Environ Monit Assess
DOI 10.1007/s10661-010-1428-1
Groundwater modeling of Saq Aquifer Buraydah Al
Qassim for better water management strategies
Ibrahim S. Al-Salamah · Yousry M. Ghazaw ·
Abdul Razzaq Ghumman
Received: 4 May 2009 / Accepted: 25 February 2010
© Springer Science+Business Media B.V. 2010
Abstract Saudi Arabia is an arid country. It has
limited water supplies. About 80–90% of water
supplies come from groundwater, which is depleting day by day. It needs appropriate management.
This paper has investigated groundwater modeling of Saq Aquifer in Buraydah Al Qassim to estimate the impact of its excessive use on depletion
of Saq Aquifer. MODFLOW model has been used
in this study. Data regarding the aquifer parameters was measured by pumping tests. Groundwater levels and discharge of wells in the area for
the year 2008 and previous record of year 1999
have been collected from Municipal Authority of
Buraydah. Location of wells was determined by
Garmin. The model has been run for different sets
of pumping rates to recommend an optimal use of
groundwater resources and get prolonged life of
aquifer. Simulations have been made for a long future period of 27 years (2008–2035). Model results
concluded that pumping from the Saq Aquifer in
Buraydah area will result into significant cones of
depression if the existing excessive pumping rates
I. S. Al-Salamah · Y. M. Ghazaw (B) ·
A. R. Ghumman
Department of Civil Engineering,
Faculty of Engineering, Qassim University,
Qassim, Saudi Arabia
e-mail: GHAZAW@yahoo.com
prevail. A drawdown up to 28 m was encountered
for model run for 27 years for existing rates of
pumping. Aquifer withdrawals and drawdowns
will be optimal with the conservation alternative.
The management scheme has been recommended
to be adopted for the future protection of groundwater resources in Kingdom of Saudi Arabia.
Keywords Saq Aquifer · Saudi Arabia · Qassim ·
MODFLOW · Groundwater · Modeling
Introduction
Sustainable management and development of water resources is a major issue all over the world.
Population in Saudi Arabia is increasing at a very
high rate. Life style is changing day by day. Water
demand for agriculture, industrial, and domestic use is becoming multifold. A survey done by
ministry of water and electricity in Saudi Arabia
(IPR 2005) shows that about 68% of the population of the country does not know that there
is scarcity in water resources in the Kingdom,
and about 99% do not know the cost of water
desalination. It shows that majority of people do
not use any water-saving tools. It is estimated
that in 2006, total water withdrawal was 23.7 km3 ,
an increase of 40% compared to 1992. The most
alarming point is that the surface and groundwater
withdrawal was 936% of total actual renewable
Environ Monit Assess
water resources in 2006 as per a survey report
(AQUASTAT 2008). Due to intensive pumping,
the cone of depression in the Buraydah area
has expanded significantly since the mid-1960s.
From 1982 to 1994, water levels dropped by about
53 m east of Buraydah in the confined area (AlSalamah 2000). Groundwater resources therefore
require careful planning and management so that
they can continue to sustain human socioeconomic development and the ecosystems (Sen and
Al-Somayien 1991).
This paper has highlighted the changes in
groundwater resources and devised appropriate
management of Saq Aquifer regime, Buraydah
Qassim, Saudi Arabia using MODFLOW Software. Although the present paper has focused on
a small area of Qassim, however, the results are
applicable to a wide range. Saq aquifer is very
large and has almost similar aquifer parameters
throughout (see Fig. 1, map of Saq Aquifer). The
pumping rates may be different in different areas, but due to the similar characteristics of the
aquifer, the results of this study are applicable
Fig. 1 Saq Aquifer (after
Ministry of Agriculture
and Water, Water Atlas
of Saudi Arabia 1984)
to a wide range of the area. Calculations at the
end of this paper regarding water consumptions in
Saudi Arabia with appropriate savings are based
on wide range of the area. It is for the whole
country of Saudi Arabia.
Groundwater modeling
Groundwater models can be physical, analogue,
or mathematical. Physical models are expensive.
Therefore, the computer-based numerical models
are widely used for this purpose (see, for example,
Ayars et al. 2006; Singh et al. 2006; Hirekhan et al.
