Modeling a Large-Scale Historic Aquifer Test:

Insight into the Hydrogeology of a Regional

Fault Zone
by Daniel R. Hadley1 , Daniel B. Abrams2 , and George S. Roadcap2

Abstract
Faults can act as flow barriers or conduits to groundwater flow by introducing heterogeneity in permeability. We examine the
hydrogeology of the Sandwich Fault Zone, a 137 km long zone of high-angle faults in northern Illinois, using a large-scale historic
aquifer test. The fault zone is poorly understood at depth due to the majority of the faults being buried by glacial deposits and its
near-vertical orientation which limits geologic sampling across faults. The aquifer test—perhaps one of the largest in terms of overall
withdrawal in North American history—was conducted in 1942 at a facility adjacent to the fault zone. More than 34,000 m3 /day
was pumped for 37 days from nine multiaquifer wells open to the stratified Cambrian-Ordovician sandstone aquifer system. We
modeled the aquifer test using a transient MODFLOW-USG model and simulated pumping wells with the CLN package. We tested
numerous fault core/damage zone conceptualizations and calibrated to drawdown values recorded at production and observation
wells. Our analysis indicates that the fault zone is a low-permeability feature that inhibits lateral movement of groundwater and that
there is at least an order of magnitude decrease in horizontal hydraulic conductivity in the fault core compared to the undeformed
sandstone. Large head declines have occurred north of the fault zone (over 300 m since predevelopment conditions) and modifying
fault zone parameters significantly affects calibration to regional drawdown on a decadal scale. The flow-barrier behavior of the
fault zone has important implications for future groundwater availability in this highly stressed region.

Introduction Fault zones typically consist of an inner fault

Faults affect groundwater movement in the shallow
crust by introducing variations in permeability. Faults
can act as groundwater flow barriers, conduits, or com-
plex barrier-conduit systems depending on the geologic
material, the type and amount of deformation, and sec-
ondary processes such as mineralization. Understanding
the hydrogeology of fault zones thus has important impli-
cations for analyzing groundwater flow patterns (Bense
and van Balen 2003; Minor and Hudson 2006; Mayer
et al. 2007; Bense et al. 2008; Burbey 2008) and for eval-
uation of regional groundwater supplies (Bredehoeft et al.
1992; Marler and Ge 2003; Roques et al. 2014).
1
Corresponding author: University of Illinois, Illinois State
Water Survey, 2204 Griffith Drive, Champaign, IL 61820; (217)
300-0402; drhadley@illinois.edu
2
University of Illinois, Illinois State Water Survey, 2204 Griffith
Drive, Champaign, IL 61820; dbabrams@illinois.edu; roadcap@
illinois.edu
Article impact statement: Modeling a large-scale historic
aquifer test determined the hydraulic conductivity of a deep vertical
fault zone in a multiaquifer system.
Received January 2019, accepted July 2019.
© 2019, National Ground Water Association.
doi: 10.1111/gwat.12922
core surrounded by a damage zone, both of which
have distinct hydrologic regimes (Caine et al. 1996;
Faulkner et al. 2010; Bense et al. 2013). Fault cores are
where most strain is accommodated and can consist of
cataclasite or gouge at the slip surface which can impede
groundwater flow normal to the fault, whereas damage
zones accommodate strain through fracturing and are areas
where preferential vertical flow may occur (Bense and van
Balen 2003; Minor and Hudson 2006; Caine and Minor
2009; Ball et al. 2010). However, for unlithified and
lithified siliciclastics (e.g., sandstones), strain is typically
accommodated by the formation of deformation bands (as
opposed to discrete slip surfaces or fractures) where grain
sliding and rotation occur, thus reducing permeability
(Antonellini and Aydin 1994; Shipton and Cowie 2001;
Fossen et al. 2007).
Many studies on the hydrogeology of fault zones take
place on the outcrop scale at individual faults or inves-
tigate fault zones at shallow depths (<100 m). Various
approaches include using field-measured permeability at
outcrops (Rawling et al. 2001; Caine and Minor 2009),
permeability measured from boreholes and cores drilled
through fault zones (Shipton et al. 2002; Nieto Camargo
and Jensen 2012), geophysical surveys to infer structure

NGWA.org Vol. 58, No. 3–Groundwater–May-June 2020 (pages 453–463) 453

Figure 1. (A) Map of hydrostratigraphic units at the bedrock surface and the Sandwich Fault Zone in northern Illinois. Dark
blue contours indicate static drawdown in the Cambrian-Ordovician sandstone system from pre-development conditions
(1863) to 2014. Additional static level decline has occurred since then (to over 300 m) at local areas within the center of the
cone of depression in Will County. Red inset box indicates study area. (B) Map of the study area showing the location of the
production and observation wells used in the Arsenal Test and contours of the top elevation of the St. Peter sandstone (m
above mean sea level) indicating the variable offset at the fault zone.

