Arab J Geosci (2016) 9:149

DOI 10.1007/s12517-015-2069-z

ORIGINAL PAPER

A spatio-temporal three-dimensional conceptualization

and simulation of Dera Ismail Khan alluvial aquifer in visual
MODFLOW: a case study from Pakistan
Anwar Qadir 1 & Zulfiqar Ahmad 2 & Tahseenullah Khan 3 & Mohammad Zafar 3 &
Atwar Qadir 4 & Mamoru Murata 5

Received: 10 February 2015 / Accepted: 10 September 2015

# Saudi Society for Geosciences 2016

Abstract Dera Ismail Khan (DIK) is situated in the Lower till 2020. The simulation revealed incremental drawdowns till
Indus Basin of Pakistan. The land use has been changed in the the end.
canal command area due to irrigation activities near the Indus
River. To check the current status and predict the groundwater Keywords Visual MODFLOW . DIK . Water table depth
levels in the area, the unconfined aquifer has been simulated in (WTD) . Indus River
Visual MODFLOW for a period of 35 years, i.e., from 1985 to
2020. The 2900-km2 area has been modeled with a grid of 500
by 500 m and the depth set to 100 m. The aquifer in the study Introduction
area has been divided vertically and laterally into three and ten
zones, respectively, for the characterization. Water wells and Visual MODFLOW aids in the understanding of flow, trans-
streams were used as the sinks and hydrologic boundaries, port of solutes, and water budgets in the groundwater system
respectively. The model was successfully calibrated in steady for effective management. The earliest analytical solutions for
and the non-steady state. The simulation revealed that the hypothetical groundwater flow systems has been elaborated in
whole simulation can be divided into two phases, i.e., before 1960s by the help of recharge and discharge areas, groundwa-
and after the construction of the Chashma Right Bank Canal ter level fluctuations, surface topography, hydrochemistry, en-
(CRBC), whereas the results were summarized in the form of vironmental isotopes, vegetations, and surface water. (Toth
water table depth maps and groundwater budget calculations. 1963, 1970, 1971, 1972). The steady state models were first
To determine the groundwater sustainability, a conjunctive use simulated for steady state modeling of hypothetically layered
scenario has been employed to simulate the aquifer dynamics aquifers putting the concepts of three-dimensional aquifer het-
erogeneity and anisotropy in the groundwater basins (Freeze
and Witherspoon 1966, 1967, 1968). The transient model was
* Anwar Qadir initiated to determine the impacts of infiltration on the water
qaq212001@hotmail.com table rise and base flow hydrographs leading to the prediction
of maximum yield, pumping pattern, and recharge and dis-
1
Department of Geology, University of Haripur, Hattar Road,
charge conditions (Freeze 1971). Following these hypotheti-
Haripur, Pakistan cal studies, a large-scale aquifer simulation program has been
2
Department of Earth and Environmental Sciences, Bahria University,
initiated during 1978 to 1995 by the name of Regional Aquifer
Shangrilla Road, Sector E-8, Islamabad, Pakistan System Analysis (RASA) program of US Geological Survey
3
Department of Earth Sciences, Quaid-e-Azam University,
(Sun and Johnson 1994). The computer modeling techniques
Islamabad, Pakistan used in most cases were the USGS 3D finite difference model
4
Department of Basic Sciences, Riphah International University,
(Trescott 1975) and the USGS MODLFOW (McDonald and
Islamabad, Pakistan Harbaugh 1988). A listing of 1105 reports was published from
5
Department of Geosciences, Faculty of Science, Naruto University of
RASA program in USGS Professional Paper numbered from
Education, National University Corporation, Sun et al. (1997). The Netherlands has also contributed in
Naruto, Tokushima 772-8502, Japan broadening the concepts and methodologies in hydrological
149 Page 2 of 9 Arab J Geosci (2016) 9:149

