ArticlePDF Available

Hydrocarbon Assessment and Prediction due to the Gulf War Oil Disaster, North Kuwait

Article

Hydrocarbon Assessment and Prediction due to the Gulf War Oil Disaster, North Kuwait

Abstract and Figures

Oil spill in the Gulf was the biggest disaster in history. The scale of damage was enormous, ranging from destruction caused by oil fires and oil spills, to economic decline for the Kuwaiti oil industry. The transport modelling of the freshwater aquifers in north Kuwait was undertaken to assess potential remediation scenarios using the MODFLOW-SURFACT numerical code. Three interlinked models were calibrated: flow, density (salinity), and transport. The model domain is a subregional area encompassing the Al-Raudhatain and Umm Al-Aish Basins. The time-variant salinity transport model was calibrated simultaneously with the transient groundwater flow system and this includes variably saturated flow and transport. This was done prior to proceeding to simulate contaminant hydrocarbon transport, as the hydraulic gradients and flow directions (and storage volume) are significant controls on contaminant migration. The results depicted after 23 years and with respect to the total area of the freshwater body at Al-Raudhatain (55.2 km2) and Umm Al-Aish (37 km2); the areal extent of the total petroleum hydrocarbon plume is estimated to be 7.3 and 8.7 km2 for the 0.1 mg/L contour, respectively. This equates to total petroleum hydrocarbon impacting 13 and 24% of the Al-Raudhatain and Umm Al-Aish freshwater bodies, respectively. The simulation indicated that even though total petroleum hydrocarbon loading was negligible in the center of the Al-Raudhatain depression up until the recent past, what has entered the groundwater system earlier from the contamination sources (pits, fringes, and lakes) is still moving toward the freshwater lenses or is potentially undetected due to lack of optimal existing monitoring bore screen and depth installations (plume diving). This implies that the environment is be the victim of war (the Gulf Crisis). The ecological and economic full impact probably will not be realized in the near future.
Content may be subject to copyright.
Hydrocarbon Assessment and
Prediction due to the Gulf War Oil
Disaster, North Kuwait
Yohannes Yihdego
1*
, Radwan A. Al-Weshah
2,3
ABSTRACT:Oil spill in the Gulf was the biggest disaster in
history. The scale of damage was enormous, ranging from
destruction caused by oil fires and oil spills, to economic decline
for the Kuwaiti oil industry. The transport modelling of the
freshwater aquifers in north Kuwait was undertaken to assess
potential remediation scenarios using the MODFLOW-SUR-
FACT numerical code. Three interlinked models were calibrat-
ed: flow, density (salinity), and transport. The model domain is
a subregional area encompassing the Al-Raudhatain and Umm
Al-Aish Basins. The time-variant salinity transport model was
calibrated simultaneously with the transient groundwater flow
system and this includes variably saturated flow and transport.
This was done prior to proceeding to simulate contaminant
hydrocarbon transport, as the hydraulic gradients and flow
directions (and storage volume) are significant controls on
contaminant migration. The results depicted after 23 years and
with respect to the total area of the freshwater body at Al-
Raudhatain (55.2 km
2
) and Umm Al-Aish (37 km
2
); the areal
extent of the total petroleum hydrocarbon plume is estimated to
be 7.3 and 8.7 km
2
for the 0.1 mg/L contour, respectively. This
equates to total petroleum hydrocarbon impacting 13 and 24%
of the Al-Raudhatain and Umm Al-Aish freshwater bodies,
respectively. The simulation indicated that even though total
petroleum hydrocarbon loading was negligible in the center of
the Al-Raudhatain depression up until the recent past, what has
entered the groundwater system earlier from the contamination
sources (pits, fringes, and lakes) is still moving toward the
freshwater lenses or is potentially undetected due to lack of
optimal existing monitoring bore screen and depth installations
(plume diving). This implies that the environment is be the
victim of war (the Gulf Crisis). The ecological and economic full
impact probably will not be realized in the near future. Water
Environ. Res.,89, 484 (2017).
KEYWORDS: hydrocarbon, pollution, contaminant, ecology,
environmental impact, health exposure.
doi:10.2175/106143016X14798353399250
Introduction
For many years, attention has been directed at contamination
from point sources such as oil spill landfills and hazardous
waste-disposal sites. The contaminants affecting water quality
are commonly spatially dispersed and may be widespread and
frequent in occurrence. The protection and enhancement of the
quality of groundwater resources is a high priority environ-
mental concern (Yihdego et al., 2016; Yihdego and Webb, 2008,
2016). Deterioration of groundwater quality may be virtually
irreversible, and treatment of contaminated groundwater can
often be expensive. Detection of groundwater contamination is
complicated by the ‘‘out-of-sight’’ nature of groundwater
(Yihdego and Webb 2007, 2009, 2010). Commonly, neither the
sources nor the effects of contamination are easily observed or
measured. Many contaminants are colorless, tasteless, and
odorless. The degree of threat posed by groundwater contam-
ination depends on many factors, including the concentrations
of the contaminants, their toxicity (individually or in combina-
tion), the volume of groundwater affected, the uses made of
water from the aquifer, the population affected by these uses,
and the availability of an alternative water supply (Dyer et al.,
2000; Haizhu et al., 2015; Rao et al., 2011; Rasa et al., 2013;
Yihdego 2010a, 2010b, 2016a).
More recently, concerns have increased about nonpoint
sources of contamination (e.g., oil fire, application of fertilizers
and wastewater sludge in agricultural areas, use of pesticides,
irrigation with wastewater, and acid rain) and about the overall
quality of groundwater resources.
The sources of components that determine groundwater
quality are numerous and diverse. These components can enter
groundwater in an aquifer through many different routes
(Giudici, 1991; Grostern and Edwards, 2006; Yihdego, 2016b),
including interactions with geologic formations; interactions of
groundwater with surface waterbodies; percolation of precipi-
tation water containing dissolved, colloidal, and suspended
materials; cross-flow between aquifers through natural frac-
1*
Snowy Mountains Engineering Corporation (SMEC), Sydney, New
South Wales 2060, Australia; e-mail: yohannesyihdego@gmail.com.
2
Department of Civil Engineering, The University of Jordan,
Amman, Jordan.
3
Kuwait Environmental Remediation Project, Kuwait City, Kuwait.
484 WATER ENVIRONMENT RESEARCH June 2017
tures; across the contact between the two aquifers; and man-
made boreholes open to more than one aquifer, direct entry
from the land surface through wells, and flow from adjacent
aquifers caused by pumping. Therefore, it is apparent that
groundwater can contain hazardous components to human
health derived from various naturally occurring sources or
introduced by human activities.
The Kuwaiti oil fires were caused by Iraqi military forces
setting fire to a reported 605 to 732 oil wells along with an
unspecified number of oil filled low-lying areas, such as oil lakes
and fire trenches, as part of a scorched earth policy while
retreating from Kuwait in 1991 due to the advances of Coalition.
When Iraqi troops withdrew from Kuwait at the end of the
Persian Gulf War in early 1991, they set fire to more than 600 oil
wells and pools of spilled oil in Kuwait, a parting shot that
exacted a significant economic toll on the country’s lucrative
petroleum industry. A massive oil spillage occurred in Kuwait as
a result of the detonation and ignition of oil wells (91.8% of the
total oil wells in Kuwait) by the retreating Iraqi troops during
the 1991 Gulf War. This has caused concerns about the
potentials of contamination of the groundwater resources by
petroleum hydrocarbons (Kwarteng et al., 2000; Maness et al.,
2012; Martin and Norman, 1992; Yihdego and Al-Weshah,
2016a). The contamination of groundwater from either natural
or anthropogenic sources is, therefore, of concern to Kuwait.
This has led to considerable interest in the design of investigative
studies and monitoring programs to describe groundwater
quality over regions that may range from tens of square
kilometers to an entire country. The public authority for
assessment of compensation for damages resulting from Iraqi
aggression has sponsored a series of studies to evaluate the
extent of this possible contamination and the steps necessary to
remediate the contaminated groundwater (SMEC 2002, 2006,
2014). Desalination plants supply the bulk of the freshwater used
in daily life in Kuwait. Brackish groundwater is, however, mixed
with desalinated water to make it potable, and is used for
irrigation and animal husbandry. Moreover, the limited reserve
of fresh groundwater in the northern part of Kuwait is bottled as
mineral water. The aim of this paper is to assess the effects of
the damaged oil wells and the spillage of crude oil on the
groundwater resources of Kuwait and predict the petroleum
hydrocarbon for determining effective remediation approach.
The data and interpretation in this paper have been extracted
from the SMEC (2014) consultant report.
Site Description
The study area is located north of Kuwait. Kuwait’s only two
freshwater aquifers, the Al-Raudhatain and Umm Al-Aish, are
located in the northern portion of the country in the Al Jarha
governate, near the Raudhatain and Sabriyah oil fields, beneath
the topograph depression associated with the Al-Raudhatain-
Umm Al-Aish drainage basin (Figure 1). The bulk of Kuwait’s
groundwater is brackish (Omar et al., 1998). These freshwater
aquifers are unique in that their recharge process reflects the
rapid infiltration of surface water runoff and their proximity to
the oil field makes them very vulnerable to contamination. The
damage caused from the destruction of oil wells during the Gulf
War of 1991 and seawater used to control and extinguish oil
fires has directly affected the aquifers.
Kuwait is an arid country, home to more than 3.5 million
people, characterized by long summers with extremely high
temperatures and high humidity, short mild winters with low
humidity, and high evaporation rates (Grealish et al., 2001), no
fresh surface water supplies and very limited renewable
groundwater. High-intensity rainstorms are common and
produce the surface water runoff through the wadi’s networks,
which provide recharge to groundwater aquifers.
Hydrogeology. The two main aquifers in Kuwait, providing
useable brackish water of salinities ranging between 2500 and
7500 mg/L, are the Mio-Pliocene clastic Kuwait Group and the
unconformably underlying Middle Eocene dolomitic limestone
of the Dammam Formation (Figure 1). An impervious zone at
the base of the Dammam Formation separates these two aquifers
from the underlying formations (Omar et al., 1998). Various
gradations of gravel, sand, silt, and clay materials with different
degrees of cementation form the Kuwait Group of rocks, with
thickness varying in relation to the structure of the underlying
Dammam Formation. The Dammam Formation consists of
dolomitic limestone of shallow marine origin with a thickness
varying between 120 and 300 m that generally increases toward
the northeast. In general, the Formation appears to have a low
effective primary porosity and very low hydraulic conductivity.
The underflow from Saudi Arabia is the main source for
recharge. Precipitation is a minor recharge source in the
depressions of northern Kuwait, where limited freshwater lenses
occur in the upper part of the Kuwait Group, especially in the
Raudhatain, Umm Al-Aish, and Abdally areas.