(Ball et al. 2010), or detailed study of fault geometries
and modeling of fracture networks (Jourde et al. 2002;
Minor and Hudson 2006). These studies determine varia-
tions in permeability for various fault components to infer
groundwater flow conditions on a larger scale.
A growing number of studies have used field
hydrogeological approaches (e.g., slug testing, aquifer
testing, and mapping of heads), analytical modeling, or
numerical modeling, to understand the effect of fault
zones on local and regional groundwater flow (see Bense
et al. 2013 for a complete review). For example, Marler
and Ge (2003) studied the permeability of the Elkhorn
Fault Zone in Colorado using slug testing at boreholes
across the trace of the fault and simulated flow using
two-dimensional numerical modeling. Ball et al. (2010)
expanded upon Marler and Ge’s study by incorporating
geophysical surveys to map heads across the fault zone
and the structure of the fault zone at depth. Both
studies indicate a complex conduit-barrier behavior of
the fault zone at a local scale (tens of meters). Mayer
et al. (2007) estimated conductance for a section of the
Mission Creek Fault Zone in California, which offsets
unconsolidated conglomerates and debris fan deposits, by
calibrating a numerical groundwater model to historic
head observations. Burbey (2008) utilized a 62-day
aquifer test to infer the influence of a small (5 m wide)
fault zone within a thick alluvial aquifer in Nevada.
Medeiros et al. (2010) also utilized pump tests to infer
hydraulic conductivity across deformation bands within
a shallow sandstone aquifer, but did not detect a strong
flow barrier despite large differences between host rock
and deformed rock permeabilities measured at the outcrop
scale. More recently, Roques et al. (2014) conducted a
63-day aquifer test in a highly permeable fault zone within
a crystalline-rock aquifer and observed changes in vertical
flow pathways under pumping and ambient conditions.
The purpose of this study is to present an analysis
of a large-scale aquifer test (referred to as the Arsenal
Test) conducted adjacent to a vertical fault system
within the highly stressed Cambrian-Ordovician sandstone
aquifer system in northeastern Illinois. To the authors’
knowledge, this is perhaps one of the largest confined
aquifer tests in North American history in terms of
overall withdrawal. The magnitude of withdrawals and
orientation of the pumping wells provides a unique
opportunity to study a faulted multiaquifer system at a
considerable depth (>450 m) on a regional scale (tens
of kilometers). We analyze the test both analytically and
numerically to determine hydraulic conductivity for 22 km
along the length of the fault zone and use our results to
update a regional groundwater model that simulates over
150 years of withdrawals. We then discuss implications
of our updated fault zone conceptualization for long-term
groundwater availability.
Geology and Hydrogeology of the Study Area
The Cambrian-Ordovician (CO) sandstone aquifer
system underlies all of northern Illinois and extends into
Wisconsin, southeastern Minnesota, and eastern Iowa. It
has been a principal groundwater source for municipal
and industrial supplies in northeastern Illinois since
the 1860s. The CO sandstone system consists of the
St. Peter sandstone and underlying Ironton-Galesville
sandstone, which are overlain by thick sequences of
carbonates, shales, and glacial deposits and thus have
limited direct recharge (Figure 1A). Since predevelopment
conditions, over 300 m of static head decline has occurred

454 D.R. Hadley et al. Groundwater 58, no. 3: 453–463 NGWA.org

in the sandstone system at the center of the cone of
depression in Will County (Figure 1A; Abrams et al.
2015). Drawdowns are minimal or nonexistent in north-
central Illinois because the St. Peter sandstone is exposed
at the land surface or bedrock surface (allowing for greater
recharge) and there is less demand (Figure 1A).
The Sandwich Fault Zone, a 137 km long zone of
parallel and en echelon faults that trend northwest to
southeast across northern Illinois (Figure 1A), affects
groundwater flow locally and regionally. As mapped
by Kolata et al. (1978), the fault zone is typically 0.8
to 3.2 km wide and is upthrown to the southwest for
most of its extent. The fault zone has a maximum
displacement of 240 m near its midpoint in DeKalb
County but displacement is much less at the ends of the
fault zone. In Will County the fault zone has variable
displacement ranging from 0 to 45 m and is typically
upthrown to the northeast (Figure 1B). The fault zone
is characterized by high-angle normal to vertical faults
that bound horst and graben structures (Kolata et al.
1978; Nelson 1995), but reverse faulting has also been
observed in borehole cuttings, outcrops, and within quarry
pits (Kolata et al. 1978). The age of the fault zone is
constrained stratigraphically by Silurian bedrock units and
Pleistocene glacial deposits. Movement likely occurred in
multiple episodes of deformation during the late Paleozoic
(Nelson 1995), possibly coincident with the Ancestral
Rockies tectonic event (Marshak et al. 2003).
The fault zone affects heads in the St. Peter and
Ironton-Galesville sandstones most notably in Kendall
County, where there is enough displacement to completely
offset the sandstones (i.e., sandstone units are not laterally
continuous) and there is a large contrast in demands on
either side of the fault zone. Both Roadcap et al. (2013)
and Abrams et al. (2015) observed sandstone heads that
were several hundred feet higher on the upthrown side
(southwest side) in Kendall County, indicating that the
fault zone acts as a flow barrier. Roadcap et al. (2013)
modeled the Sandwich fault zone as a low horizontal
hydraulic conductivity (K h ) zone of 3 × 10−5 m/day
compared to a K h of 1.5 m/day for the undeformed
sandstone units. The fault zone has also been modeled as
a no-flow barrier in contouring models of the sandstone
system (Abrams et al. 2018b). Most recently, the fault
zone has been modeled as a broad region using a
K h of 0.76 m/day (Mannix et al. 2018; Abrams et al.
2018a). In all cases, there was little constraint on K h ,
which was determined through the regional calibration
process.
Understanding the influence of the fault zone on
regional and local groundwater flow has important
implications for long-term water availability in the CO
system. The St. Peter and Ironton-Galesville sandstones
are heavily utilized in Kane, Will, and Kendall Counties
which has led to partial and complete desaturation of
the St. Peter sandstone in some areas, prompting drillers
to install wells only open to the underlying Ironton-
Galesville sandstone (Abrams et al. 2015; Mannix et al.
2018). This has led to increased head separation between
NGWA.org
the sandstone aquifers in the center of the cone (Abrams
et al. 2018b). Communities and industries are expected
to increase withdrawals and to drill additional sandstone
wells, many of which will be located near or within the
fault zone as development continues in western Will and
eastern Kendall Counties. Given that the fault zone is
buried by glacial deposits for most of its length and limited
data are available from drilling logs due to the near-
vertical orientation of individual faults, understanding the
structure and geometry of the fault zone, and thus the
hydrogeology, is extremely difficult. The Arsenal Test
provides a unique opportunity to gain insight into the
hydrogeology of a poorly understood fault system that
has significant influence on groundwater availability in a
confined aquifer system.
Arsenal Aquifer Test
Between 1940 and 1942 the Joliet Arsenal (JA) in
Will County, a World War II munitions manufacturing
facility, drilled nine multiaquifer wells open to both
the St. Peter and Ironton-Galesville sandstones (Table 1).
The wells were drilled to similar depth (around 490 m),
diameter (46 cm at the bottom) and had the same
length of casing and open interval. Wells were evenly
spaced by 762 m in the north-south direction and had
variable spacing (610-762 m) in the east-west direction
(Figure 1B). The northernmost well (JA-10) is located less
than 2 km away from the main trace of the fault zone.
In 1942 the facility, with assistance from engineers
at the Illinois State Water Survey, conducted a large-
scale aquifer test to determine whether the sandstone
aquifers could provide enough water to satisfy demands
at the height of production. Around 34,500 m3 /day was
pumped from the production wells for 37 days. Drilling
of JA-4 was completed several days after the start of the
production test and some wells were turned off for several
days due to difficulties with the turbine pumps, resulting
in withdrawals that varied for each well throughout
the duration of the test (Figure 2A). Due to drilling
difficulties, JA-5 was only finished in the St. Peter, was
never put into production, and served as an observation
well during the test. JA-8 ran prior to the start of the test
and during the recovery phase because it served as the
plant’s drinking water source at the time.
Water levels and withdrawal rates were recorded daily
at the production wells during the 37-day test. Overall
drawdowns at the production wells ranged from 68 to
95 m (Table 1, Figure 2B). Observations of recovery were
also recorded for 25 days at each of the production wells
and at JA-5. Several wells, notably JA-6 and JA-9, had
pumping events during the recovery phase (Figure 2B).
Drawdown was also recorded at sandstone wells in the
towns of Wilmington (9 km to the south), Minooka
(12 km to the northwest), and at Braidwood (15 km to the
southwest) (Figure 1B). Two of these far-field wells were
multiaquifer sandstone wells and two were only open to
the St. Peter (Table 1).
D.R. Hadley et al. Groundwater 58, no. 3: 453–463 455
 Table 1
List of Wells, Well Parameters, and Drawdown Associated with the Arsenal Test

