See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309802608

Groundwater

Article · November 2016

CITATIONS READS

2 25,365

4 authors:

Edwin Brands R. Rajagopal

University of Minnesota Morris University of Iowa
12 PUBLICATIONS   43 CITATIONS    54 PUBLICATIONS   259 CITATIONS   

SEE PROFILE SEE PROFILE

Usha Eleswarapu Peter Li

Rutgers, The State University of New Jersey Tennessee Technological University
1 PUBLICATION   2 CITATIONS    2 PUBLICATIONS   3 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Joyce Foundation Grant View project

USDA CSREES Grant 2001-05, University of Iowa View project

All content following this page was uploaded by R. Rajagopal on 15 October 2018.

The user has requested enhancement of the downloaded file.

 
Research and GIS.” In Qualitative GIS: A Mixed
Methods Approach, edited by Meghan Cope and
Sarah Elwood, 113–135. Thousand Oaks, CA:
SAGE.
Knigge, LaDona, and Meghan Cope. 2006.
“Grounded Visualization: Integrating the Analysis
of Qualitative and Quantitative Data through
Grounded Theory and Visualization.” Environment
and Planning A, 38(11): 2021–2037. DOI:10.
1068/a37327.
Pavlovskaya, Marianna. 2002. “Mapping Urban
Change and Changing GIS: Other Views
of Economic Restructuring.” Gender, Place
and Culture, 9(3): 281–289. DOI:10.1080/
0966369022000003897.
Strauss, Anselm, and Juliet Corbin. 1998. Basics of
Qualitative Research: Techniques and Procedures for
Oaks, CA: SAGE.
PR
Developing Grounded Theory, 2nd edn. Thousand
 Groundwater
ME
Edwin Brands
University of Minnesota, Morris, USA
Raj Rajagopal
University of Iowa, USA
Usha Eleswarapu
Rutgers University, USA
LU
Peter Li
Tennessee Technological University, USA
Groundwater is water that exists mainly in
subsurface pore spaces but also in defined chan-
VO
nels, such as those found in karst formations,
which are created by dissolution of soluble
rocks such as limestone. After the polar ice
caps, groundwater is the next largest reservoir
of freshwater on Earth, containing more than
100 times the volume of streams and freshwater
lakes (Shiklomanov 1993). Groundwater plays an

G RO U N DWAT E R
important role in the hydrologic cycle, in plant
growth and soil formation, and in providing
water for human activities. Since groundwater is
FS
difficult to observe and track directly, monitor-
ing, mapping, and modeling efforts are crucial to
the understanding of its storage, distribution, and
patterns in flow through the subsurface, how it
may be sustainably used, and how it may become
contaminated. Humans have used groundwater
OO
throughout history, but the demand for it and
the societal ability to consume and contaminate
it have all increased exponentially over the past
century. Uneven distribution of human popu-
lations combined with variations in availability
and accessibility to groundwater have resulted
in overexploitation of several major aquifers
serving population and agricultural produc-
tion centers in China, India, and the United
States. Significant current and future threats to
groundwater quantity and quality that require
attention from policy and planning perspectives
include climate change, fossil fuel exploration 
and extraction, natural contaminants such as
radium and arsenic, underground storage tanks
and industrial contaminants, and agricultural use
and pollution.
Groundwater: natural and physical
science perspectives
Groundwater’s place in the water cycle
More than 68% of fresh water is found in
polar ice caps and glaciers and is thus largely
unavailable for societal use (Shiklomanov 1993).
Approximately 30% of global fresh water is
groundwater, whereas only 1.2% is found in
streams and lakes. Geologic formations that
yield a significant amount of water to wells or
springs are called aquifers. An aquifer consists of
two or more permeable layers in the subsurface
separated at least locally by intervening layers
3237
 

GRO UN DWATE R

that impede groundwater movement but do not rocks, while another 18% of the subsurface is
significantly affect the regional hydraulic con- dominated by aquifers of more complex and
nectivity of the system. With the exception of heterogeneous geology (Richts, Struckmeier,

fossil aquifers, groundwater is replenished mainly
by precipitation falling within the recharge area
of an aquifer. Precipitation on land flows hor-
izontally across the surface or infiltrates into
soil and moves either horizontally or vertically
within the subsurface. Groundwater consists
FS
and Zaepke 2011, WHYMAP, Table 1). The
remaining approximately 47% of the land surface
is underlain by local and shallow aquifers along
stream valleys and lowlands. There are significant
regional variations in the distribution of aquifers.
For example, shallow local aquifers dominate

both of water that remains in the unsaturated or
vadose zone (also often termed “soil water”) and
of water that reaches the saturated zone (aquifer)
where pore spaces are completely filled.
Global distribution of groundwater
Groundwater may be found almost anywhere
on Earth if one digs deep enough, but most
accessible groundwater is generally found within
PR
1 km of Earth’s surface (Hess 2014). Beyond
 this depth, groundwater availability decreases
gradually and quality is often poor due to high
salinity and mineral concentrations. World-
ME
OO
much of North America, whereas major regional
aquifers spanning thousands of square kilometers
are the predominant aquifer type throughout
much of continental Europe (Table 1).
Currently, approximately 24% of global
aquifers are highly overexploited; the major
driving factor is irrigated agriculture for feed-
ing human and livestock populations (Gleeson
et al. 2012). Pumping from the most exploited
aquifers (Upper Ganges, North and South
Arabian, Persian, Western Mexico, and High
Plains or Ogallala) is on the order of tens of 
times the natural recharge rate, and may also
act as a limiting factor in future agricultural

wide, approximately 36% of the land surface output (Figure 1). Although 76% of aquifers
(excluding Antarctica) is underlain by major are not being significantly exploited, overex-
aquifers, primarily composed of sedimentary ploitation of a few large aquifers has led to a

Table 1 Groundwater basins and aquifers by continent.