Figure 1—Location of the Al-Raudhatain and Umm
Al-Aish aquifers in Kuwait. The location of petroleum
contamination is located within the Raudhatain and
Umm Al-Aish freshwater fields.
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 485
Extent of Hydrocarbon Pollution in Kuwait. At the
aftermath of the Gulf War of 1991, 613 oil wells were on fire,
discharging crude oil and petroleum gases; 76 wells gushed oil
but were not set on fire; and the remaining 155 wells, although
damaged, neither gushed oil nor were on fire (Al-Besharah,
1991). The total oil burnt due to oil wells on fire was about 240
310
6
m
3
. The products of combustion were spread over an area
of about 1722 km
2
, representing about 10% of the total area of
Kuwait. The total oil spilled as a result of the damaged wells
was about 3.5 310
6
m
3
. The spilled oil was collected in about
300 oil lakes, spread over an area of about 49 km
2
. Out of these,
3.3 310
6
m
3
of oil was recovered and exported. The rest was left
behind, as it was not economical to collect them. The oil that
was left behind was subjected to severe weathering over the past
25 years. The locations of the water fields of Kuwait and the
extent of oil field/polluted areas are shown in Figure 1. The
Raudhatain and the Sabriya oil fields (Figure 1), are located
adjacent to the freshwater fields of Raudhatain and Umm Al-
Aish. The ground surface in the vicinity of the damaged oil wells
in these oil fields was heavily polluted with crude oil that flowed
from the damaged oil wells, products of combustion from wells
on fire and seawater used for extinguishing the oil fires. The
crude oil and seawater flowed down the dry ‘‘wadi’’ beds and
accumulated in the surface depressions in and around the
freshwater fields, giving rise to the ‘‘oil lakes’’.
Conceptualization of the Pollutants Travel from Surface to
the Groundwater. There can be four major alternative ways by
which the hydrocarbon pollutants on the ground surface can
reach the water table. These are: (1) direct infiltration of crude
oil from the oil lakes; (2) infiltration of rainfall runoff, carrying
with it hydrocarbon pollutants leached from the surface soil; (3)
infiltration of seawater used for extinguishing oil fires that was
contaminated with oil; and (4) oil leakage from damaged oil
casings. The first three possibilities will contaminate the
unconfined aquifer (Kuwait Group: Figure 2), and both the
unconfined and the confined (Dammam Formation) aquifers
may be contaminated.
Direct infiltration of crude oil was studied and concluded that
maximum direct infiltration of crude oil in soil would not
exceed 2 to 3 m (SMEC, 2006). Considering the depth to water
table in the affected areas (15 to more than 30 m), it is most
unlikely that direct infiltration of crude oil from oil lakes will
pose any threat to groundwater pollution. Fractures and joints
that can carry the crude oil directly to the water table are not
prevalent in the unconsolidated formation that hosts the
freshwater lenses. But large trees are rare in this arid
environment and the depths of the animal burrows are expected
to be limited by the hard calcareous layers and, anyway, may
not extend to the depth to water.
Rainfall in Kuwait averages about 110 mm per year whereas
the total evaporation is in the neighborhood of 2500 mm per
year. Under this low rainfall and high evaporation rate, natural
recharge of groundwater from rainfall is considered negligible.
At Raudhatain and Umm Al-Aish areas, however, unique
morphology of the ground surface makes the natural recharge
during occasional rainstorms possible. In these areas, the ground
surface elevations decrease from 90 to 100 m above mean sea
level in the southwest to about 40 m in the middle of the
Raudhatain depression and to 30 m in the middle of the Umm
Al-Aish depression (Figure 3). The infiltration rates, measured
with double-ring infiltrometer, range from 10 cm/h at the
centers of the depressions to 20 cm/h toward the peripheries.
These infiltration rates are high compared to the evaporation
rate (3–4 mm/d during the rainy season) and reduce the chances
of loss of runoff through evaporation (Martin and Norman,
1992). Monitoring wells and augur holes drilled in this area
confirm the presence of discontinuous semipervious layers at
depths of 1 to 3 m from the ground surface (Yihdego, 2013).
Coarse sand and gravel formations exist below these layers. The
rainwater from the occasional winter rainstorms, after rapid
infiltration, follows the general slope of the impervious layers
toward the centers of the depressions (Figure 4). The wadis
concentrate the surface and subsurface runoff created at the
centers of the depressions giving rise to ‘‘playa’’ lakes.
Infiltration of the accumulated water from these playas over the
past several thousand years has given rise to the fresh
groundwater lenses, floating over the brackish to saline water
under the depressions. Hence, though rainfall is low and
infrequent, it is likely to leach some of the pollutants and carry
them to the freshwater lenses. Because the level of rainwater
recharge is significant only in the Raudhatain and Umm Al-Aish
depressions, the largest surface depressions in Kuwait, the
probability of groundwater pollution by contaminated rainwa-
ter is significant only in these two depressions. In other areas,
groundwater contamination by polluted rainwater is very low
and probably non-existent (KISR, 2013; SMEC, 2006, 2014;
Yihdego and Al-Weshah, 2016a).
Average volume of seawater used for extinguishing oil fires
was about 95 000 m
3
/d. The process of extinguishing the fire
went on for about 240 days. Considering the overall spread of
the oil wells, the maximum depth of seawater infiltration will be
less than 230 mm. In addition, most of the seawater evaporated
due to intense heat at the time of extinguishing the fires. Hence,
the extent of contamination from the seawater might not be that
significant. Infiltration might, however, take place where excess
seawater used in extinguishing the fires flowed along the wadis
and accumulated in depressions along with oil. This was
especially true for the freshwater fields in north Kuwait. Here
the seawater, with pollutants, might infiltrate to the water table.
Moreover, the salts accumulated in the near-surface soil, after
the evaporation of seawater, may act as long-time source for
pollutants that will be carried by the surface and subsurface
runoff and will reach the aquifer with infiltrating rainwater
(KISR, 2009, 2012; SMEC, 2014; Yihdego and Al-Weshah,
2016b).
The threat of oil leakage from damaged oil well casings is
difficult to determine. The damage to the casing due to oil well
fire should be mainly confined to comparatively shallow depths.
If the water table is too deep in such locations, the leaking oil,
with time, will lose the volatile parts and the heavier fractions
will form a highly viscous to semi-solid residual blocking the
pore space and thus self-limiting the leakage. In case the water
Yihdego and Al-Weshah
486 WATER ENVIRONMENT RESEARCH June 2017
table is shallow and the leakage takes place either above but
near the water table or below the water table, pollution of
groundwater is possible. Nonetheless, the threat of groundwater
pollution from damaged oil wells is expected to be low.
Hydrocarbon contamination of groundwater in Kuwait could
be expected from two major sources. In the freshwater fields of
the Raudhatain and Umm Al-Aish depressions in north Kuwait,
leaching of the pollutants by rainwater and the subsequent
infiltration of the polluted runoff from the playa lakes could
contaminate the freshwater lenses (Figures 5, 6, 7, and 8). The
leakage of oil from the damaged oil casings could contaminate
the groundwater directly in the neighborhood of such a well. In
Figure 2—Hydrostratigraphy of Kuwait (after Mukhopadhyay et al., 2006).
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 487
the absence of suitable morphometric features that could
concentrate the runoff from rare rainstorms, the possibility of
rainwater infiltration in the brackish water fields in the central
and southern Kuwait is remote. Leakage from the oil wells,
therefore, remains the most plausible anthropogenic contami-
nation source for the groundwater in these areas.
Modelling of Hydrocarbon Contamination
of the Groundwater in Northern Kuwait
The groundwater model of the Al-Raudhatain and Umm-Al-
Aish basins were undertaken using the MODFLOW-SURFACT
numerical code (HGL, 2013). Because of its ability to simulate a
wide variety of systems, its extensive publicly available
documentation, and its rigorous peer review, MODFLOW has
become the worldwide standard groundwater flow model
software.
MODFLOW-SURFACT is used for the variably saturated
three-dimensional groundwater flow and transport modelling,
and is presently considered a code that complies with industry
standards. The flow and transport models were calibrated by
using a manual iterative parameter approach and the parameter
optimizing program PEST (Doherty, 2004). Because no single
numerical technique has been effective for all transport
conditions, the combination of these solution techniques, each
having its own strengths and limitations, is believed to offer the
best approach for solving the most wide-ranging transport
problems with efficiency and accuracy.
Overcoming the difficulties and expanding the simulation
capability of MODFLOW-MT3D solutions can be done using
MODFLOW-SURFACT. This is a comprehensive simulator for
a set of equations for evaluating multicomponent contaminant
transport within multiple phases in the subsurface by using the
vadose zone flow and transport equations in unique ways to
perform the wide range of analysis needed for understanding
and managing contamination and remediation projects (Buttler,
1995; Moe et al., 2009; Panday and Huyakorn, 2008; Yihdego
and Al-Weshah, 2016c).
In this study, the model setup transport model for the total
petroleum hydrocarbons (TPH) concentration was based on the
transient flow and salinity transport models. MODFLOW-
SURFACT simulated the unsaturated-saturated domain. The
modelling activities were carried out via the transport of the
non-aqueous phase liquid (NAPL) from the oil lakes through the
unsaturated concentrated on the transport and fate of the
contaminants in the saturated zone.
Model Design. The study area, measuring 39.5 km 314.7
km at its largest extent totalling 580 km
2
was subdivided into
163 rows with 120 columns, producing a total number of 97 800
Figure 3—Ground surface elevation map of Raudhatain and Umm Al-Aish areas.
Yihdego and Al-Weshah
488 WATER ENVIRONMENT RESEARCH June 2017
cells (Figure 9). The active part of the model grid is 532 km
2
.
The model domain is a subregional area encompassing the two
basins (Figure 9). Each cell was approximately 100 m 3100 m
and the peripheral domain has a cell size of 1700 m 3757 m. The
model grid has been rotated 53.6 degrees anti-clockwise from the
east–west direction to be subparallel to the orientation of the
regional geology and groundwater contours.
Numerical modeling of the possible hydrocarbon contami-
nation of the groundwater in the Raudhatain and Umm Al-Aish
freshwater fields in northern Kuwait was attempted to assess the
possible extent of pollution and its spread with time. This
involved the development of conceptual models for the flow of
groundwater and the transport of hydrocarbons in the study
area; simulation of the groundwater flow and contaminant
transport in the study area based on these conceptual models;
and prediction of the contamination levels in the study area in
the future, should current contamination levels persist in the soil
and in the groundwater.
Model Layers. Model layer details are given in Table 1. The
active model consists of five layers with the representation of the
model presented in Table 1.
For the purpose of modelling, Layer 1, which corresponds to
Aquifer 1, is mainly unsaturated but is important for recharge
and vadose zone source migration. Aquifers 2 and 3 are
saturated with Aquifer 2 containing the main majority of the
freshwater lenses and Aquifer 3 is predominantly saline.