Observed Drawdown Skin Factor

Well Name Well Depth (m) Open Interval Well Type at End of Test (m) Adjustment

JA-1 487 SP-IG PW 74 4

JA-2 491 SP-IG PW 83 7.5
JA-3 486 SP-IG PW 81 4.5
JA-4 471 SP-IG PW 89 7
JA-5 285 SP OW 31 -
JA-6 504 SP-IG PW 68 −2.5
JA-7 504 SP-IG PW 77 −1
JA-8 496 SP-IG PW 83 7
JA-9 489 SP-IG PW 95 8
JA-10 479 SP-IG PW 82 6.5
Wilmington 1 216 SP OW 6.7 -
Wilmington 2 477 SP-IG OW 6.1 -
Minooka 2 189 SP OW 1.2 -
Braidwood 1 502 SP-IG OW 0.8 -

JA, Joliet Arsenal; SP, St. Peter; IG-Ironton-Galesville; PW, Pumping Well; OW, Observation Well

Figure 2. (A) Withdrawals from production wells during the Arsenal Test. Note that several wells turned off during the test,
most notably JA-1 and JA-2. Small withdrawals during the recovery phase (after day 37) were not recorded. The withdrawal
rate for JA-8 during the recovery phase was estimated. (B) Recorded heads at the production wells and nearby observation
well (JA-5). Note the small pumping events during the recovery phase.

Modeling Methods
Illinois Groundwater Flow Model
We used a transient 21-layer MODFLOW model
of the CO aquifer system already developed for long-
term water supply planning purposes as a base model to
simulate the Arsenal Test. This regional model (known
as the Illinois Groundwater Flow Model) consists of 12
bedrock layers and 9 unconsolidated layers on a 762 m
grid and simulates annual stress periods from 1863-
present, as well as the impacts of future demands. The
regional model covers the northern two-thirds of Illinois
and extends into southern Wisconsin, eastern Iowa, and
western Indiana. For a complete description of model
parameters, boundary conditions, and hydrostratigraphic
units, see Abrams et al. (2018a).
The St. Peter and Ironton-Galesville sandstones are
represented in the regional model with a K h of 1.5 m/day
and a vertical hydraulic conductivity (K v ) of 0.18 m/day.
Multiaquifer wells throughout the model domain are
simulated with high K v zones in the aquitards that separate
the sandstone units (following Mandle and Kontis 1992)
and utilize the Time Variant Materials package available
in Groundwater Vistas 6.0 (Rumbaugh and Rumbaugh