Major groundwater Complex hydrogeological Local and shallow
LU

basins structures aquifers

Continent (million km2 ) (%) (million km2 ) (%) (million km2 ) (%)
Africa 13.5 44.9 3.3 11.0 13.2 44.1
Asia 14.5 32.0 7.8 17.3 23.0 50.7
Australia, New Zealand 2.6 32.5 2.9 36.3 2.5 31.1
VO

Europe 5.2 53.0 1.8 18.8 2.7 28.2

Central/South America 8.4 45.0 2.0 10.9 8.2 44.1
North America 3.2 15.0 5.8 26.9 12.4 58.1
World (excl. Antarctica) 47.3 35.6 23.6 17.8 62.0 46.6
Source: Richts, Struckmeier, and Zaepke 2011. Date of the source data is 2008. With permission from WHYMAP, BGR,
Stilleweg 2, 30655 Hannover, Germany.

3238


 

G RO U N DWAT E R

FS
600

Aquifers
400

200

0–
0. .5
1– 1
5–
10 0
> 2 20
5–
0

5
1
–
0
OO
GF/AA

GF/AA
Groundwater

>20
10–20
stress

5–10
1–5
<1

Aquifer
area (AA)
Groundwater
footprint (GF)
Western
Mexico
High plains
PR
North
Arabian
Persian Upper Ganges
North
China
plain

Figure 1 Groundwater footprints of aquifers. Source: Gleeson et al. 2012. Reproduced by permission of
 Nature. 
ME

total groundwater footprint of more than 3.4
times the global area of aquifers (Gleeson et al.
2012, Table 2). If 1 and 2% of the largest pristine
or unexploited aquifers (7 and 15 observations
from 748 considered in Table 2) are removed
from consideration, the composite average of
LU
influence on the composite average of the global
groundwater footprint. In contrast, the median
of the global groundwater footprint of 0.11
is robust and is not significantly impacted by
the extreme outlying values of aquifer areas or
groundwater footprints.

the groundwater footprint rises to 4.81 and 6.34
(39 and 84% increase, respectively) indicating
that a few outlying pristine aquifers have undue
influence on the composite average of the global
groundwater footprint. Similarly, if 1 and 2% of
VO
Exploitation of aquifers also varies significantly
within aquifers. For example, the average water
level in the High Plains (or Ogallala) aquifer of
the central United States has declined by more
than 4 m since before World War II. However,

the aquifers representing the largest footprints
(7 and 15 observations from 748 considered in
Table 2) are removed from consideration, the
composite average of the groundwater footprint
falls to 1.52 and 0.93 (56 and 73% decrease,
respectively) indicating that a few outlying
aquifers with large footprints also have undue
this includes some areas where water levels have
increased by up to 20 m and others that have
decreased more than 70 m (McGuire 2013).
Total water storage in the High Plains aquifer
has decreased by approximately 8%, but storage
in 5% of the aquifer area decreased by over 50%
(McGuire 2013).

3239


 

GRO UN DWATE R

Table 2 Properties of aquifers with the largest groundwater footprints.

Aquifer Country GF (106 km2 ) AA (106 km2 ) GF/AA
Upper Ganges India, Pakistan 26.1 ± 7.5 0.48 54.2 ± 15.6

FS
North Arabian Saudi Arabia 17.3 ± 4.7 0.36 48.3 ± 13.5
South Arabian Saudi Arabia 9.5 ± 3.6 0.25 38.5 ± 14.7
Persian Iran 8.4 ± 3.7 0.42 19.7 ± 8.6
South Caspian Iran 5.9 ± 2.0 0.06 98.3 ± 32.6
Western Mexico Mexico 5.5 ± 2.0 0.21 26.6 ± 9.4

OO
High Plains USA 4.5 ± 1.2 0.5 9.0 ± 2.4
Lower Indus India, Pakistan 4.2 ± 1.5 0.23 18.4 ± 6.5
Nile Delta Egypt 3.1 ± 0.8 0.1 31.7 ± 7.9
Danube basin Hungary, Austria, Romania 2.4 ± 0.8 0.32 7.4 ± 2.6
Central Mexico Mexico 1.8 ± 0.5 0.2 9.1 ± 2.6
North China plain China 1.8 ± 0.6 0.23 7.9 ± 2.8
Northern China
North Africa
Central Valley
China
PR
Algeria, Tunisia, Libya
USA
1.4 ± 0.6
0.9 ± 0.3
0.4 ± 0.2
0.31
0.36
0.07
4.5 ± 1.8
2.6 ± 0.9
6.4 ± 2.4
Other aquifers 38.6 ± 0.8 34.17 1.1 ± 0.3
All aquifers 131.8 ± 24.9 38.27 3.45 ± 1.06
 
GF = groundwater footprint;
AA = aquifer area.
ME

Source: Gleeson et al. 2012. Reproduced with permission from Nature.

Types of aquifers have not been cemented together, and where

The unsaturated zone of an aquifer is immedi-
ately below the land surface where the soil/rock
pores contain both water and air, but are not
LU
water fills up spaces between the particles. In
contrast, water-bearing formations that have
been cemented together are termed consoli-

dated aquifers. Relatively large pore spaces in

totally saturated with water. It differs from
the zone below which is saturated with water.
The top of the saturated zone is referred to as
the water table (USGS 2014a). Aquifers may
be classified by whether they are confined or
VO
unconsolidated aquifers and solution channels in
carbonate aquifers can hold significant amounts
of groundwater with high hydraulic conductiv-
ity. In contrast, only the presence of fractures and
joints in igneous and metamorphic rocks allow

unconfined and by their geologic composition.
Confined aquifers exist between two impermeable
layers, often comprised of clay or clay-derived
rock. Unconfined aquifers lack a confining layer
above the water table, and are thus vulnerable to
pollution from the surface. Unconsolidated aquifers
consist of sand, gravel, and other materials that
water to enter and move through the formations.
Sand and gravel aquifers are generally uncon-
fined, and can be several hundred feet thick.
They are comprised of alluvial or glacial deposits
and found in valleys, depressions, and lowlands.
Alluvial aquifers were formed over hundreds to
thousands of years by water deposition of sand,