Modelling of the groundwater systems concentrated on
Aquifers 2 and 3, which remain saturated throughout the
model period.
Modelling Approach. The modeling approach was to develop
a calibrated groundwater flow model, then to add a density
(salinity)-dependent (transport) component to account for the
saline contamination that was coupled with and the transient
hydrocarbon contamination (transport) component to allow
assessment and prediction of hydrocarbon pollutants.
The initial stage involves construction of the numerical model
and calibration against known data. Calibration started with a
steady-state flow model developed to study regional flow
patterns in the aquifer and to calibrate the model’s recharge and
hydraulic conductivity among others. The steady-state flow
model looks at an average (long-term) period and allows static
calibration of, for example, hydraulic properties, to a known or
measured data set. A transient flow model was then developed
to assess seasonal recharge on groundwater and investigate their
effects on flow patterns. A transient flow model allows for time-
dependent calibration of known changes, for example, concen-
tration or water levels over time. A density (salinity)-dependent
transport model was required to be coupled to the flow model as
the density of the saline contamination (seawater) can drive flow
and contamination movement within the groundwater system.
This was then coupled with the hydrocarbon transport
components to produce the final model. The combined flow and
transport model was then used to simulate remediation
scenarios.
Figure 4—Idealized conceptual model and infiltration of rainwater runoff in the surface depressions of the
freshwater aquifer systems in North Kuwait (after Yihdego and Al-Weshah, 2016b).
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 489
The model was set up to run from March 1991 until the end of
2013. The hydrocarbon was assumed to first enter the system at
the time when the oil wells’ fire fighting began in 1991. Initial
heads for the steady-state model were set at 13 masl across the
domain. It was important to use a level slightly higher than the
highest heads expected to result from the model as it aids model
convergence by ensuring that no elements that should be
saturated would start the model as dry elements. The initial
concentration was set 0, assuming the background value of the
TPH was nil.
Boundary Condition. It was assumed that this maximum
concentration in the groundwater was reached every time
recharge to the water table carried contamination into the
aquifer. Therefore, time-variant constraint conditions were
applied to the constant mass boundary condition nodes, to
ensure they were only active at times when recharge was
entering the aquifer. The drainage pits, wadis, and fringes of
lakes were mapped and the concentrations in these areas were
set to stepped concentrations. Contaminants typically enter a
groundwater system with a fluid source (recharge) that contains
the contaminant. Consequently, a third-type solute boundary
condition (Cauchy) is logical match in which the source
concentration is specified and associated with outflow. The
concentration of the TPH was allowed to vary as a function of
time. It was assigned as 25 mg/L for the first year since the
source was added and 15 mg/L for the rest of the simulation
time. The sink is represented in the form of dispersion and
sorption. The discharge/outflow via biodegradation and chem-
ical reaction are not considered in this model.
Soil and Aquifer Properties. The aquifer properties adopted
are based on the conceptual model, results of hydraulic tests,
observations collected during purging of the monitoring wells,
and model calibration. For parameters in which no data were
available, industry-accepted values and best estimates were used.
The model was calibrated using the adopted data; the calibrated
aquifer hydraulic conductivities and variably saturated parame-
ters are summarized in Table 2 and Table 3, respectively. There
are different hydraulic conductivities applied to each freshwater
field as well as in the area of brackish water, beyond the extent of
the model aquifers. The results of the pumping test data were in
agreement with the chemistry data and showed variable hydraulic
conductivity with depth.
The Van Genuchten soil function (using upstream weighing)
was used to solve the variably saturated water flow equations,
which calculates conductance in all directions including the
vertical as per saturated conductance times a relative perme-
ability. The functional expressions used to describe the
relationship of relative permeability versus water saturation and
Figure 5—Existing Monitoring Bore Network in the Al-Raudhatain and Um Al-Aish Basins.
Yihdego and Al-Weshah
490 WATER ENVIRONMENT RESEARCH June 2017
the relationship of pressure head versus water saturation are
described in Panday and Huyakorn, (2008).
Transient-State Model. In arid environments, such as
northern Kuwait, there is no true steady-state situation. This
is because recharge only occurs in sudden intense events and the
system is, therefore, always responding to or recovering from the
last recharge event. However, the use of a steady-state model
here to approximate the initial conditions of the system is
justifiable because the available head data for calibration was an
average water level.
This model is designed to simulate contaminant transport
models beginning at the oil fire events of March 1991 to July
2013. Therefore, the transient model consists of sequential runs,
each with 1 month’s duration divided into 269 stress periods and
730 time steps of 12 hours over this time period.
The initial heads for the transient state model were the
resulting heads from the steady-state model. These were a good
starting point for the transient-state model because this model
represented the average conditions of the transient-state model.
Transient Recharge. A spatially varying recharge condition
based on simulated net infiltration was applied to the top model
boundary, with a constraint of .35 mm/mon (that is, there will
be no recharge with rainfall of ,35 mm/mon). The time-variant
recharge was applied as a percentage of average monthly
precipitation (12%) to the recharge locations.
The infiltration rate is based on the earlier discussion of the
likely ranges of rates that may be possible in the various regions
of the two catchments. The rates vary from the very rapid
infiltration/recharge experiment carried out in Raudhatain by
Parsons (1964), which showed a sustained water table response
after only 4 days (17 m/4 d ¼4 m/d), to quite low recharge
observed in the regions containing gatch. Here, the authors tried
to account for the substantial water and oil volumes that were
likely to have discharged along wadis, drained to depressions to
create lakes, and then subsequently drained to collection (open)
pits. Each of these environments had a large potential to
recharge water, and each remains drainage pathways and
collection points to this day and until remediation occurs.
Infiltration is largely a function of the intrinsic hydraulic
conductivity of the soil materials and the ponded head of water/
oil to create downward flow. The studies to-date would suggest
that 4500 mm/yr is not unreasonable in the first year. In
Figure 6—Composite TPH Concentrations from 2008 (Sample Analysis by KISR; Note: Depth of Screen Interval
or Aquifer from which the sample has been collected was not indicated). Red ¼region of highest TPH, blue¼
region of low TPH. With TPH concentration contours (iso lines) shown in the figure (after Yihdego and Al-
Weshah, 2016a).
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 491
addition, the rainfall data indicate that there was potential for
recharge every year between 1993 and 1997 with years 1993 and
1996 being exceptionally wet. Combined with the focused
drainage and recharge, these conditions would lead to rapid
leaching to the water table of hydrocarbons and salt.
Seawater Recharge. Extensive amounts of saline water were
used for firefighting the oil well fires, about 6 310
6
m
3
of
seawater or brackish water. A total of 361 seawater storage pits,
or well-head pits were excavated to support fire-fighting
operations. Limited soil analysis has shown that these areas now
have high total dissolved salts and sodium concentrations.
There was insufficient data available to model the effects on
the salinity distribution of the oil-fire related seawater recharge
of 1991. There was no data available on the volume of water, the
rates at which it was entering the area, the groundwater
concentrations over time, or the time-variant groundwater
heads. An assumed recharge rate of 1,000 m
3
/d was applied over
the entire recharge areas of the depressions for this purpose.
The model recharge flow boundaries were set up to simulate
the seawater recharge at the assumed rate of 1000 m
3
/d total
among all locations of contamination sources, between March
and November 1991. This was done to investigate the significance
of seawater recharge in terms of flow and transport model
(simultaneous calibration of the flow and salinity transport was
carried out to attain this objective). The volume of the seawater
may be significant in terms of flow direction, hydraulic and
concentration gradients, and, hence, contamination migration.
Storage. Transient groundwater models need to be able to
calculate changes in groundwater storage to simulate changes in
heads and flows over time. The storage parameters, derived
from model calibration, are tabulated in Table 4. The specific
storage is the storativity divided by the saturated thickness of the
layer. The specific yield expresses the volume of water that is
released per unit of water table drop per unit surface area.
The effective porosity of the aquifers was assumed to be 0.29,
which is higher than the estimated specific yield, as some water
is retained by the aquifer matrix against the force of gravity. The
effective porosity of the semipermeable layers was set to the
value of 0.15.
Hydrocarbon Transport Properties. During the first three
rounds of sampling and analysis carried out during 2012, only
fluorescence fingerprinting (about 20 times than that of the
cyclohexane extracts) was used to investigate the organic
contamination in groundwater samples. This measure was taken
Figure 7—Composite TPH concentrations from 2009 (Sample Analysis by KISR; Note: depth of screen interval
or aquifer from which the sample has been collected was not indicated). Red ¼region of highest TPH, blue ¼
region of low TPH. With TPH concentration contours (iso lines) shown in the figure.
Yihdego and Al-Weshah
492 WATER ENVIRONMENT RESEARCH June 2017
because in previous sampling and analysis campaigns, target
analytes such as benzene, toluene, ethylbenzene and xylenes
(BTEX) and polycyclic aromatic hydrocarbons (PAHs) were
only detected in quantities close to analytical detection limits or
not at all (using GC-MS). Also TPH measurements using GC-
FID produced only broad UCM profiles. Fluorescence finger-
printing did however provide evidence that certain parts of the
aquifer of north Kuwait are contaminated with petroleum-
related compounds (mainly polar).
The material properties for flow remained consistent with
those determined in the steady-state flow model calibration.
Hydraulic and transport models are described in the previous
sections. Oil parameters are tabulated in Table 5.
Sorption. Typically, the movement of hydrocarbons (TPH) in
soil and groundwater is slower (retarded) in relation to the
movement of groundwater or water itself. The soil/water
partitioning coefficient (Kd), is the parameter routinely used to
describe the ability of the aquifer matrix to retard the movement
of contaminants. The retardation is due in part to the sorption
of the hydrocarbons onto the aquifer matrix. The concentration
of organic carbon, typically expressed as fraction of organic
carbon (foc), on the soil matrix has a major impact on this
retardation. AACM International during their soil survey
analyzed 35 soil samples collected from all over Kuwait for foc.
Samples were collected from depths of between 1 and 60 cm
below the surface. The foc in those samples varied between 0.05
and 0.42%. The foc in soil samples collected near the freshwater
fields varied between 0.09 and 0.15%. There are no data
available for organic carbon concentrations either at the depths
greater neither than 0.6 m nor in the saturated part of the
aquifers. Considering the above, it was assumed that the organic
carbon content of the unsaturated materials was 0.1%. Sampling
of monitoring wells in May and August, 2002, indicated that
groundwater was aerobic. This implies that the organic carbon
in the aquifer is low or that the remaining concentration of
organic matter is not easily degradable. In the former case,
sorption would be low. The soil/water partitioning coefficient
(Kd) is often written as follows:
R¼1þKd 3qb=nð1Þ
where
Kd ¼foc 3Koc ð2Þ
where
Figure 8—Composite cyclohexane extract concentrations from 2012 (sample analysis by KISR; Note: depth of
screen interval or aquifer from which the sample has been collected was not indicated). Red ¼region of
highest cyclohexane, blue ¼region of low cyclohexane. With cyclohexane concentration contours (iso lines)
shown in the figure.
