptions of, 164 Regions in Northern Hemisphere, 77f Pollution,
N-Springs Site, Hanford, Washington, 497-499 Permeability, see hydraulic conductivity, Attenuation of, 373-374
Areal distribution of 90 Sr, 499f 91-94 Chemical processes, 374
Finite-difference grid of, 500f Intrinsic permeability, 91 Dilution, 374
Plan view of, 498f Range of values, 94f Filtration, 373
Subgrid of study area, 500f Permeameter, 95-96 Microbiological decomposition, 374
NWISWeb, 30-31 Constant head, 95f Sorption, 373
Falling head, 95f Distribution of underground, 375-378
Oasis Valley, Nevada, 539-541 Philip's infiltration equation, 139t Changes is plumes, 376f
Oceanic islands, 603--605 Phreatic surface, 45 Plume, 375
Geologic profile, 605f Phreatic water, 45 Plume at a disposal site, 375
Freshwater lens, 604f Piezometers, Plume at a landfill, 376f
Oceanic water, 36f Nested piezometers, 320f Plume, 375
Oil pollution of groundwater, 366f Piezometric surface, 56f, 57, 58f Hanford Site, Richland, Washington, 377-389
Distribution of saline water from, 367f Plutonic water, 35 Location of, 377f
Optimal control approach, 482, 483, 495 Poiseulle's law, 90 Distribution of carbon tetrachloride, 379f
For Barton Springs-Edwards Aquifer, Policy evaluation and allocation models, Spread of tritium contamination, 378f
Texas, 496, 496f, 497f 492-495 Pollution of groundwater, 360--408
Finite-difference grid, 497f Plume, 373-378 Agricultural uses and effect on water
Water level contours, 496f Changes is plumes, 376f quality, 368f
For groundwater remediation design, Definition of, 373 By agricultural sources, 367-368
497-499 Plume, 375 Animal wastes, 367
For N-Springs Site, Hanford, Washington, Plume at a disposal site, 375 Fertilizers, 367
497-499 Plume at a landfill, 376f Irrigation return flows, 367
Areal distribution of 90Sr, 499f PMWIN,460 Pesticides, 368
Finite-difference grid of, 500f Pollutants, mass transport of, 378-385 Soil amendments, 367
Plan view of, 498f Advection-dispersion equation, 382-384 By industrial sources, 364-367
Subgrid of study area, 500f Analytical procedure, 384 Liquid wastes, 364
Optimization methods, 482 Peclet number, 384 Mining activities, 364
Constraints, 482 Diffusion, 380, 382 Oil field brines, 366, 366f
Feasible solution, 482 Apparent (effective) coefficient 382 Pipeline leakage, 364
Linear programming, 482 Molecular diffusion, 380, 382 Tankleakage,364
Nonlinear programming, 482 Dispersion, 380, 382 By municipal sources, 361-364
Objective function, 482 Mechanical dispersion, 382-383 Lj.quid wastes, 361
Optimization problem, 482 Dispersivity, 380, 382 Sewer leakage, 361
Simplex method, 482 Dynamic dispersivity, 382 Solid wastes, 362
632 Index
Pollution of groundwater (continued) Source control with mid-plume and Mound recession, 582-583
By other sources, 368-373 down-gradient pumping, 407 Recharge rates, 556-558
Cesspools, 370 Management strategies, using, 402f Clogging effects, 558
Interchange through wells, 372 Remediation strategies, 401-403 Recharge wells, 180
Roadway deicing, 370 Simulation of, 408f Recovery test, 170-172
Saline water intrusion, 371 Variation of pulsed pumping parameters, Recovery, 170
Septic tanks, 370, 37lf 409f Recovery draw down, 171 f
Spills, 370 Pumping equipment, 232-234 Residual drawdown, 170, 171f
Stockpiles, 370 Factors for selection, 232 Relative velocity, 386
Surface discharges, 370 Pumps, 232, 233t, 235t Remediation design, 497-499
Surface water, 372 Characteristics of, 233 For N-Springs Site, Hanford, Washington,
Urbanization, 368, 369f For deep wells, 232, 235f 497-499
Contaminant release scenarios, 360f For shallow wells, 232 Areal distribution of 90Sr, 499f