2007; Hornbuckle et al. 2007; Mujtaba et al. 2008;
Banoeng-Yakubo et al. 2008; Saibi and Ehara
2008). There are a number of numerical models,
such as MODFLOW, PLASM, and AQUIFEM-1
to AQUIFEM-N (Wen-Hsing and Wolfgang 1999;
Gary et al. 2005; Anderson and Woessner 1992).
AQUIFEM-1 is a finite element model. A multilayered version called AQUIFEM-N is also
a widely used model. The MODFLOW and
Environ Monit Assess
PLASM are finite difference models, which are
extensively used these days.
Model description
The model development is based upon the
well-known three-dimensional flow equation
(Bossinesq equation).
∂
∂x
Kxx
= Ss
∂h
∂x
+
∂h
−W
∂t
∂
∂y
K yy
∂h
∂y
+
∂
∂z
Kzz
∂h
∂z
(1)
where h is head, Kxx , K yy , and Kzz are the hydraulic conductivities in x, y and z directions, respectively, Ss is the specific storage of the aquifer,
W is a sink/source, and t is time. The above Eq. 1
is a non-linear equation. Using the known value
of the aquifer thickness, the equation can be linearized and solved by finite difference method.
This approach is used in MODFLOW Model
(Anderson and Woessner 1992; Wen-Hsing and
Wolfgang 1999).
An understanding of these equations and their
associated boundary and initial conditions is necessary before a modeling problem can be formulated. The governing equations for groundwater
systems are usually solved either analytically or
numerically. Analytical models contain analytical
solution of the field equations, continuously in
space and time. In numerical models, a discrete
solution is obtained in both the space and time
domains using numerical approximations of the
governing partial differential equation. Various
numerical solution techniques are used in groundwater models. Among the most used approaches
in groundwater modeling, three techniques can
be distinguished: finite difference method, finite
element method, and analytical element method.
All techniques have their own advantages and disadvantages with respect to availability, costs, user
friendliness, applicability, and required knowledge of the user. Accordingly, there are several
commercial and research software available in the
market as mentioned earlier.
Study area
The study area is Buraydah City, which is located between latitude 26◦ 19 16 N to longitude
43◦ 57 32 E and latitude 26◦ 18 12 N to longitude 43◦ 57 59 E. It is the most important city in
Qassim area, which is famous for its agriculture.
The weather in the region is generally dry. The
temperature ranges from 43◦ C to 48◦ C during
daytime and 32–36◦ C during nighttime in summer.
In winter, it sometimes falls to 0◦ C. Buraydah has
very high rates of population growth. The population of Buraydah City is about 670,000 in year
2008, which is about 42% of the total population
of Al-Qassim. The most important aquifer in the
Qassim region is Saq Aquifer. There are some
other aquifers in Saudi Arabia, such as Minjur
Sandstone, Jilh Formation, Khuff Formation, and
Tabuk Formation.
Water demands have been increasing continuously in Buraydah due to which the pumping of
groundwater and the number of drilled wells has
increased to an alarming situation (Al-Salamah
2000). The Kingdom of Saudi Arabia began the
new development of the Al-Qassim region in
1975. A major increase in water extraction started
after 1975, due to major social, agricultural, and
constructional developments.
Important features of Saq Aquifer
Saq Aquifer starts from the Jordanian border at
latitude 24◦ 30 N to longitude 45◦ E. It has a
surface area of about 65,000 km2 . The subsurface
area is about 160,000 km2 . The Saq sandstone
crops out to the North and the West of Tabuk
Area and extends for 700 km from the Western
Edge of the Great Nafud Desert to the Jordanian
border as described above. Its thickness is up to
600 m at Jabal Saq (Saq mountain). In the Qassim
area, the Saq Sandstone thickness is gradually
increasing northward ranging from 400 m in the
southern part to 700 m in the northern part. It can
be realized from this information that Saq Aquifer
is a very large aquifer with thousands of pumping
wells.