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 493
Koc ¼an organic carbon partition coefficient,
q
b
¼the bulk density of the soil/aquifer matrix, and
n¼effective porosity.
Koc (or Log Koc) values for most of the common chemicals
are widely published. For example the Koc value for naphtha-
lene is 550 (SMEC, 2014). Previous studies presented Koc and Kd
values for some TPH compounds, however they used a foc value
of 0.0164 (1.64%) which seems high in the current simulation
and also found that for low-carbon aquifer materials that
sorption of organics was non-linear, and was difficult to
measure.
The retardation of the organic components was developed on
data of the Octanol/water partition coefficient’s and the
Karickoff KD approach (SMEC, 2014). The fraction of natural
organic carbon foc in the sediment was estimated to be 0.001.
Retardation coefficient for TPH was set to 1.2—this equates to a
distribution coefficient (Kd) value of 2.247 310
–5
m
3
/kg.
Sorption coefficient of 0.0562 was used (SMEC, 2014).
Model Calibration. The hydrocarbon transport model was
calibrated by optimizing the match between the transient state
measured and simulated TPH. The input parameters for the
flow model included the hydraulic conductivity of the aquifer,
the geometry of the recharge zones, the recharge rate from each
zone, the starting head values, and the aquifer’s top and bottom
elevations. The Parameter Estimation and Predictive Analysis
(PEST) software was used for the calibration runs. The
historically reported rainfall rates from northern Kuwait were
used as the limits of the recharge rates for the calibration. The
optimal recharge rates were used to make a final MODFLOW
run. The flow terms produced by the calibrated MODFLOW
model were used with MODFLOW-SURFACT for modelling
the movement of the hydrocarbon from the recharge areas. The
flow model was calibrated by optimizing the match between the
steady state measured and simulated hydraulic heads, and the
thickness of the freshwater lenses at observation points. A
horizontal longitudinal dispersivity of 10 m, a horizontal
transverse dispersivity of 1 m, and a vertical transverse
dispersivity of 0.2 m led to reasonable results.
Figure 9—Model domain, grid, boundary and active/
inactive cells (the black area is inactive all other
colors are active). blue cells are constant head
boundary. Purple lines represent drainage lines of
the wadis and the freshwater lenses are shown as
the shaded purple area. orange line is location of
cross section profile showing model layers.
Table 1—Model layer details.
Layer Description
Thickness
(m)
1 Aquifer 1—poorly sorted gravelly sand (blue) Up to 14 m
2 Aquitard—confining layer—calcareous
siltstone/shale (grey)
Up to 3 m
3 Aquifer 2—fine grained conglomeratic
sandstone (green)
Up to 17 m
4 Aquitard—confining layer—calcareous
siltstone/shale (yellow)
Up to 2 m
5 Aquifer 3—argillaceous and conglomeratic
sandstone (red)
Up to 18 m
Table 2—Summary of hydraulic conductivities
values adopted in the model.
Layer
Al-Raudhatain
freshwater lens
Umm Al-Aish
freshwater lens
Brackish
water
Layer 1 70 7 40 4 30 3
Layer 2 0.13 0.013 0.13 0.13 0.13 0.013
Layer 3 65 6.5 26 2.6 20 2
Layer 4 0.08 0.08 0.08 0.008 0.08 0.008
Layer 5 10 1 1 0.1 0.5 0.05
Table 3—Summary of soil parameters in the
vadose zone and variably saturated flow pa-
rameters adopted in the model.
Parameter
Phase
constants
Unsaturated flow
constants Unit
Water density 998 (at 30 8C) kg/m
3
Air density 1.770 kg/m
3
Water viscosity 0.5 Pa-s
Air viscosity 1.9383 310
–5
Pa-s
Water compressibility 1.9 310
–5
Pa
–1
Air compressibility 1.77 310
–5
Pa
–1
Atmospheric pressure 101.3 kPa
Gravitational
acceleration
9.8066 m/s
2
Van Genuchten soil
parameter (Alpha)
0.2 m
–1
Van Genuchten soil
parameter (Beta)
2.68
Residual saturation 0.45
Brooks-Corey
exponential
2
Yihdego and Al-Weshah
494 WATER ENVIRONMENT RESEARCH June 2017
Gushing oil from the oil wells, storage tanks, and pipelines
created oil lakes in the surface depressions with depths up to 2
m. In addition, the products of oil combustion covered a large
tract of the ground’s surface. Ninety-five percent of the oil from
the oil lakes was recovered, but the remaining 5% remains,
providing a significant source of contamination. As noted by
Viswanathan et al. (1998), land pollution of this extent was
probably the first of its kind.
Most of the past major spills on land were primarily refined
products such as gasoline and other fuels, the chemical
characteristics of which have been well studied. In the present
case, the residual low-volatile oil in the depressions, leachates
from the combustion products on the surface, and hydrocarbons
that have already infiltrated into the subsurface form a source of
contamination for the groundwater. The aerial extent of each of
these pollution sources is not, however, well defined. Similarly,
the physical processes and mechanisms active in the transport of
each of these pollutant types are not well understood.
Weathering processes that affect the crude oil result in the loss of
light-end components, photooxidation, and emulsification of the
remaining components. These processes result in changes in the
properties of the pollutants with time. The complexity of the
contaminant transport pathways that resulted from the unique
geomorphology and the lithology of the study area also adds to
the challenges to the modelling.
Furthermore, non-availability and inadequacy of many of the
required data and information put additional constraints on the
modeling efforts. These include information on the spatial
heterogeneity of the soil, NAPL saturation in the aquifer,
distribution of contaminant levels in the soil, details of the
contamination source terms, and detailed pumping histories of
the wells (Huling and Weaver, 1996; Hunkeler et al., 2002;
Hunkeler et al., 2005; Luciano et al., 2010).
Model Result and Analysis
Matching of the observed hydrocarbon concentrations
(expressed as TPH) with the simulated values in the monitoring
wells (P-18 and P-19) from 1991 to 2013, was used in deriving
suitable values of the source terms for prediction beyond 2013
(with and without source removal). The comparison of the
observed and the simulated concentrations of TPH in the
monitoring wells P-18 and P-19 is presented in Figure 10.
Maximum simulated TPH for the P-18 and P-19 are 6.7 and 0.8
mg/L. The simulation results shows that for P-18, TPH
concentrations peak at about 6 mg/L after about 10 years and
this concentration is not maintained since then and the
concentrations then decrease to 1.63 mg/L after an additional of
50 years. The initial concentration for the TPH was reduced to
reflect depletion of soluble material both in oil at ground surface
but also oil that has penetrated into the soil profile. For P-19, the
TPH concentrations slowly increase from about 0.1 mg/L after 8
years to peak at 0.7 mg/L at about 17 years (Figure 10) and the
concentrations increases/decreases to 1 and 0.63 mg/L after an
additional of 50 years without and with source removal,
respectively. Clearly, under the scenarios assumed here, there are
large masses of hydrocarbons stored in the soil profile, and that
will leach over time into the water table. Formation of an NAPL
lens on top of the water table was unlikely, especially when the
weathering processes that the oil would experience were taken
into account. This stems from the fact that the oil viscosity is a
very important parameter in determining the travel time through
the vadose zone. The dynamic viscosity of the oil would increase
with time as the weathering processes continued, and with the
increase, the likelihood of vertical migration of the NAPL would
be further reduced. It is unlikely that the oil migrates directly to
the water table and pollutes the groundwater, especially in the
long term. Rather, infiltrating rainwater runoff leaches the
hydrocarbon pollutants from the contaminated soil and
Table 4—Summary of storage values adopted
in the model.
Layer
Effective
Porosity
Specific
Yield
Specific
Storage (m
–1
)
Layer 1 0.29 0.14 0.00028
Layer 2 0.15 0.12 0.00013
Layer 3 0.29 0.14 0.00028
Layer 4 0.15 0.12 0.00006
Layer 5 0.29 0.14 0.00028
Table 5—Oil parameters.
Parameter Value and reason
Oil density a. qo—0.857 g/cm
3
for the light crude oil from the Kuwait north oil fields (lowest of the range of
three values, between 0.857 and 0.945 g/cm
3
, as determined by KISR; Akber et al., 2009)
Oil density b qo—0.945 g/cm
3
for the heavy crude oil from the Kuwait north oil fields (highest of the range of
three values, between 0.857 and 0.945 g/cm
3
, as determined by KISR; Akber et al., 2009)
Oil dynamic viscosity a lo—85.7 cp for the light crude oil from the Kuwait north oil fields (calculated from the above
density and kinematic viscosity of 100 mm
2
/s, (reported range of measured values 100 to 268
mm
2
/s; Akber et al., 2009)
Oil dynamic viscosity b lo—253.3 cp for the heavy crude oil from the Kuwait north oil fields (calculated from the above
density and kinematic viscosity of 268 mm
2
/s (reported range of measured values 100 to 268
mm
2
/s; Akber et al., 2009)
Aquifer residual oil saturation Sors 0.33 (assumed from reported average soil saturations; Mercer and Cohen, 1990)
Vadose zone residual oil saturation Sorv 0.15 (assumed from reported average soil saturations; Mercer and Cohen, 1990)
Oil surface tension rao 32 dyne/cm (higher value of the range for fuel oils from literature)
Oil ponding depth constant depth of 0.5 m for 100 days, then allowed to infiltrate into the soil
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 495
transfers them to the water table. For the purpose of modeling
the process in the saturated zone, it was assumed that the
groundwater in areas already affected by pollution or hydro-
carbon held at residual saturation close to the water table (i.e.,
in the smear zone) would act as constant sources of pollutants in
the future.
A non-reactive transport simulation without biodegradation
was first performed to check the resulting concentration
distribution and concentration level as well as to precalibrate
site-specific transport parameters (retardation, dispersivity, and
effective porosity). The model simulated that the concentration
trend (with biodegradation), somehow similar to the observed.
However, there are discrepancies in the matching. Large-scale
withdrawal of water from the fields might affect the prediction
results through its effects on hydrodynamic gradient, and the
transport direction and rate of movement of the pollutants. The
main reason for the discrepancy in the matching was related to
the complexity of the NAPL being the vadose zone acting as an
infinite source of hydrocarbon. The contaminant turns soluble
when it gets recharged either by rainfall or through the saline
water to extinguish the fire. Besides, the monitoring wells are
not in position to assess the NAPL. Also, the existing monitoring
wells might not be optimally located relative to sources. The
infinite source and nature of contaminant makes less to calibrate
the model. The effective porosity, which ranges from 0.2 to 0.27
is similarly sensitive to the propagation of the contaminant front
(like the dispersivity coefficients). All of the above factors tend
to limit the dependability of the modelling efforts, though every
attempt has been made to maximize the accuracy and validity of
the results obtained. An immobile depleting NAPL can be
simulated with an OS technique, NAPL components are
equilibrated with groundwater depending on their respective
concentrations in the NAPL and individual NAPL/water
partition coefficients at the beginning of each transport step.