Dissolved contaminant release, 360f Total pumping head, 234 Finite-difference grid of, 500f
DNAPL release, 360f Pump tests, Plan view of, 498f
Inorganic releases, 360f For hydraulic conductivity determination, Subgrid of study area, 500f
LNAPL release, 360f 100 Remediation of contaminated groundwater,
Domestic uses and effect on water 392-401
quality, 362f Qanats, 1-2 Cleanup goals, 393,
Industrial uses and effects on water Quasi-three dimensional approach, 429,432 Advantages and disadvantages, 393t
quality, 365f Extraction systems, 396
Land application of municipal waste- Radiation logging, 533-535 DNAPL dual--extraction system, 397f
waters, 363f Garruna-gamma logging, 534-535 LNAPL dual-extraction system, 397f
Movement of pollutants from disposal Natural-gamma logging, 533, 534f Extraction trench systems, 397f
well, 364f Neutron logging, 535, 536f Hydraulic containment, 394
Relation to water use, 359-361 Real-time data, 31 Hydraulic barriers, 395
Ponding time, 140f Collection system, 31f, 32f Nonselective physical barriers, 395
Porosity, 37-40 Transmission, 31 f Selective physical barriers, 395
Effective porosity, 37, 137, 138t Recharge, see artificial recharge Types of physical barriers, 396t
Primary porosity, 37 Definition, 16 Remediation goals, 392
Representative values of, 38t Edwards Aquifer, 70f, 71t72f And approaches, 395t
Rock interstices, 38f Annual recharge to, 71 t Remediation techniques, 395t
Secondary porosity, 37 Discharge from, 73 Plume remediation, 395t
Types of, 53t Map of, 70f, 72 Source remediation, 395f
Fracture, 53t In global hydrologic system, 15f System design, 394
Intergranular, 53t Transient recharge, 16, 18f Requirements, 394
Intergranular and fracture, 53t Recharge area, 57, 56f Steps in design, 394
Potential infiltration rate, 134 Recharge methods, 548-558 Treatment of contaminated groundwater, 396
Potentiometric surface, 57 Basin method, 549-550, 550f, 551f Aquifer restoration techniques, 398
Precipitation, Disposal wells, 554 Bioremediation, 396
In global hydrologic system, 15f Ditch-and-furrow method, 552, 552f Contaminant migration prevention, 398
In hydrologic cycle, 14 Drain wells, 554 Ex-situ techniques, 396, 398t, 399t
Pressure aquifier, 56 Flooding method, 552-553 In-situ air sparging, 399f
Publications on groundwater, Incidental recharge, 556 In-situ techniques, 396, 398t, 399t
Books, 29 Irrigation method, 553 Pump-and-treat, 399f
Journals, 29 Pit method, 553 Source zone treatment technologies, 396
U.S. Geological Survey, 27-29 Recharge well method, 554-556, 554f, Technology selection and design
Pump-and-treatment systems, 399f, 401-407 555f, 556f, 557f flowchart, 401f
Capture zone analysis, 406 Stream-channel method, 550-552 Volatization processes, 396
Capture zone, 406f Water spreading, 548 Remediation techniques, table of, 395t
Characterizing sites for, 403-405 Recharge mounds, 567-583 Plume remediation, 395t
Goals for contaminant characterization Hantush equation, 570-576 Source remediation, 395f
data, 403 Perched mounds, 567-568 Remote sensing, 510-512
Phases of, 405f Steady-state equations for, 568-570 Aerial photographs, stereoscopic
Types of data used, 404f Stormwater infiltratin basin mounds, examination of, 510
Example system, 402f 576-583 Aerial photographs, surficial features, 511 t
Extraction/injection scheme design, 407 Potential flow models, 576-582 Infared imagery, 510
Down-gradient pumping, 407 Circular basin, 578-582 Radar imagery, 510
Source control with down gradient Dupuit-Forchheimer equation, 577 Representative elementary volume (REV),
pumping, 407 Mound growth, 580-582 41-42
Index 633
Reservoirs, Oxnard Plain, Ventura, California, 609, Unconfined formation, 261-270