456 D.R. Hadley et al. Groundwater 58, no. 3: 453–463 NGWA.org

2010) to simulate the correct timing of aquifer connections
historically (Mannix et al. 2018). The regional model
is calibrated to modern pumping conditions (post-1963)
because there are limited pumping and water level records
prior to that. Calibrating to modern pumping conditions,
however, introduces two complications: (1) open intervals
of wells have changed through time (Mannix et al. 2018),
and (2) sandstone units are actively being dewatered.
Simulating the Arsenal Test within the regional model
removes these two degrees of freedom since almost all
production wells within the study area at the time of the
test were open to all CO units and the JA wells remained
fully saturated (no specific yield) during the test, allowing
us to build greater confidence in the model results near
the fault zone.
Arsenal Test Simulation
The Arsenal Test was simulated in the Illinois
Groundwater Flow Model using MODFLOW-USG (Pan-
day et al. 2013). To accommodate the Arsenal Test, we
inserted additional daily stress periods for the year 1942,
reduced the K v of the Maquoketa Shale by two orders
of magnitude where it is exposed at the bedrock surface
(Figure 1A) to better reflect historic (pre-1963) conditions
that had considerably less aquifer connections via mul-
tiaquifer wells, and then modified fault zone properties.
Given the large width of the fault zone (up to 3.2 km)
and the unknown geometry of individual faults that com-
prise the zone, we conceptualized the fault zone within the
St. Peter and Ironton-Galesville sandstones as an inner 1-
cell wide fault core surrounded by a 1-cell wide damage
zone (Figure 3). This is based on low specific capacities
of sandstone wells near the fault (Abrams et al. 2018a)
and the physical properties of faults observed elsewhere
(Caine et al. 1996; Shipton and Cowie 2001; Kim et al.
2004; Caine and Minor 2009; Ball et al. 2010). We tested
various conceptualizations of the fault zone by varying
K h of the damage zone and fault core but only present
results for three scenarios: (1) an unrestricted flow sce-
nario in which the fault zone is identical to the host
rock, (2) a restricted flow scenario in which the K h of
the damage zone is reduced by half and the K h of the
fault core is reduced by an order of magnitude, and (3)
a highly-restricted flow scenario in which the K h of the
damage zone and fault core are reduced by two orders
of magnitude (Table 2). Varying horizontal and vertical
conductivities within the fault zone for other units in the
model was beyond the scope of this study.
The original regional model is insensitive to changes
in specific storage (Ss). Consequently, we initially
analyzed the Arsenal Test with a Theis solution to ensure
a reasonable Ss for the sandstones within the study area
using: (1) an average pumping rate of 34,500 m3 /day
(calculated by averaging all of the withdrawals over the
37-day test), (2) a sandstone K h of 1.5 m/day (consistent
with Roadcap et al. 2013; Abrams et al. 2018a, 2018b),
(3) an overall aquifer thickness of 120 m (60 m for each
sandstone unit), and (4) observed drawdowns at the
Wilmington and Braidwood wells of 6.5 and 0.8 m,
NGWA.org
Figure 3. Model grid of the study area (762 m grid)
showing the K h zonation of the fault zone (fault core and
damage zone) and locations of CLN wells (observation and
production).
respectively. A Ss of 1.7 × 10−7 m−1 was determined to
be the most consistent with observed drawdowns, which
is higher than a Ss of 7.9 × 10−8 m−1 used previously
by Roadcap et al. (2013) and Abrams et al. (2018a),
but falls at the upper end of Ss ranges used in previous
regional models of the CO system (Mandle and Kontis
1992; Meyer et al. 2009; Feinstein et al. 2010).
All model runs used the Sparse Matrix Solver
Package with the XMD linear solver and Newton-
Raphson nonlinear solver (Niswonger et al. 2011) and
achieved convergence with a head change criterion for
inner and outer iterations of 1 × 10−5 and 1 × 10−3 m,
respectively.
Arsenal Test Calibration
We calibrated to daily drawdown values at the
wells instead of head, since all of the JA production
wells remained under confined conditions throughout the
test, and hence never switched between Ss and specific
yield conditions. Calibrating to heads would also have
required extensive review of historic withdrawals in the
model. We simulated the production and observation wells
transiently using the connected linear network (CLN)
package (Panday et al. 2013) (instead of using a high K v
zone and the TVM package) to calibrate to drawdown
values at the well as opposed to values in the cell
(Figure 3). The JA production wells enter the model
when they were drilled to ensure that the aquifers were
not connected via the boreholes for the entire history
of the model run. We also adjusted skin factors within
the CLN package (Konikow et al. 2009) to account for
head losses within the boreholes of the JA production
wells and only used drawdown observations at the nine
D.R. Hadley et al. Groundwater 58, no. 3: 453–463 457
 Table 2
Fault Zone Parameters of the Tested Scenarios
Scenario Fault Zone Conceptualization
Unrestricted flow Fault zone same as host rock
Restricted flow Damage zone K h reduced by
half, fault core K h reduced by
order of magnitude
Highly-restricted flow Damage zone and fault core K h
reduced by two orders of
magnitude
production wells on days with either 24 h of pumping
or 24 h of no pumping. We also used drawdowns at
surrounding municipal wells observed at the end of the test
as calibration targets. Observed drawdown curves were
compared to our simulated drawdown curves using linear
regression and by comparing the slopes of the regression
lines.
Recovery observations were recorded graphically at
the time of the test (Figure 2B), but withdrawal rates
were not recorded when individual wells turned on and
off during the recovery phase. Given the unknown rates
of withdrawals and the short duration of these pumping
events (many of which were less than 12 h), we did
not use these recovery observations as calibration targets.
Because withdrawal rates were not recorded for JA-8, but
was noted as always pumping, we calculated the average
withdrawal rate of JA-8 during the test and applied that
to the recovery phase (Figure 2A).
Arsenal Test Observations and Model Results
The cone of depression formed by the Arsenal Test
reached a peak drawdown of over 90 m at its center
and extended outward at a distance of greater than
20 km (Video S1). Review of heads, drawdown, and
recovery observations recorded during and after the test
indicates that the fault zone inhibits horizontal flow in the
sandstone units. This is supported by observations that:
(1) drawdowns at both JA-9 and JA-10 were the same
magnitude (or greater) in comparison to drawdown at JA
wells 6 to 8 (Table 2) despite lower withdrawals rates
occurring at JA-9 and JA-10 compared to JA wells 6 to
8 (Figure 2A), (2) JA-9 and JA-10 had slower recoveries
in comparison to JA-6 and JA-7 while JA-8 continued
to pump during the recovery phase (Figure 2B), and (3)
in general, recovery occurred more quickly at the wells
further away from the fault zone (JA wells 1-4) compared
to the wells closer to the fault zone (JA wells 6-10)
(Figure 2B). Full recovery did not occur at the JA wells
after 25 days and drawdown continued to occur north of
the fault zone even after this period (Video S1).
Modifying the properties of the fault zone signifi-
cantly affected calibration at production wells closest to
the fault zone, most notably at JA-10 and JA-9 (Figure 4A
458 D.R. Hadley et al. Groundwater 58, no. 3: 453–463
K h Host K h Damage K h Fault
Rock (m/day) Zone (m/day) Core (m/day)
1.5 1.5 1.5
1.5 0.76 0.15
1.5 0.015 0.015
and 4B), but had less of an effect at JA wells 1 to 4 and
6 (Figure 4C, 4D, and 4E). JA-10 achieved the best cal-
ibration in the restricted flow scenario with a root mean
square error (RMSE) of 1.5 m and a regression line slope
of 0.93 (Figure 5E), whereas in the unrestricted or highly-
restricted flow scenarios there was either not enough or too
much simulated drawdown (Figure 4A) resulting in larger
RMSEs and deviations of the regression line (Figure 5D
and 5F). The differences between simulated drawdowns
for each of the three scenarios becomes progressively less
as the distance between the production wells and the fault
zone increases (compare hydrographs for JA wells 10 and
9 versus JA wells 4 and 2 in Figure 4). The far-field
municipal wells were largely unaffected by whether the
fault zone was represented as a flow barrier or not. In all
three scenarios, the mean residual of simulated drawdown
for the far-field municipal wells was ≤0.3 m (Table 3).
Skin factors (in addition to fault zone modifications)
were needed to achieve calibration at the JA production
wells and ranged from −2.5 to 8 (Table 1). The negative
skin factor for JA-6 indicates that the well bore skin
has a higher K h than the surrounding formation, whereas
positive skin factors indicate that well bore skins have
lower K h values than the surrounding formation. A
negative skin factor is unsurprising, given that some
of the JA production wells (including JA-6) were
blasted with nitroglycerine to increase fracturing in the
Ironton Galesville. Skin factor adjustments resulted in
a decrease or increase in the magnitude of simulated
drawdown but did not result in changing the shape
of the simulated drawdown curves (Figure S1), whereas
changing fault zone parameters did affect the shape
of the drawdown curves at wells closes to the fault
(Figure 4A and 4B).
The restricted flow scenario achieved the best overall
calibration with an RMSE of 3.37 m and a slope of
the regression line nearest to 1 in comparison to the
other scenarios (Figure 5B). Overall calibration results
did not vary greatly between each of the three scenarios,
however, due to the limited influence of the fault zone
on drawdowns in the far-field municipal wells and the JA
production wells located further away from the fault zone.
While all nine JA wells were very productive at the
time of drilling, JA-5 (the dedicated St. Peter monitoring
NGWA.org
Figure 4. Hydrographs of observed and simulated drawdown for select wells used in the Arsenal Test. Note the different
drawdown scale on JA-5.
well) was the exception. JA-5 was tested prior to the
start of the Arsenal Test and determined to have a very
low specific capacity, and there was also a very subdued
recovery curve at JA-5 following the end of the test
(Figure 2B). Both the pump test and recovery curve
suggest the St. Peter is not as transmissive around JA-5
in comparison to the other production wells or that there
was an issue with the well such as formation collapse. In
all three scenarios, the largest systematic error occurred at
JA-5 (Figure 4F), likely because this local zone of reduced
permeability was not incorporated into the three model
scenarios. We thus left out the results of JA-5 in our
regression analyses.
Discussion and Implications for Groundwater
Supply
Results of our model comparisons support previous
observations (Abrams et al. 2015, 2018a) and modeling
efforts (Roadcap et al. 2013) that there is reduced fault
zone permeability within the St. Peter and Ironton-
Galesville sandstones which limits groundwater flow
across the fault zone. Decreasing the K h of the fault core
by an order of magnitude and the K h of the damage zone
by half achieved the best calibration at wells closest to
the fault zone and overall. Sections of brecciated quartzite
within the St. Peter, observed by the authors at outcrops
NGWA.org
along the fault zone in Ogle County, lends geologic
evidence to a reduced permeability within the fault zone.
Our modeling efforts determined an “effective”
hydraulic conductivity (Bense et al. 2013) for the fault
zone in Will County, but the specific mechanisms that
cause reduced flow across the fault zone are unknown.
Mechanisms could include: (1) the presence of gouge
within a fault core or series of fault cores (Rawling
et al. 2001; Shipton and Cowie 2001; Caine and Minor
2009), (2) dissolution and cementation within the core
or damage zone fracture networks (Eichhubl et al. 2009),
or (3) cataclasis in broad deformation bands bounded by
slip planes (Antonellini and Aydin 1994; Shipton et al.
2002; Fossen et al. 2007; Medeiros et al. 2010). Overall
throw along the fault zone is well defined, but there
may be many more upthrown and downthrown blocks
within the fault zone than currently mapped, resulting in
complex structural geometries. A geophysical approach
could determine some of these specific mechanisms and
structural controls (e.g., Ball et al. 2010). Clay smearing
(Bense and Person 2006; Egholm et al. 2008) could also
have a significant influence on groundwater flow in the
central portion of the fault zone (Dekalb County) since
there is enough offset to drag the Maquoketa and Eau
Claire shales to meet the sandstone units.
We analyzed how incorporating the restricted flow
fault zone conceptualization into the Illinois Groundwater
D.R. Hadley et al. Groundwater 58, no. 3: 453–463 459
Figure 5. Simulated vs. observed daily drawdown values for all production wells (top row) and only JA-10 (bottom row) for
the Arsenal Test. Gray line indicates 1:1 line.