system analysis (Engelen and Jones 1986; Engelen and storage (Ss), specific yield (Sy), and top and bottom elevation
Kloosterman 1996). In China, several regional groundwater of aquifers are incorporated in the model.
models have been constructed in recent years (Shao et al. Dera Ismail Khan (DIK) is the southernmost district of
2009; Hou and Zhang 2008; Wang et al. 2009, 2010). Khyber Pakhtunkhwa, lying between 31° 15′ and 32° 32′ north
MODFLOW evolved with time in response to its widespread latitudes and 70° 11′ and 71° 20′ east longitudes (Fig. 1). The
usage in industry. Three generations of MODFLOW can be DIK town is situated along the west bank of the Indus River. The
found, i.e., 1988, 2000, and 2005 (Harbaugh et al. 2000; study area is the Chashma Right Bank Canal (CRBC) stage-II
Harbaugh 2005; Barlow and Harbaugh 2006). Several command area, which is situated between the Indus River and
Windows-based graphic user interfaces for MODFLOW have the CRBC covering an area of 2900 km2. The Indus River forms
been developed such as Processing Modflow (Chiang and the eastern boundary of the model. The water from the Indus
Kinzelbach 2001). Visual Modflow (Waterloo Hydrogeological River is diverted in the form of canals for the irrigation purposes.
2001). Groundwater Modeling Systems (GMS) (Brigham Takwarrah Nala is a branch of Tank Zam having perennial flows.
Young University, Environmental Modeling Research Brigham Another stream Gomal Zam is a major command having peren-
Young and Environmental Modeling Research 2000). and nial flows. Sheikh Haider Zaman Nala is the southeastern bound-
Groundwater Vista (Rumbaugh and Rumbaugh 2005). ary of the model and has a negligible perennial discharge. DIK’s
An exponential growth of application of numerical model western boundary is surrounded by the mountains and hills of
flow study has been witnessed in the last 20 years in the world. South Waziristan Agency and tribal area. The highest region lies
The numerical models are used at various scales and for different toward north in the Sulaiman range about 3500 m above mean
purposes such as water management (Al-Salamah et al. 2011), sea level and the lowest point lies in the south near the Indus
surface water and groundwater interactions (Hunt et al. 2003). River and is approximately 152 m high above mean sea level.
contaminant transport (Al-Thani et al. 2004; Halford et al. 2010), The DIK district lies in the domain of hot semi-arid climate.
aquifer response determination (Clemo 2005). recharge study The model is divided into two zones by the CRBC into canal
(Korkmaz 2013; Anderson et al. 2002; Jyrkama et al. 2002). command and non-canal command. The study area consists of
and well head protection (Ahmad et al. 2010; Bolster et al. 2001). three broad hydrogeological zones: (1) the gravels and boulders
In this study, a model has been constructed in visual of the alluvial fan deposits, (2) the silty clay series of the pied-
MODFLOW 4.0 which uses the finite difference technique mont plain, and (3) the Punjab-type sand deposits of the active
to simulate heads at a block centered nodal grid. A three- and abandoned flood plains along the Indus (Hood 1970). In the
layered finite difference grid of 186 columns and 182 rows central, western, and northern regions of the area, the fill consists
was overlaid on the study area with a constant node spacing of of sediments conveyed by the streams and rivers from the con-
500 × 500 m in both x- and y-axes. The model was run initially solidated rocks surrounding the basin to the west. These deposits
with constant head and stream boundaries, but finally, the have extended to the east, where they interfinger with the
model settled down with the stream boundaries for the Indus Punjab-type sand deposits (Malik 1985; Fig. 2). The residents
River, Takwarrah Nala, and Sheikh Haider Zaman Nala of the study area utilize two sources of water: the surface water
(Fig. 1). Various hydraulic parameters including hydraulic and the groundwater. The surface water resources consist of the
conductivity (K), initial heads, recharge, porosity, specific Indus River, streams, and the ponds storing the rain water. The

Fig. 1 Location map of the DIK area showing the modeled area with all of the distinctive boundaries
Arab J Geosci (2016) 9:149 Page 3 of 9 149

