Title Author(s)

1
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Electrical Imaging for Hydrogeology

i
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Authors

ii
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

All rights reserved. This publication is protected by copyright. No part of this book
may be reproduced in any form or by any means without permission in writing from the
authors (to request permission contact [email protected]). Commercial
distribution and reproduction are strictly prohibited.
Groundwater Project (GW-Project) works are copyrighted and can be downloaded
for free from gw-project.org. Anyone may use and share gw-project.org links to
download GW-Project’s works. It is neither permissible to make GW-Project documents
available on other websites nor to send copies of the documents directly to others. Kindly
honor this source of free knowledge that benefits you and all those who want to learn about
groundwater.
Copyright
Copyright © 2022 Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and
Lee D. Slater (The Authors)

Published by the Groundwater Project, Guelph, Ontario, Canada, 2022.

Electrical imaging for hydrogeology / Kamini Singha, Timothy C. Johnson,

Frederick D. Day-Lewis, and Lee D. Slater.
76 pages.
ISBN 978-1-77470-011-2 (PDF).

Please consider signing up for the GW-Project mailing list to stay informed about new book
releases, events, and ways to participate in the GW-Project. When you sign up for our email
list, it helps us build a global groundwater community. Sign up.

APA (7th ed.) Citation:

Singha, K., Johnson, T. C., Day-Lewis, F. D., & Slater, L. D. (2022). Electrical imaging for
hydrogeology. The Groundwater Project. doi.org/10.21083/978-1-77470-011-2.

Domain Editors: Eileen Poeter and John Cherry.

Board: John Cherry, Paul Hsieh, Ineke Kalwij, Stephen Moran, Everton de Oliveira, and
Eileen Poeter.
Steering Committee: John Cherry, Allan Freeze, Paul Hsieh, Ineke Kalwij, Douglas Mackay,
Everton de Oliveira, Beth Parker, Eileen Poeter, Ying Fan, Warren Wood, and Yan
Zheng.
Cover Image: Example of time-lapse inversions showing the changes in bulk conductivity
caused by the presence of amendment. After injection, the amendment experienced
density-driven downward flow and lateral flow over the lower bounding unit
contact at approximately 10 m depth. The progression from day 93 to day 371 shows
the plume sinking and diluting.

iii
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Table of Contents
ELECTRICAL IMAGING FOR HYDROGEOLOGY ............................................................................................... I
AUTHORS .................................................................................................................................................... II
COPYRIGHT ................................................................................................................................................ III
TABLE OF CONTENTS................................................................................................................................. IV
THE GROUNDWATER PROJECT FOREWORD .............................................................................................. VI
FOREWORD ............................................................................................................................................. VII
PREFACE ................................................................................................................................................. VIII
ACKNOWLEDGEMENTS ..............................................................................................................................IX
1 INTRODUCTION .................................................................................................................................. 1
1.1 MEASUREMENT PHYSICS: THE RELATION BETWEEN DATA (VOLTAGE DIFFERENCES) AND PARAMETERS (ELECTRICAL
CONDUCTIVITY OR CHARGEABILITY) ....................................................................................................................... 3
1.2 ELECTRICAL IMAGING HARDWARE AND FIELD DEPLOYMENTS ......................................................................... 9
2 DESIGNING SURVEYS ........................................................................................................................ 12
2.1 GEOMETRIC FACTORS .......................................................................................................................... 15
2.2 SYNTHETIC MODELS ............................................................................................................................ 17
3 COLLECTION AND VERIFICATION OF FIELD DATA .............................................................................. 20
3.1 CONTACT RESISTANCE .......................................................................................................................... 20
3.2 STACKED MEASUREMENTS .................................................................................................................... 21
3.3 RECIPROCAL MEASUREMENTS ............................................................................................................... 21
3.4 ERROR CONSIDERATIONS FOR TIME-LAPSE MEASUREMENTS ....................................................................... 23
3.5 PULSE DURATION ................................................................................................................................ 24
3.6 NOTES ON FIELD CONDITIONS................................................................................................................ 24
4 DATA INVERSION .............................................................................................................................. 26
4.1 THE GOAL OF INVERSION ...................................................................................................................... 26
4.2 REGULARIZATION IN ELECTRICAL IMAGING INVERSION ................................................................................ 28
4.3 SELECTION OF INVERSION PARAMETERS TO PREVENT OVERFITTING/UNDERFITTING OF DATA............................. 30
4.4 DEFINITION OF DATA MISFIT ................................................................................................................. 34
4.5 QUANTIFICATION OF INVERSION QUALITY ................................................................................................ 34
4.6 CHECKS ON INVERSION RESULTS............................................................................................................. 35
5 CASE STUDIES ................................................................................................................................... 38
5.1 2-D WATERBORNE RESISTIVITY AND INDUCED POLARIZATION PROFILING....................................................... 38
Background ............................................................................................................................................. 38
Data Collection ........................................................................................................................................ 38
Data Processing ...................................................................................................................................... 39
Data Interpretation ................................................................................................................................. 39
5.2 4D RESISTIVITY OF A BIOSTIMULATION EXPERIMENT .................................................................................. 40
ER Monitoring System ............................................................................................................................. 42
Data Collection and Experimental Design ............................................................................................... 43
Data Quality Control and Assurance ....................................................................................................... 43
Effect of Data Weighting and Regularization Weighting on Pre-Injection Inversions ............................ 45
Time-Lapse Inversions ............................................................................................................................. 50
6 OVERVIEW AND FUTURE DIRECTIONS .............................................................................................. 52
7 EXERCISES......................................................................................................................................... 54

iv
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

EXERCISE 1 ..................................................................................................................................................... 54
EXERCISE 2 ..................................................................................................................................................... 54
EXERCISE 3 ..................................................................................................................................................... 54
EXERCISE 4 ..................................................................................................................................................... 54
EXERCISE 5 ..................................................................................................................................................... 54
EXERCISE 6 ..................................................................................................................................................... 55
EXERCISE 7 ..................................................................................................................................................... 55
EXERCISE 8 ..................................................................................................................................................... 55
8 REFERENCES ..................................................................................................................................... 56
9 BOXES............................................................................................................................................... 62
BOX 1 SCENARIO EVALUATOR FOR ELECTRICAL RESISTIVITY (SEER) ........................................................................... 62
10 EXERCISE SOLUTIONS ....................................................................................................................... 66
SOLUTION EXERCISE 1 ....................................................................................................................................... 66
SOLUTION EXERCISE 2 ....................................................................................................................................... 67
SOLUTION EXERCISE 3 ....................................................................................................................................... 68
SOLUTION EXERCISE 4 ....................................................................................................................................... 69
SOLUTION EXERCISE 5 ....................................................................................................................................... 70
SOLUTION EXERCISE 6 ....................................................................................................................................... 71
SOLUTION EXERCISE 7 ....................................................................................................................................... 72
SOLUTION EXERCISE 8 ....................................................................................................................................... 72
11 ABOUT THE AUTHORS ...................................................................................................................... 73
MODIFICATIONS TO ORIGINAL RELEASE .................................................................................................... A

v
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

The Groundwater Project Foreword

The United Nations theme for World Water Day on March 22, 2022, is
“Groundwater: making the invisible visible.” This aligns with the essence of the
Groundwater Project (GW-Project), which is aimed at raising groundwater consciousness
and strengthening groundwater expertise worldwide, and is being accomplished by
publishing books and supporting materials about “all-things-groundwater”.
The GW-Project, a non-profit organization registered in Canada in 2019, is
committed to contribute to advancement in education and brings a new approach to the
creation and dissemination of knowledge for understanding and problem solving. The
GW-Project operates the website https://gw-project.org as a global platform for the
democratization of groundwater knowledge and is founded on the principle that:
“Knowledge should be free and the best knowledge should be free knowledge.” Anonymous
The mission of the GW-Project is to provide accessible, engaging, high-quality,
educational materials, free-of-charge online in many languages, to all who want to learn
about groundwater and understand how groundwater relates to and sustains ecological
systems and humanity. This is a new type of global educational endeavor in that it is based
on volunteerism of professionals from different disciplines and includes academics,
consultants and retirees. The GW-Project involves many hundreds of volunteers associated
with more than 200 organizations from over 14 countries and six continents, with growing
participation.
The GW-Project, which began publishing books in August 2020, is an ongoing
endeavor and will continue with hundreds of books being published online over the
coming years, first in English and then in other languages, for downloading wherever the
Internet is available. The GW-Project publications also include supporting materials such
as videos, lectures, laboratory demonstrations, and learning tools in addition to providing,
or linking to, public domain software for various groundwater applications supporting the
educational process.
The GW-Project is a living entity, so subsequent editions of the books will be
published from time to time. Users are invited to propose revisions.
We thank you for being part of the GW-Project community. We hope to hear from
you about your experience with using the books and related materials. We welcome ideas
and volunteers!

The GW-Project Steering Committee

November 2021

vi
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Foreword
When considered relative to the study of the atmosphere, rivers, lakes and oceans,
the study of groundwater suffers from the disadvantage that groundwater is not visible.
Boreholes are required to obtain direct measurements of groundwater systems, but
boreholes are costly and disrupt subsurface conditions. Electrical imaging does not require
drilling and provides insight about groundwater systems. This book describes that
powerful geophysical method known as electrical imaging. It is founded on Ohm’s Law in
much the same way that understanding of groundwater flow is based on Darcy’s Law.
Darcy’s Law concerns the resistance to water flow through permeable media whereas
Ohm's Law involves the resistance of current flow through geological media. The aim of
electrical imaging is to scan the subsurface by applying an electric current to the ground
and monitoring voltage at many locations. Most often electrical imaging is employed to
learn about the geology of a study area, although it can also be used to estimate depth to
water or to find zones of saline water or oily industrial liquids. Learning about geology is
essential to understanding a groundwater system. Drilled holes reveal the geology at each
drill location but interpolation of the geology between holes is fraught with uncertainty.
Electrical imaging can reduce the number of boreholes needed to characterize the
subsurface or guide where boreholes can be most informative. Electrical imaging can also
be used to monitor progress when liquids are injected into contaminated aquifers to remove
or destroy the contaminants.
The ability of electrical imaging to provide valuable insight has increased markedly
over the past few decades as technology for measurement of electrical responses and
computing power for analyzing the responses have advanced with decreasing cost. Still,
conversion of the data into meaningful information, requires careful design and analysis as
well as consideration of multiple working hypotheses of subsurface conditions to best
explain the electrical signals. This book enhances the reader’s ability to use electrical
imaging in understanding groundwater systems.
The authors of this book: Kamini Singha, a professor of hydrogeologic science at
Colorado School of Mines; Timothy Johnson, a research geophysicist with Pacific
Northwest National Laboratory; Frederick Day-Lewis, a chief geophysicist with Pacific
Northwest National Laboratory, and Lee Slater, a professor of geophysics at Rutgers
University, have many decades of cumulative experience with research, teaching and
practice of electrical imaging. Herein, they explain, demonstrate and document best
practices for collecting and analyzing electrical imaging data. They provide guidelines for
those who decide to use this powerful geophysical method to reveal essential
hydrogeologic information that is not feasibly obtained by other means.
John Cherry, The Groundwater Project Leader
Guelph, Ontario, Canada, November 2021

vii
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Preface
Geophysical methods offer hydrogeologists unprecedented access to
understanding subsurface parameters and processes. In this book, we outline the theory
and application of electrical imaging methods, which inject current into the ground and
measure the resultant potentials. These data are sensitive to rock type, grain size, porosity,
pore fluid electrical conductivity, saturation, and temperature. Here we describe the
physical basis for electrical imaging, parallels between electrical flow equations and the
groundwater flow equation, practical considerations for field investigations, data
processing and inverse modeling of field data, and how to QA/QC (Quality
Assurance/Quality Control) data. We additionally cover two case studies, including a 2-D
waterborne survey and a 4-D dataset from a biostimulation experiment.

viii
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Acknowledgements
We deeply appreciate the thorough and useful reviews of and contributions to this
book by the following individuals:

❖ Burke Minsley, U.S. Geological Survey, Denver, Colorado, USA;

❖ Frederic Nguyen, Université de Liège, Liège, Belgium;
❖ Brady Flinchum, Clemson University, Clemson, South Carolina, USA;
❖ Damien Jougnot, Sorbonne Université, Paris, France;
❖ Eileen Poeter, Colorado School of Mines, Golden, Colorado, USA; and,
❖ Hugh Whiteley, University of Guelph, Guelph, Ontario, Canada.

We are grateful for Amanda Sills, Nithya Mani, Juliana Apolonio and Connie
Bryson of the Groundwater Project for their oversight, formatting and copyediting of this
book. We thank Eileen Poeter (Colorado School of Mines, Golden, Colorado, USA) for
editing and producing of this book.

ix
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

1 Introduction
Electrical resistivity (ER, sometimes called ERI for electrical resistivity imaging or
ERT for electrical resistivity tomography) is a direct-current (or low-frequency
alternating-current) geophysical method that can be used to estimate the spatial and, in
some applications, temporal distribution of subsurface bulk electrical resistivity, which
describes the intrinsic resistance to electric current flow in geologic media. Bulk electrical
resistivity, or its reciprocal, bulk electrical conductivity, is related to rock type, grain size,
porosity, pore fluid electrical conductivity, saturation, and temperature; these relations
underlie the utility of ER for cost-effective civil engineering and environmental studies,
including imaging of lithology, differences in water saturation below ground surface,
permafrost distribution, location of clays, and groundwater fluid conductivity, among
other properties and processes as outlined in this book.
Although water in its pure state is non-conductive, the presence of dissolved salts
in solution produces a conductive electrolyte to which ER methods are sensitive (e.g.,
Zohdy et al., 1974); hence, these techniques can be used to monitor multiple hydrogeologic
processes such as infiltration, migration of ionic tracers or chemical amendments, and
groundwater/surface water interactions. ER offers important benefits for hydrogeological
studies: (1) many features, such as clay layers, variable moisture content, high salinity,
low-porosity areas, and others, manifest as detectable electrical conductivity contrasts and
vary in space; (2) instrumentation is relatively inexpensive, robust, and easy to operate; (3)
instrumentation is mature and available commercially; and (4) ER measurements are
amenable to automation, allowing for long-term, continuous, cost-effective monitoring.
Currently, while there are standards in terms of array types (Wenner,
Schlumberger, dipole-dipole, as outlined in this book), there exist no community-accepted
standards for ER survey design (i.e., which measurements are collected), quality assurance
and quality control (QA/QC), or data analysis. In this book we review ER technology, the
physics underlying ER measurements, and modeling and inversion approaches used in
common ER software packages. We outline guidelines to ensure (1) design of robust survey
geometries; (2) selection of appropriate acquisition and inversion parameters; and (3)
documentation of data-collection configurations, QA/QC, and analysis procedures.
These practices are then demonstrated using a case study from the Defense
Reutilization and Marketing Office (DRMO) Superfund site, in Brandywine, Maryland,
USA (Johnson et al., 2014). Our objective is to document the best state-of-the-practice for an
audience of hydrogeology students and practitioners while providing sufficient details on
the mechanics of the method to relate the strengths and limitations of the data acquired.
We refer more expert readers who are interested in advanced approaches to Johnson and
others (2010), Singha and others (2014), or Binley and Slater (2020), all of which provide

1
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

valuable reviews of theory and applications of electrical imaging methods for a variety of
systems.
Early ER field measurements relied on labor-intensive methods to build up
information on the vertical (one dimensional, i.e., 1-D) variation of bulk electrical
conductivity with depth or along a profile. The concept of modern electrical imaging was
first described by Lytle and Dines (1978) and the first field demonstrations emerged in the
1990s (e.g., Griffiths et al., 1990). Over the last three decades, advances in ER hardware have
resulted in multi-channel systems capable of controlling hundreds of electrodes and
acquiring thousands of measurements per hour. During this same period, advances in
software and computing power have led to the proliferation of user-friendly programs for
ER inverse modeling (e.g., Cockett et al., 2015; Rucker et al., 2017; Blanchy et al., 2020) or,
more simply, inversion, which refers to the mathematical process of estimating unknown
subsurface parameter values from measured data. Inversion of three-dimensional (3-D)
datasets through time, often called 4-D, is now becoming commonplace. Binley and Slater
(2020) provide a recent review of ER methods that may be of interest to students moving
beyond this text.
ER imaging suffers from several limitations that include: (1) the need for direct
contact with the subsurface, which is problematic in areas with resistive surficial materials
such as highways or permafrost (the exception is capacitively coupled systems for surface
measurements, which do not require emplaced electrodes, but require resistive surficial
materials and are not discussed here); (2) significant labor for electrode array deployment,
particularly for long (many hundreds of meters) or 3-D arrays; (3) data collection can be
slow and can limit monitoring of rapid dynamic processes depending on instrumentation
and the number of electrodes; and (4) substantial user knowledge is required for processing
of data, despite commercially available code (some options are listed in Section 4.1), if
quantitative, rather than qualitative, interpretation of hydrogeologic processes is required.
Perhaps the biggest disadvantage is that, just like other geophysical techniques, we
are dealing with proxies of what we want to actually measure. Bulk electrical conductivity
has multiple dependencies that can complicate interpretation for a specific parameter or
process. In addition, choices of regularization parameters and weighting the importance or
accuracy of measurements may affect the magnitude and smoothness of ER estimates,
further complicating quantitative conversion of electrical proxies to estimates of other
physical properties (e.g., Day-Lewis et al., 2005). Collecting ER data through time, called
time-lapse imaging as described in more detail throughout this book, is one way to alleviate
this problem.
We also review an increasingly popular extension of ER in groundwater studies
known as induced polarization (IP). This method measures transient voltages that result from
temporary, reversible storage of electric current in the Earth—similar to the storage of water
in aquifer systems as defined by storativity in the time-varying groundwater flow

2
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

equation—whereas ER is defined by a steady-state flow equation, as outlined below. The

charge storage measured with IP primarily results from the electrical double layer forming
at the mineral-fluid interface or at pore constrictions that locally reduce ionic mobility. IP
measurements are consequently strongly sensitive to grain size, surface area, and pore size.
When correctly acquired, the IP dataset provides valuable additional information that helps
to constrain the hydrogeological interpretation of the subsurface relative to that possible
from ER alone. This has important implications, including improved estimation of
hydraulic conductivity from geophysical proxies (Slater, 2007). Most ER instruments allow
an IP measurement to be made and most software for the interpretation of ER datasets
supports the processing of IP measurements (if acquired) along with ER. Though IP is not
commonly included in hydrogeologic studies because the time to collect IP data in the field
is longer and processing and interpretation of ER and IP measurements combined requires
additional expertise and understanding over that needed to use ER datasets effectively, as
discussed in this book. Together, ER and IP comprise methods generally known as electrical
imaging, which is the topic of this book.

1.1 Measurement Physics: The Relation between Data (Voltage

Differences) and Parameters (Electrical Conductivity or Chargeability)
ER data acquisition systems drive an electric current that can range from milliamps
(mA) to several amps into the subsurface through galvanic contact. Current is injected
between two electrodes, a positive and a negative one (the current electrodes), and the
resultant electric potential (specifically, the voltage difference) is measured between two or
more additional electrodes (the potential electrodes, see Figure 1). The physics is
mathematically analogous to a two-well pumping test, where water is injected in one well
and removed from another, and the resultant head difference at steady-state would be
measured between two other locations. This four-electrode injection/measurement
procedure is repeated for as many combinations of current and potential electrode
positions as desired, and usually involves the acquisition of many hundreds or thousands
of multi-electrode combinations.

3
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 1 - Example of ER data collection. Multiple electrodes are installed along the
ground surface (or in boreholes), and two electrodes at a time are used to drive current
into the subsurface. The resulting voltage difference is measured between two or more
potential electrodes. Flow and equipotential lines are analogous to those estimated
from the groundwater flow equation, where current and fluid flux are mathematically
parallel, as are voltages and heads, which are both potentials.