3240


 
silt, and clay particles. These alluvial deposits
can contain large groundwater reserves that are
fairly accessible and easy to withdraw. Sand and
gravel aquifers of glacial origin may also have
large reserves of groundwater that are fairly easy
to withdraw, but hydraulic conductivity in such
aquifers may be quite variable due to lack of
sorting in glacial till.
Carbonate rock aquifers are comprised primarily
of limestone or dolostone formed in ancient
marine environments. Weak carbonic acid
in rainwater dissolves carbonate rock and over
thousands of years has in many areas formed karst
landscapes characterized by numerous solution
cavities, caverns, and sinkholes through which
water can move rapidly. Where carbonate layers
PR
are exposed to the surface, runoff from precipita-
tion or entire streams may connect directly with
groundwater, resulting in significant contamina-
tion potential. Sandstone aquifers and igneous
 and metamorphic rock aquifers typically store
and transmit water only along bedding planes,
joints, and other cracks and fractures. Hydraulic
ME
conductivity in such aquifers is low, but large
regional aquifers can yield high total volumes of
groundwater.
Groundwater flow and wells
LU
The geology of the aquifer influences water
storage capacity as well as hydraulic conductivity, or
the ability of aquifer materials to transmit water.
Factors affecting hydraulic conductivity and stor-
age capacity include porosity (proportion of pore
space) and permeability (connectedness) of the
VO
pore spaces. Groundwater flow is down-gradient
from high to low pressure, which often corre-
sponds with moving from high to low elevation.
An exception to this occurs in the case of
confined aquifers contained within tilted rock
formations where considerable pressure builds
up in the lower reaches of the aquifer. In such

G RO U N DWAT E R
Land
surface
FS
Well
Potentiometric
surface Water level Cone of
in well depression
(water table)
Aquifer
OO
Figure 2 A schematic representation of aquifers,
wells, and drawdown. Drawing by Edwin Brands.
cases, a well or a spring located at a point where
the potentiometric surface (or water table) is higher
than the land surface will flow freely without
the need for pumping; such wells or springs are
termed artesian.
Wells drilled into unconfined aquifers or con-
fined aquifers with insufficient pressure to flow
naturally require pumping to bring water to the 
surface. Some level of drawdown (or reduction in
the height of the potentiometric surface) occurs
with pumping of water from aquifers. During
pumping, a cone of depression (Figure 2) forms
around wells – the shape, size, and depth of the
cone of depression as well as the rate at which
the well water level recovers depend upon the
pumping rate and duration, aquifer character-
istics such as hydraulic conductivity, and the
presence or absence of nearby wells. Pumping
tests are done to determine whether the aquifer
will provide an adequate yield (measured in
gallons or liters per minute) for the intended
purpose of the well.
Soil formation
As it moves through soil, water dissolves
chemicals and sometimes deposits them at lower
levels. This leaching process can deplete the
topsoil of essential nutrients (Hess 2014). Fine
3241

GRO UN DWATE R
particles such as clay are picked up by water and
carried in suspension and deposited elsewhere at
greater depths. Water, thus, constantly changes
the physical, chemical, and biological makeup of
the soil as it moves downward due to gravity, or
sideways depending on available openings. When
silica is removed by dissolution leaving behind
the more insoluble iron compounds, reddish col-
ored soils such as latosols are produced in tropical
and subtropical regions of the world. In mid-
and high latitudes, water removes minerals like
oxides of aluminum and clays and forms shallow
acidic soils like podzols. Gley soils are acidic
and oxygen-poor and develop in waterlogged
conditions in cool climates. In the more arid and
semiarid regions, capillary action brings up soil
PR
moisture, which evaporates leaving behind salts
like chlorides and sulfates. Soil salinization is
exacerbated by inadequate drainage and low pre-
cipitation rates, resulting in calcic and other salt
hardpans.
 
ME
Groundwater chemistry
Groundwater quality may be affected by contam-
inants of both natural and human origin. Factors
that determine groundwater quality include
the thickness of the aquifer, aquifer structure,
mineral composition of the aquifer, presence or
LU
absence of a confining layer, presence of direct
conduits (e.g., abandoned wells or sinkholes)
to the aquifer, and in the case of unconfined
aquifers, presence of pollutants on the overlying
land surface. Unconfined aquifers are more sus-
ceptible to contamination from human activities
VO
because of their proximity to pollution sources.
The age or residence time of water in the aquifer
is also a significant factor affecting groundwater
quality as longer residence times provide more
opportunity for dissolving minerals.
Many substances may be dissolved by ground-
water as it moves through the subsurface. The
3242

most common chemical constituents in ground-
water include sodium, calcium, magnesium,
potassium, chloride, and sulfate. However,
FS
depending on mineral composition of the
subsurface and on length of exposure, high
levels of arsenic, fluoride, iron, manganese, or
radionuclides may also be found.
OO
Modeling and visualizing groundwater
The importance of modeling groundwater
Because of its intractability, understanding the
processes associated with groundwater is more
complicated relative to those associated with
surface water. Hence, monitoring, mapping, and
modeling efforts are crucial for this purpose.
As in any area of inquiry, however, the appli-
cability and accuracy of groundwater models
is highly dependent upon (i) the quality of the
input data (i.e., knowledge of initial conditions
and the existence of adequate hydrogeological
data) and (ii) the “fit” between the model-
ing approach (assumptions and equations) and
the scope and purpose of the problem being
explored.
Many groundwater models in one, two, or
three dimensions rely at least to some degree
upon Darcy’s Law (Henry Philibert Gaspard
Darcy, 1803–1858), which among other things
allows for estimating water velocity in an aquifer
as well as time required for water to travel
between points within the aquifer, and which
holds for nearly all hydrogeological condi-
tions. In brief, Darcy’s Law provides that water
flows only when there is a gradient, that water
flows down-gradient from high to low head
(pressure), that flow velocity is proportional
to head loss, and that velocity is modified
by the hydraulic conductivity of the aquifer
materials.