It is a medium to coarse sandstone, with local
areas of fine sandstone. The rock type is poorly to
Environ Monit Assess
Fig. 2 Drawdown (m)
versus time (min) on
semilogarithmic paper for
Jacob method of
analyzing of pumping test
Time (minutes)
to
1
10
100
1000
0
0.1
Draw-down (m)
0.2
0.3
0.4
Δ(ho-h )
0.5
0.6
0.7
0.8
0.9
1
well sorted quartz sandstone. According to Sharaf
and Hussein (1996) and Abdel-Aal et al. (1997),
the electrical conductivity has an increasing trend
(it increased from 1.93 dS/m in 1983 to 2.76 dS/m
in 1987). They found that the total dissolved
solids (TDS) increased from 1,395 mg/L in 1983
to 1,992 mg/L in 1987. Saq Aquifer mineralized
to reasonable limits. According to Sharaf and
Hussein (1996), the TDS concentrations ranged
between 300 and 1,000 mg/L.
Fig. 3 Locations of
pumping wells in the
modeled area
The unique feature of this aquifer is that it
has known boundaries at its ends only, and the
boundary conditions are changing day by day due
to pumping from the aquifer. There is negligible recharge to the aquifer (AQUASTAT 2008;
Hussein et al. 1992). To incorporate the real natural boundaries, one has to collect a huge amount of
data spread over a long period. One has to collect
the data of thousands of pumping wells. Therefore, the only alternative is to work with a small
30000
25000
Y-direction (m)
20000
15000
10000
5000
0
0
5000
10000
15000
X-Direction (m)
20000
25000
30000
Environ Monit Assess
Fig. 4 Extensive mesh at
center of the modeled
area and well numbers
16000
10
5
Y-direction (m)
12
9
6
7
1
8
15000
3
13
2
14000
14
15
16
17
13000
13000
14000
15000
16000
X-Direction (m)
area with unknown boundaries. A general head
boundary has been considered in present work.
used to estimate the transmissivity T, hydraulic
conductivity K, and coefficient of storage S.
Hydraulic properties of Saq Aquifer
T=
For the major Saq Sandstone and associated
aquifers in Saudi Arabia, covering an area of
about 370,000 km2 , the data point frequency is
only about one value per 10,000 km2 (Lloyd 2007;
Bureau de Recherches Geologiques et Minieres
(BRGM) and Abunayyan Trading Corporation
2006). Hence, the hydraulic properties of the
aquifer in study area were determined by pumping
tests performed by the authors. The results are
shown in Fig. 2. The following equations were
2.25Tto
r2
K=T B
S=
(2)
(3)
(4)
where Q is well discharge, r is distance between
pumping and observation wells, to is the intercept
of a straight line on the x-axis and (ho − h) is
drawdown for one log time cycle are shown in
Fig. 2.
530.0
Groundwater elevation (m)
Fig. 5 Historic depletion
in groundwater elevation
monitored in observation
wells bottomed in Saq
Aquifer
2.3Q
4π ho − h
525.0
520.0
515.0
510.0
505.0
500.0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Year
Environ Monit Assess
520
Measured water levels (m)
515
510
505
500
495
490
485
480
480
485
490
495
500
505
510
515
520
Simulated water levels (m)
Fig. 6 Simulated and observed water levels at different
observation wells (year 2008)
The hydraulic conductivity of the aquifer was
calculated by dividing the transmissivity over the
saturated thickness of the aquifer. The hydraulic
conductivity value was found to be 3.6 m/day.
The specific storage was calculated by dividing the
storage coefficient by the saturated thickness of
the aquifer. The specific storage was found to be
6e−7 m−1 .
Model formation
Fig. 7 Hydrographs
showing average
drawdown in
groundwater level in Saq
Aquifer comparing with
groundwater level in 2008
(case of increasing rate of
pumping)
drawdon in Groundwater
elevation (m)
In MODFLOW, an aquifer system is replaced by a
discretized domain consisting of an array of nodes
and associated finite difference blocks (cells). In
this study, the area is represented by 60 columns
and 60 rows; more columns and rows have been
added at the locations of wells to minimize the error effects of large elements of grid. The location
of wells and extensive mesh are shown in Figs. 3
and 4, respectively. Three layers of 600, 250, and
250 m were considered. Upper 600 m layer represents the impermeable cover to the 500 m thick
aquifer. The aquifer was divided into two layers
each of 250 m thickness to model the partially penetrated wells accurately. The length and breadth
of area are 30,000 and 30,000 m covering the
modeling domain of 900 km2 . The layer type is
taken as “CONFINED.” The effective average
transmissivity (T) for the purpose of estimating
regional decline due to pumping in this area is
taken as 1.25 m2 /min.