The measurements presented by ER instrumentation are voltage differences in volts

(equivalent to differences in hydraulic heads in the analogy to groundwater flow), which
are divided by the applied current to calculate resistance values (from Ohm’s law, where
R = ∆V/I) in ohm (Ω), sometimes referred to as transfer resistance. The physical
measurement of resistance is not an intrinsic property of a material (in our case rock or soil)
because it depends on the geometry of, or distance between, electrodes used for
measurement. A well-known example is the resistance of a length of wire. The resistance
across 100 m of homogeneous copper wire would be 100 times the resistance across 1 m of
the same wire. The wire has an intrinsic resistivity (in this case, the resistivity would be
very low or the conductivity very high as copper is an excellent conductor) that limits
current flow as evident from an observed voltage drop. Thus, in ER we measure a resistance
that must be converted to a relevant intrinsic property, electrical resistivity, ρ (or its
reciprocal, electrical conductivity, σ), from knowledge of the electrode locations. In the flow
analogy, hydraulic conductivity is treated as an intrinsic property, whereas transmissivity
is a function of geometry. In SI units, resistivity has units of ohm-meter (Ω-m) and
conductivity has units of Siemens/meter (S/m). Each measured resistance is a function of
the electrode locations, as noted above, and the electrical properties of both solids and
liquids in the system.
The physics underlying ER measurements are described by the Poisson equation
(Equation 1), which is a simplification of the Maxwell equations. This is also the equation
that describes steady-state groundwater flow.

𝛻 ∙ 𝜎(𝑥, 𝑦, 𝑧)𝛻𝑉(𝑥, 𝑦, 𝑧) = −𝐼δ(𝑥 − 𝑥𝑠 , 𝑦 − 𝑦𝑠 , 𝑧 − 𝑧𝑠 ) (1)

4
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

subject to boundary conditions, where:

𝛻 = gradient operator
𝛻 ∙ = divergence operator
σ = electrical conductivity, an intrinsic property of the material (Siemens/meter)
V = electric potential (Volts), where V can be used to determine the voltage
differences between two potential electrodes for a given current injection
I = electric current source magnitude, otherwise known as the current injected
in the field (Amperes)
δ = Dirac delta function
x, y, z = spatial position vectors (meters)
xs, ys, zs = spatial coordinates of the current source (meters)
As with the groundwater flow equation, Equation 1 can be simplified to an equation
for two-dimensional (2-D) systems, and/or to an equation where the electrical conductivity
is considered to be homogeneous and isotropic (see Section 7 of the Groundwater Project
book “Hydrogeologic Properties of Earth Materials and Principles of Groundwater
Flow”). 3-D data acquisition and inversion methods are increasingly practical and
appropriate, although many practitioners still use 2-D inversion. Commonly, commercially
available software for 2-D inversion invokes what is known as a 2.5-D assumption for
computational efficiency. Under this assumption, inverse modeling is performed for a 2-D
parameterization (e.g., to identify the best-fit cross section), while forward modeling of the
electrical measurements is performed in 3-D. The 2-D structure is assumed to extend
infinitely into the third dimension (e.g., Dey and Morrison, 1979; LaBrecque et al., 1996; see
Section 0.2 for more details). The 2.5-D approximation thus combines a 2-D
parameterization with 3-D physics. Conceptually this is similar to classical pumping test
analysis, where hydrogeologists combine a 1-D parameterization (layers) and 2-D
(axisymmetric) flow.
Equation 1 combines conservation of charge and Ohm's law. Ohm’s law, shown in
continuous form by Equation 2, is the constitutive relation analogous to Darcy’s law,
linking electrical (as opposed to hydraulic) potential gradients and fluxes.

𝑗 = −𝜎𝛻𝑉 (2)

In geophysics, this link is assumed to be linear. Here, j is the electric current density
(Amperes/m2) in the ground in response to the external current source ( I) and is analogous
to the specific discharge in Darcy’s Law. In adopting Equation 1, we make some important
assumptions about our measurement. Equation 1 assumes equilibrium or steady-state
electrical conditions and includes neither transient (i.e., induced polarization) effects nor
current sources other than what is injected through the current electrodes (e.g., no
spontaneous potentials), which are assumed to act at a single point (i.e., xs, ys, zs). The
5
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

validity of these assumptions depends in part on instrument settings, as explained in

Section 0, below. Equations 1 and 2 describe the link between the subsurface potential
distribution (V), which determines the voltage differences between two potential electrodes
(∆V), the current injection magnitude and location, and the electrical conductivity (σ).
In some software packages, the data are specified in terms of resistance (∆V/I)
where both I and ∆V are reported by the instrumentation. Other packages expect data in
the form of apparent resistivity, ρa, which is the resistivity that the subsurface would have if
it were homogeneous and isotropic and is calculated as shown in Equation 3.

ρa = Kg (ΔV/I) (3)

Kg is a geometric factor (in meters) accounting for electrode configuration. In our earlier
example of the copper wire, this geometric factor is simply the cross-sectional area of the
wire divided by its length. Apparent resistivity is sometimes preferred over resistance
because it scales the data to have the same units and magnitude as the intrinsic property
being estimated (electrical resistivity), and thus it is more intuitive. Note that the resistance
measurements can be both positive and negative, as geometric factors can be positive or
negative. It is important to note that intrinsic subsurface electrical conductivity cannot be
negative, and neither are the magnitudes of injected currents. However, the sign of the
measured potential difference is purely dependent on which electrode we use as our
reference electrode, and thus negative voltages (and resistances) can be recorded. Although
the apparent resistivity is typically positive, negative apparent resistivities are also
possible. A measurement that would be positive under homogeneous subsurface
conditions may be negative under certain heterogeneous subsurface conditions (e.g.,
Wilkinson et al., 2008; Jung et al., 2009). For this reason, it is crucial to collect signed voltage
differences in the field, rather than the absolute value of the voltage difference between
electrodes. The use of apparent resistivity can be helpful in assessing measurement errors
when compared to examining resistance values, given that the apparent resistivity values
are of similar magnitude to one another. In the field, geophysicists used to plot
pseudosections of apparent resistivity, which assign the volumetric measurements to a point
location in x-z space based on the measurement-electrode locations (e.g., Hallof, 1957);
these plots are still generated by many inversion programs. In general, plotting
measurements prior to inversion is important for visualizing trends that may be indicative
of certain subsurface objects or to identify obvious errors as in the case of malfunctioning
electrodes.
Induced polarization measurements record the effect of temporary charge storage
on the electric field. One way to observe storage effects is to capture the transient voltage
decay that occurs on abrupt termination of the applied electric current when the Earth
stores charge. This time-domain IP effect is quantified as an integral of Vm, the voltage decay

6
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

curve, from t1 to t2 divided by the primary or total voltage (VT) as shown by Equation 4 and
in Figure 2a.
𝑡
1 ∫𝑡 2 𝑉𝑚
𝑀𝑎 = 1 (4)
(𝑡2 − 𝑡1 ) 𝑉𝑇
Ma is the apparent chargeability. Similar to apparent resistivity, the apparent chargeability
depends on the electrode locations and is different from the intrinsic chargeability of the
subsurface, which describes the polarization strength of a geologic material. It is worth
noting also that sampling period is not standard between different instruments and may
affect measurements; thus, instrument settings must be consistent to allow for meaningful
comparisons of measurements between surveys. A number of conventions have been
proposed, e.g., integration over one log cycle or a specified time window (Sumner, 1976).
Apparent and intrinsic chargeability are both unitless although typically expressed as
mV/V.

Figure 2 - Different ways to measure induced polarization: a) a time-domain measurement, where voltage
decay is recorded following abrupt current termination, and b) a frequency-domain measurement, where
magnitude and phase of a sinusoidal voltage with a period T (related to frequency ω recorded between two
electrodes lags behind the current recorded across a reference resistor placed in series with the Earth by a
time ∆t. Note that in ER measurements, voltages are only measured at the plateau of the injected current
(i.e., VT), not during the decay.

The intrinsic chargeability or phase of the Earth must be positive. However, the
measured apparent chargeability or measured apparent phase over a heterogenous Earth
can be positive or negative, depending on the location of chargeable objects relative to the
sensitivity pattern of the electrode array, described in more detail below. Any array will
have some regions of negative sensitivity, where in ER, an increase in subsurface resistivity
will counterintuitively be observed as a decrease in measured resistance. When a highly
chargeable object is located within this region of negative sensitivity, a negative apparent
chargeability can be recorded with IP (Dahlin and Loke, 2015). When expressed with
respect to a complex impedance or complex resistivity, the phase should normally be
negative, being consistent with Figure 2 where the voltage lags in time behind the current
7
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

waveform. However, the measured apparent phase recorded over a heterogeneous Earth
can sometimes be positive (Luo and Zhang, 1998; Wang et al., 2020) as a result of the
sensitivity patterns of an array. A practitioner may be tempted to discard negative apparent
chargeabilities or positive (for complex impedance or resistivity) apparent phase as data
errors. Error checks, described in Section 3.3, can help to differentiate between errors and
negative apparent chargeabilities that inform on the subsurface structure.
The physics of induced polarization can be incorporated into Equation 1 by
representing the conductivity and potential gradient terms as complex numbers, as shown
in Equation 5.

𝛻 ∙ 𝜎 ∗ (𝑥, 𝑦, 𝑧)𝛻𝑉 ∗ (𝑥, 𝑦, 𝑧) = −𝐼δ(𝑥 − 𝑥𝑠 , 𝑦 − 𝑦𝑠 , 𝑧 − 𝑧𝑠 ) (5)

Here, σ* is known as the complex conductivity. The real and imaginary components of the
complex conductivity separate out the electrical conduction and polarization properties of
the subsurface. In the mathematical analogy to groundwater flow, chargeability is related
to parameters controlling storage (e.g., specific storage or storativity), and Equation 5
resembles the transient groundwater flow equation with a single complex-valued
parameter, σ*, where the real part relates to resistance (or its reciprocal, conductance) and
the imaginary part to reactance (or its reciprocal, susceptance). Indeed, the electrical
analogy for the transient problem was the basis for simulating non-equilibrium
groundwater flow using resistor-capacitor networks prior to the advent of digital
computing (Freeze and Cherry, 1979).
Frequency-domain IP considers measurements in terms of the frequency of
waveforms (Figure 2b), which are made by using a sine-wave current source, and
measuring the magnitude and phase (Ø) of the complex resistance (ΔV/I)* or of the complex
apparent resistivity ρa*. The phase refers to the phase shift between the injected current and
the measured voltage and is the frequency-domain measure of the IP effect. In the absence
of current storage (either in non-polarizing materials or because we do not measure the
time-varying piece), Ø = 0 and Equation 5 simplifies to Equation 1. Frequency-domain
measurements are popular in the laboratory, and some instruments exist to perform
field-scale frequency-domain acquisition. However, it is often simpler to measure the field
IP effect using time-domain IP (Figure 2a). The time-domain and frequency-domain IP
effects are theoretically equivalent, and one can be determined from the other through a
Fourier transformation (the Fourier transform of a time series is a complex valued function
of frequency).
Regardless of whether IP data are being collected, the goal of data collection is to
create a cross section or volume distribution of subsurface electrical conductivity in x-y (or
x-y-z) space, which requires the process of inversion, described in detail in Section 0.
Inversion software solves the forward model problem using Equation 1 or 5, which takes
assumed model parameters (electrical conductivity or resistivity and chargeability or

8
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

phase) and produces model predictions that can be compared with observed data—
resistances or apparent resistivity and apparent chargeability or apparent phase. ER
inversion is commonly done using finite-difference or finite-element techniques for solving
partial differential equations, where Equation 1 (or Equation 5 in the case of IP) is solved at
spatially distributed discrete locations corresponding to the centers of finite-difference grid
cells or to nodes in a finite-element mesh, with the accuracy of the solution depending on
the level of discretization, as outlined in Section 2.2. Finite-difference codes are not
frequently seen for surface data collection due to complexities with topography. Depending
on the survey geometry (i.e., number and placement of electrodes), inversion can produce
1-, 2- or 3-D tomograms, reconstructed images that show the estimated subsurface
distribution of electrical resistivity or conductivity. The term tomography refers to the image
reconstruction process using ER measurements and is described in Section 0.

1.2 Electrical Imaging Hardware and Field Deployments

To inject electric current and collect the requisite resistance measurements in the
field (Figure 3), a series of electrodes are attached to an ER control unit, which consists of a
regulated current source (for example, a 12-V deep-cycle battery or a generator), voltage
meter, current meter, and multiplexers (units for switching between electrodes) for
multi-channel data collection. Despite the advent of multi-node cables (i.e., cables with
connection points, called takeouts, for many tens of electrodes, as shown in Figure 3),
considerable labor is still required for the initial deployment of cables. However, once the
instrumentation deployment is complete, it can be left in place for many hours to years to
explore changes in electrical conductivity through time associated with processes, which is
a reason these methods show strength for time-lapse monitoring. Commercially available
multi-channel electrical imaging systems commonly accept input files listing the selected
four-electrode current-potential pairs, called quadripoles. These files are variously referred
to as sequence, command, or schedule files. The control unit then drives current on two
electrodes and measures potential differences between two or more electrodes based on the
input sequence file.

9
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 3 - Electrical imaging instrumentation,

including the control unit in the foreground, electrodes,
and cables that connect the control unit to the
electrodes. The batteries and multiplexers are inside of
this control unit. From Binley and Slater (2020).

Commercial systems on the market today generally are constant-voltage systems,

i.e., they establish a constant voltage difference over the current electrodes and measure the
resulting current. Thus, the amount of current driven is influenced by the conductivity of
the Earth, where lower input currents may be needed in resistive settings where potential
differences are larger, and higher input current is needed in conductive settings. Typical
currents are usually in the mA to A range. Subsequent measured voltages are usually in the
μV to V range. Electrodes are commonly composed of stainless steel or graphite. Steel
electrodes tend to be more durable, whereas graphite electrodes resist corrosion and thus
are preferred in marine deployments. In applications where spontaneous potential (SP) data
(based on naturally occurring voltages in the earth) are collected in addition to ER,
non-polarizing electrodes (e.g., copper/copper-sulfate) are necessary for making voltage
measurements. For a review on various electrode types, see Morris and others (2004) and
LaBrecque and Daily (2008). SP is beyond the scope of this book, but it is worth noting that
SP data are complementary to ER and IP data, can help to reduce uncertainty in
interpretation of hydrogeologic systems, and can be used to interpret flow directions in
some applications.

10
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

IP data acquisition uses the same hardware as ER, but additional considerations are
needed to acquire reliable measurements. Any hydrogeologist considering the use of IP
should be aware that it is substantially more challenging than ER surveying alone, and
more time consuming to collect in the field. This is because the signal-to-noise ratio of the
IP measurements (Ø, Ma) is typically 2.5 to 3 orders of magnitude smaller than the
magnitude of the resistance recorded with ER, and accurate transmitter-receiver
synchronization is needed. In addition, electromagnetic and capacitive coupling between the
different wires used to connect the current injection and voltage recording electrodes is
manifest as spurious charge storage effects that may corrupt the response from the earth.
Field procedures have been developed to alleviate these concerns, including separating the
wires that connect to the voltage-receiving pairs from those that connect to the
current-injection pairs (e.g., Dahlin and Leroux, 2012).
Regardless of the type of electrode used or whether ER or ER and IP data are being
collected, it is important to record locations of the electrodes accurately in the field, as well
as electrode and transect elevations, which will be incorporated into the inversion
procedure to get the correct topography for the upper boundary (see Section 3.6). The
accuracy required for surveying and georeferencing of electrodes is highly dependent on
the survey design, with greater accuracy required for smaller electrode spacing or in areas
of more topographic relief.

11
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

2 Designing Surveys
ER and IP data can be collected on the earth surface—including in water bodies, in
boreholes, or both. The volume of subsurface sampled, sometimes called the depth or
distance of penetration of the current, depends on the (unknown) electrical conductivity
structure of the subsurface and the spacing of the electrodes and is, therefore, difficult to
quantify or predict prior to data collection and analysis (e.g., Daily and Ramirez, 1995),
which partially accounts for variable practices of data collection. Many surface studies
successfully image electrical conductivity to depths of a few tens of meters below ground
surface. The depth to which a particular survey can image effectively depends on the
survey geometry, measurement errors, and the subsurface conductivity structure. The
depth of penetration can be interpreted from sensitivity or resolution maps (e.g., Figure 4),
or Oldenburg and Li (1999) provided an inversion-based approach (see more on inversion
in Section 0) to predicting what they instead termed the depth of investigation (DOI). For DOI
calculations, two inversions are performed using two reference models which differ by
orders of magnitude, and the resulting images are compared. Depending on the type of
regularization used (see Section 0.2 for definitions and details), the images are either
differenced or cross-correlated to determine the depth to which the inversion is strongly
affected by the reference models, i.e., below this DOI the data provide negligible
information. This approach is supported in some inversion software. Simpler approaches
to predicting DOI (e.g., Barker, 1989) are based on calculating the measured signal versus
depth for a homogeneous half space and identifying the depth corresponding to the
maximum, mean, or median signal contribution. These approaches produce simple rules
of thumb for various array types and provide practical guidance for survey design. For
example, for the popular dipole-dipole array, the median depth of investigation is on the
order of 1/5 the maximum electrode spacing in an array (Roy and Apparao, 1971). The
important messages here are 1) that some parts of the tomogram will be better resolved
than others; 2) there are multiple tools to assess where resolution is expected to be high,
and these methods are not absolute measures of accuracy; and 3) there is limit to how far
from the electrodes ER and IP can see, which is impossible to determine in advance as it is
dependent on the electrical conductivity of the earth and the geometry used to collect data,
described below.

12
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 4 - Cumulative squared sensitivity maps (a proxy for resolution) for surface a) and crosswell b) ERT
arrays. These maps are the sum of squared sensitivity (the diagonal of J*J’) where J is the Jacobian
matrix and J’ is its transpose. The Jacobian matrix is a matrix of first-order partial derivatives that shows
the sensitivity of the model parameters to the data (more details in Section 4.2). c) The absolute
sensitivity for a single measurement (i.e., a single row of J); cool colors are negative sensitivity and
warm colors are positive (on c only).

Sensitivity is generally highest near the electrodes (Figure 4), whether the electrodes
are on the surface or in boreholes. Practitioners are faced with a tradeoff between resolution
and spatial coverage. Resolution improves with smaller electrode spacing, but smaller
spacing (or well offsets) for a fixed number of electrodes reduces the volume of the
subsurface studied. Consequently, when designing a survey, it is important to keep in mind
the depth and size of targets. In field surveys and in the presence of heterogeneity, the
volume of earth sampled by a particular resistance measurement is unknown—not unlike
estimating the volume of earth sampled by a pumping test—and conversion from
resistance to electrical conductivity requires inverse modeling. Information on inversion
and image reconstruction is outlined in Section 0.
Historically, ER and IP data were collected on the surface using a fixed set of
electrode geometries where the two current and two potential electrodes were moved by
hand. However, such fieldwork was highly labor intensive. Modern systems are almost
always automated using tens to hundreds of electrodes in an array. Some standard array
types are often used in the field, such as Wenner, dipole-dipole, or Schlumberger arrays

13
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

(Figure 5). Some geometries (e.g., Wenner) are favored for their sensitivity to vertical
contrasts in electrical conductivity, whereas other geometries (e.g., dipole-dipole) are
favored for sensitivity to lateral changes in electrical conductivity. IP measurements benefit
from arrays where the voltage pair are nested between the current pair, as with the Wenner
and Schlumberger arrays, because of their high signal-to-noise ratio, although this comes
at the expense of additional electromagnetic coupling effects relative to non-nested arrays
such as dipole-dipole. By restricting data collection to simple geometries, analytic methods
could be used to estimate the subsurface electrical conductivity without numerical
modeling and inversion (for example, Zohdy et al., 1974). While selection of an ideal
geometry has been the subject of past research, the ability to resolve subsurface structure is
dependent not only on the geometry used, but on the electrical-conductivity structure of
the subsurface, which is unknown. Optimized sets of measurements based on arbitrary
array geometries can now be designed based on considerations of the expected subsurface
structure (e.g., Stummer et al., 2004). Modern inversion software is capable of processing
ER and IP data in minutes on a low-end PC and does not require that the electrode
arrangement corresponds to any of the traditional, standard array types.