 
Modeling approaches and data needs
There are three types of groundwater models:
analytical element, finite element, and finite dif-
ference. The latter two are data driven and are
calibrated or fitted with extensive field data. The
simplicity of the analytical element method is in
its assumptions. It uses points, lines, and polygons
to spatially represent contaminant sources, rivers,
flow barriers, and groundwater zones. Assump-
tions such as flat water tables, uniform aquifer
thickness, and constant conductivity are typical
in such modeling efforts. However, such models
may not accurately represent the complex geom-
etry and heterogeneity of the real world of sub-
surface groundwater (Bennett 2014).
Depending on the scope and purpose of
PR
modeling efforts, data of varying quality at com-
patible scales may be required. Generally, the data
required includes hydrologic measurements (e.g.,
precipitation, infiltration, evapotranspiration),
 initial and boundary conditions (e.g., poten-
tiometric surface height, presence of confining
layers), subsurface characteristics (e.g., thick-
ME
ness, composition, hydraulic conductivity), and
human conditions (e.g., presence and pumping
rate of wells, surface modifications). There is
much more detailed hydrologic and subsurface
data available in developed nations than in
developing nations. For example, in the United
LU
States, vast amounts of hydrologic data may be
obtained from federal government agencies such
as the US Geological Survey and the National
Oceanic and Atmospheric Administration and
most states require submission of well boring
logs to permitting authorities for every well that
VO
is drilled, resulting in a slowly increasing store of
knowledge of the subsurface. In recent decades,
satellite technology has also been employed to
search for signs of groundwater (e.g., landforms
such as ancient lake beds) or overexploitation of
groundwater (e.g., changes in surface elevation)
over time.

G RO U N DWAT E R
Software use in groundwater modeling
Many commercial and open-source software
packages exist for purposes of modeling and
FS
displaying various aspects of groundwater
quantity and quality, including flow, transport,
and geochemical reactions. A selected list of
open-source software for groundwater models is
provided by the US Geological Survey (USGS
OO
2014b).
MODFLOW is the USGS’s three-dimensional
(3-D) finite-difference groundwater model
(USGS 2013) that is considered an interna-
tional standard for simulating and predicting
groundwater conditions and the interactions that
take place between ground and surface water.
Originally developed and released in 1984 as
a computer code for simulating groundwater
flow, its modular structure provides a robust
framework for integrating additional features.
Current capabilities include simulations of cou-
pled groundwater and surface-water systems, 
solute transport, variable-density flow (includ-
ing saltwater), aquifer-system compaction and
land subsidence, parameter estimation, and
groundwater management (USGS 2013).
To enable users to make appropriate soft-
ware selections, Kumar (2012) has provided
a review of the capabilities and limitations of
available groundwater modeling software. 3-D
groundwater computer software uses borehole
stratigraphy, geologic layers, cross sections,
and terrain data (for example, digital elevation
models) to construct 3-D models of the sub-
surface (Strassberg, Maidment, and Jones 2014).
With its ability to provide preprocessing data
input to MODFLOW, and store results from
MODFLOW, the Arc Hydro Groundwater data
model of the flagship software company ESRI
of California (Strassberg, Maidment, and Jones
2014) has created a very rich and comprehensive
environment for groundwater modeling and
visualization.
3243
 

GRO UN DWATE R

FS
Discharging
playa Dry playa Dry playa Discharging

FA
UL
playa

T
Springs
Local
Water Table
Basin fill
Local Basin fill
Basin fill Intermediate

OO
Low-permeability rock Permeable rock Low-permeability rock
Regional
Undrained Partly drained Drained Partly drained
Sink
Closed basin Closed basin Closed basin Tributary area

Single-valley system Regional system

Phreatophytes
PR
Approximate location of local,
intermediate, and regional systems
Groundwater flow

Faults

Figure 3 Schematic block diagram of Death Valley and other basins illustrating the structural relations between
mountain blocks, valleys, and groundwater flow. Source: Faunt, D’Agnese, and O’Brien 2010. Reproduced by
 permission of USGS. 
ME

Groundwater models and geovisualization, (e.g., tracing historical contamination to its

selected examples source) scales. Two examples of 3-D ground-
water modeling efforts are provided below and
Hydrogeologists often use diagrammatic repre-
include predicting the migration of a uranium
sentations (Figure 3) to characterize and visualize
plume (Figure 4) and assessing risk of encounter-
the subsurface flow systems and the influence
ing high arsenic concentrations at various depths
of geology and climate on such systems over
(Figure 5).
LU

time. The movement of groundwater in the

subsurface of the Death Valley Regional Flow
System (DVRFS) is captured and depicted in
Figure 3 (Faunt, D’Agnese, and O’Brien 2010).
It shows the flow paths of groundwater move-
ment under the influence of hydraulic gradient
VO
To construct 3-D images for visualizing
concentrated plumes of contaminants such
as uranium in subsurface sediments of waste
sites, as shown in Figure 4, scientists at the US
Department of Energy (2009) have developed

through varying zones of permeability, from
recharge to discharge areas in a regional context.
In addition to testing local aquifers to deter-
mine well or aquifer yields, and using well data
to describe subsurface flow patterns, models are
also used to understand groundwater both at
spatial (local to regional aquifers) and temporal
groundwater models. Such models enable the
prediction of movement of contaminants such
as uranium over tens to hundreds of meters on
temporal scales ranging from hours to years.
Arsenic contamination of shallow groundwater
and its impact on the health of affected popula-
tions is one of the most serious environmental

3244


 

G RO U N DWAT E R

UO22+ concentration (M)

Saturation: 0.05 0.25 0.45 0.65 0.85
5E-07

FS
1E-07
8E-08 N

110

OO
105

Z (m)
100

95

1000
PR 600
800

0
)
200 400 (m
Y
 
400
200
X (m 600
)
ME

800 0 Time: 0.42 y (May)

Pressure (Pa): 10 000 50000 90000 130000 170000 210000

Figure 4 Three-dimensional plot of the uranium plume in groundwater beneath a portion of the Hanford
300 Area (a nuclear reservation site in Washington, USA), using PFLOTRAN model. Source: US Department
LU

of Energy 2009.

problems in developing countries. In Figure 5, Social, political, and planning aspects of

the authors of a study (Winkel et al. 2011) groundwater
on groundwater contamination of the Red
VO