Table 3 mean error, absolute mean error, and RMSE of head

Results of Simulated vs. Observed Drawdown targets in both sandstone units for both 1980 and 2014
Values at the End of the Test for Far-Field (Table 4). We then incorporated our updated Ss and
Municipal Observation Wells Maquoketa Shale K v values (in addition to updating the
fault zone). This changed calibration results only slightly
Simulated Overall (Table 4), indicating that fault zone modifications are
Drawdown (m) more important for CO sandstone calibration within the
Observed cone of depression.
Overall A reduced hydraulic conductivity in the fault zone
Drawdown Un- Highly means less recharge available from southern Will and
(m) restricted Restricted Restricted
Kendall counties to satisfy demands north of the fault
Wilmington 1 6.7 4.9 5.2 5.5 zone. Municipal wells north of the fault zone within
Wilmington 2 6.1 5.2 5.5 5.9 the current cone of depression are at risk of partially
Minooka 2 1.2 2.8 2.6 2.7 desaturating the Ironton-Galesville by 2040 (Mannix et al.
Braidwood 1 0.8 0.9 1.0 1.0 2017). Given the findings in this study, coupled with
Mean Residual 0.3 0.1 −0.1 newly estimated water use demands for the region, this
time frame is likely less. This shortened time frame is
exacerbated by the expectation that new demands will be
Flow Model affects calibration to post-1963 heads in met by wells near or within the fault zone. The properties
the CO sandstones. We compared head targets for the of the fault zone in this multiaquifer sandstone system
years 1980 and 2014 (see Abrams et al. 2015 for a have implications for other areas in similar geologic
description of these measurements) in a zone centered settings if groundwater use occurs near faults and flow
on the modern day cone of depression (Figure 1A) barriers are reached. The flow-barrier behavior of the
that includes the heavy municipal users northeast of Sandwich Fault Zone may also mean a slower recovery
the fault zone and the heavy industrial users south of (e.g., see Marshall et al. 2019) of the CO system if
the fault zone in Will County. Incorporating only the communities are eventually forced to switch to Lake
restricted flow fault zone conceptualization improved Michigan water as groundwater is depleted.