Figure 5 - Some common surface electrode geometries for ER and IP, including Wenner, dipole-dipole, and
Schlumberger arrays, which have different positioning of current (A,B) and potential (M,N) electrodes, as
defined by spacings a and b and n, an integer.

For n electrodes, the number of fully independent 4-electrode measurements (i.e.,

quadripoles) is n(n-3)/2 (Xu and Noel, 1993). Collecting all possible combinations of
measurements is often unrealistic in the field given the memory constraints of ER meters
and the time required to collect the data. Depending on the speed of data acquisition (i.e.,
instrument capabilities and measurement times) and whether time-lapse data are required
(which may constrain the time allowed for measurements), an appropriate number of
quadripoles can be selected. The choice of which quadripoles to collect in the field can be
determined by two criteria: (1) geometric factors and (2) signal-to-noise ratios (dependent
on the site). Larger electrode spacings will require larger source voltages to get sufficient
current into the ground to ensure good data quality. Injected current is limited by the
equipment and the resistivity of the ground. With respect to the speed of data collection,
new multi-channel instrumentation, capable of multiple voltage measurements at once for a
given current pair—one voltage measurement at a time per channel—leads to faster data
collection than older single-channel tools, although this functionality may only be
applicable for certain array types depending on the instrument. To fully capitalize on
14
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

multi-channel data acquisition, surveys can be designed to minimize the number of unique
current injections collected during a given survey.
In borehole surveys, selected quadripoles would ideally combine in-well and
cross-well dipoles, i.e., with current pair in one well and potential pair in a second, as well
as with current and (or) potential pairs split between wells. In-well dipoles are sensitive to
targets located near boreholes, but do not provide much information farther from
boreholes. Cross-well dipoles are more sensitive to targets located farther from wells. To
collect quality cross-well data, the boreholes should be approximately at least 1.5 times as
deep as they are far apart. For much larger offsets, resolution between boreholes becomes
highly degraded.

2.1 Geometric Factors

We introduced the concept of the geometric factor (Equation 3) as the parameter
that converts a measured resistance to apparent resistivity. For surface arrays, the
underlying math to calculate the geometric factor is fairly simple. Assuming a
homogeneous and isotropic half space (meaning the same electrical conductivity in the earth
to infinite distance below a surface boundary) without any electrical sources, the geometric
factor Kg for every quadripole can be calculated for surface arrays using Equation 6a.
2𝜋
𝐾𝑔 = (6a)
1 1 1 1
̅̅̅̅̅ − ̅̅̅̅
𝐴𝑀 𝐴𝑁 𝐵𝑀
− ̅̅̅̅̅ + ̅̅̅̅
𝐵𝑁
̅̅̅̅̅ ̅̅̅̅̅ 𝐵𝑀,
𝐴𝑀, 𝐴𝑁, 𝐵𝑁 are the distances between electrodes A and M, A and N, B and M, and
̅̅̅̅̅ and ̅̅̅̅
B and N, respectively. Current electrodes are defined as A and B and potential electrodes
are M and N. Current electrodes are defined as A and B and potential electrodes are M and
N (Figure 5). These electrodes are often also called C+, C-, P+, and P-, respectively, in other
literature, however the older A, B, M, N standard is used in this book. The geometric factor
accounts for the arrangement of electrodes and allows one to calculate an apparent
resistivity (Equation 3).
For borehole geometries, the electrodes are located within the half-space rather than
at the boundary at the Earth’s surface. In this case, use of Equation 6a is inappropriate, as
it does not account for the no-flow boundary at Earth’s surface. To account for the effect of
the boundary on cross-well measurements, the method of images from optics is invoked. This
approach is analogous to the use of image wells in groundwater hydrology for analytical
modeling of aquifer response to pumping. Here, imaginary image current electrodes are
introduced on the other side of the boundary, equidistant from the real current electrodes,
to mathematically produce a no-flow condition at the Earth’s surface and calculate a
geometric factor for borehole arrays as shown in Equation 6b.
4𝜋
𝐾𝑔 =
1 1 1 1 1 1 1 1 (6b)
̅̅̅̅̅ + ̅̅̅̅̅̅̅̅̅̅̅̅
𝐴𝑀 𝐴𝑖𝑚𝑎𝑔𝑒 𝑀 ̅̅̅̅
− −
𝐴𝑁 ̅̅̅̅̅̅̅̅̅̅̅
− ̅̅̅̅̅ − ̅̅̅̅̅̅̅̅̅̅̅ + ̅̅̅̅ + ̅̅̅̅̅̅̅̅̅̅̅
𝐴𝑖𝑚𝑎𝑔𝑒 𝑁 𝐵𝑀 𝐵𝑖𝑚𝑎𝑔𝑒 𝑀 𝐵𝑁 𝐵𝑖𝑚𝑎𝑔𝑒 𝑁

15
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Here, “image” indicates the image current electrode. When the electrodes are all on the
boundary, Equation 6b simplifies to Equation 6a. Also, limited burial is necessary before
Equation 6b simplifies to twice the result of Equation 6a, as the distances from the potential
electrodes to the true current electrodes and their images are approximately equal.
Quadripoles with large geometric factors may produce small voltage differences,
which are prone to measurement errors due to a lower signal-to-noise ratio. These are
manifest (via propagation of errors) as higher relative errors in apparent resistivity data. A
critical geometric-factor cutoff can be determined based on the average expected electrical
conductivity of the subsurface and the instrument specifications. Based on Equation 3, for
a given geometric factor and expected instrument error (in terms of voltage, inserted as the
potential difference), we can calculate the expected error in apparent resistivity. Figure 6
illustrates how error in measured potential difference translates into error in calculated
apparent resistivity as a function of Kg. In this example, we consider an applied current of
50 mA and assume a 1-microvolt (μV) instrument accuracy (note the logarithmic scale). In
practice, accuracy may be less. As evident in Figure 6, for large geometric factors or small
assumed apparent resistivity, errors are larger relative to measurements.

Figure 6 - Apparent resistivity error, as a percentage of the true apparent

resistivity, as a function of geometric factor for three different values of
resistivity (50, 500, and 5,000 Ω-m), assuming the voltage accuracy is
1 microvolt and the applied current is 50 milliamperes, where apparent
resistivity error is given by Kg*Verror/I, which is then compared to the
assumed apparent resistivity in a relative sense. An error in the measured
voltage translates into error in calculated apparent resistivity as a function
of Kg. In practice, larger errors may occur in the field.

16
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

2.2 Synthetic Models

Forward modeling is a powerful tool in the design and simulation of electrical
imaging surveys. Forward modeling codes are commonly based on finite-difference or
finite-element solutions of the Poisson equation (Equation 1 for ER only and Equation 5 for
ER and IP). Forward-modeling codes can simulate “synthetic data” for different spatial
distributions of electrical conductivity (and phase or intrinsic chargeability for IP), survey
geometries, and random noise levels for a given discretization of Equation 1 or 5
(Figure 7a). Common software packages allow users to set the size of finite-difference cells
or finite elements. These predicted data can be inverted to generate tomograms (Figure 7b
and Section 0). Such exercises provide insight into the resolving power of different survey
geometries, noise levels, and inversion approaches.

Figure 7 – a) Hypothetical electrical resistivity cross section, and b) resulting tomogram. Because
of the limited resolution of the survey and regularization in the inversion routine, the tomogram is a
blurred, blunted version of reality.

Forward-model accuracy is thus another factor to consider in designing survey

geometries. We should eliminate quadripoles that are expected to have large modeling
errors from our measurement sequence. We can quantitatively evaluate the forward-model
accuracy by modeling the apparent resistivity for all candidate quadripoles assuming a
homogeneous Earth (i.e., a single conductivity) and comparing them to the apparent
resistivities based on the geometric factors calculated analytically above. Quadripoles that
do not model well (i.e., the apparent resistivity calculated numerically differs substantially

17
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

from that computed analytically) should not be collected in the field, or the discretization
of the mesh and the location of the boundary conditions should be refined if problems exist
or too many data are eliminated in this manner. This exercise can only be performed in the
simple case where an analytical solution to Equation 1 exists. For example, in the presence
of a heterogenous subsurface or topography, analytical solutions are not generally
available, and assessment of numerical model accuracy is cumbersome for thousands of
quadripoles. Hence test criteria are based on simple analytical models, assumptions of
homogeneity, or criteria based on experience. As a rule of thumb, grid spacing near
electrodes where potential gradients are large should be finer than one quarter of the
electrode spacing. Grid spacing can be coarser further from electrodes where voltage
gradients are smaller. Commonly, grid spacing is increased by a factor of less than 1.5 from
one grid row or column to the next, deeper or neighboring row in finite-difference models.
In finite-element models, unstructured meshes are typically refined about the electrodes
with similar discretization.
In the example of Figure 7, a cross-well ER survey is conducted for a cross section
containing a single 25-cm fracture zone and no other heterogeneity. Assuming a low-noise
dataset, 2 percent random, normally distributed errors are added to the data to introduce
noise as might be expected in the field. The resulting tomogram provides only a blurry and
blunted image of the true electrical conductivity distribution, and interpretation of the
location and extent of the fracture zone is complicated by the limited resolution. If another
heterogeneity existed in the cross section (for example, lithologic or porosity variation), or
if the fracture zone was a small discrete fracture (perhaps 2.5 mm instead of 25 cm), it might
not be possible to identify the fracture at all. By considering different input models, or
targets, it is also possible to gain insight into how resolution varies spatially over a
tomogram. Indeed, conducting such synthetic modeling exercises prior to field surveys
represents a best practice. Many ER modeling and inversion software packages can be used
for this purpose. Open-source and free solutions for testing scenarios include:
• the United States Geological Survey spreadsheet-based Scenario Evaluator for
Electrical Resistivity (SEER), which approximates the inversion of ER data and
is particularly aimed at the non-expert (see Figure 8 and also Terry et al., 2017);
and,
• ResIPy (Blanchy et al., 2020), an intuitive graphical user interface for the family
of ER and IP inversion codes written by Andrew Binley (Lancaster University).

18
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 8 – Synthetic modeling workflow. The steps are: 1) assign best-guess physical properties for the
hypothetical subsurface model; 2) forward model, i.e., calculate the data that would result from the assumed
‘true’ model entered by the user in the first step and corrupt the data with random errors for realism,
generating ‘synthetic data’; 3) analyze the synthetic data by inverse modeling to produce an image, or
tomogram; and 4) compare the inverted synthetic image with the assumed true model. If the synthetic image
does not sufficiently resolve the target sought, i.e., a light non-aqueous phase liquid plume in this schematic,
field implementation of the method will likely fail and should be discouraged. After Day-Lewis and others
(2017) and Terry and others (2017).

Box 1 of this book provides an opportunity for readers to experiment with the
electrical resistivity modeling software SEER and explore the impact of some survey design
parameters on the resulting electrical image.

19
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

3 Collection and Verification of Field Data

In addition to quadripoles collected, it is highly advisable to collect (1) electrode
contact resistance measurements, (2) stacked or repeated measurements, and (3) reciprocal
measurements. As described below, these measurements allow the user to evaluate the
quality of the data both in the field and later during analysis back in the office. Accurate
assessment of measurement errors is often critical for obtaining an optimal image from the
inversion procedure.

3.1 Contact Resistance

Contact resistance refers to the resistance to current flow between the current
electrodes and the surrounding medium (soil, water, rock). Recorded contact resistances
are the sum of the resistances of the contacts between each current electrode and the ground
and the resistance of the formation between the current electrodes. Resistance checks
should be run on the electrodes prior to data collection to assure that contact resistances are
not too large. This test is the key quality control in the field. Cutoffs on the order of tens of
kiloohm (kΩ) may help determine where limited current will be injected for an electrode
pair. It is difficult to come up with absolute numbers for this metric because contact
resistance measurements vary with geology. For example, crystalline rock may have high
contact resistance measurements due to the formation resistivity, but quality data are still
possible to collect. Contact resistances should be recorded manually if not automatically
recorded by the instrument software. Contact resistance measurements commonly can be
made automatically with modern ER instrumentation, which applies a small voltage to the
current injection electrodes, measures the resulting current with the ER instrumentation,
and computes the resistance by dividing the applied voltage by the injected current.
Alternatively, manual contact resistance measurements for a given electrode pair can be
made with a voltmeter by measuring the resistance between corresponding pins in the head
of the electrode cable. Contact resistance values can provide a basis for editing data
associated with particular electrodes that exhibit poor contact with the formation, allowing
corrections to be made prior to a survey. Low-contact resistances are critically important
for the collection of reliable IP datasets because the signal-to-noise ratio of IP measurements
is typically 2.5 to 3 orders of magnitude smaller than resistance measurements. In fact,
contact resistance is often the limiting factor preventing acquisition of meaningful IP data
(Zarif et al., 2017), as is also true for ER. For more advanced reading on contact resistances,
we point readers to the 2013 paper by Hördt and others.
In surface arrays, it is possible to add a small amount of saltwater around electrodes
to improve contact resistance, but this generally is not possible for cross-well arrays. Note
that the introduction of saltwater would be a poor idea if one were interested in monitoring
salinity or saturation changes, in which case a metallic anti-seize paste or ultrasound gel
may also work to increase the electrical contact of the electrodes with the ground around

20
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

them. If the soil drains too quickly to add water, these materials might also be helpful, or
electrodes can be placed in bentonite or saturated sponges. Below the water table, borehole
electrodes are generally in good contact with the formation as a function of the presence of
water. In the vadose zone, electrode surface areas may need to be larger to provide good
coupling with the subsurface. Contact resistance generally decreases notably as the size of
the electrode increases. However, care must be taken not to violate the point-source
approximation made by most processing codes—where electrodes are assumed to be
infinitesimally small points in numerical modeling codes. The actual sizes and shapes of
large electrodes may require explicit representation in the numerical model used by the
inversion software. A common rule is that the size of the electrode should not exceed
10 percent of the distance between electrodes for the point-source assumption to be
approximately valid (e.g., Rücker and Günther, 2011).

3.2 Stacked Measurements

It is common practice for most commercial systems to collect each quadripole
several times in a row immediately one after another and average the results. This
procedure is referred to as stacking. Although collection of stacked measurements
nominally increases the duration of the survey, stacking improves the signal-to-noise ratio
because noise, if random, is averaged out. In addition, the standard deviation of the stacked
measurements (i.e., the stacking error) provides a means to quantify error and define data
weights for inversion (although reciprocal measurements described next are preferred).
Stacking errors are useful in QA/QC and can be used to weight data in inversion or form a
basis for editing datasets prior to inversion. The number of stacked measurements should
be recorded on field data collection forms if not recorded by the software. On some
instruments, the number of stacked measurements is determined by calculating the
running stacking error, with more stacks collected for measurements with larger error.

3.3 Reciprocal Measurements

A reciprocal measurement involves swapping current and voltage electrode pairs
(Figure 9). In other words, electrodes A and B are used as potential electrodes while
electrodes M and N are used as current electrodes; consequently, Kg remains the same.
Theoretically, the reciprocal measurement should yield the same resistance (and hence
apparent resistivity) no matter the distribution of conductivity in the ground as the regular
measurement provided that (1) no sources are present, (2) there are no effects due to contact
resistance, and (3) that Ohm’s law is linear, with any difference in values indicating
instrument error and/or effects in violation of Equation 1 or 5. These measurements should
be collected to quantify error in measurements and to check for poorly performing
quadripoles, which may then be removed. Apparent resistivities generally should match
between stacked or reciprocal datasets within a few percent, if not better. Data

21
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

corresponding to very low applied currents (e.g., < 10 mA or another cutoff), may also be
removed if needed.

Figure 9 - Reciprocal measurements involve swapping the position of

the potential and current electrodes, and can be used to identify
problems because the reciprocal measurement will yield the same
resistivity as the original measurement unless there is a problem or
interference.

Different users evaluate reciprocal measurements in different ways, the most

common being reciprocal standard error and reciprocity. If the normal and reciprocal
measurements of apparent resistivity are ρa,1 and ρa,2, and the average resistivity is
𝜌𝑎,1 +𝜌𝑎,2
𝜌𝑎,𝑎𝑣𝑒 = 2
, the reciprocal standard error srecip and reciprocity r are given by:

(𝜌𝑎,1 − 𝜌𝑎,𝑎𝑣𝑒 )2 + (𝜌𝑎,2 − 𝜌𝑎,𝑎𝑣𝑒 )2

𝑠𝑟𝑒𝑐𝑖𝑝 = √ (7)
2

and
𝜌𝑎,1 − 𝜌𝑎,2
𝑟 = 𝑎𝑏𝑠 ( ) (8)
𝜌𝑎,𝑎𝑣𝑒
Both the reciprocal error and the reciprocity are useful. Reciprocal errors are used
to set data weights in the inversion as explained in Section 0.3. The reciprocity is a
dimensionless measure of relative error and, when multiplied by 100, gives the percent
error in the reciprocal measurement. It is a useful measure of error for data filtering (e.g.,
reject all data with a reciprocity larger than 0.10) as is shown in the example in the Data
Quality Control and Assurance subsection of Section 5.2.
In general, reciprocal errors are larger than stacking errors, and it is commonly
thought that reciprocal errors provide a more comprehensive quantification of noise than
stacking errors (Binley et al., 1995). For QA/QC, either reciprocity or reciprocal errors may
be used instead of, or in addition to, stacking errors. Depending on the array type,
reciprocal measurements may take more time than the regular measurements on
multi-channel instruments, which, as mentioned earlier, collect multiple potential
22
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

measurements for each current pair. If n potential measurements are made for a single
current pair, then n reciprocal measurements must be made, each with a different current
pair. However, a reciprocal survey for a dipole-dipole array will take exactly the same
length of time as the regular survey; this is achieved by running the sequence in the reverse
direction of the line. A reciprocal survey for a Wenner array will also take exactly the same
length of time as the regular survey as each Wenner measurement requires a separate
current injection, so the multi-channel functionality cannot be used for this array. The
commonly used Schlumberger array type is an example where reciprocals take much
longer, but it is also an example where reciprocity can speed up data collection on a
multi-channel instrument (i.e., the inverse Schlumberger array). However, arbitrarily
designed sequences often take longer to run as reciprocals when multiple current injections
in the reciprocal dataset are needed to reciprocate a set of measurements in the regular
survey acquired with a single current injection.
A reciprocal measurement should not be collected immediately after its associated
regular measurement, as any residual charge up (polarization) of the current electrodes will
affect the voltages recorded between these electrodes with the reciprocal measurement.
Such effects generally dissipate in a few seconds, although it may take minutes or longer
depending on injection. Reciprocal measurements are best collected either interleaved
throughout the measurement sequence file or following regular data collection, depending
on whether time-lapse processes are being considered and the subsequent time lag between
the regular and reciprocal measurement. For investigations of time-varying processes,
collecting a subset of data for reciprocals is preferred because otherwise meaningful
temporal changes could appear as error. More information on how to use these data to
build an error model can be found in the 2017 paper by Lesparre and others.
The principle of reciprocity also applies to IP datasets. Similar to other aspects of IP
data acquisition, acquiring reciprocal IP datasets is more challenging than acquiring
reciprocal ER datasets. The polarization of a recently used current electrode may severely
corrupt the reciprocal IP measurement. IP errors can instead be quantified through analysis
of the shape of the decay curves following current shut off (Flores Orozco et al., 2018).
However, reciprocation of IP datasets can be done with careful attention to data acquisition
to ensure dissipation of IP effects, resulting in significantly improved confidence in IP
datasets, which are traditionally susceptible to errors and misinterpretation (Slater and
Binley, 2006; Zarif et al., 2017).