Fig. 2 The fence diagram of the DIK modeled area showing the variation in subsurface rock assemblages

two main diversions that drain the modeled area are the CRBC adopted from Anderson and Woessner (1992). Visual
and the Paharpur canals. The CRBC command area located in MODFLOW uses MODFLOW 2000 to describe groundwater
the Indus basin receives its water from the Indus River at flow of constant density under non-equilibrium conditions in a
Chashma Barrage and irrigates an area of 0.470 million acres heterogeneous and anisotropic medium according to the fol-
of the DIK (Ahmed 1972). This canal was completed in three lowing equation:
stages, i.e., in 1987, 1992, and 2002.      
∂ ∂h ∂ ∂h ∂ ∂h ∂h
To predict the situation of groundwater levels in this area, a K XX þ K YY þ K ZZ −W ¼ S s ð1Þ
∂x ∂y ∂y ∂t ∂z ∂z ∂t
numerical model integrated in a Geographic Information System
(GIS) has been simulated for 30 years, i.e., from 1985 to 2020. where KXX, KYY, and KZZ are the hydraulic conductivity along
The most important tasks undertaken were to conceptualize the the x, y, and z coordinate axes (Lt−l); h is the potentiometric
natural system of the subsurface aquifer system and then the head (L); W is a volumetric flux per unit volume and repre-
simulation to predict the groundwater level changes in the longer sents sources and/or sinks of water (t−l); SS is the specific
term, i.e., 2020. This model will work as a tool for understand- storage of the porous material (L−l); and t is the time (t). In
ing the interaction between the surface and the groundwater and general, SS, KXX, KYY, and KZZ may be functions of space
take necessary measures to prevent the deteriorating effects on (Ss = Ss (x,y,z), KXX = KXX (x,y,z), etc.) and W may be a func-
land and water resources of the study area (Katpatal et al. 2014). tion of space and time (W = W(x,y,z,t)).
Further, this model may be integrated with the climate change Initially, data from Public Health Engineering Department
models to simulate the long-term effects for the future research. (PHED) and Water and Power Development Authority
Consequently, it will provide a baseline for the understanding of (WAPDA) were obtained. Most of the details of the field con-
future groundwater flow modeling and the problem solution ditions are also addressed during the number of field works
regarding the well head protection, water logging status deter- made in the area. On the basis of the data and the relevant
mination, and the determination of optimum well discharges in research, a conceptual model is prepared (Figs. 2 and 3). The
the resembling set of conditions. hydrogeological system of the study area is modeled as multi-
layered unconfined aquifer with variable thickness gently
sloping from mountain regions in NNW to Indus River in
Methodology SSE (Fig. 3). Based on the subsurface depositional unit cor-
relations and the pump screen settings in Fig. 3, the model is
The methodology followed for the current study involved divided into three layers (Table 1). Initial data of aquifer layer
software Visual MODFLOW 4.0 with standard procedures, dimensions, i.e., elevations and lateral extents, have been
149 Page 4 of 9 Arab J Geosci (2016) 9:149

Fig. 3 The conceptual model of

the simulated area showing slices
of aquifer on the basis of various
groundwater tapping depths and
general conditions of flow in
DIK. The blue and red arrows
mark the volume of water moving
in and out of the system

imported in the digitized vector format prepared in Global area. Using the data of annual precipitation of the DIK
Mapper 7.0 and Rockworks 2002 (Fig. 4). meteorological station from Naqvi (1977) and Malik (1985).
The boundaries are assigned in the software by digitizing the recharge was calculated. The rest of the recharge is
and attributing pertaining to the flows, channel widths, stages, through the infiltration from the Indus River, CRBC, and other
and vertical hydraulic conductivities. The area has been perennial streams. Discharge components include the evapo-
subdivided into a three-layered finite difference grid of 186 transpiration and the pumping wells data that are incorporated
columns and 182 rows with a constant node spacing of in this model. The evapotranspiration and pumping wells data
500 × 500 m both in x- and y-axes. The model was run initially have been acquired from the previous literature (Naqvi 1977;
with constant head and stream boundaries, but finally, the Malik 1985) and measured the recent discharge data of selec-
model settled down with the stream boundaries for the Indus tive wells in the study area. The recharge in this model is taken
River, Takwarrah Nala, and Sheikh Haider Zaman Nala. The for the rainfall (Kijne 1996) whereas the recharge from the
CRBC canal has also been modeled as a stream boundary after canals and rivers is addressed separately in the stream bound-
the 1987. Subsequently, ten test wells data obtained from the aries (Fig. 4). A conjunctive use scenario has been simulated
previous reports of Naqvi (1977) and Malik (1985) were used such as the usage of canal water and the groundwater to pre-
to interpolate the properties of the modeled area, e.g., hydrau- dict the groundwater balance in the study area.
lic conductivities, specific yield, recharge, and porosity, using After importing and assigning the respective features in the
Theissen polygon method (Theissen 1911). software, the model has been simulated and calibrated in
The precipitation, river flows, canals, and irrigation steady state in 1985 by modifying the hydraulic conductivities
practices were considered as the recharge source in the study and recharge values in a single time step during several runs