460 D.R. Hadley et al. Groundwater 58, no. 3: 453–463 NGWA.org

 Table 4
Calibration Results from the Illinois Groundwater Flow Model for Cambrian-Ordovician Sandstone Water
Levels in 1980 and 2014 (Years of Mass Measurements) Within the Cone of Depression in Will and Kendall
Counties

1980 2014
OFZ RFZ RFZ + Ss OFZ RFZ RFZ + Ss

Ironton Galesville1
Mean error (m) −3.85 0.87 −0.14 −8.36 3.09 −0.92
Absolute mean error (m) 13.12 11.69 12.51 16.29 12.30 13.23
RMSE (m) 17.69 15.36 15.73 19.74 14.96 15.81
St. Peter2
Mean error (m) −8.87 −8.31 −8.35 −23.64 −22.19 −19.93
Absolute mean error (m) 20.54 17.21 18.62 27.50 24.80 24.55
RMSE (m) 24.26 21.19 22.67 31.98 28.76 28.56

OFZ, original fault zone conceptualization; RFZ, restricted flow fault zone conceptualization; RFZ + Ss, restricted flow fault zone conceptualization plus updated Ss
and Maquoketa Shale K v values.
1 n = 62 for 1980, n = 42 for 2014.
2 n = 47 for 1980, n = 16 for 2014.

Conclusions
We used a numerical modeling approach to simulate
a large-scale historic aquifer test that occurred adjacent to
a poorly understood near-vertical fault zone system. The
fault zone system is structurally complex with variable
offset and numerous fault traces bounding upthrown and
downthrown blocks. Observations of head offset and low
specific capacities near the fault zone indicate that the fault
zone is an important structural control on groundwater
movement in the CO sandstone aquifer system.
The magnitude of groundwater withdrawals and the
duration of the Arsenal Test allowed insight into the
hydraulic properties for a large area of the fault zone
(>20 km) at a considerable depth (>450 m) for a confined
multiaquifer sandstone system. We conceptualized the
fault zone in the sandstone units as a low-permeable fault
core surrounded by a damage zone based on previous
research of deformation bands in faulted sandstones. The
K h of the fault core and damage zone were modified until
the magnitude and shape of the simulated drawdown
curves matched observed records at the production
wells and at surrounding observation wells. We presented
results from three scenarios: an unrestricted flow scenario,
a restricted flow scenario, and a highly-restricted flow
scenario.
Our study determined an effective hydraulic conduc-
tivity for the fault zone and indicates that there is at least
an order of magnitude reduction in K h of the fault core
compared to the undeformed sandstones. We achieved
the best calibration to drawdown levels (overall and at
wells closest to the fault) under the restricted flow sce-
nario, in which the K h of the undeformed sandstones,
damage zone, and fault core were set to 1.5, 0.76, and
0.15 m/day, respectively. The unrestricted flow scenario
and the highly-restricted flow scenario either allowed too
much or too little flow across the fault zone to achieve
calibration at the production wells closest to the fault.
Production wells further from the fault zone, and the
far-field observation wells, were less affected by modi-
fying fault zone properties.
We then analyzed how the restricted flow fault
zone conceptualization affects calibration to heads in
the CO sandstone system post-1963, when there is
considerably more uncertainty in the system because of
multiaquifer wells and desaturated conditions. Updating
the Illinois Groundwater Flow Model with the restricted
flow scenario significantly improved calibration to
CO sandstone water levels for years in which mass
measurement of sandstone water levels took place (1980
and 2014). The flow-barrier behavior of the fault zone
has important implications for groundwater availability
in the highly stressed CO sandstone aquifer system.
Communities near the fault zone continue to rely on the
deep sandstone system despite increasing drawdown, and
are expected to increase withdrawals before alternative
supplies can be utilized. Projected desaturation of the
Ironton-Galesville sandstone may occur sooner than the
previously estimated year of 2040 given our updated
fault zone conceptualization, and many wells could face
serious issues under pumping conditions by 2030. Results
from this study may inform the hydraulic properties of
other faulted sandstone aquifers in confined settings and
help in long-term groundwater management.
Acknowledgments
This project was funded in part by the Illinois
Department of Natural Resources under the purview of
water supply planning. We thank Chris Langevin, John
Quinn, and an anonymous reviewer for their positive and
helpful comments that improved this manuscript.
Authors’ Note
The authors do not have any conflicts of interest or
financial disclosures to report.