3.4 Error Considerations for Time-Lapse Measurements

In the case of time-lapse imaging, collecting multiple background datasets can be
useful for determining systematic errors and is generally well worth the time, as the most
time-consuming part of ER data collection usually is deploying the electrodes and cables,
and understanding the background conditions is important to interpreting changes. In
systems where a hydrogeologic change is being introduced (infiltration or a tracer test, for
23
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

instance), collecting multiple background datasets provides some measure of natural

variability before the introduction of that change. In cases where fast time-lapse data are
needed, collecting a subset of reciprocals—or collecting repeat measurements to get errors
between two complete data sequences, one collected right after the other—may be a
valuable way to get another measure of error beyond stacks. An example of time-lapse
monitoring is provided in Section 0.2.

3.5 Pulse Duration

Some time-domain instruments allow the user to select the duration of the current
injection. Commonly, pulse duration varies from 250 milliseconds (ms) to several seconds.
Lower pulse duration results in shorter data acquisition time, but longer pulse duration
may be key to achieving steady state voltages and thus data more consistent with the
physics assumed for analysis (e.g., the Poisson Equation). Similar issues arise in
hydrogeology when performing hydraulic analysis based on steady-state assumptions
(e.g., Thiem-Equation based analysis), where pumping continues until a quasi-steady state
is achieved and then measurements are made. Otherwise, transient analysis is necessary,
for example, Theis-Equation-based analysis in the groundwater analogy and IP in electrical
imaging. Pulses on the order of 250 ms may be acceptable when exclusively collecting ER
data in conductive, low-clay media. In the presence of clays and other chargeable media,
however, longer durations may be required to achieve equilibrium voltages. Acquisition of
IP datasets generally requires longer pulse durations (1 s or greater is recommended) so
that the sufficient discharge (or charge up) occurs to be reliably recorded. As an
approximate guide, the acquisition of high-quality IP data is likely to increase the survey
time two to three-fold over acquisition of only ER data. The length of the pulse duration
can be varied, and surveys repeated, to determine the minimum duration necessary to
achieve good data.

3.6 Notes on Field Conditions

Notes on the location of electrical lines, radio transmission towers, known metallic
objects, topography, and weather should be collected at each field site, as should decisions
about the QA/QC procedures above. It is important to check and record: battery voltages,
x-y-z positions of the electrodes, how their position was determined (meter tape, GPS, laser
level, etc.), and how the electrodes were deployed and/or built. The integrity of multi-core
cables should be assessed, particularly around individual electrode takeouts that are most
prone to damage during use. For cross-well measurements, information on well
construction (casing length, total depth, borehole integrity, presence of monitoring
equipment, etc.) should be known prior to installing cables and injecting current. Some
commercially available systems provide a test resistor to verify the correct operation of the
instrument. Results of this test, if performed, should be recorded. Although outside the
scope of this book, safety is an important consideration when dealing with any electrical
24
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

instrumentation, and equipment users should develop protocols to ensure that no one
touches energized electrodes during data collection. These issues do not prevent collection
of quality data, but should be recorded and considered when evaluating data quality.

25
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

4 Data Inversion
4.1 The Goal of Inversion
Once electrical imaging data are collected, they are inverted to obtain a spatially
discretized (i.e., gridded or meshed) distribution of the electrical properties of the
subsurface. In the case of ER measurements alone, it is just the electrical conductivity
structure that is estimated by the inversion. When IP datasets are acquired, both the
electrical conductivity and the intrinsic chargeability (or the intrinsic phase) structure are
estimated, and IP data cannot be inverted without ER data. With IP datasets, images can be
presented in terms of the real and imaginary components of the complex conductivity.
Irrespective of what data have been acquired—whether ER data alone, or ER and IP data
combined—the inversion of electrical imaging datasets involves a number of common key
steps/concepts. For simplicity, we describe the inversion process from the perspective of an
ER dataset alone, but the mechanics and considerations introduced in this section apply
equally to combined ER and IP datasets.
The goal of inversion of an ER dataset is to recover a subsurface distribution of
electrical conductivity (Figure 10), σx,y,z (note for frequency domain, IP this would be σ*x,y,z)
that could have produced the observed data (step 2 and 3 below). The general procedure
for inverse modeling consists of the following steps:
1. Start with a distribution of electrical conductivity (typically a homogeneous
starting model corresponding to the average apparent conductivity
measured in the field);
2. Use a forward simulator which, for the given distribution of electrical
conductivity, calculates predicted data using Equation 1 (or Equation 5 for
frequency-domain IP, and time-domain IP with caveats as noted below);
3. Calculate the misfit between the predicted and the observed data and also a
measure of the complexity (e.g., roughness) of the electrical conductivity
distribution; and
4. If the misfit is less than our stopping criteria, stop and accept the current
subsurface distribution of electrical conductivity as the final result. If not,
modify the model to improve the fit, and return to step 2.

26
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 10 - A simplified overview of the inversion process.

In inversion codes, forward models are called repeatedly within an optimization

framework to calculate predicted data for comparison with observed data. The
optimization iteratively updates the model to improve the match between predictions and
observations. Data processing and inversion can be performed with commercially available
software packages such as AGI EarthImager (LaBrecque and Yang, 2001), Res2DInv (Loke
and Barker, 1996), or MPT ERTLab, and public-domain or freeware packages such as
RESINVM3D (Pidlisecky et al., 2007), E4D (Johnson et al., 2010), ResIPy (Blanchy et al.,
2020), pyGIMLIi (Rucker et al., 2017), or SimPEG (Cockett et al., 2015). The majority of these
codes also permit the inversion of IP datasets. It is not our goal to review or compare these
codes, but most follow similar inversion approaches, which are based on Gauss-Newton,
quasi-Newton or steepest-descent algorithms (Tarantola, 1987), with some codes
supporting use of multiple inversion algorithms. Depending on selection of modeling and
inversion parameters, these codes generally can be made to produce similar results. Default
values differ greatly, however, and it is not always clear how parameters are used within
the inversion. Selection of many inversion settings can be somewhat subjective and should
be guided by prior knowledge of the site geology or the nature of the targets. For example,
in a layered system, one might choose to apply anisotropic smoothing, which will result in
a tomogram that has a layered character. For results to be reproducible, it is critical to (1)
report all parameter selections including default values, (2) document the algorithm used
by the software, and (3) archive a copy of the software code or executable. Justifications of
parameter choices should also be documented. An example of how inversion settings affect
tomograms is demonstrated in Section 0.2. The process for selecting inversion settings
should be guided by prior information, and many hydrogeologists will find it useful to
work with a geophysicist while learning inverse techniques for these data.

27
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Ideally, the inversion should result in the true distribution of electrical

conductivities of the subsurface. In practice, this is impossible for several reasons. First, the
electrical conductivity estimates are commonly produced for blocks with dimensions of
tens of centimeters to several meters on the side, while earth conductivity varies on much
smaller scales. Thus, the best one can hope for is to identify block conductivities that
represent some sort of weighted, spatial averages. Additionally, there commonly is not
enough information in the data to uniquely determine all the block bulk conductivity
parameter values. Unlike medical imaging, where it is possible to acquire a 360-degree
view around the target, ER is usually limited to surface and borehole electrodes unless
working on soil cores or experimental tanks, which is partly why medical imaging offers
higher resolution compared to geophysical imaging. The result of this limited available
information is called ill-conditioning, which is a property of matrix-inversion problems
where limited data sensitivity to parameter values causes uncertainties in the data (e.g.,
resistance measurements) and can lead to large errors in the parameter estimates (e.g.,
electrical conductivity values). This ill conditioning is typically rectified through model
regularization, which imposes additional constraints on the estimated model to get a unique
set of stable solutions, as described subsequently. Additionally, as noted above, depending
on the discretization of the finite-difference grid or finite-element mesh used for numerical
approximations of Equation 1 or 5, some quadripoles may model poorly. For example, the
inverse model would not be able to accurately match these data. Finally, accurate
representation of data errors (for example using reciprocal measurements or stacking
described above) is needed to ensure quality inversion results that do not over- or under-fit
measurements.
Resistivity software packages that incorporate time-domain IP measurements
model the distribution of the intrinsic chargeability in addition to electrical conductivity.
In contrast, frequency-domain measurements are processed with algorithms that model the
complex electrical conductivity distribution per Equation 5. The electrical conductivity
magnitude and phase, or the real and imaginary parts of the complex electrical
conductivity, can be imaged. The measured apparent chargeability and measured apparent
phase are directly proportional to each other, although the proportionality constant will
vary depending on how a time-domain instrument is configured (Slater and Lesmes, 2002).
Consequently, time-domain measurements can be modeled using Equation 5 if this
proportionality constant is defined (Mwakanyamale et al., 2012).

4.2 Regularization in Electrical Imaging Inversion

Most tomographic problems in geophysics are solved with an excess number of
model parameters and use regularization to create a mathematically stable solution (e.g.,
Constable et al., 1987). Due to these issues, this non-linear problem is solved using iterative
inversion (Tripp et al., 1984; Daily and Owen, 1991). The solution to the ER inverse problem
is a cross section or volume of electrical conductivity values and is usually based on
28
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

non-linear least-squares minimization of a two-part objective function, F, which is given by

Equation 9a.
2
𝐹 = ‖𝐶𝐷−0.5 (𝑑𝑜𝑏𝑠 − 𝑔(𝑚))‖ + 𝜀‖𝐷𝑚‖2 (9a)

This can also be written as Equation 9b.

𝑇
𝐹 = (𝑑𝑜𝑏𝑠 − 𝑔(𝑚)) 𝐶𝐷−1 (𝑑𝑜𝑏𝑠 − 𝑔(𝑚)) + 𝜀𝑚𝑇 𝐷 𝑇 𝐷𝑚 (9b)

where (for the case of ER inversion alone):

dobs = vector of electrical resistance (∆V/I) or apparent resistivity

measurements, where the vector length is equal to the total number of
quadripoles.
g( ) = forward model for electric potential (Equation 1 or 5), which produces
the simulated estimates of the measurements.
m = vector of parameter estimates, (log electrical conductivity).
CD = covariance matrix of data uncertainty or errors (often where the diagonal
of the matrix is defined by reciprocal or stacked measurements, and the
off-diagonal values are zero), which defines how certain we are in our
measurements. This matrix often includes some measure of modeling
errors, as described in Section 4.3 below.
ε = regularization parameter that determines the importance given to the
smooth appearance of the electrical conductivity field relative to the
misfit between calculated and observed resistances. An overly small ε
will minimize the residual error between measured and modeled
resistances and may overfit the data, producing spurious heterogeneity
in the solution. In contrast, an overly large ε will identify an overly
smooth electrical conductivity field that may not fit the measured field
data (resistances) well (see, for example, Tikhonov and Arsenin, 1977).
D = model-weighting regularization matrix, which can either be defined by
a discretized spatial-derivative operator or be based on the covariance of
the model parameters (Tarantola, 1987; Gouveia and Scales, 1997;
Kitanidis, 1997; Vasco et al., 1997; Day-Lewis et al., 2003). This matrix
defines how each pixel is related to one another.
The first part of the objective function is the data misfit term, which minimizes the
discrepancy between field resistance data (or resistance and IP data for IP inversion) and
the computed resistances (or resistances and IP measurements for IP inversion) based on
Equation 1 or Equation 5, within measurement errors. The second part is the regularization
term, often called the model roughness if smoothing is sought, which typically minimizes the
roughness (or maximizes the smoothing) of the electrical conductivity field and allows for
a well-posed inverse problem. This term is required due to the overparameterization of the
29
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

inverse problem, meaning that the information provided by the measurements cannot
uniquely resolve each of the conductivity parameters. It is possible to be creative with the
regularization term, depending on prior information available to develop a conceptual
model (e.g., Caterina et al., 2014; Nguyen et al., 2016; Hermans et al., 2016).
In a non-linear inverse problem, model parameters are updated iteratively by
repeated solution of a linearized system of equations for Δm at successive iterations. Such
an approach results in the regularization changing throughout the iterative process. This
process makes it difficult to map the effect of regularization throughout the inversion
process, and consequently impairs quantitative inference from the images. The update
appears as shown in Equation 10.

[𝐽𝑇 𝐶𝐷−1 𝐽 + 𝜀𝐷 T 𝐷]∆𝑚 = JT 𝐶𝐷−1 (𝑑𝑜𝑏𝑠 − 𝑔(𝑚𝑘−1 )) − 𝜀𝐷 T 𝐷𝑚𝑘−1 (10a)

𝑚𝑘 = 𝑚𝑘−1 + ∆𝑚 (10b)

where:
𝜕𝑑𝑠𝑖𝑚,𝑖
J = Jacobian matrix at iteration k, with elements 𝐽𝑖𝑗 = ⁄𝜕𝑚
𝑗

mk = vector of parameter estimates after updating in iteration k

Δm = vector of parameter updates for iteration k
Although tomographic inversion with regularization is useful for imaging
large-scale (low-spatial-frequency) structures, it yields poor results when attempting to
infer quantitative values from the recovered images (e.g., Binley et al., 2002; Singha and
Gorelick, 2005; Day-Lewis et al., 2007) due to uncertainty in the inversions for reasons
discussed in the next section.
3-D acquisition and inversion are increasingly possible and appropriate, although
many practitioners still use 2-D inversion. As noted earlier, commercially available
software for 2-D inversion commonly invokes the 2.5-D assumption for computational
efficiency, where 2-D ER cross sections are constructed by simulating 3-D current flow
under the assumption of 2-D heterogeneity (Dey and Morrison, 1979; LaBrecque et al.,
1996). The 2.5-D assumption amounts to assuming that all heterogeneity in the 2-D imaged
cross section extends infinitely out the 2-D plane. Clearly, this assumption is violated in the
presence of strong 3-D heterogeneity, which requires 3-D acquisition and inversion
approaches to image accurately.

4.3 Selection of Inversion Parameters to Prevent

Overfitting/Underfitting of Data
Data should generally be weighted based on their measurement (reciprocal or
stacking) uncertainty (Sections 3.2 and 3.3) to prevent overfitting or underfitting of the data.
In other words, the inversions should generally match the data within the uncertainties

30
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

(sometimes called errors) quantified by reciprocal or stacking errors. However, there are
often systematic data or model sources of error that are not manifested in reciprocal or
stacking errors. For example, systematic sources of data error can arise from electrode
position errors or temporal variation in subsurface electrical conductivity over the course
of the survey. Systematic model errors include coarse-grid error (i.e., the inability of the
model to simulate the effects of sub-grid heterogeneity), or violation of the 2-D
heterogeneity assumption in 2-D inversions as noted above. We usually cannot quantify
these errors and therefore cannot account for them via data weighting; their existence can
produce inversion artifacts if they are not accounted for. The sources and magnitudes of
these errors are often unknown and therefore require a subjective estimation of how to
approach error weighting by the user.
One common approach to estimating data error is given by Equation 11.

𝑠𝑖 = 𝑎𝑖 (abs(𝑅𝑖 )) + 𝑏 (11)

where:

si = standard deviation of resistance measurement i

ai = unitless scaling factor giving the error in terms of the magnitude of the
measured resistance Ri
b = small resistance indicating the precision of the voltmeter
As written, Equation 11 allows for a different value ai for each resistance
measurement i, where the a values initially could be based on the reciprocal or stacking
error expressed as a decimal fraction. These values can be increased until the inversion
converges to a reasonable conductivity structure, at which point additional sources of error
would presumably be accounted for. This approach assumes that standard or reciprocal
errors are proportional to total error. Alternatively, a single a is often used to establish
representative weights for all resistance measurements in cases where individual error
weights are too low. In this case, it is common to first calculate average reciprocal errors for
a range of bins that divide up the reciprocals in terms of increasing resistance (Figure 11).
The reciprocal error for any resistance value is then determined from the equation of the
line. One way to test whether reciprocal and stacking errors adequately represent the true
data error is to first invert the data under the hypothesis that reciprocal/stacking error
adequately quantifies all sources of error. If the inversion (1) does not converge (i.e., cannot
fit the data) in a reasonable number of iterations, (2) results in an overly heterogeneous
electrical conductivity structure, or (3) produces unrealistic electrical conductivity values,
then it is likely that the reciprocal and stacking errors do not adequately capture the true
error. In this case, the user must increase the error to produce smoother or more uniform
tomograms. Use of the correct errors will result in convergence within a reasonable number

31
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

of iterations, typically 3 to 10, although this will depend strongly on the inversion
algorithm.

Figure 11 - Example reciprocal error plot for an electrical resistivity dataset. Blue
crosses are individual reciprocal errors whereas orange dots are average values for bins
defined in terms of increasing resistance. The linear fit of an estimated reciprocal error
(si) as a function of resistance Ri is based on the binned values. Plot created with ResIPy
(Blanchy et al., 2020).

There are several approaches commonly used to set the relative weighting between
the model roughness and data misfit in the inversion (Equation 9a and b), i.e., the value of
ε. This issue is not trivial, as the tradeoff between terms of the objective-function controls
the variability of estimated electrical conductivity and, ultimately, whether the information
in the data is optimally utilized. If too much weight is ascribed to the model roughness
term, underfitting occurs—thus the inversion does not capitalize on all the information
provided by the data, resulting in an overly smooth tomogram. On the other hand, if too
much weight is given to the data misfit term, overfitting results—the data are fit so well
that the inversion reproduces noise, resulting in an overly complex tomogram with
spurious structure and possibly unrealistic electrical conductivity values. While these
models are mathematically viable and may even fit data better than other models, they are
geologically unrealistic and this is where the art of inversion and prior knowledge of the
system are important.
The tradeoff parameter, ε, is analogous to a contrast knob, which if set incorrectly
results in an image that is washed out at one extreme or noisy at the other. The simplest
approach to identify ε is subjective selection by the user, such that the resulting tomogram
is qualitatively consistent with existing knowledge of the range of subsurface electrical
conductivity and geologic structure. More objective approaches include an Occam’s
inversion (Constable et al., 1987), the L-curve (Hansen and O’Leary, 1993), and generalized
cross-validation (GCV) (e.g., Haber and Oldenburg, 2000; Farquharson and Oldenburg,
2004). In general, the three techniques produce similar results for most datasets. These three
approaches are:
32
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

• In Occam’s inversion, ε is determined as part of the optimization to achieve the

smoothest model that matches the data to the desired misfit.
• In the L-curve approach, the inversion is repeated for a number of values of ε,
and a plot of model complexity versus data misfit (i.e., the second term of
Equation 9a and b versus the first term of Equation 9a and b) is constructed. The
optimal ε is taken as the value at the elbow of the resulting L-shaped curve.
• In GCV, the tradeoff parameter is identified based on a procedure analogous to
inverting the dataset repeatedly, leaving one measurement out at a time, and
finding the value for ε that minimizes the average prediction error for the data
eliminated.
An inversion converges when the data misfit is reduced to some set value. Criteria
for computing the data misfit commonly are based on the absolute weighted error (AWE),
sometimes called the normalized error or the percent error (PE), which for measurement i are
given respectively as Equations 12a and 12b.
((𝑑_(𝑜𝑏𝑠, 𝑖) − 𝑑_(𝑠𝑖𝑚, 𝑖) ))
𝐴𝑊𝐸𝑖 = (12a)
√(𝐶_(𝑑, 𝑖) )
100% (𝑑𝑜𝑏𝑠,𝑖 − 𝑑𝑠𝑖𝑚,𝑖 )
𝑃𝐸𝑖 = (12b)
𝑑𝑜𝑏𝑠,𝑖
where:

dobs,i and dsim,i = observed and simulated values of datum i

Cd,i = variance of datum i as determined by the data noise estimate
AWE and PE = used to compute the normalized chi-squared (χ2) value and the
root-mean-squared error (RMS) value defined by Equation 13a and 13b
respectively.