With elapsed transport simulation time, water-soluble compo-
nents are depleted from the NAPL and source concentrations in
groundwater gradually decrease (Guvanasen et al., 2009).
Because of the depletion of the more soluble components, the
concentrations of the less-water-soluble components relatively
increase within NAPL in the course of simulation time.
The simulation result (Figure 11) shows that the TPH should
be detected in the aquifer in reasonable concentration. Predictive
analysis using mathematical models (KISR, 2009; SMEC, 2014)
underscored the possibility of large-scale contamination of the
groundwater resources in the Umm Al-Aish basin, caused by
ongoing movement of hydrocarbon pollutants toward the
supply bores. Interestingly, TPH has not been traced signifi-
cantly in the monitoring wells at Al-Raudhatain until recently
and this may be related to the timing required to reach the water
table, without ruling out the possibility of short-circuit path
toward the water table in few spots. Recent studies (SMEC,
2014) suggested that the groundwater in the Umm Al-Aish area
has been significantly affected by the hydrocarbon pollutants.
Figure 11 suggest that significant portion of the Umm Al-Aish
depression has been already contaminated and the extent of the
polluted area will increase with time. It is likely that these
pollutants would move toward the Al-Raudhatain freshwater
lens if no preventive and/or remedial measures were taken in the
near future. The predictions and estimates of plume extents
should be viewed as very approximate, but based on best
Figure 10—Simulated concentration profile for well
P-18 and P-19 over the next 73-year spatial historical
matching after 23 years and prediction for 50 years,
starting from year 2013 (with and without source
removal for monitoring well P-19).
Yihdego and Al-Weshah
496 WATER ENVIRONMENT RESEARCH June 2017
estimates of hydraulic and transport parameters. Significant
plume growth is approximately 500 m from the source regions
after 12 years. Groundwater velocities are of the order of 40 m/
yr. Concentrations of TPH were simulated to be very high over
the initial 23-year period. From 0 to 23 years, the TPH
concentration increases and plumes continue to expand, so no
apparent steady state has been achieved.
After 23 years and with respect to the total area of the
freshwaterbody at Al-Raudhatain (55.2 km
2
), the areal extent of
the total petroleum hydrocarbon plume is estimated to be 7.3
km
2
for the 0.1-mg/L contour. This equates to total petroleum
hydrocarbon affecting 13% of the freshwater body.
After 23 years and with respect to the total area of the
freshwater body at Umm Al-Aish (37 km
2
), the areal extent of
the total petroleum hydrocarbon plumes is estimated to be 8.7
km
2
for the 0.1-mg/L contour. This equates to total petroleum
hydrocarbon impacting 24% of the freshwater body.
The shape of the contours in Figures 11 indicates that
contaminant transport is affected by transverse dispersion,
which can make it difficult to locate the advective front of TPH
plume. In assessing the degree to which a plume may dive, the
vertical dispersion characteristics of an aquifer should be
considered. An aquifer with high vertical dispersivity is likely to
cause a higher degree of plume thickening relatively close to its
source, whereas an aquifer with low vertical dispersivity is likely
to result in thinner plumes with a greater degree of overlying,
accumulated clean recharge water. Factors controlling the plume
(and dive) includes hydraulic factor—horizontal flow is
dominant in the aquifer and vertical flow in the aquitard (Layers
2 and 4), geological factors and biogeochemical factors.
Discussion and Conclusion
The modelling of the Al-Raudhatain and Umm Al-Aish
freshwater aquifer has been undertaken to assess the oil spill
resulting from the Gulf War using MODFLOW-SURFACT
numerical code. Using these codes overcame the limitations of
separate modelling of the two freshwater lenses. This model is
able to combine both freshwater lenses in one model domain
simulating the vadose zone (that is, unsaturated zone) together
with the saturated zone.
The hydrocarbon transport model was set up to run from
March 1991 to July 2013 for historical matching, simultaneously
with the transient flow model. The time-variant hydrocarbon
transport model was calibrated simultaneously with the
transient groundwater flow system (for the period March 1991
to July 2013) and this included variably saturated flow and
transport. This was done prior to proceeding to simulate
contaminant (hydrocarbon) transport as the hydraulic gradients
and flow directions (and storage volume) are significant controls
on contaminant migration. The purpose was to allow 50 years of
predictive remediation simulation time from the end of 2013.
This helps to assess the contamination risk, mitigation options
and set strategy for a remedial implementation.
The numerical model was used to investigate the potential for
oil present on the ground surface to migrate vertically through
the unsaturated zone as NAPL and to reach the water table. The
results suggested that after 23 years and with respect to the total
area of the freshwater body at Al-Raudhatain, the areal extent
of the total petroleum hydrocarbon plume is estimated to be 7.3
km
2
for the 0.1 mg/L contour. This equates to total petroleum
hydrocarbon affecting 13% of the freshwater body. After 23
years and with respect to the total area of the freshwater body at
Umm Al-Aish, the areal extent of the total petroleum
hydrocarbon plumes is estimated to be 8.7 km
2
for the 0.1 mg/L
contour. This equates to total petroleum hydrocarbon affecting
24% of the freshwater body.
The magnitude of the pollution problem in Kuwait due to this
incident is unparalleled in the history of mankind. Conse-
quently, no comparisons can be made and no conclusions can be
derived from similar experiences elsewhere. The environmental
and economic hardships imposed on Kuwait, and to the Gulf
Region in general, both during and after the Gulf War will be
endured for many years to come. The destruction caused by the
oil fires and the oil spills were devastating, and at times were
fatal. The result indicates that even though the present-day total
petroleum hydrocarbon loading was negligible in the center of
Figure 11—Predicted distribution of pollutants (sim-
ulated TPH concentration) after year 23 (2013) (i.e.,
in 23 years from the start of the pollution event) in
Layer 1. Shaded color shows TPH concentrations
with values greater than 0.01 mg/L, where red
shading shows high concentrations and blue shad-
ing shows low concentrations.
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 497
the Al-Raudhatain depression, what has entered the ground-
water system earlier from the contamination sources (pits,
fringes, and lakes) is still moving toward the freshwater lenses or
is potentially undetected due to lack of optimal existing
monitoring bore screen and depth installations referred as plume
diving. In assessing the degree to which a plume may dive, the
vertical dispersion characteristics of an aquifer should be
considered.
Not only did humans suffer the consequences of the conflict,
but the ecosystem and the atmosphere were also innocent
victims. Additionally, numerous parties involved also experi-
enced the economic impact that the Gulf War had on the region.
Even today, some Arab nations, such as Iraq, are still repaying
debt, while others such as Kuwait, are rebuilding both the
economy and the ecosystem in the aftermath of the conflict. The
current study will be of interest to wider community to
appreciate the recent past man-made crises and its consequence
to restore the environment for decades to come, due to the
complexity of the problem domain. Therefore, great care needs
to be taken and attention is highly sought after generations as a
typical examples.
Acknowledgments
The authors would like to thank the Kuwait National Focal
Point for Environmental Projects, Ministry of Electricity and
Water in Kuwait—Groundwater Sector and SMEC Interna-
tional for providing access to some of their data. The authors
will acknowledge the comments from the anonymous reviewers.
Submitted for publication September 27, 2016; revised
manuscript submitted November 26, 2016; accepted for
publication December 5, 2016.
References
Akber, A.; Al-Murad, M.; Mukhopadhyay, A.; Rashid, A. T.; Al-Qallaf, H.; Al-
Haddad, A.; Bhandary, H.; Marzouk, F.; Al-Salman, B. (2009) Long-Term
Monitoring and Remediation Strategy for Hydrocarbon Pollutants in the
Groundwater of Raudhatain and Umm Al-Aish Fields; Report No.
KISR9902; Kuwait Institute for Scientific Research: Kuwait.
Al-Weshah, R.; Yihdego Y (2016) Flow Modelling of Strategically Vital Freshwater
Aquifers in Kuwait. Environ. Earth Sci., 75, 1315. doi:10.1007/s12665-016-
6132-1. http://link.springer.com/article/10.1007/s12665-016-6132-1 (ac-
cessed Dec 15, 2016).
Buttler, W. A.; Bartlett, C. L. (1995) Taking Advantage of Natural Biodegradation.
In Intrinsic Bioremediation; Hinchee, R. E., Wilson, J. T., Downey, D. C.,
Eds.; Battelle Press: 505 King Avenue, Columbus, Ohio 43201, USA.
Doherty, J. (2004) PEST: Model-Independent Parameter Estimation, User Manual,
5th ed.; Watermark Numerical Computing: Brisbane, Australia.
Dyer, M.; Heiningen, E.; Gerritse, J. (2000) In Situ Bioremediation of 1,2-
dichloroethane under Anaerobic Conditions. Geotech. Geol. Eng., 18, 313–
334.
Giudici, M. (1991) Identifiability of Distributed Physical Parameters in Diffusive-
Like Systems. Inverse Probl., 7, 231–245.
Grealish, G.; Omar, S.; Quinn, M. (2001) As Sabriyah and Ar-Rawdatayn Oil
Affected Area Soil Survey, Assessing Damage Magnitude and Recovery of
the Terrestrial Eco-System / Follow Up of Natural and Induced Desert
Recovery; AACM and KISR, Project FA015C.
Grostern, A.; Edwards, E. A. (2006) Growth of Dehalobacter and Dehalococcoides
spp. during Degradation of Chlorinated Ethanes. Appl. Environ. Micro-
biol., 7, 428–436.
Guvanasen, V.; Himml, M.; Buechler, B. (2009) Simulation of Kinetic-Limited
Degradation of Multi-Component Hydrocarbons with Particular Consid-
eration on the Supply of Electron Acceptors. In The Challenge of Quantity
and Quality for Sustainable Future, International Symposium on Efficient
Groundwater Resources Management; Bangkok, Thailand.
Hu, H.; Mao, X.; Barry, D. A.; Liu, C.; Li, P. (2015) Modeling Reactive Transport
of Reclaimed Water Through Large Soil Columns with Different Low-
Permeability Layers. Hydrogeol.J., 23 (2), 351–364.
HGL (2013) MODFLOW-SURFACTeVersion 4.0; User’s Manual and Guide;
HydroGeoLogic, Inc.: Reston, Virginia.
Huling, S. G.; Weaver, J. W. (1996) Dense Non Aqueous Phase Liquids. in: EPA
Environmental Assessment Sourcebook; Ann Arbor Science Publishers:
Chelsea, Michigan.
Hunkeler, D.; Aravena, R.; Berry-Spark, K.; Cox, E. (2005) Assessment of
Degradation Pathways in an Aquifer with Mixed Chlorinated Hydrocarbon
Contamination Using Stable Isotope Analysis. Environ. Sci. Technol., 39,
5975–5981.