NGWA.org D.R. Hadley et al. Groundwater 58, no. 3: 453–463 461

Supporting Information
Additional supporting information may be found online
in the Supporting Information section at the end of
the article. Supporting Information is generally not peer
reviewed.
Figure S1 Plot of simulated drawdown at JA-10 under
the restricted flow scenario using different skin factors
for calibration. Note that skin factor adjustments change
the magnitude of drawdown but not the shape of the
drawdown curve.
Video S1 Animation of drawdown and subsequent
recovery in the St. Peter sandstone during the Arsenal
Test using the restricted flow fault zone conceptualization.
Note that drawdown values are cell values. Actual
drawdown at the wells were much greater during the test.
References
Abrams, D.B., D.H. Mannix, D.R. Hadley, and G.S. Roadcap.
2018a. Groundwater flow models of Illinois: Data, pro-
cesses, model performance, and key results. Illinois State
Water Survey Contract Report 2019-04. Champaign, Illi-
nois.
Abrams, D.B., G.S. Roadcap, and D.H. Mannix. 2018b.
Developing potentiometric surfaces and flow fields with a
head-specified MODFLOW model. Groundwater 56, no. 1:
8–17. https://doi.org/10.1111/gwat.12540
Abrams, D.B., D.R. Hadley, D.H. Mannix, G. Roadcap, S.C.
Meyer, K.J. Hlinka, K.L. Rennels, K.R. Bradbury, P.M.
Chase, and J.J. Krause. 2015. Changing groundwater levels
in the sandstone aquifers of northern Illinois and southern
Wisconsin: Impacts on available water supply. Illinois
State Water Survey Contract Report 2015–02. Champaign,
Illinois.
Antonellini, M., and A. Aydin. 1994. Effect of faulting on
fluid-flow in porous sandstones—Petrophysical properties.
AAPG Bulletin 78, no. 3: 355–377.
Ball, L.B., S. Ge, J.S. Caine, A. Revil, and A. Jardani.
2010. Constraining fault-zone hydrogeology through inte-
grated hydrological and geoelectrical analysis. Hydrogeol-
ogy Journal 18, no. 5: 1057–1067. https://doi.org/10.1007/
s10040-010-0587-z
Bense, V.F., T. Gleeson, S.E. Loveless, O. Bour, and J. Scibek.
2013. Fault zone hydrogeology. Earth-Science Reviews
127: 171–192. https://doi.org/10.1016/j.earscirev.2013.09
.008
Bense, V.F., M.A. Person, K. Chaudhary, Y. You, N. Cremer,
and S. Simon. 2008. Thermal anomalies indicate pref-
erential flow along faults in unconsolidated sedimentary
aquifers. Geophysical Research Letters 35, no. 24: L24406.
https://doi.org/10.1029/2008GL036017
Bense, V.F., and M.A. Person. 2006. Faults as conduit-barrier
systems to fluid flow in siliciclastic sedimentary aquifers.
Water Resources Research 42, no. 5. https://doi.org/10
.1029/2005WR004480
Bense, V., and R. van Balen. 2003. Hydrogeological aspects
of fault zones on various scales in the Roer Valley
Rift System. Journal of Geochemical Exploration 78-79:
317–320. https://doi.org/10.1016/S0375-6742(03)00031-1
Bredehoeft, J.D., K. Belitz, and S. Sharp-Hansen. 1992. The
hydrodynamics of the Big Horn Basin; a study of the role
of faults. AAPG Bulletin 76, no. 4: 530–546.
Burbey, T.J. 2008. The influence of geologic structures on
deformation due to ground water withdrawal. Ground Water
46, no. 2: 202–211. https://doi.org/10.1111/j.1745-6584
.2007.00395.x
462 D.R. Hadley et al. Groundwater 58, no. 3: 453–463
Caine, J.S., and S.A. Minor. 2009. Structural and geochemical
characteristics of faulted sediments and inferences on the
role of water in deformation, Rio Grande Rift, New Mexico.
Geological Society of America Bulletin 121, no. 9–10:
1325–1340. https://doi.org/10.1130/b26164.1
Caine, J.S., J.P. Evans, and C.B. Forster. 1996. Fault zone
architecture and permeability structure. Geology 24,
no. 11: 1025–1028. https://doi.org/10.1130/0091-7613
(1996)024<1025:Fzaaps>2.3.Co;2
Egholm, D.L., O.R. Clausen, M. Sandiford, M.B. Kristensen,
and J.A. Korstgard. 2008. The mechanics of clay smearing
along faults. Geology 36, no. 10: 787–790. https://doi.org/
10.1130/G24975a.1
Eichhubl, P., N.C. Davatzes, and S.P. Becker. 2009. Structural
and diagenetic control of fluid migration and cementation
along the Moab fault, Utah. AAPG Bulletin 93, no. 5:
653–681. https://doi.org/10.1306/02180908080
Faulkner, D.R., C.A.L. Jackson, R.J. Lunn, R.W. Schlische, Z.K.
Shipton, C.A.J. Wibberley, and M.O. Withjack. 2010. A
review of recent developments concerning the structure,
mechanics and fluid flow properties of fault zones. Journal
of Structural Geology 32, no. 11: 1557–1575. https://doi
.org/10.1016/j.jsg.2010.06.009
Feinstein, D.T., R.J. Hunt, and H.W. Reeves. 2010. Regional
groundwater-flow model of the Lake Michigan Basin in
support of Great Lakes Basin water availability and use
studies. U.S. Geological Survey Scientific Investigations
Report 2010–5109. Reston, Virginia: USGS.
Fossen, H., R.A. Schultz, Z.K. Shipton, and K. Mair. 2007.
Deformation bands in sandstone: A review. Journal of the
Geological Society 164, no. 4: 755–769. https://doi.org/10
.1144/0016-76492006-036
Jourde, H., E.A. Flodin, A. Aydin, L.J. Durlofsky, and X.H.
Wen. 2002. Computing permeability of fault zones in eolian
sandstone from outcrop measurements. AAPG Bulletin 86,
no. 7: 1187–1200.
Kim, Y.-S., D.C.P. Peacock, and D.J. Sanderson. 2004. Fault
damage zones. Journal of Structural Geology 26, no. 3:
503–517. https://doi.org/10.1016/j.jsg.2003.08.002
Kolata, D.R., T.C. Buschbach, and J.D. Treworgy. 1978. The