River Delta in Vietnam show the probability of

Arsenic (As) concentration exceeding 10 μg/l
at 10 m depth intervals. Such analyses resulting
from models enable us to study the connections
between large-scale pumping of groundwater
and the resulting vertical migration of arsenic in
the subsurface.
Historical access to and use of groundwater
Throughout history, humans have accessed
groundwater in several different ways, likely
beginning with locations where seeps or artesian
upwellings occurred, progressing to hand-dug
wells and wells accessed via windlass and bucket,

3245


 

GRO UN DWATE R

Hanoi
N
Haiphong

FS
Thaibinh

0m
Depth

0–1

OO
m
20
10–

m
30
Probability 20–
As >10 μgL−1
0.8–0.9 m
40
0.7–0.8 30–
0.6–0.7
0.5–0.6
0.4–0.5
0.3–0.4
PR 40–
50
m

m
0.2–0.3 60
0.1–0.2
50–
 
Figure 5 Risk of arsenic pollution of groundwater in Vietnam plotted in three dimensions and at 10m depth
intervals. Source: Winkel et al. 2011. Reproduced by permission of the Proceedings of the National Academy of
ME

Sciences.

and finally tube wells with electric or fossil
fuel-powered pumps.
Qanats, or hand-dug tunnels with numerous
vertical shafts that access groundwater and deliver
LU
demanded a cohesive community to successfully
maintain channels. Flow in qanats varies with
the level of the water table in the aquifer, which
ensured sustainable use of the aquifer but also

water by gravity for irrigation or domestic use,
have been in use for more than eleven centuries
throughout much of the Middle East, western
Asia, southern Europe, northern Africa, and
Central America. This technology is believed
VO
may have limited use for crops (and therefore
crop yields) or other uses during periods of low
recharge. Although qanats are still in use in many
regions, their importance for domestic supply
and irrigation has decreased due to increasing

to have originated in Persia and spread via
trade routes and expansion of Arab and Roman
empires. Water from qanats was historically
viewed as a community resource to be shared,
and the maintenance and cleaning of qanats was
also a laborious and challenging exercise. Thus
the use of qanats for accessing water in effect
use of the same aquifers by powered pumps
or other hydrological modifications that may
have reduced recharge (e.g., as in the case of
Morocco’s Tafilalet oasis and aquifer system).
Rapidly growing urban populations with
increasing water use and consumption during
the twentieth and twenty-first centuries have

3246


 
stressed aquifers with relatively high withdrawal
rates. In the case of coastal megalopolises such as
Houston, New York, and Chennai, the increas-
ing water use rates have outstripped recharge
rates, thus enabling saltwater intrusion into
coastal aquifers. Human population and agricul-
tural centers have not all been developed in close
proximity to locations with ample surface or
groundwater resources. Major river systems or
areas of significant precipitation often give rise to
the highest yielding alluvial freshwater aquifers.
Large human settlements in arid and semiarid
regions and areas overlying saline groundwater
have long been water stressed.
Groundwater rights
Modern groundwater rights are most often
PR
determined by ownership of overlying land, or,
where legally permissible, by leasing or purchas-
 ing the rights to groundwater residing beneath
lands owned by others. The pumping or abstrac-
tion rate may be determined by principles such
ME
as absolute ownership (pumping without regard
to depletion or conflict with others), correlative
rights (considering the rights of other users), or
a permit that specifies a maximum pumping rate.
Historically, groundwater pumping rates have
not been of significant concern, except in water
LU
scarce regions.
In recent decades, with increasing ground-
water consumption for irrigation and other
purposes, and with more complete information
about aquifer extents, capacities, and depletion
rates, pumping has become increasingly reg-
VO
ulated in semiarid regions. In some situations
(e.g., Texas, United States), groundwater rights
may be separated from the overlying lands and
may be purchased or leased and exploited by
individuals or entities who do not own the land.
Groundwater rights in some locations are also
affected by the distinction between percolating

G RO U N DWAT E R
groundwater (e.g., as in a sandstone aquifer) and
groundwater flowing as a subterranean stream in
a defined channel (e.g., as in a karst formation).
FS
Groundwater, conflict, and transboundary
aquifers
The destruction, deliberate draining, or poison-
ing of wells has been a feature of many conflicts
OO
throughout history; instances include Assyria’s
use of rye ergot to poison enemy wells in the
sixth century BCE, Saladin’s sanding of wells in the
defeat of the Crusaders in the twelfth century CE,
poisoning of wells in German Southwest Africa
by retreating German troops in 1915, Japan’s
use of biological weapons to contaminate wells
in China in 1939–1942, Serbian cut-off of well
water to Bosnians in Sarajevo in 1992, Botswana
troops’ destruction of Khoisan (Bushmen) wells
in 2002, and intentional contamination of Darfur
wells in 2004 (Gleick et al. 2012).
There are numerous transboundary aquifers 
worldwide, some of which are shared by more
than two nations. The nature of aquifer sharing
includes several different types of situations, for
example:
• two or more nations overlie an unconfined
aquifer;
• two or more nations overlie a fossil aquifer
with no meaningful recharge;
• one nation has the recharge zone and part
of an unconfined aquifer within its territory,
whereas down-basin nations have access only
to the water in the aquifer (Eckstein and
Eckstein 2005).
With some exceptions, and despite the
catastrophic potential impacts of aquifer contam-
ination and depletion, there has been relatively
little progress on the international management
of groundwater. In 2008, the United Nations
adopted Resolution A/RES/63/124 on the Law
3247