Hunkeler, D.; Aravena, R.; Cox, E (2002) Carbon Isotopes as a Tool to Evaluate
the Origin and Fate of Vinyl Chloride: Laboratory Experiments and
Modeling of Isotope Evolution. Environ. Sci. Technol., 36, 3378–3384.
KISR (2009) Long-Term Monitoring and Remediation Strategy for Hydrocarbon
Pollutants in the Groundwater of Al-Raudhatain and Umm Al-Aish Fields
(Volume-1), WM016C; Hydrology Department Water Resources Division,
Kuwait Institute for Scientific Research: Kuwait.
KISR (2012) Assessment of Usable Groundwater Reserve in Northern Kuwait,
WM027C; Hydrology Department Water Resources Division, Kuwait
Institute for Scientific Research: Kuwait.
KISR (2013) Interim Report of Deliverable Tms 4.1.9 Results of the 1st Year
Groundwater Quality Monitoring the First Four Rounds of Groundwater
Surveillance Monitoring Management Support and Technical Supervision
of Kuwait Environmental Remediation Program for Kuwait National Focal
Point (Sp006c); Kuwait Institute for Scientific Research: Kuwait.
Kwarteng, A. Y.; Viswanathan, M. N.; Al-Senafy, M. N.; Rashid, T. (2000)
Formation of Fresh Groundwater Lenses in Northern Kuwait. J. Arid
Environ., 46 (2), 137–155.
Luciano, A.; Viotti, P.; Papini, M. P. (2010) Laboratory Investigation of DNAPL
Migration in Porous Media. J. Hazard. Mater., 176, 1006–1017.
Maness, A. D.; Bowman, K. S.; Yan, J.; Rainey, F. A.; Moe, W. M. (2012)
Dehalogenimonas spp. Can Reductively Dehalogenate High Concentra-
tions of 1,2-Dichloroethane, 1,2-Dichloropropane, and 1,1,2-Trichloroeth-
ane. AMB Express,2,54.
Martin, J.; Norman, M. (1992) Assessment of Groundwater Resources in Kuwait
Using Remote Sensing Technology; Report No. WH-002, Volume IV,
Hydrometeorology; Kuwait Institute for Scientific Research: Kuwait.
Moe, W. M.; Yan, J.; Nobre, M. F.; Costa, M. S.; Rainey, F. A. (2009)
Dehalogenimonas Lykanthroporepellens gen. nov., sp. nov., a Reductively
Dehalogenating Bacterium Isolated from Chlorinated Solvent-Contami-
nated Groundwater. Int. J. Syst. Evol. Microbiol., 59, 2692–2697.
Mukhopadhyay, A.; Rashid, T.; Al-Haddad, A. (2006) Technical Assistance to the
Monitoring and Assessment of the Environmental Damage to the
Groundwater Resources of Kuwait, Progress Report No. 3; Report No.
KISR8054; Kuwait Institute for Scientific Research: Kuwait.
Omar, S. A.; Al-Sdirawi, M. A.; Razzaque, M. A. (1994) Agricultural and
Environmental Laws, Policies and Regulations; Report No. KISR4523;
Kuwait Institute for Scientific Research: Kuwait.
Omar, S. A. S.; Shahid, S.; Al-Ali, E.; Al-Qallaf, A. M.; Minkara, M. H.; Al-Shatti,
K.; Elgawad, M. A.; Grose, C.; Burke, K.; King, P. D.; Grealish, G. J. (1998)
Soil Survey and Associated Activities for the State of Kuwait, Technical
Report No. KISR FA006C; Kuwait Institute for Scientific Research: Kuwait.
Panday, S.; Huyakorn, P. S. (2008) MODFLOW SURFACT: A State-of-the-Art
Use of Vadose Zone Flow and Transport Equations and Numerical
Techniques for Environmental Evaluations. Vadose Zone J., 7 (2), 610–631.
Parson Corporation (1964) Groundwater Resources of Kuwait. Ministry of
Electricity and Water: Kuwait; Vols I, II and III.
Rao, G. T.; Rao, V. V. S.; Ranganathan, K.; Surinaidu, L.; Mahesh, J.; Ramesh, G.
(2011) Assessment of Groundwater Contamination from a Hazardous
Dumpsite in Ranipet, Tamil Nadu, India. Hydrogeol.J., 19, 1587–1598.
Rasa, E.; Bekins, B. A.; Mackay, D. M.; De Sieyes, N. R.; Wilson, J. T.; Feris, K.
P.; Wood, I. A.; Scow, K. M. (2013) Impacts of an Ethanol Blended Fuel
Yihdego and Al-Weshah
498 WATER ENVIRONMENT RESEARCH June 2017
Release on Groundwater and Fate of Produced Methane: Simulation of
Field Observations. Water Resour. Res., 49, 4908–4926.
SMEC (2002) GD1.2—Overview of Groundwater Quality/Pollution in Kuwait Up
to 2002. Program for the Monitoring and Assessment of the Environmental
Consequences of the Iraqi Aggression in Kuwait; Snowy Mountains
Engineering Corp.: Sydney, New South Wales, Australia.
SMEC (2006) GD 5.12—Groundwater Damages and Remediation of the
Raudhatain and Umm Al-Aish Freshwater Aquifers of Kuwait. Program for
the Monitoring and Assessment of the Environmental Consequences of the
Iraqi Aggression in Kuwait; Snowy Mountains Engineering Corp.: Sydney,
New South Wales, Australia.
SMEC (2014) Groundwater Modelling of the Raudhatain and Umm Al-Aish
Freshwater Aquifers of Kuwait. Program for the Monitoring and
Assessment of the Environmental Consequences of the Iraqi Aggression in
Kuwait; Snowy Mountains Engineering Corp.: Sydney, New South Wales,
Australia.
Viswanathan, M. N.; Quin, M. F.; Al-Senafy, M. N.; Akber, A. (1998) Rainfall
Induced Transport of Crude Oil Contaminants from Unsaturated Zones;
Final Report, KISR Project WH007C; Report No. KISR5374; Kuwait
Institute for Scientific Research: Kuwait.
Yihdego, Y. (2010a) Modelling of Lake Level and Salinity for Lake Purrumbete in
Western Victoria, Australia: A Co-Operative Research Project Between La
Trobe University and EPAVictoria, Australia. http://www.epa.vic.gov.au/
~/media/Publications/Yihdego%202008%20Lake%20Purrumbete%
20report%20updated%202010.pdf (accessed Dec 23, 2016).
Yihdego (2010b) Modelling Bore and Stream Hydrograph and Lake Level in
Relation to Climate and Land Use Change in Southwestern Victoria,
Australia. PhD thesis, Faculty of Science, Technology and Engineering,
Melbourne, La Trobe University, Australia.
Yihdego, Y. (2013) Data Visualization Tool as a Framework for Groundwater
Flow and Transport Models. Proceedings of the International Conference
on ‘‘MODFLOW and More 2013’’ , Translating Science into Practice,
Maxwell, Hill, Zheng & Tonkin—igwmc.mines.edu. June 2013, Integrated
Groundwater Modelling Centre (IGWMC); Colorado School of Mines
University: Golden, Colorado.
Yihdego, Y. (2016a; in press) Drought and Pest Management Initiatives. In
Handbook of Drought and Water Scarcity (HDWS): Vol. 3: Management
of Drought and Water Scarcity; Eslamian, S., Eslamian, F.A., Eds.; Francis
and Taylor, CRC Group: Boca Raton, Florida; Chapter 11. https://www.
crcpress.com/Handbook-of-Drought-and-Water-Scarcity-Environmental-
Impacts-and-Analysis/Eslamian-Eslamian/p/book/9781498731041(accessed
Dec 23, 2016).
Yihdego, Y. (2016b; in press) Drought and Groundwater Quality in Coastal Area.
In Eslamian, S.; Eslamian, F.A., Eds.; Handbook of Drought and Water
Scarcity (HDWS): Vol. 2: Environmental Impacts and Analysis of Drought
and Water Scarcity; Francis and Taylor, CRC Group: Boca Raton, Florida;
Chapter 15. https://www.crcpress.com/Handbook-of-Drought-and-Water-
Scarcity-Environmental-Impacts-and-Analysis/Eslamian-Eslamian/p/book/
9781498731041 (accessed Dec 23, 2016).
Yihdego, Y.; Al-Weshah, R. (2016a) Gulf War Contamination Assessment for
Optimal Monitoring and Remediation Cost–Benefit Analysis, Kuwait.
Environ. Earth Sci., 75, 1234. doi: 10.1007/s12665-016-6025-3. http://link.
springer.com/article/10.1007%2Fs12665-016-6025-3 (accessed Dec 15,
2016).
Yihdego, Y.; Al-Weshah, R. (2016b) Assessment and Prediction of Saline Sea
Water Transport in Groundwater Using Using 3-D Numerical Modelling.
Environ. Processes J. doi: 10.1007/s40710-016-0198-3. http://link.springer.
com/article/10.1007/s40710-016-0198-3 (accessed Dec 15, 2016).
Yihdego, Y.; Al-Weshah, R. (2016c) Engineering and Environmental Remediation
Scenarios Due to Leakage from the Gulf War Oil Spill Using 3-D Numerical
Contaminant Modellings. J. Appl. Water Sci., DOI: 10.1007/s13201-016-
0517-x. http://link.springer.com/article/10.1007/s13201-016-0517-x/
fulltext.html (accessed Dec 23, 2016).
Yihdego, Y.; Webb, J. A. (2007) Hydrogeological Constraints on the Hydrology of
Lake Burrumbeet, Southwestern Victoria, Australia. In 21st VUEESC
conference September 2007 Victorian Universities Earth and Environmental
Sciences Conference; Sept 2007; Hagerty, S.H., McKenzie, D.S. and
Yihdego, Y., Eds.; Geological Society of Australia Abstracts No 88.
Yihdego, Y.; Webb, J. A. (2008) Modelling of Seasonal and Long-term Trends in
Lake Salinity in Southwestern Victoria, Australia. Proceedings of Water
Down Under April 2008, 994–1000, Adelaide. Engineers Australia. Casual
production, 2008: 994–1000. ISBN: 0858257351. http://search.informit.
com.au/documentSummary;dn¼574001033603140;res¼IELENG (accessed
Dec 23, 2016).
Yihdego, Y.; Webb, J. A. (2009) Characterizing Groundwater Dynamics in
Western Victoria, Australia Using Menyanthes Software. Proceedings of the
10th Australasian Environmental Isotope Conference and 3rd Australasian
Hydrogeology Research Conference December 2009, Perth.