Sandwich Fault Zone of Northern Illinois. Illinois State
Geological Survey Circular 505. Urbana, Illinois.
Konikow, L.F., G.Z. Hornberger, K.J. Halford, and R.T. Hanson.
2009. Revised Multi-Node Well (MNW2) package for
MODFLOW ground-water flow model. U.S. Geological
Survey Techniques and Methods 6-A30. Reston, Virginia:
USGS.
Mandle, R.J., and A.L. Kontis. 1992. Simulation of regional
ground-water flow in the Cambrian-Ordovician aquifer
system in the Northern Midwest, United States. U.S.
Geological Survey Professional Paper 1405-C. Reston,
Virginia: USGS.
Mannix, D.H., D.B. Abrams, D.R. Hadley, and G.S. Roadcap.
2018. Conceptualizing leakage and storage contributions
from long open interval wells in regional deep basin flow
models. Hydrological Processes 33, no. 2: 271–282. https://
doi.org/10.1002/hyp.13324
Mannix, D.H., D.B. Abrams, G.S. Roadcap, D.R. Hadley, and
W.R. Kelly. 2017. Groundwater Depletion in Chicago’s
Southwestern Suburbs. Champaign, Illinois: University of
Illinois at Urbana-Champaign.
Marler, J., and S. Ge. 2003. The permeability of the Elkhorn
Fault Zone, South Park, Colorado. Ground Water 41,
no. 3: 321–332. https://doi.org/10.1111/j.1745-6584.2003
.tb02601.x
Marshak, S., W.J. Nelson, and J.H. McBride. 2003. Phanerozoic
strike-slip faulting in the continental interior platform of
the United States: Examples from the Laramide Orogen,
Midcontinent, and Ancestral Rocky Mountains. Geological
Society, London, Special Publications 210, no. 1: 159–184.
https://doi.org/10.1144/gsl.sp.2003.210.01.10
NGWA.org
Marshall, S.K., P.G. Cook, A.D. Miller, C.T. Simmons, and
S. Dogramaci. 2019. The effect of undetected barriers
on groundwater drawdown and recovery. Groundwater.
https://doi.org/10.1111/gwat.12856
Mayer, A., W. May, C. Lukkarila, and J. Diehl. 2007. Estimation
of fault-zone conductance by calibration of a regional
groundwater flow model: Desert Hot Springs, California.
Hydrogeology Journal 15, no. 6: 1093–1106. https://doi
.org/10.1007/s10040-007-0158-0
Medeiros, W.E., A.F. do Nascimento, F.C.A. da Silva, N. Destro,
and J.G.A. Demétrio. 2010. Evidence of hydraulic connec-
tivity across deformation bands from field pumping tests:
Two examples from Tucano Basin, NE Brazil. Journal of
Structural Geology 32, no. 11: 1783–1791. https://doi.org/
10.1016/j.jsg.2009.08.019
Meyer, S.C., G.S. Roadcap, Y.F. Lin, and D.D. Walker. 2009.
Kane County water resources investigations: Simulation of
groundwater flow in Kane County and northeastern Illinois.
Illinois State Water Survey Contract Report 2009–07.
Champaign, Illinois.
Minor, S.A., and M.R. Hudson. 2006. Regional survey of struc-
tural properties and cementation patterns of fault zones
in the northern part of the Albuquerque Basin, New
Mexico—Implications for ground-water flow. U.S. Geo-
logical Survey Professional Paper 1719. Reston, Virginia:
USGS.
Nelson, J.W. 1995. Stuctural features in Illinois. Illinois State
Geological Survey Bulletin 100. Champaign, Illinois.
Nieto Camargo, J.E., and J.L. Jensen. 2012. Analysis of fault
permeability using mapping and flow modeling, Hickory
Sandstone Aquifer, Central Texas. Natural Resources
Research 21, no. 3: 395–409. https://doi.org/10.1007/
s11053-012-9181-5
Niswonger, R.G., S. Panday, and M. Ibaraki. 2011. MODFLOW-
NWT, A newton formulation for MODFLOW-2005. U.S.
Geological Survey Techniques and Methods 6-A37. Reston,
Virginia: USGS.
NGWA.org
Panday, S., C.D. Langevin, R.G. Niswonger, M. Ibaraki, and J.D.
Hughes. 2013. MODFLOW–USG version 1: An unstruc-
tured grid version of MODFLOW for simulating ground-
water flow and tightly coupled processes using a control
volume finite difference formulation. U.S. Geological Sur-
vey Techniques and Methods 6-A45. Reston, Virginia.
Rawling, G.C., L.B. Goodwin, and J.L. Wilson. 2001. Inter-
nal architecture, permeability structure, and hydrologic
significance of contrasting fault-zone types. Geology
29, no. 1: 43–46. https://doi.org/10.1130/0091-7613(2001)
029<0043:Iapsah>2.0.Co;2
Roadcap, G.S., S.C. Meyer, W.R. Kelly, H.A. Wehrmann, and
Y.F. Lin. 2013. Groundwater studies for water supply
planning in Kendall County, Illinois. Illinois State Water
Survey Contract Report 2013–05. Champaign, Illinois.
Roques, C., O. Bour, L. Aquilina, B. Dewandel, S. Leray,
J.M. Schroetter, L. Longuevergne, T. Le Borgne,
R. Hochreutener, T. Labasque, N. Lavenant, V. Vergnaud-
Ayraud, and B. Mougin. 2014. Hydrological behavior
of a deep sub-vertical fault in crystalline basement and
relationships with surrounding reservoirs. Journal of
Hydrology 509: 42–54. https://doi.org/10.1016/j.jhydrol
.2013.11.023
Rumbaugh, J.O., and D.B. Rumbaugh. 2010. Groundwater Vistas
6.0 . Reinholds, Pennsylvania: Environmental Simulations
Inc.
Shipton, Z.K., J.P. Evans, K.R. Robeson, C.B. Forster, and
S. Snelgrove. 2002. Structural heterogeneity and perme-
ability in faulted eolian sandstone: Implications for sub-
surface modeling of faults. AAPG Bulletin 86, no. 5:
863–883. https://doi.org/10.1306/61EEDBC0-173E-11D7-
8645000102C1865D
Shipton, Z.K., and P.A. Cowie. 2001. Damage zone and
slip-surface evolution over μm to km scales in high-
porosity Navajo sandstone, Utah. Journal of Structural
Geology 23, no. 12: 1825–1844. https://doi.org/10.1016/
S0191-8141(01)00035-9
D.R. Hadley et al. Groundwater 58, no. 3: 453–463 463