Table 1 Hydraulic conductivity zonation in the three-layered aquifer model

Layers 1 2 3

Properties K (m/s) Sy POR K (m/s) Sy POR K (m/s) Ss POR

Z O N AT I O N Zone 1 1.30 × 10−5 0.15 0.3 2.02 × 10−5 0.15 0.32 3.20 × 10−5 1.20 × 10−4 0.35
Zone 2 1.00 × 10−5 0.13 0.33 1.20 × 10−5 0.13 0.34 1.50 × 10−5 1.00 × 10−4 0.35
Zone 3 1.03 × 10−6 0.07 0.37 1.50 × 10−6 0.05 0.36 2.00 × 10−6 2.50 × 10−3 0.38
Zone 4 5.00 × 10−7 0.01 0.4 3.00 × 10−7 0.02 0.403 1.00 × 10−7 2.50 × 10−2 0.412
Zone 5 1.51 × 10−5 0.16 0.31 2.76 × 10−5 0.17 0.32 3.90 × 10−5 1.00 × 10−3 0.31
Zone 6 3.20 × 10−5 0.19 0.3 2.40 × 10−5 0.17 0.33 3.00 × 10−5 2.00 × 10−4 3.60 × 10−1
Zone 7 1.30E-05 0.15 0.34 1.50 × 10−5 0.15 0.36 2.00 × 10−5 1.20 × 10−4 3.90 × 10−1
Zone 8 1.50 × 10−6 0.09 0.39 1.70 × 10−6 0.09 0.4 1.00 × 10−5 1.70 × 10−3 3.70 × 10−1
Zone 9 2.00 × 10−6 0.03 0.41 3.70 × 10−6 0.03 0.405 3.50 × 10−6 2.70 × 10−2 4.32 × 10−1
Zone 10 1.21 × 10−5 0.15 0.31 3.00 × 10−5 0.16 0.3 3.31 × 10−5 1.10 × 10−3 2.90 × 10−1

The K, Sy, and POR describes the hydraulic conductivity, specific yield, and porosity
Arab J Geosci (2016) 9:149 Page 5 of 9 149

analyzing the drawdown, water table depth calculated maps,

and the analysis of the water balance in the model exported
and interpolated in software Surfer 7.