The hydraulic conductivity of the area in X, Y,
and Z directions was taken as K = T/B, where B
is 500 m the thickness of aquifer, hence Kx = Ky =
Kz = 3.6 m/day. No recharge from the surface was
taken. Boundary conditions of study area were
unknown and changing with time. These were
calibrated.
Calibration of model
Calibration is the most important part of groundwater modeling (McCuen 2008). Data of water
levels in the study area for the year 1997 was used
as initial water levels. The model was calibrated
and validated for steady-state condition. Then, the
model was calibrated for unsteady conditions for
the observed data of water levels in years 1999
and 2008. Mean error and model efficiency
(Abulohom et al. 2002; Walpole and Myers 1985)
were used to check the performance of model.
The most difficult and challenging part of this
research was setting the right boundary conditions as the boundary conditions were not known,
and these are changing with time. General Head
40
30
20
10
0
2008
2010
2015
2020
Year
2025
2030
2035
Environ Monit Assess
Table 1 Average
drawdown in
groundwater level in Saq
Aquifer comparing with
groundwater level in 2008
Year
Case of constant
rate of pumping
Case of increasing
rate of pumping
Case of decreasing
rate of pumping
2008
2010
2015
2020
2025
2030
2035
0.003
2.443
7.523
12.483
17.433
22.363
27.303
0.003
2.443
7.523
14.733
20.183
27.783
33.493
0.003
2.443
7.523
10.163
14.673
16.933
21.103
Boundary package was used to calibrate the
boundary conditions.
A preliminary analysis of 17 wells for the area
of investigation showed that significant depletion
in the groundwater level has occurred. Figure 5
shows an average historic observed depletion of
25 m in the levels of groundwater with effect from
1993 to 2008. Groundwater elevation data were
not available from 2001 to 2007.
Results and discussion
Effect of different pumping alternatives on
groundwater levels
Fig. 8 Hydrographs
showing average
drawdown in
groundwater level in Saq
aquifer comparing with
groundwater level in 2008
(case of constant rate of
pumping)
drawdon in Groundwater
levels (m)
The simulated and observed hydraulic heads for
the year 2008 are shown in Fig. 6. Mean error was
observed to be 0.03 m. After calibration of model,
three pumping scenarios have been considered
and simulations have been made for 27 future
years (2008–2035).
The model developed for simulation of
Buraydah area was run for various scenarios to
assess the effects of different pumping alternatives
on future groundwater levels and drawdowns in
the study area.
First scenario of pumping rates
In this plan, it is assumed that there is an increase
in the present water extraction rates by 10% every
10 years until the end of year 2035. The simulated
hydraulic heads and drawdowns for the years of
2010, 2015, 2020, 2025, 2030, and 2035 have been
shown in Fig. 7. The highest average drawdown is
observed to be nearly in the center of the study
area where there are many domestic wells for
the city of Buraydah. The average hydraulic head
in the central area is 501.2 m, and the resulting
change of the hydraulic head (drawdown) from
the current levels (year 2008) in years of 2010,
2015, 2020, 2025, 2030, and 2035 are shown in
Table 1.
Second scenario of pumping rates
In this plan, it is assumed that the present trend
of increase in the water extraction rates is not
allowed, and only the present rate of pumping is
allowed for future. The simulated drawdowns for
the years of 2010, 2015, 2020, 2025, 2030, and 2035
have been shown in Fig. 8. The lowest hydraulic
head is observed nearly in the center of the study
40
30
20
10
0
2008
2010
2015
2020
Year
2025
2030
2035
Fig. 9 Hydrographs
showing average
drawdown in
groundwater level in Saq
aquifer comparing with
groundwater level in 2008
(case of decreasing rate of
pumping)
drawdon in Ground water
elevation (m)
Environ Monit Assess
40
30
20
10
0
2008
2010
2015
2020
2025
2030
2035
Year
area where there are many domestic wells for
the city of Buraydah. The hydraulic head in the
central area is 500 m in 2008, and the resulting
change of the hydraulic head (drawdown) from
the current levels in years of 2010, 2015, 2020,
2025, 2030, and 2035 are shown in Table 1.
a drawdown of 33.49 m up to the year 2035. This
may result in serious environmental effects and
crises in groundwater resources in future. Hence,
some management plans must be implemented.