𝑁
1
𝜒 = ∑ 𝐴𝑊𝐸𝑖2
2
(13a)
𝑁
𝑖=1

𝑁
1
𝑅𝑀𝑆 = √ ∑ 𝑃𝐸𝑖2 (13b)
𝑁
𝑖=1

where:
N = number of data

The normalized χ2 value is a linear scaling of the first term of the objective function
(Equation 9a and b). It is a useful measure of data misfit because it gives a direct indication
of what the inversion is trying to minimize (in addition to the regularization term) and
includes the covariance of the data (the error weights) directly. When the data misfit in the
numerator of the AWE is consistent with the data error estimate in the denominator of the

33
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

AWE, then χ2 = 1. Assuming our data are appropriately weighted, then χ2 = 1 is the target
value we are aiming for at convergence. That is, we would ideally fit our observed data
with our simulated data in a manner consistent with the uncertainty in the measurements.
The RMS (root-mean-squared error) value (as defined in Equation 13b) is equivalent to the
standard deviation of the PE distribution, and therefore provides an intuitive measure of
the total data misfit in terms of percent error, with no covariance term. Also, in contrast to
the χ2 value, the RMS value is independent of data weighting. Consequently, it is possible
to have a χ2 close to 1 but a very large RMS error if the covariances are large.
Some software packages assume the inversion has converged when the data misfits
are within the limits specified by the data error, as outlined above. Other packages support
Occam’s inversion or use of the L-curve approach. Yet other packages leave it to the user
to decide when the inversion has converged, placing the burden of balancing the tradeoff
between the model and data misfit on the scientist’s subjective judgment. In any case, all
selections of inversion parameters should be recorded and reported. Some examples of
under- and overfitting are provided in the case studies in Section 0.

4.4 Definition of Data Misfit

In Equation 9a and b, the data misfit is based on an L2 norm (i.e., as the sum of
weighted, squared differences between predicted and observed values), which is the first
term in the equations. The L2 norm is highly sensitive to outlier data, hence the need to
carefully edit datasets and remove data corresponding to electrodes with poor electrical
contact or faulty channels. Many inversion packages also support minimization of the L1
data misfit, which has been referred to as robust inversion. The L1 norm is based on absolute
differences instead of squared differences and is therefore less sensitive to outliers
(Claerbout and Muir, 1973). The L1 norm can also be applied to the regularization term of
Equation 9a and b instead of the L2 norm, which tends to produce blocky inversion results
with sharp boundaries. Implementation of the L1 norm on either the data misfit or
regularization term of the objective function require re-weighting of CD and/or D
respectively at each outer iteration of an iterative least-squares solution. This is commonly
referred to as Iteratively Reweighted Least Squares inversion (Farquharson and Oldenburg,
1998).

4.5 Quantification of Inversion Quality

Several approaches are commonly used to gain insight into the reliability of
tomograms. For small inverse problems, it is possible to calculate the model resolution
matrix (e.g., Menke, 1984) and present the diagonals, rows, and columns of these matrices
as cross-sectional images. Conceptually, the model resolution matrix is the lens or filter
through which the inversion sees the study region. For a linear inverse problem, the
parameter estimates are expressed by Equation 14.

34
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

𝑚 = [𝐽𝑇 𝐶𝐷−1 𝐽 + 𝜀𝐷 T 𝐷]−1 JT 𝐶𝐷−1 𝑑𝑜𝑏𝑠 ≈ [𝐽T 𝐶𝐷−1 J + 𝜀𝐷 T 𝐷]−1 𝐽T 𝐶𝐷−1 𝐽𝑚𝑡𝑟𝑢𝑒 (14)

In this case, the model resolution matrix R is defined as shown in Equation 15.

𝑅 = [𝐽T 𝐶𝐷−1 J + 𝜀𝐷 T 𝐷]−1 𝐽T 𝐶𝐷−1 𝐽 (15)

Consequently, the parameter estimates are the product of the true parameter values
and the resolution matrix as shown in Equation 16.

𝑚 = 𝑅𝑚𝑡𝑟𝑢𝑒 (16)

For linear problems, where J is independent of mtrue, R can be calculated prior to

data collection. Given an estimate of measurement errors, the model resolution matrix can
be calculated using Equation 15 and used as a tool to assess and refine hypothetical survey
designs and regularization criteria. In interpreting inversion results, R is useful for
identifying likely inversion artifacts (Day-Lewis et al., 2005). The model resolution matrix
quantifies the spatial averaging inherent to tomography; hence, it gives insight into which
regions of a tomogram are well resolved versus poorly resolved. This information is
valuable if tomograms are to be converted to quantitative estimates of porosity,
concentration, or other hydrogeologic parameters. Calculation of resolution matrices,
however, remains computationally prohibitive for many problems, particularly those
involving 3-D inversion. Hence, few commercially available software packages support
calculation of R, and it is instead more common to look at an inverse problem’s cumulative
squared sensitivity vector (S) as shown in Equation 17.

𝑆 = 𝑑𝑖𝑎𝑔(𝐽𝑇 𝐽) (17)

Here, J is the sensitivity matrix defined in Equation 10a and diag( ) indicates the
diagonal elements of a matrix. The sensitivity matrix can be used to gain semi-quantitative
insight into how resolution varies spatially over a tomogram. Pixels with high values of
sensitivity are relatively well informed by the measured data, whereas pixels with low
values of sensitivity are poorly informed. It is important to note that, in contrast to R, S does
not account for the effects of regularization criteria (as contained in D) or measurement
error (as contained in CD). Rather, S is based only on the survey geometry and measurement
sensitivity. An example sensitivity map is provided in the case study in Section 5.2 and
qualitatively in Figure 4. Another question is whether inversion results are consistent with
our conceptual models of the site—this is a different definition of inversion quality. A good
review exploring this idea is presented by Linde (2014).

4.6 Checks on Inversion Results

Tomographic inversion results are strongly affected by selected inversion
parameters and regularization criteria, especially in the presence of large measurement
errors. It is instructive, therefore, to run multiple inversions to gain insight into the effects
35
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

of different software settings, which is the philosophy of the depth of investigation (DOI)
analysis seen earlier. Rarely are default inversion settings appropriate and the inversion
should be guided by prior information. Prior information that may be useful includes past
geophysical results, (hydro)geologic maps, and drillers’ logs. If inverted electrical
conductivity cross sections are inconsistent with such prior information, this could indicate
that settings are suboptimal or that assumptions (e.g., 2-D heterogeneity) are violated.
Table 1 lists some common problems and their associated symptoms and solutions. We
emphasize that Table 1 is by no means exhaustive in terms of the symptoms, problems,
and solutions and relationships between them. Rather, this is meant as a starting point for
practitioners to begin thinking about the roles of various inverse settings.

Table 1 - Common problems with inversion settings, and the associated symptoms and solutions.
Symptom Possible problems Solution

• Stop inversion at an earlier iteration or

The inversion may be
Minimum/maximum estimated increase the assumed measurement
overfitting the data
electrical conductivity too error
low/high compared to
expected values Non-random outlier data may • Check the dataset for outliers and edit
be present • Try the L1 norm for data misfit

• Stop inversion at an earlier iteration or

The inversion may be
increase the assumed measurement
overfitting the data
Tomogram is speckly or looks error in an Occam’s inversion
like a checkerboard
Non-random outlier data may • Check the dataset for outliers and edit
be present • Try the L1 norm for data misfit

• Change optimization tolerances

The optimization algorithm may • Increase number of iterations
be caught in a local minimum • Update the Jacobian more frequently
• Change the starting model

The finite-difference or
The inversion cannot match
finite-element grid may be too • Refine the grid or mesh
the data to within the
coarse
reciprocal error

The inversion grid may be too • Increase the number of inversion

coarse parameters

Non-random outlier data may • Check the dataset for outliers and edit
be present • Try the L1 norm for data misfit

• Try robust model misfit instead of L2

Regularization criteria may be
model misfit
smoothing/blunting the
• Use anisotropic regularization
The tomogram does not look tomogram too much
• Try different regularization criteria
like expected geology

Electrical conductivity may not • Another geophysical technique may

correlate with lithology be needed

36
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Symptom Possible problems Solution

• Another geophysical technique may

Resolution may be above the
be needed or array type and/or
scale of the pertinent
electrode spacing used may need to
heterogeneity.
be reconsidered

• Use an inversion package with a

Electrical anisotropy forward model that allows for
electrical conductivity anisotropy

Two (or more) tomograms

Outlier data are present in at • Check the dataset for outliers and edit
that share a borehole appear
least one dataset • Try the L1 norm for data misfit
inconsistent at the borehole

Parsing data into individual

• Invert all data at once, rather than
tomograms when they should
breaking it into pieces
be considered together

• Create synthetic or hypothetical

Tomograms show vertical forward models of the experiment to
Resolution may vary greatly
streaking, or high or low evaluate resolution and likely artifacts
from the sides to the middle of
electrical conductivity patches • Examine plots of the inversion’s
the tomogram
only at boreholes sensitivity or resolution matrix
• Explicitly model the boreholes

37
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

5 Case Studies
5.1 2-D Waterborne Resistivity and Induced Polarization Profiling
Background
Here, we outline the case study of 2-D electrical resistivity and induced polarization
imaging reported in the 2010 paper by Slater and others, which focused on improving
understanding of the hydrogeological framework regulating exchange of groundwater
with surface water of the Columbia River at the United States Department of Energy
Hanford 300 Facility, Richland, Washington, USA. The basic hydrogeological setting
consists of a coarse-grained aquifer (the Hanford Formation) underlain by a lower
permeability, fine-grained confining unit (the Ringold Formation). A legacy of nuclear
waste processing and disposal at the site extending through the Cold War era resulted in
significant potential for radionuclide-contaminated groundwater to discharge into the
Columbia River. The existence of relict paleochannels incised into the Ringold Formation
had previously been proposed to provide preferential flow paths promoting rapid
transport of contaminants from the aquifer into the river. The risks of radionuclide
contamination led to a high cost of drilling at this site, encouraging the use of geophysical
surveys to understand the structure of the region of interaction between surface water and
groundwater. Waterborne surveys have been successfully used to investigate coastal
processes and groundwater-surface water exchange in other systems (e.g., Day-Lewis et al.,
2006).

Data Collection
Two-dimensional resistivity and induced polarization imaging surveys were
performed to improve estimates of the spatial variability in the depth to the contact
between the Pleistocene Hanford formation and the Pleistocene Ringold Formation. The
surveys were designed to explore for evidence of incisions into the Ringold Formation that
might represent the location of high-permeability paleochannels. The rationale for the
application of IP was a suspected strong contrast in polarizability between the
coarse-grained, Hanford sediments (low polarizability) and the fine-grained Ringold
sediments (high polarizability). The acquisition of ER measurements alone would have
been less informative because of the expected influence of variations in the groundwater
electrical conductivity due to variable surface water-groundwater interaction on the
electrical images.
To rapidly image a long reach of the river corridor, measurements were acquired
using a floating array of 13 graphite electrodes spaced at 5 m intervals pulled behind a boat.
In this study, ER and IP measurements were performed on approximately 30 km of 2-D line
profiles in water depths varying from 2 m to 18 m. Data were collected in July 2008 from a
Gregor aluminum-hull jet boat, using a 10-channel time-domain ER/IP instrument (Syscal

38
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Pro, Iris Instruments, France) as shown in Figure 12. This time-domain instrument records
the apparent integral chargeability (Equation 4) determined from the decay curve after
current shutoff. Measurements were recorded every 0.5 to 3.0 m depending on survey
speed, resulting in more than 65,000 measurements over the 30 km of line.

Figure 12 - Photographs of deployment of waterborne resistivity and induced polarization data acquisition at
the Hanford site. a) Gregor jetboat used for data acquisition, b) deployed floating array concept (yellow
symbols depict 8 of 13 electrodes, remaining 5 are closer to boat).

Data Processing
The waterborne resistivity measurements were inverted for an estimated
subsurface distribution of electrical conductivity and chargeability using the commercially
available RES2DINV package (Loke et al., 2003). The variable-thickness water layer was
constrained to a uniform conductivity and zero chargeability (water is non-polarizable at
low frequencies). The dataset was treated as a series of near-parallel 2D lines for individual
inversions. Electrode locations were calculated from a GPS located on the boat and
knowing the length of the electrode takeouts on the cable pulled behind the boat. Each line
was then inverted using the conventional smooth regularization constraint except for the
fixed/known surface water layer.

Data Interpretation
Figure 13 shows the 2-D inversion of one line (Line 20 approximately 20 m from the
shore). The inverted electrical resistivity distribution (lower panel) and the inverted
normalized chargeability (Mn = M/ρ) (upper panel) are shown. This normalization of the
chargeability by the resistivity provides a direct measure of the polarizability that is
unaffected by variations in pore fluid conductivity of the groundwater and therefore is
exclusively related to the physical properties of the sediments (Slater and Lesmes, 2002).
The riverbed is shown as a black line.

39
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 13 - 2D resistivity and induced polarization image acquired along a 3 km section of the Columbia
River, WA, USA corridor. Black solid line is base of water layer and white dashed line is interpreted contact
between aquifer and underlying confining unit. Crosshairs denote suspected paleochannels incised below
the contact of the aquifer with the confining unit. Modified from Slater and others (2010).

Beneath the riverbed, the resistivity image is primarily composed of a higher

resistivity layer underlain by a low resistivity layer that appears to come into contact with
the riverbed between 1100-1500 m along the line. The upper resistive layer was interpreted
as the coarse-grained, Hanford formation sediments with the lower, more conductive layer
representing the finer-grained Ringold Formation sediments. The IP image (normalized
chargeability) provides a much clearer picture of this hydrogeological structure. The water
layer and the coarse Hanford formation sediments are both low polarizability whereas the
fine-grained Ringold sediments are highly polarizable. From this IP image, the variation in
the depth to the Hanford-Ringold interface along this portion of the river corridor is
highlighted (white dashed line). The IP image reveals strong evidence for the location of at
least two coarse-grained paleochannels incised into the Ringold sediments (black
crosshairs). One of these paleochannels was subsequently mapped with ER and IP
measurements performed inland (Mwakanyamale et al., 2012). Overall, this 2-D ER and IP
survey notably improved understanding of the hydrogeological framework along this
important river corridor.

5.2 4D Resistivity of a Biostimulation Experiment

Time-lapse inversion is frequently used to map changes in electrical conductivity
through time, as might be expected from processes such as infiltration of water into dry
soils or tracer transport (Figure 14). Here, we demonstrate the steps taken for data
collection in the field and the impacts of inversion decisions using a case study at the former
Brandywine Defense Reutilization and Marketing Office (DRMO), which is an inactive 3-
hectare facility administratively controlled by Andrews Air Force Base (AAFB) and located
approximately 13 kilometers south-southeast of AAFB in Brandywine, Maryland, USA. The
Brandywine DRMO yard was used from 1943 to 1987 for temporary storage of scrap
materials and hazardous waste generated from various Department of Defense facilities in
the region and is currently classified as a Superfund site. The primary groundwater
contaminant at the former DRMO is trichloroethylene, which has spread beyond the AAFB

40
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

property into adjacent commercial and residential properties (United States Air Force,
2006). We note that this case study shows the state of the art in terms of data collection,
inversion, and analysis, especially for a 4-D system, rather than the state of the practice,
which remains 2-D static inversions at the time of this writing.

Figure 14 - Schematic of a cross-well inversion from a system with active

changes—in this case, the movement of an electrically conductive tracer.
By looking at a series of inversions through time, we would see changes in
conductivity at particular pixels, as demonstrated by the “pixel
breakthrough curves” (e.g., Slater et al., 2000) on the right.

To remediate groundwater contaminants at the site, a biostimulation effort

commenced in 2008. This effort involved injecting amendments into the subsurface to
enhance microbial activity associated with contaminant biodegradation. An autonomous
cross-well ER monitoring system was installed at the site to demonstrate and validate the
utility of geophysical measurements for providing timely and actionable information on
the spatiotemporal behavior of amendments injected into the subsurface.

41
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

ER Monitoring System
The Brandywine ER configuration consists of seven boreholes with 15 electrodes
each for a total of 105 electrodes (see Figure 15 and also Johnson et al., 2014). The borehole
electrodes are arranged to optimize imaging of the time-lapse distribution of changes in
bulk electrical conductivity caused by the injection, migration, and reaction of the
biostimulant.

Figure 15 - Configuration of ER boreholes and electrodes at the former Brandywine Defense Reutilization and
Marketing Office site, Maryland, USA. Red lines (E1 through E7) represent borehole electrode arrays, blue dots
represent electrodes (15 per well), and black lines (I1 and I2) show bioamendment injection locations.

42
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Data Collection and Experimental Design

An MPT-ER1 instrument was used to collect 10,939 dipole-dipole measurements
in a variety of cross-well and in-well configurations, including full reciprocal
measurements (i.e., 21,878 measurements total). The MPT-ER system uses an external
voltage source for the current injection electrodes, which was set at 150 V for this survey.
Current source and potential measurement electrodes were chosen according to a
user-supplied measurement schedule file and switching between electrodes was completed
through a control unit and a series of multiplexers attached to the electrode cables. Baseline
data sets were collected before amendment injections and equivalent data sets were
collected twice daily for approximately three years after injections. The amendment
consisted of a proprietary lactate-based bio-stimulant with fluid conductivity significantly
greater than groundwater conductivity. Two injections occurred using direct-push
methods (i.e., a pipe with an opening at the end is pushed into the subsurface), one at
location I1 and one at location I2 in Figure 15. For each injection, the direct-push pipe was
advanced to approximately 12 m below ground surface, where approximately 113 L of
amendment was injected. The pipe was then retracted by approximately 0.3 m and another
113 L was injected, repeating to approximately 1 m below ground surface. ERT data were
not collected during injections but commenced shortly after the injections were completed.
Importantly, the amendment formula was modified between injections such that the
amendment injected into I1 had greater fluid conductivity than the amendment injected
into I2.

Data Quality Control and Assurance

As previously mentioned, data quality is a function of injected current, which
depends on the voltage applied to the current electrodes, electrode contact resistance,
electrical conductivity of the medium, and distance between electrodes. Data quality is also
a function of random and systematic error. As outlined in Section 0, random error can result
from a number of factors such as: (1) high geometric factors, which result in a poor
signal-to-noise ratio; (2) high contact resistance, which occurs when electrodes are in poor
contact with the formation; (3) insufficient (i.e., low relative to what is needed) current
injection; and (4) random ambient electrical noise. As discussed in Sections 3.2 and 3.3,
random errors can be quantified from stacked and/or reciprocal measurements. It is
important to eliminate (or appropriately weight) data with large random errors prior to
inversion.
Histograms and summary statistics of the pre-filter reciprocity and apparent
resistivity distributions for the Brandywine data are shown in Figure 16. We eliminated
data with stacking errors larger than 3 percent, reciprocity larger than 10 percent, or applied

1 Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by
the U.S. Government.
43
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

currents less than 10 mA. We also eliminated data with apparent resistivities less than zero
or greater than 7000 ohm-m, which would well exceed the resistivity expected for saturated
medium sands at the site. Apparent resistivities above 7000 ohm-m (71 measurements)
make up the tail of the histogram and were assumed to be outliers in this case (Figure 16).
This data filtering was performed outside of the inversion software. The editing filters for
current, reciprocity, and apparent resistivity removed 90, 58, and 751 measurements,
respectively, leaving 10,040 of 10,939 measurements for a total data reduction of
approximately 9 percent. Note that this type of data filtering is not required; it is possible
to invert datasets with all collected data, especially if weighting each measurement
individually by its data error. However, it can be useful to remove particularly poor data—
the criteria for which will change with every field system such that the values here should
not be assumed to be appropriate at all sites—as there is no need for the model to work to
fit them.