Results and discussion

The successful calibration between the observed and the cal-

culated hydraulic head values led to the steady state model of
the groundwater flow in DIK. The root-mean-square (RMS)
for the correlation is 0.65 with correlation coefficient 1.0 and
mean residual around 0.055 m (Fig. 5). The model was sim-
ulated with all the boundaries set to stream package of the
Visual MODFLOW.
The equipotential map of the year 1985 shows a general
gradient from northwest to southeast, i.e., from the mountain
ranges of the west to the Indus River (Fig. 6). An overall head
drop of 110 m is observed with a higher value of 260 m in the
extreme northwest to 150 m in the extreme southeast. The
Fig. 4 Location ofwater wells used to establish the subsurface main recharge tends to exist from the western rivers of
depositional unit correlation, calibration points for hydraulic heads in
steady state Gomal Zam and Tank Zam. The results have been matched
with the previous equipotential map of the study area to reach
the final conclusion (Naqvi 1977; Malik 1985).
(Thangarajan 2004). The parameter estimation and testing
The sensitivity analysis has been carried out for hydraulic
(PEST) has been used to manage the calibration with the min-
conductivities, recharge, and specific yield by multiplying
imum and maximum ranges of 30 hydraulic conductivities
with the pre-defined factors and observed the changes in the
and recharge values as described by Hill (1998). Later on,
mean, root-mean-square, and standard deviation of the simu-
on the basis of successful calibration, the sensitivity analysis
lated hydraulic heads (Thangarajan 2004). The results of the
is performed. The sensitivity of the model was tested by uni-
analysis show that model is more sensitive to recharge as
formly multiplying recharge and hydraulic conductivity by the
compared to hydraulic conductivity in the steady state. The
factors of 0.2, 0.4, 0.8, 1.2, 1.5, 2.0, 2.5, and 3.0 throughout
calibration in the transient state has been followed by the
the model’s interior nodes, and then the model was rerun. The
sensitivity analysis for the specific yield, and model has been
process is to be carried individually with one parameter and
found to be less sensitive.
later with the other. First, the analysis was carried out for
The results of the simulation are summarized in Fig. 7. The
hydraulic conductivities by multiplying with the pre-defined
water table depth maps show that there is a continuous draw-
factors and observed the changes in the mean, root mean
down in the groundwater table from 1985 to 1995 (Fig. 7a–c).
square, and standard deviation of the simulated hydraulic
Initially, there is a sharp decline in water table; however, after
heads. Following the sensitivity analysis, the transient calibra-
1995, i.e., the time of inception of CRBC, there was
tion has been carried out to define the storage coefficients and
to test the reliability of parameters (Guvanasen et al. 1998;
Akhter 2002). The transient state model has been calibrated
using water levels and drawdowns measured in ten test wells
over a 6-month period of the Rabi season (agricultural crops
sown in winter and harvested in the spring) from April 15,
1985 to September 15, 1985 using steady state hydraulic
heads as the initial condition. Therefore, the PEST has been
utilized to assess the automatic parameter estimation by cali-
brating it with the specific yield and specific storage (Table 1).
The steady state model was further enhanced for transient
conditions and simulated for 25 years more up to 2020. The
results thus acquired by the simulations were compared in
different maps such as equipotential maps and water table Fig. 5 The steady state calibration results of the DIK groundwater flow
depth maps. The modeling results have been summarized by model
149 Page 6 of 9 Arab J Geosci (2016) 9:149

irrigation activities. According to Kijne (1996) and Qureshi

(2008). there is a rise of water tables in the canal command
areas of Pakistan. However, in this simulation, it is clearly
visible that a long-term pumping of the groundwater will af-
fect the wells and there will be a decline of water table depths
in the aquifers in the order of tens of meters till 2020.
The drawdowns calculated in the ten zones have been used
to assess the potential of aquifers for the future exploitation in
hypothetical observation wells corresponding to the zones of
the model (Fig. 8). It has been found that the 35 years of
drawdown simulation curves can easily be divided into two
phases, i.e., before CRBC from 1985 to 1995 and after CRBC
from 1995 to 2020. The predictive period is from 2013 to
2020. In pre-CRBC phase, the drawdown curves show a mod-
Fig. 6 The steady state equipotential map of the DIKalluvial aquifer after erate steep rise in the curve and may be attributed to the initial
calibration and sensitivity analysis
stress conditions in the area. The post-CRBC phase shows a
gentle gradient of drawdown curves that is the result of the
stabilization of water table hence leading to a gradual decline stable situation. In predictive period, it can be found that there
(Fig. 7). This change may be attributed to the CRBC and is again a change in the situation as the drawdown curves are

Fig. 7 The water table depth maps of the years 1985, 1990, 1995, 2000, 2005, 2010, 2015, and 2020 (a–h), showing the changes in the water table depth
Arab J Geosci (2016) 9:149 Page 7 of 9 149