Storage relations, 470f 610f, 611f Bouwer and Rice method, 262-268
Subsurface, Fresh-salt water interface, 592-600 Dagan method, 268-270
Advantages of, 465t Biscayne aquifer, Florida, 596f KGS model, 271-272
Disadvantages of, 465t Flow pattern, 593f, 596f Well skin effects, 273
Surface, Ghyben-Herzberg model, 593 Soil aquifer treatment (SAT) systems, 560-566
Advantages of, 465t Relative salinity, 597, 597f Components of, 561f, 561-562
Disadvantages of, 465t Shape of, 592-595 Nitrogen removal efficiency, 562, 563f
Residual drawdown, 170, 171f Structure of, 596-597 Pathogen removal efficiency, 562, 563f
Residual moisture content, 137 Ghyben-Herzberg relation, 590-592 Total organic removal efficiency, 562, 563f
Resistivity-concentration curves, 531 f Heads for nonequilibrium conditions, 592f Definition of, 560-561
Resistivity curves, 529 Karst terrains, 605-606 Dynamics of, 562
Resistivity logging, 529-532 Mechanisms responsible for, 589-590 Modeling of, 565-567
Electric log of a well, 529 Oceanic islands, 603-606 N1TRINFIL, 566
Electrode arrangements, 529f Geologic profile, 605f Nitrogen cycle, 565
Field-formation factor, 531 Freshwater lens, 604f Operation conditions, 566f, 567f
Hydrologic conditions and resistivity Occurrence of, 589-590 Operation of, 563-565
curves, 532f Sketch of in unconfined aquifer, 59lf Application time, 565
Resistivity-concentration curves, 531 f Saline water, sources of, 589 Cycle time, 565
Resistivity curves, 529 Upconing, 600-603 Draining period, 565
Spontaneous potential, 530f Diagram of, 600 Drying period, 565
Specific conductance, 531 Effect of pumping rate, 602f Operation processes, 564f
Temperature correction factors, 53lf Maximum permissible pumping rate, 601 Optimization of, 565
Resource Conservation and Recovery Act Scavenger wells, 602 Processes, 564f
(RCRA), 386 Well water salinity curves for, 601 f Purification processes, 563
Response matrix approach, 492-495 Saline water, sources of, 589 Adsorption/desorption, 563
Algebraic technological function, 492 Saltwater intrusion, see saline water intrusion Chemical precipitation/dissolution, 563
Santa Clara-Calleguas Basin, California, Sandstone, 55 Denitrification, 563
499-504 SAT, see soil aquifer treatment (SAT) systems Disinfection, 563
Groundwater model, 502 SATOM,566 Infiltration, 563
High chloride areas, 502f Saturated flow, 125 Ion exchange, 563
Map of, 501f Saturation zone, 1, 45, 50-52 Nitrification, 563
Model formulation, 503 Divisions of subsurface water, 47f Organic biodegradation, 563
Study area, 500-501 Saturation percentage of a sample, 39 Recovery and recharge systems, 562f
Water management facilities, 503f SCS infiltration method, 139t Removal mechanisms, 565
Retardation, 385 Seawater intrusion, 500 Adsorption-desorption, 565
Retardation equation, 385-386 See saline water intrusion Biodegradation, 565
Reuse, 559-560 Seepage,67 Soil classification, 40-41
Reynold's number, 90 Seepage areas, 67 Based on particle size, 41t
Richard's equation, 128 Seepage face, 148 Particle-size distribution graph, 41
Rock interstices, 38f Semiconfined aquifer, 57 Uniformity coefficient, 40
Rotary drill bits, 220f Sewage effluent, 560 Uniform material, 40
Carbide button bit, 220f Seismic refraction method, 517-520 Well-graded material, 40
Cone-type rock bit, 220f For depth to water table, 519f Soil moisture, 14, 125
Fishtail bit, 220f Seismic velocity of geologic materials, 517f In hydrologic cycle, 14
Simulation models, see Models for simulation Soil texture, 41-42
Safe yield, 469 Sinkhole formation, 317 Soil texture triangle, 42f
Saline water intrusion, 589-614 Slug tests, 250-273 Soil water diffusivity, 128
Control of, 606 Conditions for, 254f Soil water retention,
Artificial recharge, 606 Confined formation, 255-262 Relationships for, 130t
Extraction barrier, 606, 607f Cooper, Bredehoeft, Papadopoulos Soil water zone, 46f
Injection barrier, 607, 607f method, 255-259 Capillary water, 46
Methods for, 606t Hvorslev method, 259-261 Gravitational water, 46
Pumping pattern modification, 606 Definition of, 250 Hygroscopic water, 46
Subsurface barrier, 607 Design guidelines for, 250-251 Solute transport equation, 455-456
Effect of wells on, 597-600 High conductivity formations, 272-273 Assumptions of, 455
Strack relationship, 597 Methods for analyzing, 253-272 Solute transport modeling, 455-459
Examples of, 608-610 Performance guidelines, 252-253 Laurel Bay, Florida application of
Long Island, New York, 608 Performance of, 251-253 MOC3D,458f
Los Angeles, California, 608-609 Schematic of, 251f, 252f Simulation of benzene results, 458f