Yihdego, Y.; Webb, J. A. (2010) Characterizing Groundwater Dynamics Using
Transfer Function-Noise and Auto-Regressive Modelling in Western
Victoria, Australia. Proceedings of the 5th IASME/WSEAS International
Conference on Water Resources, Hydraulics and Hydrology (WHH ’10),
February 2010; University of Cambridge: UK. ISBN: 978-960-474-160-1
(ISSN: 1790-5095).
Yihdego, Y.; Webb, J. A. (2016) Assessment of Wetland Hydrological Dynamics in
a Modified Catchment Basin: Case of Lake Buninjon, Victoria, Australia.
Water Environ. Res. J., 89. https://doi.org/10.2175/
106143016X14798353399331 (accessed Dec 23, 2016).
Yihdego, Y.; Webb, J. A.; Leahy P. (2016) Response to Parker: Rebuttal: ENGE-D-
13-00994R2 ‘‘Modelling of Lake Level Under Climate Change Conditions:
Lake Purrumbete in South Eastern Australia’’. Environ. Earth Sci. J., 75 (1),
1–4. DOI: 10.1007/s12665-014-3669-8. http://link.springer.com/article/10.
1007/s12665-015-4808-6 (accessed Dec 23, 2016).
Yihdego and Al-Weshah
WATER ENVIRONMENT RESEARCH June 2017 499
... Additional factors for consideration are the location of the site in relation to urban areas, agricultural areas or drinking water sources, also in combination with legal and financial constraints. The choice of remediation technology is therefore a case-by-case decision and different parameters have to be accounted for (Yihdego and Al-Weshah, 2016a). ...
Article
Pump-and-treat technology is among the most used technologies for groundwater remediation. While conventional, vertical wells (VRWs) are well-known and used from long time, horizontal wells (HRWs) have been explored for remediation technologies only in last few decades. HRWs have shown to outperform vertical wells in terms of versatility, productivity and clean-up times under certain conditions. In this paper, the efficacy of an innovative pump-and-treat (P&T) configuration for groundwater remediation obtained by adopting either VRWs or HRWs technology is comparatively tested. A 3D transient finite element model of an unconfined aquifer containing a hexavalent chromium (Cr(VI)) contamination plume is considered to compare a single horizontal well configuration vs a range of spatially-optimised arrays containing vertical wells. A sensitivity analysis aimed at finding the best configuration to minimise the remediation time and the related cost is carried out by comparing different well diameters, D, pumping rates, Q, and position of wells. A comparative cost analysis demonstrates that, for the examined case-study, a single HRW achieves the clean-up goals in the same time span as for a greater number of vertical wells, but at higher price due to the excavation costs.
... Since the liberation of Kuwait in 1991, a number of survey and assessment projects have been conducted and a great deal of data has been collected characterising the oil pollution in the air, water and soil. Despite the monitoring and assessment efforts during the past years, the long-term effects of the released pollutants, their fate and pathways, as well as the potential risks associated with their existence are still not fully understood (AACM and KISR, 2001;Kwarteng et al., 2000;SMEC, 2002SMEC, , 2014Yihdego and Al-Weshah, 2016b). In order to determine the full nature and extent of the environmental damages caused by Iraq and to fully quantify the Kuwait environmental claims before the United Nations Compensation Commission (UNCC), a program for the monitoring and assessment of the environmental consequences of the Iraqi aggression in Kuwait has been implemented. ...
Article
This paper investigated approaches to treat the polluted groundwater in selected aquifers in Kuwait, which is the most extensive and second to none in terrestrial world case history. The selected aquifers are susceptible to pollution by oil spills during the first Gulf War. Experimental samplings of polluted soils were analysed. Results showed that granulated activated carbon is very effective in removing petroleum hydrocarbons from contaminated water. A pump-and-treat remediation scheme has been suggested for the affected parts of the freshwater fields. The cost estimates indicated that a treatment process involving carbon adsorption to remove petroleum hydrocarbons, followed by reverse osmosis to remove salt was highly likely to be the most cost-effective treatment system. The current treatment design and parametric costing approach can be applied elsewhere for the role of bioremediation in the treatment of petroleum contaminated environment, hazardous effects of petroleum hydrocarbon and genetic engineering in bioremediation.
... Since the liberation of Kuwait in 1991, a number of survey and assessment projects have been conducted and a great deal of data has been collected characterising the oil pollution in the air, water and soil. Despite the monitoring and assessment efforts during the past years, the long-term effects of the released pollutants, their fate and pathways, as well as the potential risks associated with their existence are still not fully understood (AACM and KISR, 2001;Kwarteng et al., 2000;SMEC, 2002SMEC, , 2014Yihdego and Al-Weshah, 2016b). In order to determine the full nature and extent of the environmental damages caused by Iraq and to fully quantify the Kuwait environmental claims before the United Nations Compensation Commission (UNCC), a program for the monitoring and assessment of the environmental consequences of the Iraqi aggression in Kuwait has been implemented. ...
Article
Full-text available
This paper investigated approaches to treat the polluted groundwater in selected aquifers in Kuwait, which is the most extensive and second to none in terrestrial world case history. The selected aquifers are susceptible to pollution by oil spills during the first Gulf War. Experimental sampling of polluted soils were analysed. Results showed that granulated activated carbon is very effective in removing petroleum hydrocarbons from contaminated water. A pump-and-treat remediation scheme has been suggested for the affected parts of the freshwater fields. The cost estimates indicated that a treatment process involving carbon adsorption to remove petroleum hydrocarbons, followed by reverse osmosis to remove salt was highly likely to be the most cost-effective treatment system. The current treatment design and parametric costing approach can be applied elsewhere for the role of bioremediation in the treatment of petroleum contaminated environment, hazardous effects of petroleum hydrocarbon and genetic engineering in bioremediation.
Article
Full-text available
Freshwater groundwater resources at north Kuwait were contaminated by infiltrated oil as well as sea water that was used to fight the oil-well fires during the liberation of Kuwait in 1991. This paper investigates the feasibility of five remediation options to restore the polluted aquifers. These options include: (i) pump and treat of contaminated groundwater; (ii) cleaning the aquifer formation; (iii) construction of additional desalination plant; (iv) constructing additional storage tanks, and (v) development of artificial aquifer recharge schemes. The basis for this assessment study is to supply minimum basic drinking water to Kuwait City at a total rate of 50,000 m³/day in an emergency for up to one year based on essential basic need of 32 liters per capita per day. To compare these options, a decision matrix to select suitable remediation options using Multiple Criteria Decision Analysis (MCDA) approach is developed. The cost was given a relative weight of 20 whereas other criteria are given weight of 10. Based on these MCDA scores, it was found that option 3, namely, establishing an additional water desalination plant, is the most feasible option followed by option 5, artificial recharge of aquifers.
Article
Full-text available
FOR CITATIONS: Yohannes Yihdego, Hilmi S. Salem, Bediaku G. Kafui & Zarko Veljkovic (2018): Economic geology value of oil shale deposits: Ethiopia (Tigray) and Jordan, Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 40(17): 2079-2096. DOI: https://doi.org/10.1080/15567036.2018.1488015 URL:https://www.researchgate.net/publication/326274780_Economic_geology_value_of_oil_shale_deposits_Ethiopia_Tigray_and_Jordan ABSTRACT: Oil shale is an organic-rich, fine-grained sedimentary rock, containing kerogen, from which liquid hydrocarbons (called shale oil) can be produced. The oil shale deposits in the Tigray region are found in the northern parts of Ethiopia, Eastern Africa. They are of Upper Paleozoic in age, existing as remnants of the Cretaceous erosion period, underlain by tillites and overlain by sandstones. They were formed during the glacial retreat followed by marine deposition of shales in a basin created by the enormous load of the glaciers. The Ethiopian-Tigray oil shale deposits cover an area extending over approximately 30 km 2 , with an average mineable bed-thickness of 55 m, showing on the upper part inter-beds and laminations of shaley limestones. The oil shale resources in this region are estimated to be approximately 4 billion tonnes.
Article
Full-text available
In mining operations carried out below the water table, mine area could potentially affect the surrounding. With further deepening of the mine and quarry, the drawdown can impact on water supply wells and base flow. The variation in radius of influence under confined and unconfined aquifer condition is assessed using a hypothetical case study with three operational quarry sites. Average pit-wise radius of influence (R0) for the pit mine, under unconfined and confined aquifer conditions is 963 m and 61 km, respectively, whereas effective radius (Re) for total (cumulative) excavated mine area is 1.7 and 146 km, respectively. It means that the maximum and minimum value of overall impact/influence for the unconfined aquifer lies in-between 963 m and 1.7 km. Similarly that the maximum and minimum value of overall impact/influence for the confined aquifer lies in-between 61 and 146 km. Meanwhile, Re or effective radius for total (cumulative) excavated area seems more appropriate from evaluation viewpoint due to the overlapping (superposition) concentric mining quarries. Thus, assessment of quarry-wise and total radius of influence is of importance to fulfil targeted production, economically with minimum interruptions as part of the mine planning. Moreover, it helps us to assess, monitor and regulate the impact of a mine-dewatering program in the area. Also, the pumping test is often used to estimate radius of influence; it can trigger the inflow or radius of influence. Estimate of the radius of influence using analytical equation mainly independent of discharge/abstraction from the aquifer stands as a preferred choice for the prediction of radius of influence at the quarry/mine pit. There is a strong rationale that policies should be informed by such analytical assessment because it helps us to prevent or minimize negative impacts as part of regulation and pit-mining management and are useful tools for practitioners to design economically efficient and cost-effective in situ groundwater remediation systems, to contain contaminant plumes, to evaluate the surface–subsurface water interaction and to verify numerical models. http://dx.doi.org/10.1080/23249676.2017.1287022
Article
Full-text available
A method is presented to analyze the interaction between groundwater and Lake Linlithgow (Australia) as a case study. A simplistic approach based on a “node” representing the groundwater component is employed in a spreadsheet of water balance modeling to analyze and highlight the effect of groundwater on the lake level over time. A comparison is made between the simulated and observed lake levels over a period of time by switching the groundwater “node “on and off. A bucket model is assumed to represent the lake behavior. Although this study demonstrates the understanding of Lake Linlithgow’s groundwater system, the current model reflects the contemporary understanding of the local groundwater system, illustrates how to go about modeling in data‐scarce environments, and provides a means to assess focal areas for future data collection and model improvements. Results show that this approach is convenient for getting first‐hand information on the effect of groundwater on wetland or lake levels through lake water budget computation via a node representing the groundwater component. The method can be used anywhere and the applicability of such a method is useful to put in place relevant adaptation mechanisms for future water resources management, reducing vulnerability and enhancing resilience to climate change within the lake basin.