Fig. 8 The drawdowns

calculated during the complete
simulation of the model for
30 years. The general trend of
drawdowns show an increase in
the extraction of groundwater;
however, the well PM 1(4) is
showing a variation in the general
trend because of its location in a
low-conductivity zone

gaining the higher angles and continue to rise till 2020. In The construction of CRBC in 1987 has divided the whole
general, the drawdowns tend to descend in order from PM 1 region into two parts, i.e., the western bank and the eastern
(8) (water well dried up during the early stage of pumping), bank area. Based upon the concepts of the regional flows, it
(4), (5), (3), (7), (2), (6), (10), (9), and (1). The drawdowns are obliterated the whole groundwater flow system (Toth 1970).
lesser than 5 m in five wells that are located in areas near the The groundwater movement shows that the western side of the
western and eastern parts of the model (Fig. 5). These zones CRBC is an effluent part whereas the eastern bank is an influ-
are prospective zones for the exploitation of groundwater in ent part in most of the cases. The central zone lies adjacent to
the future till 2020. the west bank of the CRBC. So the sensitivity of this zone to

Fig. 9 The groundwater budget

analysis of the DIK groundwater
flow model simulation
149 Page 8 of 9 Arab J Geosci (2016) 9:149

the pumping is an obvious one due to the drainage of ground- prediction results prove the affordability of the short-term in-
water in that direction. crease in groundwater extraction in the DIK aquifer mostly in
Following the drawdown analysis, the groundwater budget the northwestern and southeastern extremities.
in the study area has been analyzed and divided into two
characteristic temporal zones, i.e., pre- and post-CRBC as Acknowledgments The Higher Education Commission is highly ac-
knowledged for providing all the necessary funding for this research
described earlier. In pre-CRBC conditions, the graphs for the
under the Indigenous Scholarship Scheme (No. B 064).
groundwater balance show a moderate rise in the curves for
the storage, recharge, stream leakage, evapotranspiration, and
well discharge (Fig. 9). The storage out shows a constant trend
References
as compared to the storage in that shows a small rise after
2010, i.e., the prediction period. The evapotranspiration is
Ahmad Z, Kausar R, Ahmad I (2010) Highlighting the implications of
found to be greater than the well discharges. The curves of selenium dispersion from disposal of Kahota industrial triangle area,
stream leakage in and out comparison revealed that most of Islamabad, Pakistan using three dimensional solute transport model.
the streams have an influent as well as effluent character. On Environ Monit Assess 167:565–579
the basis of groundwater budget analysis of the system, it can Ahmed N (1972) Groundwater resources of Pakistan. (revised and en-
larged ed.). Shahzad Nazir, Lahore
be inferred that the water moving into the system, i.e., re-
Akhter GM (2002) Three dimensional numerical groundwater flow
charge, is lesser than the discharge, i.e., water moving out of modeling of lower Bari Doab area, Punjab, Pakistan. Thesis Ph.D
the system (Fig. 9). In the previous studies of the groundwater (Unpublished), Quaid-e-Azam University, Islamabad, Pakistan
budgets, it has been reported that the CRBC command area is Al-Salamah IS, Ghazaw YM, Ghumman AR (2011) Groundwater model-
having more of the recharge, whereas this study revealed a ing of Saq Aquifer Buraydah Al Qassim for better water manage-
ment strategies. Environ Monit Assess 173(1–4):851–860
lesser recharge (Qadir et al. 2014) owing to a larger area where Al-Thani AA, Beaven RP, White JK (2004) Modelling flow to leachate