634 Index
Solute transport modeling (continued) Storage coefficient, 58-59 Geologic log, 524f
Simulation of MTBE results, 458f Definition of, 58 Well cuttings, 525f
MOC3D, 456--459 Illustrative sketches of, 59f Water level measurement, 526-527
Application illustration, 457f Storavity, 58 Air-line method, 526
Finite difference grid for, 456f Stormwater infiltratin basin mounds, 576-583 Electric water-level sounder, 526
Retardation factor, 455 Potential flow model, 576-582 Rock technique, 526
Solute transport equation, 455 Circular basin, 578-580 Suction force, 127
Assumptions of, 455 Dupuit-Forchheimer equation, 577 Suction gradient, 127
Sorption, 385 Mound growth, 580-582 Suction head, 127
Sorption isotherm, 385 Mound recession, 582-583 Surface investigations, 511-522
Source zone treatment technologies, 396 Streamflow, 284-290 Electrical resistivity method, 513-517
Specific capacity, 196, 248 Bank storage, 285-287 Electrical circuit for, 514f
Relation to storage coefficient, 249f As function of time, 287f Electrode arrangements for, 515f
Relation to transmissivity, 249f Baseflow, 287, 288f Horizontal profile, 516f
Well efficiency, 248-249 Variation of, 289f Interpretation of, 515f
Specific discharge, 88 Gaining stream, 284f, 285f Representative ranges of, 513f
Specific retention, 50 Hyporheic zone, 286f Resistivity equations, 513-514
Specific surface, 42 Losing streams, 284f, 285f Schlumberger arrangement, 514, 515f
Specific water capacity, 127 Recession curve, 289 Wenner arrangement, 514, 515f
Specific yield, 50, 125 Rising water, 284 Geologic methods, 509-510
Function of time of drainage, 125f Subsurface barrier, 608 Geophysical exploration, 512
Measurement of, 51 Subsurface flow, 14, 15, 125 Gravity method, 520
Representative values of, 5lt In global hydrologic system, 15f Magnetic method, 520
Spontaneous potential logging, 533 In hydrologic cycle, 14 Remote sensing, 510-512
Electrochemical potentials, 533 Subsurface investigations, 523-546 Aerial photographs, stereoscopic
Spontaneous potentials equation, 533 Acoustic logging, 538-539 examination of, 510
Springs,67-73,513t Caliper logging, 536, 536f Aerial photographs, surficial features, Silt
Classification of, 68 Hole caliper, 536f Infared imagery, 510
Comal Springs, 70, 72f Case study, Oasis Valley, Nevada, 539-541 Radar imagery, 510
Definition of, 67 Geophysical logging, 527-528 Seismic refraction method, 517-520
Fissure springs, 67 Definition of, 527 For depth to water table, 519f
Gravity springs, 67 Schematic diagram of, 528f Seismic velocity of geologic
Artesian springs, 67, 511 t Summary of applications, 527t materials, 517f
Contact springs, 67, 511 t Fluid-conductivity logging, 537 Surface runoff,
Depression springs, 67, 511 t Fluid conductivity log, 537f In global hydrologic system, 15f
Fracture springs, 67 Fluid-velocity logging, 537 In hydrologic cycle, 14
Impervious springs, 67 Example of, 537f Subsurface water
Tubular springs, 67 Radiation logging, 533-535 Divisions of, 47f
Thermal springs, 73-75 Gamma-gamma logging, 534-535
Fumerole, 74-75 Natural-gamma logging, 533-534, 534f Telescopic mesh refinement, 440-443
Geyser, 73 Neutron logging, 535, 536f Application of, 442f, 443f
Hot springs, 73 Resistivity logging, 529-532 Definition of, 440
Mudpot, 74 Electric log of a well, 529 Illustration of, 441f
Warm springs, 73 Electrode arrangements, 529f MODTMR,440
Volcanic springs, 67 Field-formation factor, 531 Temperature logging, 535
Springflow, 13 Hydrologic conditions and resistivity Test drilling, 523-525
In hydrologic cycle, 13 curves, 532f Drilling-time log, 525f
Steady flow to a well, 152-161 Resistivity-concentration curves, 531f Geologic log, 524f
Confined aquifer, 152-156 Resistivity curves, 529 Well cuttings, 525f
Equilibrium equation, 153 Spontaneous potential, 530f Theis equation, 163-164
Thiem equation, 153 Specific conductance, 531 Assumptions of, 164
Unconfined aquifer, 156-159 Temperature correction factors, 531 f Theis method, 164-167
Unconfined flow with uniform recharge, Spontaneous potential logging, 533 Theis type curve, 173
159-161 Electrochemical potentials, 533 Thermal springs, 73-75
Steady unidirectional flow, 146-152 Spontaneous potentials equation, 533 Fumerole, 74-75
Base flow to a stream, 149-152 Television logging, 538 Geyser, 73
Confined aquifer flow, 146-147 Example of, 538f Hot springs, 73
Dupuit equation, 148 Temperature logging, 535 Mudpot, 74
Unconfined aquifer flow, 147-149 Test drilling, 523-525 Warm springs, 73
Step-drawdown pumping test, 243 Drilling-time log, 525f Thiem equation, 153