Article
Full-text available
The transport groundwater modelling has been undertaken to assess potential remediation scenarios and provide an optimal remediation options for consideration. The purpose of the study was to allow 50 years of predictive remediation simulation time. The results depict the likely total petroleum hydrocarbon migration pattern in the area under the worst-case scenario. The remediation scenario simulations indicate that do nothing approach will likely not achieve the target water quality within 50 years. Similarly, complete source removal approach will also likely not achieve the target water quality within 50 years. Partial source removal could be expected to remove a significant portion of the contaminant mass, but would increase the rate of contaminant recharge in the short to medium term. The pump–treat–reinject simulation indicates that the option appears feasible and could achieve a reduction in the area of the 0.01 mg/L TPH contour area for both Raudhatain and Umm Al-Aish by 35 and 30%, respectively, within 50 years. The rate of improvement and the completion date would depend on a range of factors such as bore field arrangements, pumping rates, reinjection water quality and additional volumes being introduced and require further optimisation and field pilot trials. URL: https://link.springer.com/article/10.1007/s13201-016-0517-x
Article
Full-text available
The common method to estimate lake levels is the water balance equation, where water input and output result in lake storage and water level changes. However, all water balance components cannot always be quickly assessed, such as due to significant modification of the catchment area. A method that assesses general changes in lake level can be a useful tool in examining why lakes have different lake level variation patterns. Assessment of wetlands using the dynamics of the historical hydrological and hydrogeological data set can provide important insights into variations in wetland levels in different parts of the world. A case study from a saline landscape, Lake Buninjon, Australia, is presented. The aim of the present study was to determine how climate, river regime, and lake hydrological properties independently influence lake water levels and salinity, leaving the discrepancy, for the effect of the non-climatic/ catchment modification in the past and the model shows that surface inflow is most sensitive variable. The method, together with the analysis and interpretation, might be of interest to wider community to assess its response to natural/anthropogenic stress and decision choices for its ecological, social, scientific value, and mitigation measures to safe guard the wetland biodiversity in a catchment basin. https://onlinelibrary.wiley.com/doi/abs/10.2175/106143016X14798353399331 Publisher: ohn Wiley & Sons, Inc.
Article
Full-text available
This paper presents the application of a three-dimensional numerical model to the the Al-Raudhatain and Umm A-Aish fresh water aquifers located in north Kuwait. The two aquifers have been polluted by saline sea water imported to extinguish the oil well fires during the Gulf war. A time-variant salinity transport model was calibrated simultaneously with the transient groundwater flow system to assess the impact of saline sea water. Variably saturated flow and transport were modelled. The results of the salinity transport model suggest that although the fresh water–saline water interface, as defined by the 1,500 mg/L contour, has moved towards the centre of the lens, in some areas up-gradient of the freshwater recharge, the extent of the freshwater lens has actually increased slightly in the down-gradient areas. After 23 years (simulation period 1990-2013), the areal extent of the total dissolved solids plume is estimated 27 % and 29 % of the Al-Raudhatain and Umm A-Aish fresh water bodies, respectively. Under the scenarios assumed, there are large masses of salts stored in the soil profile that will leach over time to the water table. The total dissolved solids concentrations are predicted by the model to decrease to 4,500 mg/L from 7,800 mg/L, after 73 years (simulation period 1990-2063) from the moment the saline sea water was added. The predicted total dissolved solids concentration simulation provides a worst-case scenario of the likely extent of contaminant movement in groundwater in the two fresh water fields. Solute transport modelling has become increasingly important tool for interpreting groundwater quality data and processes relevant to natural and contaminant aquifer systems to a wide range of real-world groundwater quality problems. Further data, from drilling sampling and other testing and experimentations should help clarify these assumptions and assist in updating the solute transport modelling effort which helps to provide insights into the past and present behavior, and allows to predict water quality management scenarios.
Article
Full-text available
An oil flow from the oil wells damaged during the 1991 Gulf War and the sea water used for extinguishing the resulting oil fire have contaminated the freshwater aquifers of Raudhatain and Umm Al-Aish water fields in north Kuwait. The numerical flow modelling of the aquifers was undertaken to create a calibrated and validated model that could be used in the future to explore a viable remediation strategy for the aquifers. The Environmental Visualization Software (EVS-pro) 3-D data presentation program was used to construct a conceptual model as a preliminary step. A 3-D numerical model was developed using the MODFLOW-SURFACT code that overcame the limitations of the classical MODFLOW. This model was able to combine both freshwater lenses in one model domain simulating the vadose zone together with the saturated zone. The model domain covered an area of 580 km2 encompassing the Al-Raudhatain and Umm Al-Aish basins. A steady-state model was developed first to study regional flow patterns in the aquifer. A transient model was developed subsequently to assess seasonal recharge on groundwater and investigate their effects on flow patterns. Analysis of the calibrated steady-state model output indicated that the model simulated the groundwater elevation and flow direction across the model domain to an acceptable level. The calibrated transient flow model is of significant importance to assess the vertical and lateral plume migration in the area and helps to come up with a plausible remediation strategy.
Article
Full-text available
Site characterization was performed on an area of 580 km2 around the strategically vital freshwater aquifers of the Al-Rawdhatain and Umm Al-Aish to assess the status of groundwater pollution as the result of Iraq invasion to Kuwait in 1991. Advanced data analysis and visualization software (EVS-Pro) was used for groundwater contamination assessment analytes: total petroleum hydrocarbon (TPH) and total dissolved solids (TDS). This will reduce the number of samples needed (saves time and money) and provide a superior assessment of the analytes distribution. Based on the “minimum–maximum plume technology” analysis, the nominal plume area with a threshold of 0.031 mg/kg TPH is estimated at about 0.47 km2. This is the difference between the maximum and minimum predicted plume sizes. EVS-Pro also computed 3.3775 × 109 and 4.0788 × 106 for the plume volumes and masses (dollars per volume and mass), respectively. Also, new sampling locations were determined for further detailed site assessments based on the confidence and uncertainty analysis, which is more defensible and cost-optimized approach. This will reduce the number of samples needed (saves time and money) and provide a superior assessment of the analytes distribution. These tools prove to be effective in assessing remediation costs of clean-up versus benefits obtained and in developing a cost-effective monitoring programme for insights into processes controlling subsurface contaminant transport that impact water quality.
Article
Full-text available
The efficacy of different proportions of silt-loam/bentonite mixtures overlying a vadose zone in controlling solute leaching to groundwater was quantified. Laboratory experiments were carried out using three large soil columns, each packed with 200-cm-thick riverbed soil covered by a 2-cm-thick bentonite/silt-loam mixture as the low-permeability layer (with bentonite mass accounting for 12, 16 and 19 % of the total mass of the mixture). Reclaimed water containing ammonium (NH4 +), nitrate (NO3 −), organic matter (OM), various types of phosphorus and other inorganic salts was applied as inflow. A one-dimensional mobile–immobile multi-species reactive transport model was used to predict the preferential flow and transport of typical pollutants through the soil columns. The simulated results show that the model is able to predict the solute transport in such conditions. Increasing the amount of bentonite in the low-permeability layer improves the removal of NH4 + and total phosphorous (TP) because of the longer contact time and increased adsorption capacity. The removal of NH4 + and OM is mainly attributed to adsorption and biodegradation. The increase of TP and NO3 − concentration mainly results from discharge and nitrification in riverbed soils, respectively. This study underscores the role of low-permeability layers as barriers in groundwater protection. Neglect of fingers or preferential flow may cause underestimation of pollution risk.
Article
Full-text available
Lake Purrumbete is located in western Victoria, Australia, and is highly regarded for its ecological, social, economic and scientific values. Recently, many lakes in the region have been dry or at their lowest level in recorded history, due to a drought that broke in 2010. For this study, a modified difference water budget method was employed to estimate net groundwater flux through the difference between the level of the lake and the water table, along with the specific yield and area of the aquifer. This model successfully modelled the lake level fluctuations. In recent years, Lake Purrumbete has fallen below the outflow level; however, because of its large volume, changes in salinity to date are minor and do not affect its freshwater status. An understanding of how these systems will behave in the future and how they may be best managed in a drying climate is considered an important step to adapt to climate change. Postulated future climatic changes in the region of the lake were used to project the lake level fluctuations to 2030 using the water budget model, and showed that dry conditions would cause the lake level to remain below its outlet elevation, but wet conditions would result in a significant recovery in the lake level. If the level of Lake Purrumbete continues to fall, the main threat to its ecological status will be the potential loss of some significant areas of fringing wetland habitat. The lowering in lake level could cause a shift in the abundance of algal plankton and so influence the whole lake food web. This may reduce the invertebrate diversity of the lake.
Article
Full-text available
In a field experiment at Vandenberg Air Force Base (VAFB) designed to mimic the impact of a small-volume release of E10 (10% ethanol and 90% conventional gasoline), two plumes were created by injecting extracted groundwater spiked with benzene, toluene, and o-xylene, abbreviated BToX (No-Ethanol Lane) and BToX plus ethanol (With-Ethanol Lane) for 283 days. We developed a reactive transport model to understand processes controlling the fate of ethanol and BToX. The model was calibrated to the extensive field dataset and accounted for concentrations of sulfate, iron, acetate, and methane along with iron-reducing bacteria, sulfate-reducing bacteria, fermentative bacteria, and methanogenic archaea. The benzene plume was about 4.5 times longer in the With-Ethanol Lane than in the No-Ethanol Lane. Matching this different behavior in the two lanes required inhibiting benzene degradation in the presence of ethanol. Inclusion of iron reduction with negligible growth of iron-reducers was required to reproduce the observed constant degradation rate of benzene. Modeling suggested that vertical dispersion and diffusion of sulfate from an adjacent aquitard were important sources of sulfate in the aquifer. Matching of methane data required incorporating initial fermentation of ethanol to acetate, methane loss by outgassing, and methane oxidation coupled to sulfate and iron reduction. Simulation of microbial growth using dual Monod kinetics, and including inhibition by more favorable electron acceptors, generally resulted in reasonable yields for microbial growth of 0.01-0.05.
Article
MODFLOW SURFACT is a state-of-the-art simulator that utilizes vadose zone flow and transport equations to provide practical solutions to the analysis of flow and contaminant transport at various levels of complexity and sophistication as needed for site evaluation and closure. The variably. saturated flow equation can be solved with standard retention functions or with bimodal or multimodal relative permeability curves for unsaturated flow in porous and fractured systems. The equation can further be I solved with pseudo-soil retention functions for confined-unconfined simulations and for use in wellbore hydraulics. Finally, the equation can be cast in terms of air phase flow to analyze subsurface air flow behavior. The variably saturated transport equation can be solved for an unsaturated medium or can be used for confined-unconfined situations. The passive phase of flow can be included in the equation to include both air and water phases in the transport situation. An immobile multicomponent nonaqueous phase liquid (NAPL) phase can further be included in the transport simulation with equilibrium partitioning providing mass transfer between phases, which adjusts NAPL saturations. Dual domain equations can be condensed into the transport equation to provide capabilities for analyzing transport in fractured media. General reaction capabilities provide analyses of complex environmental and geochemical interactions. Two examples are provided to demonstrate the value of a comprehensive simulation capability for site investigations.