there is less calculated recharge as compared to the discharge. wells in landfills. Waste Manag 24(3):271–276
With the technological and demographic changes, it is per- Anderson M and Woessner W (1992) Applied groundwater modeling. (S.
tinent that all the groundwater basins of Pakistan have to be D. Inc., Ed.) Academic Press, p 381
modeled for the better understanding and management in the Anderson MP, Hunt RJ, Krohelski JT, Chung K (2002) Using high hy-
draulic conductivity nodes to simulate seepage lakes. Ground Water
forecoming era of climatic changes and uncertainities to en- 40(2):117–122
sure the sustainability of aquifers and food security in this Barlow PM, Harbaugh AW (2006) USGS directions in MODFLOW
area. development. Ground Water 44(6):771e774
Bolster CH, Genereux DP, Saiers JE (2001) Determination of specific
yield for the Biscayne aquifer with a canal-drawdown test. Ground
Water 39(5):768–777
Conclusions Brigham Young University, Environmental Modeling Research
Laboratory (2000) Groundwater modeling system (GMS), GMS
The simulation results confirmed the nature of most of the 3.1 tutorials
streams to be of influent character toward northwest. The Chiang WH, Kinzelbach W (2001) 3D-groundwater modeling
northwestern part of the region is attributed to be the recharge withPMWIN: a simulation system for modeling groundwater flow
and pollution. Springer, New York, 346 pp
area, whereas the southeastern part consists of the effluent Clemo T (2005) Improved water table dynamics in MODFLOW. Ground
nature of streams in this model. The recharge rate is much less Water 43(2):270–273
than the combined rate of extraction and natural discharge. Engelen GB, Jones GP (1986) Developments in the analysis of ground-
The central part of the modeling represents a low-velocity water flow systems. IAHS Publication, Wallingford, pp 163–365
zone, and a moderate-velocity zone lies in the extreme Engelen GB, Kloosterman FH (1996) Hydrological system analysis:
methods and application. In: water science and technology library,
southeast. vol. 20. Kluwer Academic Publishers, p 149
Two temporal phases of groundwater flow are found, i.e., Freeze RA (1971) Three-dimensional, transient, saturated-unsaturated
the pre-CRBC, a sharp increase in water level decline, and flow in a groundwater basin. Water Resour Res 7(2):e347–e366
post-CRBC, showing a gradual decline due to an added re- Freeze RA, Witherspoon PA (1966) Theoretical analysis of regional
charge in the command area. The simulation also revealed that groundwater flow: 1. Analytical and numerical solutions to the
math-ematical model. Water Resour Res 2(4):e641–e656
at present extraction rates, the maximum groundwater decline
Freeze RA, Witherspoon PA (1967) Theoretical analysis of regional
will be found in the central low-velocity zone, i.e., located on groundwater flow: 2. Effect of water table configuration and subsur-
the west bank of the CRBC that has an influent nature. The face permeability variation. Water Resour Res 3(2):e623–e634
prospective zones for the groundwater exploitation in the Freeze RA, Witherspoon PA (1968) Theoretical analysis of regional
study area exist near the Indus River and near the western groundwater flow: 3. Quantitative interpretations. Water Resour
Res 4(3):e581–e590
mountain ranges. The transition zone between the piedmont Guvanasen V, Wade SC, Barcelo MD (1998) Application of
and the floodplain deposits lying away from the canal com- MODFLOW-SURF to simulate regional groundwater flow in
mand area toward west is sensitive to the pumping. The Hernando county, Florida. Ground Water 38(5):772–783
Arab J Geosci (2016) 9:149 Page 9 of 9 149