Index 635
Total volume of voids, 39 Cooper-Jacob method, 167-169 Metamorphic water, 35, 36f
Tracer, 119-122 Nonequilibrium equation, 163-164 Meteoric water, 35, 36f
Distribution of, 121f Theis equation, 163 Oceanic water, 36f
Groundwater tracer, 122 Theis method, 164-167 Plutonic water, 35
Lateral dispersion of, 121 f Recovery test, 170--171 Volcanic water, 35, 36f
Longitudinal dispersion of, 120f Unconfined aquifer, 172-177 Water law and policy, 477~79
Tracer tests, 98 Delayed yield, 172 Appropriation law, 477
For hydraulic conductivity determination, 98 Delay index, 173 Arizona's groundwater management code,
Transmissivity, 92 Upconing, 600--603 477~78
Transpiration, Diagram of, 600f Goals of, 478
In global hydrologic system, 15f Effect of pumping rate, 602f Doctrine of prior appropriation, 477
In hydrologic cycle, 14 Maximum permissible pumping rate, 601 Groundwater law, 477
Transport of reactive pollutants, 385-386 Scavenger wells, 602 Surface water law, 477
Desorption, 385 Well water salinity curves for, 601 f Water level measurement, 526-527
Distribution coefficient, 385 Urbanization effects on groundwater, 306f Air-line method, 526
Freundlich isotherm, 385 U.S. EPA Center for Exposure Assessment Electric "Yater-level sounder, 526
Linear isotherm, 385 Modeling (CEAM), 459 Rock technique, 526
Relative velocity, 386 Center for Subsurface Modeling Support, Water quality, see groundwater quality
Retardation, 385 (CSMoS), 459 Water quality criteria, 346-348
Retardation equation, 385-386 U.S. Geological Survey Drinldng water standards, 346
Sorption, 385 Internet locations, 27 Industrial water criteria, 346
Sorption isotherm, 385 Computer models, Irrigation water criteria, 347
Travel time, MODFLOW, 431 Laws regulating, 346t
Of groundwater, 16 MODFLOW-GUI PIE, 460 Water spreading, 550
Treatment of contaminated groundwater, 396 MODPATH, 449 Water table, 16-19,45, 56f, 57f, 58f
Aquifer restoration techniques, 398 MOC3D,455 Configuration of, 18f
Bioremediation, 396 SUTRA,459 Definition, 16
Contaminant migration prevention, 398 ZONEBUDGET, 459 In hydrologic cycle, 14f
In-situ air sparging, 399f Publications, 27-29 Near hyporheic zone, 19f
In-situ techniques, 396, 398t, 399t On-line reports, 28t-29t Perched, 57f
Ex-situ techniques, 396, 398t, 399t Types of, 28t Phreatic surface, 45
Pump-and-treat, 399f Water use, 5-13
Source zone treatment technologies, 396 Vadose water, 45, 47f Freshwater source, use and disposition, 1Of
Technology selection and design flow- VIEWLOG, 460 Groundwater withdrawals by water-use, 13t
chart, 401f Visual MODFLOW, 460 Offstream water use in U.S., 6t
Volatization processes, 396 Void ratio, 39 Offstream water use in U.S. by State, 11t-12t
Tubular springs, 67 Volcanic rock, 55 Trends in U.S., 8t
Type curve, 164 Volcanic springs, 67 Water withdrawals, 5-9
Volcanic water, 35, 36f By water region in U.S., 7f
Unconfined aquifer, 56 Volatization processes, 396 Groundwater withdrawals by water-use, 13t
Steady radial flow in, 156-159 Volumetric water content, 39 Offstrearn water use in U.S. by State, llt-12t
Steady unidirectional flow in, 147-149 Trends in U.S., 8f, 9f
Unsteady radial flow in, 172-189 Warm springs, 73 WATSTORE, 30
Unconfined well function, 174 Wastewater recharge, 561-562 Well completion, 222
Uniform flow, 161-162 Irrigation method, 562 Casings, 222-223
Unifonnity coefficient, 40, 227t Overland flow method, 562 Recommended diameters, 223t
Unifonn material, 40 Recharge well method, 562 Cementing, 223
Unsaturated flow, 125-133 Spreading method, 562 Hydraulic jetting, 230
Control volume equation for continuity, 126 Water-bearing formation, 36 Screens, 223-230
Richard's equation, 128 Water budget, see hydrologic budget Well construction, 206-242
Unsaturated hydraulic conductivity, 127-133 Water capacity, 128 In unconsolidated formations, 215f
Hysteresis effect, 131 Water content, 49 Methods for, 206-214
Relationships for, 130t Measurement of, 49 Augers, 210f
Brooks and Corey, 130t Gravimetric method, 49 Bored wells, 209-211
Campbell, 130t Neutron probe, 49 Construction methods, 207t
Van Genuchten, 130t Tensiometer, 49f Driven wells, 211-212, 211f
Unsaturated zone, 1 Water, genetics types of Dug wells, 208, 209f
Unsteady radial flow in, Connate water, 35, 36f Hollow-stem auger drilling, 21 Of
Confined aquifer, 163-172 Juvenile water, 35, 36f Jetted wells, 212-213, 213f
Chow method, 169-170 Magmatic water, 35, 36f Jetting drill bits, 213
636 Index
Well construction (continued) Leaky aquifers, 177-180, Incrustation, 238