Figure 16 - a) Reciprocal error histogram showing 10 percent data

rejection limits and associated statistics. A total of 58 measurements
were rejected due to excessive reciprocal error; b) Apparent resistivity
histogram and rejection limits. A total of 1316 measurements were
rejected due to excessive or negative apparent resistivity.

44
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

As discussed in Section 3.3, reciprocal measurements commonly are considered to

provide a meaningful quantification of random errors, incorporating instrument error,
error arising from nearby anthropogenic current sources, electrical storms, and/or random
physical effects in violation of Equation 1 (or Equation 5). Here, however, reciprocal errors
are small and have a standard deviation of 1.8 percent (Figure 16a). As shown in the next
section, the reciprocal error does not represent the total error (both random and systematic
as discussed in Section 4.3) and is inappropriate for use as a stopping criterion in the
inversions.

Effect of Data Weighting and Regularization Weighting on Pre-Injection Inversions

Inversions were performed using a 3-D finite-element ER modeling and inversion
code (E4D) developed at Pacific Northwest National Laboratory. E4D was implemented
in parallel to facilitate inversion of large 3D electrical conductivity data sets. However, the
Brandywine DRMO data presented here could easily be accommodated by commercially
available serial inversion codes on a standard workstation. The principles and effects of
data and model weighting (i.e., data noise estimation and regularization) demonstrated
here are also valid for commercially available codes.
We stress that for most ER studies, the choices of data weighting and regularization
will strongly affect results, and it is critical to experiment with different approaches to help
distinguish artifacts from geologic features. Features common to a suite of tomograms
inverted from the same dataset with different settings should inspire some degree of
confidence for interpretation, whereas features that result only for a specific inversion
setting or regularization are more likely to be artifacts. To demonstrate this concept, we
inverted the Brandywine data under a variety of choices for data error and regularization.
For the base case, we used Equation 11 to estimate data noise with a = 0.15 and
b = 0.10 ohms. This choice was the outcome of trial and error using several different noise
estimates, including direct reciprocal errors, and comparing inversion results with one
another and with the general geology obtained from core samples. For the base-case
regularization, we used an isotropic nearest-neighbor smoothing constraint, whereby
neighboring elements are encouraged to be equivalent in bulk conductivity regardless of
their orientation with respect to one another. The inversion started with a large ε value, and
ε was decreased only when the χ2 between iterations decreased by less than 5 percent. This
constraint and ε ‘cooling’ procedure was chosen to ensure that any heterogeneity
introduced into the inverse solution was required to fit the data (i.e., to ensure a
parsimonious, or simple, solution).
To demonstrate the effects of stopping criteria on the inversion results, we show
examples of inversions with different data misfits in Figure 17 and Figure 18. Note that for
most inversion packages, the stopping criteria are based on some measure of data misfit as
discussed previously (e.g., RMS or χ2). Because the stacking and reciprocal errors do not
represent the true error (i.e., the error estimated by Equation 11), determination of the
45
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

appropriate misfit is somewhat subjective. Thus, judgment and prior information are
critical for determining an appropriate data fit, as indicated by the amount and distribution
of heterogeneity in the inverse solution. For instance, in this case, core samples taken during
installation of the ER boreholes indicated an upper layer of fine sands and silts, an
intermediate zone of coarse sands and gravels constituting the aquifer, and a lower
fine-grained confining unit. The approximate contact depths correspond well to electrical
conductivity changes in the ‘appropriately’ fit tomogram in Figure 17b, suggesting an
appropriate data weighting and stopping criteria for this inversion.

Figure 17 - A demonstration of data overfitting and underfitting. Here the data noise is estimated by
Equation 11 using a = 0.15 and b = 0.1 ohm. a) Excessively smooth inverse model due to data underfitting.
b) An inverse model with appropriately fit data. c) An inverse model with excessive heterogeneity due to data
overfitting.

46
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 18 - L-curve plot for the Brandywine inversion. Each point and label on the plot represent an
iteration of the inversion and the corresponding regularization weighting value (ε) used for that
iteration. As ε decreases, more heterogeneity is introduced into the solution to decrease the χ2 value,
resulting in an increase in the regularization misfit. Representative visualizations of the inverse
solution at selected iterations are shown to illustrate model complexity as the inversion progresses.

Figure 18 shows the L-curve representation of the inverse solution. Each point on
the L-curve represents an iteration of the inversion, with the regularization term of
Equation 9a and b on the horizontal axis and the χ2 value (data misfit) on the vertical axis.
Each point on the L-curve represents a single iteration of the inverse solution. At the
homogeneous starting model, the χ2 value is high, but the regularization misfit is zero. As
the inversion progresses, heterogeneity is introduced into the solution to decrease the χ2
value, causing an increase in the regularization misfit. When the decrease in the χ2 value is
small between iterations (as noted by two points close together on the L-curve), the
regularization weighting (i.e., the ε value) is decreased by 50 percent, enabling the inversion
to include more heterogeneity and decrease χ2 at the next iteration. At the tail end of the
L-curve, each decrease in ε results in a large increase in heterogeneity, but only a small
decrease in the χ2 value. At this point, it is assumed that the appropriate stopping point has
been exceeded and the inversion is fitting noise, thus introducing artifacts into the solution.
Given the data error model used in this example, a normalized χ2 value of 1.08 was reached
at ε = 6.25, resulting in the solution shown in Figure 17b.
In addition to stopping criteria, the data weighting scheme can have profound
effects on the inversion results. Figure 19 shows two inversions with comparable data
misfits [in terms of root-mean-square (RMS) error] but different data weighting schemes.
47
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 19a shows results for the same data weighting scheme and regularization as the
inversions in Figure 17b. Figure 19b uses data standard deviations set to a constant value
of 0.1 ohm. Other than the data weighting, all inversion parameters for tomograms in
Figure 19a and Figure 19b are equal, and the inversion is allowed to iterate to an RMS error
of ~0.37. The high 𝜒 2 value in Figure 19b suggests that the constant standard deviation of
0.1 ohm underestimates the true error. That is, fitting the data to a χ2 value of 1 in this case
would overfit the data, cause artifacts, and thus yield less meaningful inversion results.
However, the similarity between the tomograms for the same RMS error value suggests
that the constant error estimate is useful as a relative data weight in this case.

Figure 19 - Effects of data weighting on the inverse solution. a) Data standard deviations are estimated with
Equation 11 using a = 0.15 and b = 0.10. b) Data standard deviations are set to a constant value of
0.10 ohm. Each inversion was fit to approximately the same RMS value (Equation 13b). The χ2 large value
(Equation 13a) of inversion B suggests the data errors used for B underestimate the actual field-based
errors. The differences between the inversions in a) and b) are due to the different emphasis placed on
different data by the data weighting.

Inverse results also can be affected significantly by the regularization scheme,

particularly in cases where the data are noisy and/or the estimated parameters are not well
constrained by the data. That is, the effects of regularization are dependent on the ability
of the data to resolve the electrical conductivity in all parts of the model, with resolving
power varying spatially (e.g., Figure 4). Regularization will have a stronger influence on
parts of the model less resolved by the data. To demonstrate the effects of regularization,
we show three inversions with different regularization schemes in Figure 20. Although
there are significant differences, the major features of the inversion are similar, suggesting
these features are relatively well resolved by the data. It is good practice to invert data with
several different regularization schemes, as this gives insight into what features are
constrained by the data and what features may be controlled by regularization.

48
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Determination of which inversion features are constrained by the data and which are likely
regularization-driven can also be made through model appraisal. Most model appraisal
approaches require computation of the resolution matrix (Equation 15), which is only
feasible for relatively small inverse problems. However, an indication of resolution is given
by the sensitivity vector S (Equation 17), which shows the overall sensitivity of each region
of the model to the data. The sensitivity vector S for the inverse solution in Figure 17b is
shown in Figure 21. Note that the sensitivity is greater near the electrodes and in lower
electrical conductivity regions. That is, a change in electrical conductivity in the resistive
part of the model will have a greater influence on the data than a change in electrical
conductivity in the conductive part.

Figure 20 - a) Inverse solution with isotropic nearest-neighbor smoothing constraints. b) Inversion with
anisotropic nearest-neighbor smoothing constraints, encouraging homogeneity in the horizontal plane (i.e.,
layered structure). c) Hybrid regularization that promotes blocky structure at sharp conductivity boundaries.
Data were weighted equivalent to the weighting used in Figure 14 and fit to a normalized χ2 of ~1.0 and an
RMS value of ~0.30 in each case.

49
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Figure 21 - Cumulative squared sensitivity distribution (S in

Equation 17) for the inverse solution in Figure 17b. Larger sensitivities
are found near electrodes and in lower electrical conductivity regions.

Time-Lapse Inversions
For the time-lapse inversions, the data weighting and convergence criteria for each
time-lapse data set were equivalent to that shown in Figure 17b. However, the
regularization constraints were modified to encourage temporal changes in bulk
conductivity with respect to the baseline inversion (Figure 17b) to vary smoothly in space.
To implement those constraints, Equation 18a was added to the regularization constraint
matrix, with one equation for each neighboring pair of elements in the computational mesh.

𝜖(𝑚𝑖,𝑡 − 𝑚𝑖,𝑟𝑒𝑓 ) = 𝜖(𝑚𝑗,𝑡 − 𝑚𝑗,𝑟𝑒𝑓 ) (18a)

Here, mi,t is the log conductivity of computational mesh element i at time t, mj,t is a
neighboring element, mi,ref represents the pre-injection, baseline log conductivity of element
i, and ϵ represents the regularization weighting parameter. Noting that:

𝑚𝑖,𝑡 = 𝑚𝑖,𝑡−1 + ∆𝑚𝑖,𝑡 (18b)

(and similarly for mj,t) Equation 18a can be re-arranged and written as Equation 18c.

50
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

𝜖(∆𝑚𝑖,𝑡 − ∆𝑚𝑗,𝑡 ) = 𝜖(𝑚𝑖,𝑟𝑒𝑓 − 𝑚𝑗,𝑟𝑒𝑓 + 𝑚𝑖,𝑡−1 − 𝑚𝑗,𝑡−1 ) (18c)

The left- and right-hand side of Equation 18c are incorporated into the matrix
equations in the left- and right-hand sides of the inversion formulation in Equation 10a,
respectively.
Figure 22 shows four representative examples of the time-lapse inversion results
spanning approximately one year. The elevated fluid conductivity of the amendment
injected at I1 compared to I2 is immediately evident on day 5, as the increase in bulk
conductivity at I1 is greater than at I2. Over time, the amendment plume spreads laterally,
which is important for effective coverage and treatment. After day 93, the plume sinks and
spreads over the lower flow-bounding unit, likely due to density-driven flow. By day 371,
the plume is diluted relative to day 93, as expressed by a decrease in bulk conductivity.

Figure 22 - Example time-lapse inversions showing the changes in bulk conductivity

caused by the presence of amendment. After injection, the amendment experienced
density-driven downward flow and lateral flow over the lower bounding unit contact
at approximately 10 m depth (Figure 17b). The progression from day 93 to day 371
shows the plume sinking and diluting.
51
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

6 Overview and Future Directions

Instrumentation and software for electrical imaging are evolving rapidly, as are the
applications to hydrogeology. Advances in multi-channel systems now allow for the
collection of large 3-D datasets from systems involving hundreds of electrodes, and
advances in computing and inversion software allow analysis of such datasets on desktop
computers, which opens up possibilities for these methods to be used in more
hydrogeologic systems.
Despite these advances, experimental design and effective data analysis require
considerable care and scientific insight to ensure meaningful results and interpretation for
hydrogeologic processes and parameters. Practitioners are faced with numerous choices
for hardware settings (e.g., stacking error cutoffs, applied current, pulse duration) and
inversion settings (e.g., regularization, measurement weights), many of which can strongly
affect reconstructed images. In cases where quantitative information is to be extracted from
tomograms (i.e., rock physics models are applied to tomograms, estimation of hydrologic
or geochemical processes changing through time), careful selection of inversion parameters
is critical. Data overfitting can result in spurious structures and unrealistic estimates of
geophysical and, thus, unrealistic hydrogeologic parameters of interest. On the other hand,
data underfitting may result in tomograms that underpredict the degree of spatial
variability and, thus, the variability of hydrogeologic parameters. Although multiple
strategies to prevent overfitting and underfitting (e.g., L-curve, Occam inversion, and GCV)
are discussed in the literature, application of these techniques is less common in practice
and not supported by all commercially available electrical imaging inversion software.
Because inversion settings can strongly affect resulting tomograms, it is important to
document and justify choices. Without such documentation, reproduction of results is
problematic.
The goal of this book is to demonstrate and document best practices for electrical
imaging data collection and analysis for hydrogeology students and practitioners. In
summary, we provide guidelines in seven areas:
1) Survey geometry design: Numerical modeling, i.e., synthetic experiments or
‘pre modeling’ (e.g., Terry et al., 2017), should be employed prior to collecting
geophysical field data to determine the best survey geometry, based on a best
estimate of subsurface heterogeneity or processes of interest and the amount of
time available to collect data (a particular issue when monitoring time-lapse
processes, where temporal smearing will need to be minimized). Note that the
maximum offset between electrodes is limited by the power of the electrical
imaging unit and the unit’s ability to inject sufficient current to achieve a good
signal-to-noise ratio. Good quadripoles can be selected, in part, by choosing
geometries where the geometric factor is small. If boreholes are to be used, they
should be spaced such that electrode strings are at least 1.5 times as long
52
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

vertically as their horizontal separation distance. Survey geometries should

capitalize on the sensitivity of each measurement and maximize the coverage of
the tomogram.
2) Standard procedures for data collection: Data collection and quality assurance
and control should be documented using standardized forms and procedures
that include how field equipment was set up and deployed, electrode locations,
how errors and topography were measured in the field, weather, battery
voltages, filenames, contact resistances and the locations of any infrastructure
that could influence the measurements.
3) Quantification of measurement error: Stacked (at minimum), reciprocal, and/or
repeated errors should be collected in the field to assess the quality of the data,
inform the editing of datasets, and calculate minimum measurement weights
for the inversion.
4) Selection of inversion parameters: Existing data and (hydro)geologic insight
should be used to inform selection of inversion parameters or to develop prior
information to inform the inversion process, like imposing known layers or
contacts. For 2-D datasets involving multiple planes, it is useful to apply
multiple approaches to data from one plane, compare results, and design a
consistent approach to inversion for data from across the site. Investigation of
alternative inversion settings can aid in distinguishing artifacts from
hydrogeologic features.
5) Checks on inversion results: Tomograms should be evaluated for likely
inversion artifacts and the effects of bad data. The practitioner should look at
the range of estimated electrical conductivity values for plausibility; pixelated
(checkerboard) appearance of tomograms; artifacts such as streaking,
anomalous blocks, or diagonal patterns; and goodness of fit and convergence of
the inversion.
6) Resolution assessment: Results of tomograms should be compared with the
sensitivity and/or resolution matrix to assess the general quality of the
tomogram.
7) Comparison to other information: Tomograms and our interpretations of them
should be considered given existing information, including hydrogeologic
maps, lithology, borehole logs, hydraulic tests, and even tables of expected
properties (given the wealth of published ER and IP data that exist for both
unconsolidated and consolidated materials). Borehole logs can provide
information to help interpret features seen in tomograms and to help correlate
estimated electrical properties with lithology and/or hydraulic properties.

53
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

7 Exercises
Exercise 1
Describe the difference between resistivity and resistance. How is apparent
resistivity different than both resistivity and resistance?
Click for solution to exercise 1

Exercise 2
Examine the mathematical parallels between Darcy’s law and Ohm’s law.
Click for solution to exercise 2

Exercise 3
Select and describe the hydrologic equivalent of a:
a. Dead battery?
b. Short circuit?
c. Open circuit?
d. Voltmeter?
Click for solution to exercise 3

Exercise 4
Sketch the field layout for a 1D resistivity sounding, showing how electrodes are
placed and moved to vary survey depth.
Click for solution to exercise 4

Exercise 5
Calculate the apparent resistivity in the following arbitrary electrode configuration
when a 5 mA current is injected between electrodes A and B and 80 mV is measured
between M and N.

Click for solution to exercise 5

54
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Exercise 6
Calculate the geometric factors, measured voltage differences and signal-to-noise
ratio (SNR, the ratio of the field measured voltage difference to the noise) for a Wenner and
dipole-dipole array based on the information given (fill out the table below).
Current injected [I]: 2 mA
Apparent resistivity [ρ]: 250 ohm-m
Noise level: 15 mV

Array a(m) n Kg V(mV) SNR

Wenner 5 -

Dipole-dipole 3 3

Click for solution to exercise 6

Exercise 7
Explain, with the help of an illustration, how the ER inversion process works.
Highlight [1] the starting model, [2] the field data, and [3] the decision to update the model.
Click for solution to exercise 7

Exercise 8
Hypothesize how the map of inverted conductivity of the earth below an array of
electrodes is likely to deviate from reality. Explain factors that cause this deviation.
Click for solution to exercise 8

55
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

8 References
Barker, R.D, 1989, Depth of investigation of collinear symmetrical four-electrode arrays.
Geophysics, volume 54, pages 1031-1037, doi: 10.1190/1.1442728.
Binley, A., G. Cassiani, R. Middleton, and P. Winship, 2002, Vadose zone flow model
parameterization using cross-borehole radar and resistivity imaging. Journal of
Hydrology, volume 267, pages 147-159, doi: 10.1016/S0022-1694(02)00146-4.
Binley, A., A. Ramirez, and W. Daily, 1995, Regularized image reconstruction of noisy
electrical resistance tomography data, in Proceedings of the 4th Workshop of the
European Concerted Action on Process Tomography, editors, M.S. Beck et al.,
Bergen, Norway, pages 401-410.
Binley, A. and L. Slater, 2020, Resistivity and Induced Polarization: Theory and
Applications to the Near-Surface Earth. Cambridge University Press, 408 pages, ISBN
9781108492744.
Blanchy, G., S. Saneiyan, J. Boyd, P. McLachlan and A. Binley, 2020, ResIPy, an intuitive
open-source software for complex geoelectrical inversion/modeling. Computers &
Geosciences, volume 137, page 104423, doi: 10.1016/j.cageo.2020.104423.
Caterina, D., T. Hermans and F. Nguyen, 2014, Case studies of incorporation of prior
information in electrical resistivity tomography: comparison of different approaches.
Near Surface Geophysics, volume 12, issue 4, pages 451-465, doi: 10.3997/1873-
0604.2013070.
Claerbout, J.F. and F. Muir, 1973, Robust modeling with erratic data. Geophysics,
volume 38, issue 5, pages 826–844, doi: 10.1190/1.1440378.
Cockett, R., S. Kang, L.J. Heagy, A. Pidlisecky and D.W. Oldenburg, 2015, SimPEG: An
open-source framework for simulation and gradient based parameter estimation in
geophysical applications. Computers & Geosciences, volume 85, part A,
pages 142-154, doi: 10.1016/j.cageo.2015.09.015.
Constable, S.C., R.L. Parker and C.G. Constable, 1987, Occam's inversion: a practical
algorithm for generating smooth models from electromagnetic sounding data.
Geophysics, volume 52, issue 3, pages 289-300, doi: 10.1190/1.1442303.
Dahlin, T. and V. Leroux, 2012, Improvement in time‐domain induced polarization data
quality with multi‐electrode systems by separating current and potential cables. Near
Surface Geophysics, volume 10, issue 6, pages 545-565, doi: 10.3997/1873-
0604.2012028.
Dahlin, T. and M.H. Loke, 2015, Negative apparent chargeability in time-domain induced
polarization data. Journal of Applied Geophysics, volume 123, pages 322-332,
doi: 10.1016/j.jappgeo.2015.08.012.
Daily, W., and E. Owen, 1991, Cross-borehole resistivity tomography. Geophysics,
volume 56, number 8, pages 1228-1235, doi: 10.1190/1.1443142.