GRO UN DWATE R
of Transboundary Aquifers, which provides a
set of guidelines and recommendations to assist
nations in establishing bilateral and multilateral
agreements. More recently, the UN has discussed
establishing a more formalized convention on
transboundary aquifers. Key issues in the man-
agement of transboundary aquifers include prior
use patterns, military conflict, economic growth,
population growth, irrigated agriculture in arid
or semiarid lands, land use, climate change, and
the proportion of overlying land or recharge area
controlled.
There are several notable regional conflicts
in which groundwater plays an important
role. Groundwater is interwoven into the
Israeli–Palestine conflict because the Mountain
PR
Aquifer provides a significant proportion of the
domestic and agricultural water to Israeli terri-
tory, but the recharge zone lies nearly entirely
within the West Bank. The lack of stable and
durable peace in the region makes it difficult
 
to manage and use the resource efficiently and
also construct and operate water and wastewater
ME
treatment facilities in the West Bank so as to
protect the highly vulnerable aquifer.
The Nubian sandstone aquifer is a fossil
aquifer in northern Africa shared by Libya,
Egypt, Sudan, and Chad. In 2013, these four
nations adopted the Regional Strategic Action
Plan on the Nubian Sandstone Aquifer in
LU
response to declining water levels, increased
salinization in the Post-Nubian System (north of
the Nubian System), and the declining health of
dependent ecosystems.
VO
Key issues related to planning and future
sustainability of groundwater resources
Water quantity: climate change
Climate change is already impacting groundwa-
ter and will continue to impact it in at least two
3248

important ways: (i) groundwater may become a
more valuable resource in a changing climate,
where precipitation becomes less predictable
FS
and more “flashy,” and (ii) aquifer recharge
may also be altered by changes in precipitation
patterns and evaporation rates. Climate change
impacts on groundwater resources may be most
significant in areas that have scarce precipitation
or which already are water stressed, as well as in
OO
monsoon regions.
Water quantity: over abstraction
In addition to the 15 highly overexploited
aquifers of the world (shown in Table 2),
over-abstraction or mining of groundwater for
agriculture (e.g., San Joaquin Valley, California,
United States), industrial use, and drinking water
supply (e.g., Osaka City, Japan) has occurred in
numerous other aquifers worldwide. In many
locations, such overuse has caused permanent
compaction and aquifer capacity reduction and
accelerated sinkhole formation in karst regions.
Resulting land subsidence of several tens of
meters in some cases (e.g., Houston, Texas,
United States) have increased the vulnerability
of such areas to flooding. Due to approximately
8 m of subsidence in the last century, numerous
buildings in Mexico City have been damaged
and significant repairs have been necessary to
much of the subsurface infrastructure, including
subways, sewers, and potable water pipes.
Groundwater quality issues
Arsenic (arsenicosis), fluoride (fluorosis), and
radionuclides (mutations, cancer) are all found
in groundwater in some regions and have sig-
nificant human health impacts. For example,
relatively low concentrations of arsenic can be
found in groundwater in many regions; how-
ever, the highest concentrations consumed as

 
potable water appear to be found in the southern
Asian nations of Bangladesh, India, and Vietnam
(Mukherjee et al. 2006; Winkel et al. 2011).
Other groundwater constituents tend to have
impacts that are less health-related but significant
nonetheless. For example, the main impacts of
calcium and magnesium are the formation of
scaly deposits on pipes and reduced performance
of soaps and detergents. Sulfur compounds typi-
cally affect the smell and taste of water, whereas
high levels of iron can leave reddish stains on
fabrics washed with groundwater.
Groundwater contaminants of human origin
include industrial discharge, landfills and legacy
waste disposal sites, agriculture, toxic chem-
PR
ical spills, leaking underground storage tanks
(LUSTs), and domestic sewage discharge. The
behavior and toxicity of contaminants varies
widely, as does the ability to remediate and
restore groundwater quality. Nitrate, the most
 common agricultural groundwater contaminant,
is highly soluble in water, but its presence in
ME
affected aquifers may be significantly reduced by
source reduction or prevention. Other contam-
inants that are less soluble and less mobile may
be extremely difficult and expensive to remove
from aquifers. Dense nonaqueous phase liquids
(DNAPLs) such as trichloroethylene (TCE)
tend to sink toward the bottom of aquifers
LU
where they may slowly degrade and release toxic
by-products. Gasoline from leaking underground
storage tanks is perhaps the most common light
nonaqueous phase liquid (LNAPL). Benzene,
toluene, ethylbenzene, and xylene (BTEX, four
VO
of the major hydrocarbons found in petroleum
products) spread in plumes at different rates
throughout the top of the aquifer. Microbial
contaminants may be introduced to aquifers
by improperly sited septic tanks as well as by
untreated or inadequately treated municipal
discharges above unconfined surficial aquifers.

G RO U N DWAT E R
Significant and growing threats to
groundwater: agriculture
Agricultural productivity has increased signifi-
FS
cantly over the past several decades. Yields have
risen due to increased use of pesticides, fertilizers,
and irrigation (often from groundwater sources),
as well as to expansion of cultivated areas. But
gains in productivity have been accompanied by
OO
increasing groundwater degradation and deple-
tion. For example, herbicides such as atrazine and
alachlor (used respectively on cereals and fruit
trees) have been extensively detected in ground-
water in Europe and North America; and in
South Asia, commonly used insecticides includ-
ing carbofuran and lindane have been widely
detected in groundwater (Mateo-Sagasta and
Burke 2011). Elevated nitrate levels, sometimes
exceeding the US Environmental Protection
Agency (EPA) maximum contaminant level of
10 mg/L are nearly ubiquitous in unconfined

shallow aquifers in major breadbaskets such as
the Midwestern United States.
Significant and growing threats to
groundwater: fossil fuel exploration and
extraction
Hydraulic fracturing or “fracking” is a drilling
technique used to obtain natural gas and oil
from shale rock and other nonconventional
sources (Bigham 2013). For every fracking well,
between 8 and 20 million liters of water along
with sand and hundreds of compounds (e.g.,
for lubricating, anticorrosion) are injected to
fracture deep tight rock formations, releasing gas
and oil from small pores within them (Figure 6).
Groundwater contamination associated with
fracking operations appears to be associated
mainly with (i) the disposal of “produced” water
and infiltration into shallow unconfined aquifers,
and (ii) well barrier and integrity failures, which
3249
 

GRO UN DWATE R

Fracking for fuel

Hydraulic fracturing is used to access oil and gas

FS
resources that are locked in nonporous rocks.

Water recovery tanks

Methane gas Polluted flowback water
escapes during the may be injected into a deep
mining process. storage well, recycled, or
sent to a treatment plant

OO
Blowouts
are possible.