Halford KJ, Stamos CL, Nishikawa T, Martin P (2010) Arsenic manage- Qadir Anwar, Ahmad Zulfiqar, Khan Tahseenullah, Zafar Muhammad
ment through well modification and simulation. Ground Water (2014) Causes and effects of temporal groundwater level change
48(4):526–537 in the alluvial aquifer of Dera Ismail Khan area, Pakistan. JHES,
Harbaugh AW (2005) MODFLOW-2005, the U.S. geological survey’s NCEG University of Peshawar, Pakistan, pp 185–191
modular ground water flow modele the groundwater flow process. Qureshi AS (2008) Managing salinity and water logging in the Indus
Techniques and methods 6eA16. U.S. Geological Survey, Reston, Basin of Pakistan. Agric Water Manag 95:5
Virginia Rumbaugh J and Rumbaugh D (2005) Groundwater vistas 5.0,
Harbaugh AW, Banta ER, Hill MC, McDonald MG (2000) The U.S. 1996e2005. Environmental Simulation, Inc
geological survey’s modular ground water flow modele user guide Shao JL, Zhao ZZ, Cui YL, Wang R, Li CQ, Yang QQ (2009)
to modularization concepts and the ground water flow process. U.S. Application of groundwater modeling system to the evaluation of
geological survey. Open-file report 00e92 groundwater resources in North China plain. Resour Sci 313:
Hill MC (1998) Methods and guidelines for effective model calibration. 361e367 (in Chinese)
U.S. geological survey water resources investigations report 98– Sun RJ and Johnson RH (1994) Regional aquifer system analysis pro-
4005, 90 p gram of the U.S. geological survey, 1978e1992. U.S. geological
Hood JW (1970) Water resources and related geology of Dera Ismail survey. Circular 1099
Khan District, West Pakistan with reference to the availability of Sun RJ, Weeks JB and Grubb HF (1997) Bibliography of regional aquifer
groundwater for development: U. S. geological survey water supply system analysis program of the U.S. geological survey, 1978e1996.
paper 1608-K, p 74 U.S. geological survey water resources, Austin, Texas. Investigation
Hou GC, Zhang MS (2008) Exploration and study of groundwater in the report 97–4074
Ordos Basin. Geological Publishing House, Beijing, p 563, in Thangarajan M (2004) Regional groundwater modeling. Capital
Chinese Publishing Company, New Delhi, India
Theissen AH (1911) Precipitation average for large areas. Mon Weather
Hunt RJ, Haitjema HM, Krohelski JT, Feinstein DT (2003) Simulating
Rev 39:1082–1084
ground water-lake interactions: approaches and insights. Ground
Toth J (1963) A theoretical analysis of groundwater flow in small drain-
Water 41(2):227–237
age basins. J Geophys Res 68(16):4795e4812
Jyrkama MI, Sykes JF, Normani SD (2002) Recharge estimation for
Toth J (1970) A conceptual model of the groundwater regime and the
transient ground water modeling. Ground Water 40(6):638–648
hydrogeological environment. J Hydrol 10:164e176
Katpatal YB, Pophare AM, Lamsoge BR (2014) A groundwater flow Toth J (1971) Groundwater discharge: a common generation of diverse
model for over exploited basaltic aquifer and Bazada formation in geological and morphological phenomena. Bull Int Assoc Sci
India. Environ Earth Sci 72(11):4413–4425 Hydrol 16:9e24, 1.3
Kijne WJ (1996) Water and salinity balances for irrigated agriculture in Toth J (1972) Properties and manifestations of regional groundwater
Pakistan. International Irrigation Management Institute (IIMI), Sri movement. In: Proceeding of 24th international geological congress,
Lanka Montreal, section 11, 153e163
Korkmaz S (2013) Transient solutions to groundwater mounding in Trescott PC (1975) Documentation of a finite difference model for sim-
bounded and unbounded aquifers. Ground Water 51(3):432–441 ulation of three dimensional ground-water flow. U.S. geological
Malik YM (1985) Technical report on Kulachi-Tank Dera Ismail Khan survey, open-file report 75–438, p 48
District NWFP. TNO-DGV institute of applied geosciences. Wang LY, Han JP, Liu JR, Ye C, Zheng YJ, Wan LQ, Li WP, Zhou YX
Netherlands, WAPDA. Peshawar: hydrgeology section, planning (2009) Modeling of regional groundwater flow in Beijing plain. J
directorate Hydrogeol Eng Geol 36(1):11e19 (in Chinese with English
McDonald MG, Harbaugh AW (1988) A modular three-dimensional fi- abstract)
nite-difference ground-water flow model. Open-file report 83e875. Wang H, Lu CY, Qin DY, Sang XF, Li Y, Xiao WH (2010) Advances in
U.S. geological survey. Book 6 method and application of groundwater numerical simu-lation.
Naqvi AH (1977) Groundwater conditions in Paharpur area, Dera Ismail Earth Sci Front 17(6):1e12 (in Chinese with English abstract)
Khan district, NWFP. Dera Ismail Khan: hydrogeology section plan- Waterloo Hydrogeological (2001) Visual MODFLOW Pro, 3D ground-
ning directorate WAPDA water flow and contaminant transport modeling, V.3.1