56
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Daily, W. and A. Ramirez, 1995, Electrical resistance tomography during in-situ

trichloroethylene remediation at the Savannah River Site. Journal of Applied
Geophysics, volume 33, issue 4, pages 239-249, doi: 10.1016/0926-9851(95)90044-6.
Day-Lewis, F.D., Y. Chen, and K. Singha, 2007, Moment inference from tomograms.
Geophysical Research Letters, volume 34, issue 22, doi: 10.1029/2007GL031621.
Day-Lewis, F.D., J.W. Lane Jr., J.M. Harris, and S.M. Gorelick, 2003, Time-lapse imaging of
saline tracer tests using cross-borehole radar tomography. Water Resources Research,
volume 39, issue 10, page 1290, doi:10.1029/2002WR001722.
Day-Lewis, F.D., K. Singha, and A.M. Binley, 2005, Applying petrophysical models to
radar traveltime and electrical resistivity tomograms – Resolution-dependent
limitations. Journal of Geophysical Research, volume 110, issue B8,
doi:10.1029/2004JB003569.
Day-Lewis, F.D., L.D. Slater, J. Robinson, C.D. Johnson, N. Terry, and D. Werkema, 2017,
An overview of geophysical technologies appropriate for characterization and
monitoring at fractured-rock sites. Journal of Environmental Management,
volume 204, pages 709-720, doi: 10.1016/j.jenvman.2017.04.033.
Day-Lewis, F.D., E.A. White, C.D. Johnson, J.W. Lane Jr., and M. Belaval, 2006,
Continuous resistivity profiling to delineate submarine groundwater discharge—
Examples and limitations. The Leading Edge, volume 25, issue 6, pages 724-728,
doi: 10.1190/1.2210056.
Dey, A., and H.F. Morrison, 1979, Resistivity modeling for arbitrarily shaped
three-dimensional structures. Geophysics, volume 44, issue 4, pages 753-780,
doi: 10.1190/1.1440975.
Farquharson, C.G. and D.W. Oldenburg, 1998, Non-linear inversion using general
measures of data misfit and model structure. Geophysical Journal International,
volume 134, issue 1, pages 213-227, doi: 10.1046/j.1365-246x.1998.00555.x.
Farquharson, C.G., and D.W. Oldenburg, 2004, A comparison of automatic techniques for
estimating the regularization parameter in non-linear inverse problems. Geophysical
Journal International, volume 156, issue 3, pages 411-425, doi: 10.1111/j.1365-
246X.2004.02190.x.
Flores Orozco, A., J. Gallistl, M. Bücker and K.H. Williams, 2018, Decay curve analysis for
data error quantification in time-domain induced polarization imaging. Geophysics,
volume 83, issue 2, pages 1MA-Z8, doi: 10.1190/geo2016-0714.1.
Freeze, R.A. and J.A. Cherry, 1979, Groundwater. Prentice Hall, pages 360-364, 604,
https://gw-project.org/books/groundwater/.
Gouveia, W.P. and J.A. Scales, 1997, Resolution of seismic waveform inversion–Bayes
versus Occam. Inverse Problems, volume 13, pages 322-349,
https://iopscience.iop.org/article/10.1088/0266-5611/13/2/009/meta.

57
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Griffiths, D.H., J. Turnbull, and A.I. Olayinka, 1990, Two-dimensional resistivity mapping
with a computer-controlled array. First Break, volume 8, issue 4, pages 121–129,
doi: 10.3997/1365-2397.1990008.
Haber, E. and D. Oldenburg, 2000, A GCV based method for nonlinear ill-posed
problems. Computational Geosciences, volume 4, pages 41-63,
doi: 10.1023/A:1011599530422.
Hallof, P.G., 1957, On the interpretation of resistivity and induced polarization
measurements. Doctor of Philosophy Dissertation, Massachusetts Institute of
Technology, http://hdl.handle.net/1721.1/54400.
Hansen, P.C., and D.P. O'Leary, 1993, The use of the L-curve in the regularization of
discrete ill-posed problems. Society for Industrial and Applied Mathematics (SIAM)
Journal of Scientific Computing, volume 14, issue 6, pages 1487-1503,
doi: 10.1137/0914086.
Hermans, T., A. Kemna and F. Nguyen, 2016, Covariance-constrained difference
inversion of time-lapse electrical resistivity tomography data. Geophysics, volume 81,
issue 5, pages 311-E322, doi: 10.1190/geo2015-0491.1#.
Hördt, A., P. Weidelt and A. Przyklenk, 2013, Contact impedance of grounded and
capacitive electrodes. Geophysical Journal International, volume 193, issue 1,
pages 187-196, doi: 10.1093/gji/ggs091.
Johnson, T.C., R.J. Versteeg, A. Ward, F.D. Day-Lewis, and A. Revil, 2010, Improved
hydrogeophysical characterization and monitoring through parallel modeling and
inversion of time-domain resistivity and induced polarization data. Geophysics,
volume 75, issue 4, pages 1JA-Z98, doi: 10.1190/1.3475513.
Johnson, T.C., R.J. Versteeg, F.D. Day‐Lewis, W. Major, and J.W. Lane Jr., 2014, Time‐lapse
electrical geophysical monitoring of amendment‐based biostimulation. Groundwater,
volume 53, issue 6, pages 920-932, doi: 10.1111/gwat.12291.
Jung, H. K., D.J. Min, H.S. Lee, S. Oh, and H. Chung, 2009. Negative apparent resistivity
in dipole-dipole electrical surveys. Exploration Geophysics, volume 40, issue 1,
pages 33-40, doi: 10.1071/EG08111.
Kitanidis, P.K., 1997, The minimum structure solution to the inverse problem. Water
Resources Research, volume 33, issue 10, pages 2263-2272, doi: 10.1029/97WR01619.
LaBrecque, D. and W. Daily, 2008, Assessment of measurement errors for
galvanic-resistivity electrodes of different composition. Geophysics, volume 73,
issue 2, pages 1MA-Z29, https://doi.org/10.1190/1.2823457.
LaBrecque, D.J., M. Miletto, W. Daily, A. Ramirez, and E. Owen, 1996, The effects of noise
on Occam’s inversion of resistivity tomography data. Geophysics, volume 61, issue 2,
pages 538-548, doi: 10.1190/1.1443980.
LaBrecque D.J., X. Yang, 2001, Difference inversion of ERT Data: A fast inversion method
for 3-D in situ monitoring. Journal of Environmental and Engineering Geophysics,
volume 6, issue 2, pages 83–89, doi: 10.4133/JEEG6.2.83.
58
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Lesparre, N., F. Nguyen, A Kemna, T. Robert, T Hermans, M. Daoudi, and A. Flores

Orozco, 2017, A new approach for time-lapse data weighting in ERT. Geophysics,
volume 82, issue 6, pages 325-E333, doi: 10.1190/geo2017-0024.1.
Linde, N, 2014, Falsification and corroboration of conceptual hydrological models using
geophysical data. WIREs Water, volume 1, pages 151-171. doi: 10.1002/wat2.1011.
Loke, M.H., R.I. Acworth, and T. Dahlin, 2003, A comparison of smooth and blocky
inversion methods in 2D electrical imaging surveys. Exploration Geophysics,
volume 34, pages 182-187, doi: 10.1071/EG03182.
Loke, M.H. and R.D. Barker, 1996, Rapid least-squares inversion of apparent resistivity
pseudosections by a quasi-Newton method. Geophysical Prospecting, volume 44,
issue 1, pages 131–152, doi: j.1365-2478.1996.tb00142.x.
Luo, Y. and G. Zhang, 1998, Forward Theory of Spectral Induced Polarization, in Theory
and Application of Spectral Induced Polarization. Society of Exploration
Geophysicists, volume 2, pages 13-90, doi: 10.1190/1.9781560801856.ch2.
Lytle, R.J. and K.A. Dines, 1978, Impedance camera: A system for determining the spatial
variation of electrical conductivity, Report No. UCRL-52413. Lawrence Livermore
Lab., page 11, https://www.osti.gov/servlets/purl/6753686.
Menke, W., 1984, The resolving power of cross-borehole tomography: Geophysical
Research Letters, volume 11, issue 2, pages 105-108, doi: 10.1029/GL011i002p00105.
Morris, G., A.M. Binley, and R.D. Ogilvy, 2004, Comparison of different electrode
materials for induced polarization measurements. Symposium on the Application of
Geophysics to Environmental & Engineering Problems (SAGEEP2004), volume 17,
pages 573, doi: 10.4133/1.2923371.
Mwakanyamale, K., L. Slater, A. Binley, and D. Ntarlagiannis, 2012, Lithologic imaging
using complex conductivity: Lessons learned from the Hanford 300 Area.
Geophysics, volume 77, issue 6, pages E397-E409, doi: 10.1190/geo2011-0407.1.
Nguyen, F., A. Kemna, T. Robert, and T. Hermans, 2016, Data-driven selection of the
minimum-gradient support parameter in time-lapse focused electrical imaging.
Geophysics, volume 81, issue 1, pages 1JF-Z7, https://doi.org/10.1190/geo2015-
0226.1.
Oldenburg, D.W. and Y. Li, 1999, Estimating depth of investigation in dc resistivity and IP
surveys. Geophysics, volume 64, issue 2, pages 403–416, doi: 10.1190/1.1444545.
Pidlisecky, A., E. Haber, and R. Knight, 2007, RESINVM3D: A MATLAB 3-D Resistivity
Inversion Package, Geophysics, volume 72, issue 2, pages H1-H10,
doi: 10.1190/1.2402499.
Roy, A. and A. Apparao, 1971, Depth of investigation in direct current methods.
Geophysics, volume 36, issue 5, pages 943-959, doi: 10.1190/1.1440226.
Rücker, C. and T. Günther, 2011, The simulation of finite ERT electrodes using the
complete electrode model. Geophysics, volume 76, issue 4, pages 1JA-Z99,
doi: 10.1190/1.3581356.
59
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Rücker, C., T. Günther, and F.M. Wagner, 2017, pyGIMLi: An open-source library for
modelling and inversion in geophysics. Computers & Geosciences, volume 109,
pages 106-123, doi: 10.1016/j.cageo.2017.07.011.
Singha, K., F.D. Day-Lewis, T. Johnson and L.D. Slater, 2014, Advances in interpretation
of subsurface processes with time-lapse electrical imaging. Hydrological Processes,
page 28. doi: 10.1002/hyp.10280.
Singha, K. and S.M. Gorelick, 2005, Saline tracer visualized with electrical resistivity
tomography–Field scale moment analysis: Water Resources Research, volume 41,
W05023, doi: 10.1029/2004WR003460.
Slater, L., 2007, Near surface electrical characterization of hydraulic conductivity: from
petrophysical properties to aquifer geometries - a review, Surveys in Geophysics,
volume 28, pages 169-197.
Slater, L., A.M. Binley, W. Daily and R. Johnson, 2000, Cross-hole electrical imaging of a
controlled saline tracer injection. Journal of Applied Geophysics, volume 44,
issues 2-3, pages 85-102, doi: 10.1016/S0926-9851(00)00002-1.
Slater, L. and A. Binley, 2006, Synthetic and field-based electrical imaging of a zerovalent
iron barrier: Implications for monitoring long-term barrier performance. Geophysics,
volume 71, issue 5, pages 1SO-viii, doi: 10.1190/1.2235931.
Slater, L.D. and D. Lesmes, 2002, IP interpretation in environmental investigations.
Geophysics, volume 67, issue 1, pages 77-88, doi: 10.1190/1.1451353.
Slater, L.D., D. Ntarlagiannis, F.D. Day‐Lewis, K. Mwakanyamale, R.J. Versteeg, A. Ward,
C. Strickland, C.D. Johnson, and J.W. Lane Jr., 2010, Use of electrical imaging and
distributed temperature sensing methods to characterize surface water–groundwater
exchange regulating uranium transport at the Hanford 300 Area, Washington. Water
Resources Research, volume 46, issue 10, doi: 10.1029/2010WR009110.
Stummer, P., H. Maurer, and A.G. Green, 2004, Experimental design – Electrical
resistivity datasets that provide optimum subsurface information. Geophysics,
volume 69, issue 1, doi:10.1190/1.1649381.
Sumner, J.S., 1976, Principles of Induced Polarization for Geophysical Exploration.
Elsevier, New York, page 277, ISBN 9780444599872.
Tarantola, A., 1987, Inverse problem theory – Methods for data fitting and model
parameter estimation. Elsevier, Amsterdam, page 613, ISBN 978-0444427656.
Terry, N., F.D. Day‐Lewis, J.L. Robinson, L.D. Slater, K. Halford, A. Binley, J.W. Lane, and
D. Werkema, 2017, Scenario evaluator for electrical resistivity survey pre‐modeling
tool. Groundwater, volume 55, issue 6, pages 885-890, doi: 10.1111/gwat.12522.
Tikhonov, A.N. and V.Y. Arsenin, 1977, Solutions of ill-posed problems. John Wiley &
Sons, New York, page 258, ISBN 978-0470991244.
Tripp, A.C., G.W. Hohmann, and C.M. Swift Jr., 1984, Two-dimensional resistivity
inversion. Geophysics, volume 49, issue 10, pages 1708-1717, doi: 10.1190/1.1441578.

60
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

United States Air Force, 2006, Interim Record of Decision, Site SS-01, Brandywine DRMO
Andrews Air Force Base, Maryland, page 112.
Vasco, D.W., A. Datta-Gupta, and J.C.S. Long, 1997, Resolution and uncertainty in
hydrologic characterization. Water Resources Research, volume 33, issue 3,
pages 379-397, doi: 10.1029/96WR03301.
Wang, C., A. Binley, L.D. Slater, 2020, On negative induced polarization in frequency
domain measurements. Geophysical Journal International, volume 225, issue 1,
pages 342-353, doi: 10.1093/gji/ggaa581.
Wilkinson, P. B., J.E. Chambers, M. Lelliott, G.P. Wealthall, and R.D. Ogilvy, 2008,
Extreme sensitivity of crosshole electrical resistivity tomography measurements to
geometric errors. Geophysical Journal International, volume 173, issue 1, pages 49-62,
doi: 10.1111/j.1365-246X.2008.03725.x.
Xu, B. and M. Noel, 1993, On the completeness of data sets with multielectrode systems
for electrical resistivity survey. Geophysical Prospecting, volume 41, issue 6,
pages 791-801, doi: 10.1111/j.1365-2478.1993.tb00885.x.
Zarif, F., P. Kessouri, and L. Slater, 2017, Recommendations for field-scale induced
polarization (IP) data acquisition and interpretation. Journal of Environmental and
Engineering Geophysics, volume 22, issue 4, pages 395-410,
doi: 10.2113/JEEG22.4.395.
Zhody, A.A.R., G.P. Eaton, and D.R. Mabey, 1974, Application of surface geophysics to
ground-water investigations. U.S. Geological Survey Techniques of Water-Resources
Investigations, book 2, chapter D1, page 86, https://pubs.usgs.gov/twri/twri2-d1/.

61
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

9 Boxes
Box 1 Scenario Evaluator for Electrical Resistivity (SEER)
As discussed in Section 2, synthetic modeling provides useful insight for 1)
designing geophysical surveys, and 2) understanding the ability of a given survey to
resolve a hypothetical target. Synthetic modeling provides a basis for ‘go/no-go’ decisions
on geophysical field campaigns, i.e., whether a survey can adequately resolve expected
targets so is worth the expense. Many public-domain and commercially available off-the-
shelf software packages for electrical imaging provide synthetic modeling capabilities. In
this exercise, we consider the public-domain tool SEER, the Scenario Evaluator for Electrical
Resistivity. Available for free from the U.S. Geological Survey, SEER is a spreadsheet-based
tool for synthetic modeling of resistivity experiments. SEER is user friendly but limited in
its functionality.
To get started with the SEER software, proceed through the steps described in the
following bullets, then set up a simple model as described and continue with the activities.
• Download the zip file containing SEER from the USGS repository at
https://doi.org/10.5066/F7028PQ1. The README file on the web page provides a
brief overview and background information about the software.
• Extract the contents of the zip file into a folder on your computer. This will produce
the SEER.xlsm spreadsheet, the SEERhelp.chm file which is an electronic user’s
manual that can be accessed by double clicking on the file, and a folder titled
“ResponseArrays” that contains the results of previously executed simulations.
There is also an instructional video that describes electrical resistivity imaging and
explains how SEER can be useful to designing electrical resistivity surveys.
• Open the spreadsheet. You will need to enable macros when requested to do so by
Excel. After reading the INTRO worksheet, navigate to the “Survey” worksheet.
There are four template models available from the Scenario drop-down menu in cell
B1 of the worksheet. These include 1) DNAPL pool, 2) LNAPL pool, 3) underground
storage tank (UST), and 4) block targets (BLOCKS); these can all be modified by the
user. In cells B4 through B7, the user may adjust survey parameters including the
number of electrodes, type of survey, measurement error levels, and whether
borehole electrodes will be used. Familiarize yourself with the operation of the
spreadsheet and the three template models, by making selections and clicking the
“Simulate” button.
To explore how to make a custom model by selecting a template and modifying the
parameters. Begin by selecting the UST template model. Starting from this template of a
layered system, you can create a homogeneous background. Be sure to unclick the ‘Using
specified scenario checkbox’ to enable a custom model. You will assess the ability of
electrical imaging to resolve a water filled cavity. Change the background for the UST

62
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

model to 500 ohm-m. Assume 1-m electrode spacing, 10% measurement error, and a dipole-
dipole survey. Next. explore what you can do with the SEER software through the
following activities.

Activity 1) Assume the electrical conductivity of the water-filled cavity is 200 micro-S/cm.
Convert the fluid conductivity to resistivity for input to the spreadsheet. (Hint: resistivity
is the reciprocal of conductivity, and 10 S is equivalent to 0.1 ohms.)
Solution to activity 1

Activity 2) Assume the cavity occupies the space from cells in columns AL to AO and rows
13 to 15, and assign the resistivity you calculated to these cells. Press the ‘Simulate’ button
and the spreadsheet will produce the predicted inversion result for your hypothetical
model. Evaluate how well your predicted inversion result compares to your true cavity
model. Consider its resolution of the top, sides, and bottom of the cavity, as well as how
well the estimated resistivity compares to the true resistivity.
Solution to activity 2

Activity 3) Explore how changing the measurement error to 1% and pressing the ‘Simulate’
button changes the result.
Solution to activity 3

Activity 4) Explore how changing the survey type to ‘Combined’ and pressing the
‘Simulate’ button changes the result.
Solution to activity 4

Activity 5) Explore how adding borehole electrodes and pressing the ‘Simulate’ button
changes the result.
Solution to activity 5

Having gained familiarity with how to change both subsurface properties and survey
parameters, we encourage you to explore other subsurface systems and combinations of
survey configurations. Given the level of importance of precisely defining subsurface
properties for a given project, simulating the resistivity survey before undertaking the
field work provides the information needed to design the field survey by balancing the
cost of conducting the survey against the expected resolution in the result. Of course, one
needs to have an estimate of the size, depth, and relative resistivity of the features of
interest.
Return to where the text linked to Box 1

63
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution to Activity 1
To calculate the cavity resistivity based on the conductivity of 200 micro-S/cm, first convert
to S/m.
200 𝑚𝑖𝑐𝑟𝑜⎻𝑆 1 𝑚𝑖𝑐𝑟𝑜⎻𝑆 100 𝑐𝑚 𝑆
= 0.02
𝑐𝑚 1𝑥106 𝑆 1𝑚 𝑚

1 1
𝑅𝑒𝑠𝑖𝑠𝑡𝑖𝑣𝑖𝑡𝑦 = = = 50 𝑜ℎ𝑚 − 𝑚 → 𝑡ℎ𝑒 𝑖𝑛𝑣𝑒𝑟𝑠𝑒 𝑜𝑓 𝑆𝑒𝑖𝑚𝑒𝑛𝑠 𝑖𝑠 𝑜ℎ𝑚
𝐶𝑜𝑛𝑑𝑢𝑐𝑡𝑖𝑣𝑖𝑡𝑦 0.02 𝑆
𝑚
Return to activity 1

Solution to Activity 2
The predicted inversion result is blurry and blunt compared to the true model. The top is
relatively well resolved, the sides less so, and the bottom is poorly resolved. The estimated
resistivity does not capture all of the contrast between the background resistivity (500 ohm-
m) and the cavity resistivity (50 ohm-m); the minimum value estimated inside the cavity is
about 370 ohm-m compared to 50 ohm-m.