Poorly treated

Water
table
Well
PR
Possible flow
Leakage of fracking
fluids from the pipe
has not been seen.
flowback water
has leaked into
drinking water.

of methane.

Leakage of
 fracking fluid 
Cement
Holes in the well from the
casing
casing allow fluid fracture zone
ME

to exit and gas to is highly

enter. unlikely.
Steel pipe

High-pressure fracturing fluid

High-pressure Gas flows

fracking fluid from the
opens networks fractures
LU

Shale of fractures in the into the

shale. Sand props pipe.
the fractures open.

Fissures
VO

Figure 6 Hydraulic fracturing. Source: Howarth, Ingraffea, and Engelder 2011. Reproduced by permission
of Nature.

3250


 
may allow migration of methane and fracking
fluids to aquifers used for drinking and irrigation.
Davies et al. (2014) indicate well failure rates
are highly variable (1.9–75%) depending upon
when and where they have been drilled. Other
recent studies indicate that groundwater in close
proximity to fracking operations is much more
likely to be contaminated with produced water
and methane, and that in some cases the chem-
ical composition of methane found in drinking
water aquifers was quite similar to the gas found
in the much lower shale formations ( Jackson
et al. 2013; Gordalla, Ewers, and Frimmel 2013).
To facilitate rapid exploration and recovery
of crude oil and natural gas, changes have been
made to existing legislation such as the United
PR
States’ Safe Drinking Water Act to exclude
hydraulic fracturing from the definition of
underground injection, a practice that was until
the early 2000s largely prohibited or tightly
regulated. Within the United States, fossil fuel
 exploration is mainly regulated at the state level,
but the US EPA has recently undertaken a
ME
nationwide study on fracking and may issue
nationwide regulations in the future. Hydraulic
fracturing is already being employed or planned
in Germany, the United Kingdom, India, and
China (Gleick et al. 2012), and numerous suit-
able formations for exploration and extraction of
resources via hydraulic fracturing exist on every
LU
populated continent.
Strategies for protecting and replenishing
aquifers
At least in part due to difficulties in directly
VO
observing groundwater, efforts to conserve and
protect the quality of groundwater resources have
lagged behind such measures for surface water. In
addition to international cases noted above, this is
also true at national scales. For example, whereas
the Clean Water Act is aimed at protecting
surface water, there are no federal environmental

G RO U N DWAT E R
laws focused on groundwater in the United
States. Much of the regulation of groundwater is
then left to individual states. But since aquifer and
FS
state boundaries rarely coincide, this leads or has
the potential to lead to problems similar to those
faced with international aquifers. One major
policy need, therefore, is to “scale up” policy
related to protecting and conserving aquifers.
Numerous strategies have been proposed for
OO
situations in which aquifers are stressed due
to over abstraction from irrigated agriculture,
including substituting dryland-adapted crops
(e.g., wheat, sorghum, millet) in place of rice
or maize, developing drought-tolerant strains
of rice or maize, reducing tillage, reestablishing
native grasslands, limiting drilling of new wells,
and using moisture sensors and ultra-efficient
irrigation technologies. In urban areas where
aquifers are depleted, strategies such as aquifer
recharge via infiltration ponds or galleries and
injection of surface water or treated municipal
wastewater are being increasingly employed. 
SEE ALSO: Aquifers; Environmental
uncertainty; Hydrologic cycle; Soil water; Waste
and waste management; Water conflicts; Water:
drinking; Water and human rights; Water
quality; Water rights
References
Bennett, Ray R. 2014. “Analytic vs. Numeric
Ground Water Models.” Colorado Division of
Water Resources. http://www.cwi.colostate.edu/
southplatte/files/Presentations/Glove-Modflow-
SPDSS.pdf (accessed February 5, 2016).
Bigham, R. 2013. “Fracking Features.” Pollution Engi-
neering, 45(6): 39–41.
Davies, Richard J., Sam Almond, Robert Ward,
et al. 2014. “Oil and Gas Wells and Their
Integrity: Implications for Shale and Uncon-
ventional Resource Exploitation.” Marine and
Petroleum Geology, 56: 239–254. DOI:10.1016/j.
marpetgeo.2014.03.001.
3251

GRO UN DWATE R
Eckstein, Yoram, and Gabriel Eckstein. 2005. “Trans-
boundary Aquifers: Conceptual Models for Devel-
opment of International Law.” Ground Water,
43(5): 679–690. DOI:10.1111/j.1745-6584.2005.
00098.x.
Faunt, Claudia C., Frank A. D’Agnese, and Grady
M. O’Brien. 2010. “Hydrology.” In Death Val-
ley Regional Groundwater Flow System, Nevada and
California – Hydrogeologic Framework and Transient
Groundwater Flow Model, edited by Wayne R.
Belcher and Donald S. Sweetkind. USGS Profes-
sional Paper 1711. http://pubs.usgs.gov/pp/1711/
(accessed February 5, 2016).
Gleeson, Tom, Yoshihide Wada, Marc F.P. Bierkens,
and Ludovicious P.H. van Beek. 2012. “Water Bal-
ance of Global Aquifers Revealed by Groundwater
Footprint.” Nature, 488: 197–200. DOI:10.1038/
nature11295.
PR
Gleick, Peter, Lucy Allen, Michael Cohen, et al. 2012.
The World’s Water: The Biennial Report on Freshwater
Resources, vol. 7. Washington, DC: Island Press.
Gordalla, B.C., U. Ewers, and F.H. Frimmel. 2013.
 “Hydraulic Fracturing: A Toxicological Threat for
Groundwater and Drinking-Water?” Environmen-
tal Earth Sciences, 70: 3875–3893. DOI:10.1007/
ME
s12665-013-2672-9.
Hess, D. 2014. McKnight’s Physical Geography: A Land-
scape Appreciation, 11th edn. New York: Pearson
Publishers.
Howarth, Robert W., Anthony Ingraffea, and Terry
Engelder. 2011. “Natural Gas: Should Frack-
ing Stop?” Nature, 477: 271–275. DOI:10.1038/
LU
477271a.
Jackson, Robert B., Avner Vengosh, Thomas H.
Darrah, et al. 2013. “Increased Stray Gas Abun-
dance in a Subset of Drinking Water Wells
Near Marcellus Shale Gas Extraction.” Proceed-
ings of the National Academy of Sciences, 110(28):
VO
11250–11255. DOI:10.1073/pnas.1221635110.
Kumar, C.P. 2012. “Groundwater Modeling Soft-
ware – Capabilities and Limitations.” IOSR Journal
of Environmental Science, Toxicology and Food Tech-
nology, 1(2): 46–57. http://www.academia.edu/
2050290/Groundwater_Modelling_Software_-_
Capabilities_and_Limitations (accessed February
5, 2016).
3252