Methods for drilling deep wells, 214-222 Unconfined aquifers, 172-177 Sanitary protection, 234, 236f, 237f
Air rotary method, 220--221 Well flow for special conditions, 198 Well rehabilitation, 237-238
Cable tool method, 218-219 Well flow in unifonn flow, 161-162 Methods for, 239t
Performance, 214t Well flow near aquifer boundaries, 180-182 Well skin effect, 273
Reverse-circulation rotary method, 222 Boundary locations, 191-192 Well testing, for yield, 232
Rotary drill rig, 218f Near impenneable boundaries, 186-189 Wetness gradient, 127
Rotary method, 220 Near other boundaries, 189-191 Wetting front, 133, 134f, 135
Rotary-percussion method, 221 Near streams, 180-185 Wetting zone, 134f
Test holes for, 206 Well installation, Wilting point, 50
Well point system, 212f Collector wells, 240f
Well efficiency, 248-250 Domestic wells, 238f Yield, 469-473
Well flow, see well hydraulics Drilled wells, 238f Deferred perennial yield, 470
Well functions, unsteady radial flow, 164, 165t, Grout seal, 237f Maximum perennial yield, 471
174, 175f, 178f, 179f Infiltration gallery, 240f Mining yield, 469
Confined conditions, l65t, 167f Sanitary protection, 236f Perennial yield, 469
Leaky conditions, l78f, 179f Well logs, 206 Calculation of, 472
Unconfined conditions, 174, 175f Well loss, 242-247 Evaluation of, 471-473
Well-graded material, 40 Definition of, 242 Factors governing, 471-472
Well hydraulics, 152-198 Evaluation of, 243-247 Maximum, 471
Multiple well systems, 192-195 Relation to drawdown, 243f Variability of, 472-473
Partially penetrating wells, 195-197 Relation to well discharge, 243f
Steady radial flow, 152-161 Step-drawdown pumping test, 243 Zone of aeration, see aeration zone
Confined aquifers, 152-159 Well-point system, 212f Zone of saturation, see saturation zone
Unconfined aquifers, 159-161 Well protection, 234-237
Unsteady radial flow, 163-180 Abandonment, 237
Confined aquifers, 163-172 Frost protection, 236-237