Return to activity 2

Solution to Activity 3
By assuming a smaller measurement error of 1%, the general pattern is the same, but the
estimated resistivity is closer to the actual resistivity of the target, with a minimum
estimated resistivity of ~320 ohm-m.
Return to activity 3

64
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution to Activity 4
Changing the survey type to ‘Combined’ results in the same general pattern of estimated
resistivity, so one concludes that adding Wenner measurements alone does not improve
the results sufficiently to warrant the additional field effort.
Return to activity 4

Solution to Activity 5
The addition of borehole electrodes and associated crosshole measurements greatly
improves the resolution of the cavity. The boundaries of the target are more precisely
located, perhaps to +/- 1 m, and the magnitude of the anomaly is better captured, with an
estimated resistivity showing a minimum of about 100 ohm-m.

Return to activity 5

Return to where the text linked to Box 1

65
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

10 Exercise Solutions
Solution Exercise 1
Resistivity is an intrinsic property of a material and describes how difficult it is for
electrical current to flow through that material. Electrical resistivity is the reciprocal of
electrical conductivity. Electrical resistivity is therefore analogous to the reciprocal of
hydraulic conductivity (K).
Resistance is also a measure of the difficulty with which electrical current can flow,
but it is dependent on geometry. It is similar to transmissivity, which is dependent on
hydraulic conductivity and layer thickness, but when resistance to current flow is
measured, it depends on resistivity of the material, the distance between the measurement
points, and the cross-sectional area through which current is flowing between those points.
In 1-D current flow (e.g., in a wire, or column of uniform properties), resistance is directly
proportional to length between the ends of the column and inversely proportional to the
cross-sectional area of the column. The equivalent hydraulic parameter would be the
hydraulic resistance, or L/(KA), where L is the length of the flow path and A is the
cross-sectional area through which water flows. Consequently, because of this geometric
dependence, resistance values can be small or large—even in a system with homogeneous
resistivity.
Apparent resistivity applies a geometric factor to the measured resistance to
estimate what the resistivity of the system would be, assuming a homogeneous Earth
extends infinitely with depth and lateral distance. Only in the case of a homogeneous Earth
will the apparent resistivity equal the true resistivity. The utility of apparent resistivity is
that it allows a practitioner to look at their field data to see if the values are reasonable and
to rapidly locate anomalies. Inversion of the data is required to obtain actual resistivity
values.
Return to Exercise 1

66
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution Exercise 2
Darcy’s law is defined by q = –K∇h, where:
q = specific discharge (length/time)
∇h = head gradient (length/length)
K = hydraulic conductivity (length/time)
Ohm’s law is defined by J = –σ∇V, where:
J = current density (Ampere/length2)
∇V = voltage gradient (voltage/length)
σ = electrical conductivity (Seimens/length)
The electrical current density quantifies the flow of electric charge and is thus
equivalent to the flow of water. Head and voltages are both measures of potential (or
energy). Electrical conductivity controls how easily current can flow through the medium,
like water with hydraulic conductivity. When combined with conservation of mass or
charge, both of these equations become diffusion equations.
Return to Exercise 2

67
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution Exercise 3
a. Dead battery: Batteries are sold by the potential difference they maintain and by
the amount of electricity (charge) they can deliver. This is equivalent to the
height of water behind a dam, and the volume of water that can be released. A
dead battery is thus like an empty reservoir.
b. Short circuit: A short circuit occurs when there is a highly electrically conductive
pathway across a circuit, shortcutting current flow. The closest analogy in
hydrology might be a conductive fracture, which could shortcut flow across an
otherwise extremely low permeability rock.
c. Open circuit: An open circuit exists when electric current cannot traverse the
entire circuit e.g., if there is a break in the power cord leading to a hair dryer.
This would be equivalent to a truly impermeable block of material that water
could not move through.
d. Voltmeter: A voltmeter measures a drop in electrical potential in an electric
circuit. Heads are the equivalent potential term in hydrology and are measured
by a water-level tape (physical or electrical), pressure transducer, a sonic device,
or something similar. Of course, a voltmeter and a water-level tape aren’t
entirely equivalent. A water-level tape just measures the head at one location.
With electricity, we are always looking at differences in voltage. A voltmeter is
the electrical equivalent of measuring the head difference between two wells.
While in both cases, differences in potential are what drive flow, with heads, we
can set a datum and make individual measurements relative to the datum, while
with current we need to capture the voltage difference in one measurement.
Return to Exercise 3

68
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution Exercise 4
Field layout for a 1D resistivity sounding:

and so on, with incrementally larger distances.

Return to Exercise 4

69
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution Exercise 5
Calculate the apparent resistivity in the following arbitrary electrode configuration when a
5 mA current is injected between electrodes A and B, and 80 mV is measured between M
and N.
2𝜋
𝐾𝑔 =
1 1 1 1
̅̅̅̅̅ − ̅̅̅̅
𝐴𝑀 𝐴𝑁 𝐵𝑀
− ̅̅̅̅̅ + ̅̅̅̅
𝐵𝑁

2𝜋
=
1 1 1 1
2𝑚 − 3𝑚 − 4𝑚 + 3𝑚

= 25.1 m

ρa = Kg(ΔV/I)
= 25.1 m * 80 mV / 5 mA
= 400 ohm-m
Return to Exercise 5

70
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution Exercise 6
Current injected [I]: 2 mA
Apparent resistivity [ρ]: 250 ohm-m
Noise level: 15 mV
2𝜋
𝐾𝑔 =
1 1 1 1
̅̅̅̅̅ − ̅̅̅̅
𝐴𝑀 𝐴𝑁 𝐵𝑀
− ̅̅̅̅̅ + ̅̅̅̅
𝐵𝑁

2𝜋
𝐾𝑔,𝑊𝑒𝑛𝑛𝑒𝑟 =
1 1 1 1
− − +
5𝑚 10𝑚 10𝑚 5𝑚

= 31.4 m

For a positive geometric factor with a dipole-dipole array, you need to have the electrodes
configured as A B N M (this also gives a positive resistance). If you configure the electrodes
as A B M N (as often written) the geometric factor is negative (so is the resistance). Using A
B N M the geometric factor is:

2𝜋 2𝜋
𝐾𝑔,𝑑𝑖𝑝𝑜𝑙𝑒−𝑑𝑖𝑝𝑜𝑙𝑒 = = = 56.5 𝑚
1 1 1 1 1 1 1 1
̅̅̅̅̅ − − + − − +
𝐴𝑀 ̅̅̅̅ 𝐴𝑁 ̅̅̅̅̅
𝐵𝑀 ̅̅̅̅
𝐵𝑁 9𝑚 6𝑚 6𝑚 3𝑚

If you do the calculation as A B M N, you get a geometric factor of -56.5.

We then need to calculate the apparent resistivity:

ρa = KgΔV/I
31.4 m 𝑉 mV
ρa_Wenner = 250 ohm-m = 2 mA
solving for V yields V = 15.9 mV

56.5 m 𝑉 mV
ρa_dipole-dipole = 250 ohm-m = 2 mA
solving for V yields V = 8.8 mV
A negative voltage would be correct for A B M N configuration, as the geometric
factor is negative.
In summary:

Array a(m) n Kg V(mV) SNR

Wenner 5 - 31.4 15.9 1.06

Dipole-dipole 3 3 56.5 8.8 0.59

Return to Exercise 6
71
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Solution Exercise 7
The ER inversion process, involves estimating parameter values for the starting
model and simulating the field measurements, then comparing the simulated values to the
field data, and if there is a good fit then the inversion process is stopped and the current
parameter values are used, while if not, the parameter values are updated and the process
is repeated until an acceptable fit is reached.

Return to Exercise 7

Solution Exercise 8
There are multiple reasons why the model for the conductivity of the earth below
an array of electrodes is likely to deviate from reality. Four reasons are:
1. Measurement physics: electrical conduction is described by a diffusion
equation; small-scale heterogenieties are often not captured by these equations
2. Parameterization and regularization: as part of the inverse problem, the
mathematics describing electrical flow is discretized, leading to some errors,
and regularization is added to the inverse problem that often serves to smooth
the final image
3. Limited resolution: there is insufficient information in the data to capture all the
heterogeneity of the subsurface
4. Measurement errors: due to a variety of often unavoidable issues in the field
Return to Exercise 8

72
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

11 About the Authors

Dr. Kamini Singha is a Professor in the Department of Geology
and Geological Engineering and Associate Dean of Earth and
Society Programs at Colorado School of Mines in Golden,
Colorado, USA. She has a Ph.D. in Hydrogeology from the
Department of Geological and Environmental Sciences at
Stanford University in 2005 and a Bachelor of Science degree with
Honors in Geophysics from the Department of Geology and
Geophysics at the University of Connecticut in 1999.
Her research interests rely on fluid flow and solute transport in porous media and
fractured rock as well as groundwater-surface-water interactions; integration of hydrologic
and geophysical techniques; quantification of the “geophysical footprint” to constrain the
operation of parameters controlling fluid flow and solute. She has worked a wide variety
of multidisciplinary problems including imaging tree water use with electrical geophysics;
mapping water quality degradation and its connections to human behavior; imaging
moisture dynamics in desert ecosystems; exploring changes in infiltration after fire; and
quantifying kinetics of iron reduction given electrical measurements.
Dr. Singha is an award-winning teacher, a recipient of a U.S. National Science
Foundation CAREER award and the Early Career Award from the Society of
Environmental and Engineering Geophysics, a Geological Society of America Fellow, and
a former Fulbright Scholar. She served as the U.S. National Groundwater Association's
Darcy Lecturer in 2017, and is a current Editor at Water Resources Research.
She currently teaches Groundwater Engineering, Geological Data Analysis, and
Field Methods in Hydrology at the Colorado School of Mines. For a full, up-to-date list of
her publications, please click here.

Dr. Timothy C. Johnson is a computational scientist in subsurface

geophysical imaging and interpretation related to complex
environmental challenges and energy applications. He is
nationally and internationally recognized for his work in electrical
resistivity tomography (ERT) as a characterization and monitoring
technology. He pioneered the development of E4D-RT, a real-time,
four-dimensional subsurface imaging software that allows
scientists to “see” subsurface processes and solutions in real-time.
He and his team received a prestigious R&D 100 Award in 2016 for this tool. He joined
Pacific Northwest National Laboratory in 2010 as a senior research scientist. He worked at
Idaho National Laboratory from 2007-2010 and, before that, was a staff engineer at

73
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

American Geotechnics. He is focused on joint inversion of multiple geophysical techniques

using parallel computing to improve time-lapse imaging.
Dr. Johnson has a Doctorate degree in Geophysics in 2006, a Bachelor's degree in
Civil Engineering in 2001, and a Bachelor's degree in Geophysics in 1999, all from Boise
State University. He has received awards and recognitions, including: Laboratory
Director's Award for Exceptional Engineering Achievement in geophysical imaging, 2017;
R&D 100 Award for Real-Time Four-Dimensional Subsurface Imaging Software
(E4D-RT - (https://e4d.pnnl.gov) in 2016; and R&D Magazine honor for inventors of 100
most technologically significant products each year. He has the U.S. Patent No. 9,772,423,
September 26, 2017, "Method Of Imaging The Electrical Conductivity Distribution Of A
Subsurface”. The full list of his publications can be found here.
The bulk of Dr. Johnson’s research has been commissioned by the U.S. Department
of Energy and Department of Defense, focusing on developing joint geophysical methods
for investigating shallow subsurface contamination issues, for understanding stress and
fluid flow in deep crystalline rock systems, and for understanding the effects of
underground explosions.

Dr. Frederick D. Day-Lewis a Chief Geophysicist and Laboratory

Fellow with the Pacific Northwest National Laboratory’s Earth
Systems Science Division. He received a Ph.D. in Hydrogeology
from Stanford University (2001) and a Bachelor of Science and
Bachelor of Arts in Hydrology and English from the University of
New Hampshire (1994). Prior to starting at Pacific Northwest
National Laboratory, he was a Research Hydrologist at the U.S.
Geological Survey, where he developed and applied
hydrogeophysical methods in groundwater studies across the U.S. Dr. Day-Lewis conducts
applied research on characterization, monitoring, and understanding of hydrologic
processes, with current emphases on groundwater/surface-water interactions and
fractured rock. Recent activities include studies to improve the understanding of hyporheic
zone processes and measurements of properties controlling contaminant storage in
geologic materials. He has experience using electrical, radar, and electromagnetic
geophysical methods as well as distributed temperature sensing. In this work, he also
develops software for improved analysis and visualization of related data (1DTempPro,
DTSGUI, FLASH-R). As part of his technology transfer efforts, Dr. Day-Lewis collaborates
with state and federal partners including SERDP, ESTCP, DOE, EPA, and others to support
informed decision-making on the selection and application of hydrogeophysical tools and
methods. He also develops decision support tools (SEER, FRGT-MST) to help practitioners
and environmental site managers identify near-surface geophysical methods appropriate
for a given project’s goals and site conditions.

74
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Dr. Day-Lewis co-edited the American Geophysical Union monograph “Subsurface

Hydrology: Data Integration for Properties and Processes” (2007). He is a Fellow of the
Geological Society of America, past President of the American Geophysical Union's Near
Surface Geophysics Section, and past Vice President of the Environmental and Engineering
Geophysical Society. He has served as an Associate Editor for Water Resources Research,
Groundwater, Hydrogeology Journal, and Geosphere.

Dr. Lee D. Slater, after study in the United Kingdom, a

postdoctoral appointment, and initial tenure-track faculty
position in the U.S., joined Rutgers University in 2002 to continue
performing research in hydrogeophysics. He is currently a Henry
Rutgers Professor in Geophysics at Rutgers University Newark
and an internationally recognized expert in near-surface
geophysics and hydrogeophysics. His research is primarily
focused on the use of electrical geophysical technologies for
subsurface characterization and monitoring. He has published
extensively, including 150 papers in peer-reviewed international journals of hydrogeology
and geophysics, as well as being a co-author of the recent book “Resistivity and Induced
Polarization: Theory and Applications to the Near Surface Earth”.
Dr. Slater has served as principal investigator on multiple research and technology
demonstration projects funded by the U.S. Department of Defense, U.S. Department of
Energy, U.S. Department of Agriculture, U.S. National Parks Service, and National Science
Foundation. He has also served in prominent leadership roles in the academic geophysical
community, including Chair of the Near Surface Geophysics Focus Group of the American
Geophysical Union (AGU), Chair of the AGU Hydrogeophysics Technical Committee, and
President of the Environmental and Engineering Geophysical Society (EEGS).
Dr. Slater currently serves as Associate Editor of Water Resources Research (WRR)
and he recently edited a new volume on Near Surface Geophysics published in the 2nd
Edition of the Treatise on Geophysics, part of the Elsevier Major International Reference
series. He has served on multiple advisory boards for large interdisciplinary
hydrogeological research projects in Europe.
He was the recipient of the 2013 Harold B. Mooney award of the Society of
Exploration Geophysicists (SEG). His numerous Ph.D. graduates have mostly gone onto
academic positions and are now making their own contributions to advancing research in
hydrogeophysics. In 2018 Dr. Slater was elected Fellow of the American Geophysical Union
(AGU), with the citation: "for visionary experimentation in near-surface geophysics that
has advanced understanding of subsurface hydrogeological and biogeochemical
processes".

75
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Please consider signing up to the GW-Project mailing list and stay informed about new
book releases, events and ways to participate in the GW-Project. When you sign up to our
email list it helps us build a global groundwater community. Sign up.

76
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

Modifications to Original Release

Changes from the Original Version to Version 2

Original Version: December 2021, Version 2: August 2022

General changes:

Seven links to Figures in the original book showed an error message instead of the figure
number. These were corrected and the table of contents was updated to accommodate any
page number changes that occurred given the substitution of the figure label for the error
statement.

Changes from Version 2 to Version 3

Version 3: 9 March 2023

General changes:

removed Table of Figures as per updated Groundwater Project book format

major sections now start on a new page as per updated Groundwater Project book format

parameters that were not in italic font were changed to italic font

units that were in italic font were changed to standard font

updated Table of Contents

Specific Changes:

page numbers refer to the pdf of version 2

page ii, added Version 3

page iii, updated copyright format and added doi, updated number of pages, changed to
APA 7th edition format for citation, added “of” after “example in description of the cover
image

page ix, corrected two misspellings of imaging from imagining in the foreword

page 4, 1st paragraph, 4th line from bottom, units Ω.m corrected to Ω-m

A
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

page 5, after Equation 1, first letter of parameter definitions changed to lower case unless
proper nouns as per Groundwater Project book format

page 5, added units of meters to definition of xs, ys, zs

page 8, Equation 5, a close parenthesis ‘ ) ’ was added to the end of the righthand side

page 12, caption of Figure 4, J' is now defined as J's transpose

page 27, in equation 9a, CD-1 was corrected to CD-0.5

page 27, deleted ‘is the’ in definition of D

page 28, removed variable dsim,i from variable definitions because it does not appear in
Equation 10a or 10b

page 29, definitions after equation 11, variable a changed to ai

page 31, Equation 12b, 100* changed to 100%

page 34, end of third to last sentence of Section 4.5, Figure 2 changed to Figure 4 and the
errant image of Figure 2 at that location was removed

page 34, last sentence of Section 4.5, added () around 2014

page 54, cCaterina changed to Caterina

page 64, title of Section 10, Exercises changed to Exercise

page 65, definition of Darcy's Law, variable k changed to upper case K

page 65, definition of K dimensions changed from (length time) to (length/time)

page 69, added overbars to AM, AN, BM, BN in equation for Kg,dipole-dipole

page 69, added m after 56.5 at end of equation for Kg,dipole-dipole

page 69, presented ρa_Wenner and ρa_dipole-dipole as stacked fractions to facilitate reading

Changes from Version 3 to Version 4

Version 4: 15 March 2023

Specific Changes:

B
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.
Electrical Imaging Kamini Singha, Timothy C. Johnson, Frederick D. Day-Lewis, and Lee D. Slater

page numbers refer to the pdf of version 3

page ii, added Version 4

page 13, in the caption of Figure 4, changed:

and J’ is its transpose, which is
to:
and J’ is its transpose. The Jacobian matrix is

C
The GROUNDWATER PROJECT ©The Authors Free download from gw-project.org
Anyone may use and share gw-project.org links. Direct distribution of the book is strictly prohibited.