Mateo-Sagasta, Javier, and Jacob Burke. 2011. “Agri-
culture and Water Quality Interactions: A Global
Overview.” SOLAW Background Thematic
FS
Report TR08. Food and Agriculture Organi-
zation of the United Nations. http://www.fao.
org/fileadmin/templates/solaw/files/thematic_
reports/TR_08.pdf (accessed February 21, 2016).
McGuire, Virginia. 2013. “Water-Level and Stor-
age Changes in the High Plains Aquifer, Prede-
OO
velopment to 2011 and 2009–11.” United States
Geological Survey Scientific Investigations Report
2012–5291. http://pubs.usgs.gov/sir/2012/5291/
sir2012-5291.pdf (accessed February 5, 2016).
Mukherjee, A., M.K. Sengupta, M.A. Hossain, et al.
2006. “Arsenic Contamination in Groundwater: A
Global Perspective with Emphasis on the Asian.”
Journal of Health, Population and Nutrition, 24(2):
142–163.
Richts, A., W.F. Struckmeier, and M. Zaepke. 2011.
“WHYMAP and the Groundwater Resources of
the World 1:25,000,000.” In Sustaining Ground-
water Resources, edited by J.A.A. Jones, 159–173. 
London: Springer.
Shiklomanov, I. 1993. “World Freshwater
Resources.” In Water in Crisis: A Guide to the
World’s Fresh Water Resources, edited by Peter H.
Gleick. New York: Oxford University Press.
Strassberg, G., D.R. Maidment, and N.L.
Jones. 2014. “Arc Hydro Groundwater
Data Model.” http://www.crwr.utexas.edu/
gis/gishydro05/ArcHydroGroundwater/
ArcHydroGroundwaterESRIUC2005.htm
(accessed February 5, 2016).
US Department of Energy. 2009. “Subterranean
Blues: Models Track Contamination from Reser-
vation to River.” http://ascr-discovery.science.
doe.gov/synchronized/hanford1.shtml (accessed
February 5, 2016).
USGS. 2013. “MODFLOW and Related Programs.”
http://water.usgs.gov/ogw/modflow/index.html
(accessed February 5, 2016).
USGS. 2014a. “Glossary of Hydrologic Terms.”
http://or.water.usgs.gov/projs_dir/willgw/
glossary.htm (accessed February 5, 2016).

 
USGS. 2014b. “Water Resources Groundwater
Software.” http://water.usgs.gov/software/lists/
groundwater/ (accessed February 5, 2016).
Winkel, Lenny H.E., Pham Thi Kim Trang, Vi Mai
Lan, et al. 2011. “Arsenic Pollution of Groundwater
in Vietnam Exacerbated by Deep Aquifer Exploita-
tion for More Than a Century.” Proceedings of the
National Academy of Sciences, 108(4): 1246–1251.
DOI:10.1073/pnas.1011915108.
Further reading
Domenico, Patrick A., and Franklin W. Schwartz.
1997. Physical and Chemical Hydrogeology, 2nd edn.
New York: John Wiley & Sons.
Holden, Joseph, ed. 2014. Water Resources: An Inte-
grated Approach. New York: Routledge.
Growth machines
PR
 For urban geographers, the growth machine
Andrew E.G. Jonas
Hull University, UK
ME
The concept of the growth machine refers to
systemic properties operating within capitalist
cities which generate growth around the inten-
sification of land use. This often happens despite
pressures and political interests opposed to
such growth. The term “growth machine” can
LU
also refer to coalitions of business and local
government interests seeking to attract inward
investment, promote redevelopment, and market
the city. The growth machine concept was first
proposed by the American urban sociologist
Harvey Molotch in the 1970s (Molotch 1976).
VO
It was subsequently refined into a thesis about
the tensions between use and exchange value
arising from unregulated capitalist forms of
urban growth (Logan and Molotch 1987). By
connecting urban political agency to certain
systematic properties of capitalism, the growth
machine thesis was able to transcend some of
View publication stats

GROWTH MAC HIN ES
the shortcomings of social scientific analysis of
urban power structures in the USA.
Logan and Molotch (1987) suggested that
FS
the activities of rentiers – land development
interests – lie at the crux of understanding the
behavior of growth machines. Rentiers prepare
the ground for land-use intensification and
profit-making in the city. Often this is to the
benefit of a wide range of business interests
OO
and organizations, such as corporations, con-
tractors, the local media, universities, utilities,
sports franchises, chambers of commerce, trades
unions, and so forth. Local government is not
a disinterested party in this process as it strives
to influence land-use outcomes in ways that
enhance the local tax base and foster a sense of
community solidarity. Growth coalitions tend
to converge around those levels of government
having fiscal resources and land-use powers
necessary for delivering growth and jobs.

thesis has provided fertile territory for critique
and further refinement (Jonas and Wilson 1999).
Some critiques have highlighted how its focus
on the tension between use and exchange value
subverts knowledge of class tensions and progres-
sive urban politics. Others examine the discur-
sive tools deployed by growth machines and their
effects on marginalizing the voices of minorities
in the city. Then there are those who have sought
to apply the thesis to settings outside the USA,
finding that the national state is a more powerful
driver of urban development than local growth
coalitions. However, the jury is still out in respect
of whether the growth machine thesis is con-
ducive to the international comparison of urban
development politics.
Urban scholars today argue that knowledge
of power structures operating in cities requires
a shift in focus away from territorial growth
coalitions and towards relational processes and
networks operating beyond the city limits.
3253