Groundwater Hydrology: Third Edition

Uploaded by

Copyright:

Available Formats

Groundwater Hydrology: Third Edition

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

Groundwater Hydrology: Third Edition

Uploaded by

Copyright:

Available Formats

Groundwater Hydrology

David Keith Todd

ISBN 0-471-05937-4 (cloth)

Printed in the United States of America

Intended Audience and Introduction

New to this edition

Student and Instructor Resources

2.6.5 Igneous and Metamorphic Rocks 55

3 Groundwater Movement 86 4.5

3.10.3 Vertical and Horizontal Flows 133

4 Groundwater and Well Hydraulics 146

5 Water Wells 206

6 Groundwater Levels and Environmental Influences 279

6.3.1 Evaporation Effects 291

7 Quality of Groundwater 329

8 Pollution of Groundwater 359

8.2.2 Liquid Wastes 361

9 Groundwater Flow Modeling Techniques 413

9.4.2 Finite Difference Equations 417

10 Management of Groundwater 464

10.7 .2 Optimization Methods 482

11 Surface Investigations of Groundwater 509

12 Subsurface Investigations of Groundwater 523

13 Artificial Recharge of Groundwater 547

14 Saline Water Intrusion in Aquifers 589

applications. Therefore, groundwater in petroleum reservoirs will not be treated specifically in

1.2 HISTORICAL BACKGROUND

1.2.2 Groundwater Theories

' Superscript numbers refer to references at the end of the chapter.

1.2.3 Recent Centuries

Per Withdrawals (Mgal/d) (includes irrigation conveyance losses) Consump- n

.~ ~~- ···- -" ·.:;:"...

"<\.. f. ·· J·ALASKA "' ~

SOURCE USE DISPOSITION

Table 1.4.1 Total Offstream Water Use by State, 1995

Missouri 5,324 1,320

Per Withdrawals (Mgal/d) (includes irrigation conveyance losses) Consump- n

_ _..._ _lliiillll_ _ _ _ _ _ _ _ _ _ _......_..._ _ _ _ _ _ _ _ _ _............._ ....................__"'""-..........-............

New England 335 168 64 47 6.4 53 0 2.9 0 48 725 0

1.5 GROUNDWATER IN THE HYDROLOGIC CYCLE

~< ¢= ~< 0;1'

A hydrologic system is defined as a structure or volume in space, surrounded by a bound- Fi

Surface Runoff to streams

und- Figure 1.5.2. Block-diagram representation of the global hydrologic system. 22

Conditions before recharge event- Some groundwater

Initial effects of recharge event- Water table mounds near

later effects of recharge event- Eevated water table mimics

Conditions after effects of recharge event- Groundwater

Direction of groundwater flow- Ught arrows are associated with shallow,

Boundary of major surface-water

1.6 HYDROLOGIC BUDGET

Figure 1.6.1. Components of hydrologic

Surface Water System Hydrologic Budget

Groundwater Sysum Hydrologic Budget

Sysum Hydrologic Budget

Oi•charge Figure 1.6.2. Diagrams illustrating water

;;;: -2- 43 +53 - 1 + 173

OVERALLPREDEVELOPMENT GROU NO-WATER PREDEVELOPMENT

The components of hydrologic budgets are either measured, calculated, or estimated.

__ sz Piezometric surface