Management plan
Third scenario of pumping rates
In this plan also, it is assumed that the present
trend of increase in the water extraction rates
is not allowed, rather a decrease in the present
rate by 10% per 10 years is implemented for
future. The simulated draw downs for the years
of 2010, 2015, 2020, 2025, 2030, and 2035 have
been shown in Fig. 9. As shown in the figure, the
lowest hydraulic head is nearly in the center of the
study area where there are many domestic wells
for the city of Buraydah. The resulting change of
the hydraulic head (drawdown) from the current
levels in years of 2010, 2015, 2020, 2025, 2030, and
2035 are shown in Table 1.
It is observed from Figs. 7, 8, and 9 and Table 1
that if the pumping rate is increasing, there will be
70
Population (million persons)
Fig. 10 Population
growth of Saudi Arabia
Data shows that there is water consumption at the
rate of about 250 L/day per capita. It is nearly
double than that of the normal water consumption
in developed countries. Population is increasing
at the rate of nearly 2% per year (see Fig. 10).
Hence, there is possibility of decreasing the pumping rates for optimal use of groundwater resources
(see Table 2). In 2035, the population will grow
to 47.7 million as compared to 28.7 million today
in 2009. On these bases, if the water consumption
is assumed to be 150 L per capita per day, then
the pumping rates worked out on the basis of
a decrease of 10% each 10 years may fulfill the
demand and hence increasing the life of aquifer
by 50% as compared to the present trend of
increase.
Series1
60
y = 0.6315x - 1240.6
Linear (Series1)
50
40
30
20
10
0
1980
1990
2000
2010
Year
2020
2030
2040
Environ Monit Assess
Table 2 Water consumptions in Saudi Arabia with appropriate savings
Year
Population
Total water consumption
at 250 L/day without
appropriate management
Total water consumption
at 150 l/day with appropriate
management
Groundwater
share
From
desalination
2008
2010
2015
2020
2025
2030
2035
28.7
29.8
32.74
35.97
39.52
43.42
47.71
2.62
2.72
2.99
3.28
3.61
3.96
4.35
1.57
1.63
1.79
1.97
2.16
2.38
2.61
1.52
0.53
0.69
0.87
1.06
1.28
1.51
1.1
1.1
1.1
1.1
1.1
1.1
1.1
How to achieve the management plan
Conclusions
The abovementioned optimal use of water can be
achieved by the following measures.
Groundwater of Buraydah has been investigated
using numerical model MODFLOW. It is concluded that Saq Aquifer is depleting at the rate
of about 1.1 m per year. If no precautionary
measures are made, it may bring a serious damage to water quality and environment. Impacts
of different pumping rates have been studied.
It is observed that if the increase in pumping
rates prevails (10% each 10 years), then there
will be a drawdown of about 33.5 m in only
27 years. A decrease in pumping rate by 10%
each 10 years may result in prolonging the aquifer
life by about 50%. The present rate of pumping will result in drawdown of about 27.3 m,
which may help prolonging the life of aquifer by
about 22% in comparison to the increased rates
of pumping.
Educating the people The government, Mosque
Khatibs, and social bodies may offer programs
to educate the public about the scarcity of water
resources in the country.
Installation of modern water saving devices Modern water saving devices at public places and in
houses may be introduced to reduce the wastage
of water.
Pricing water Putting appropriate price for water
may reduce its wastage.
Reuse of wastewater For irrigation and industries, the water quality is slightly different from
drinking water. Therefore, wastewater can be
used after the treatment for this purpose.
Recharging groundwater by deep wells Precipitation up to more than 100 mm has been recorded
in various years in the Qassim area. Runoff from
precipitation is usually not being taken care of.
This runoff can be injected to aquifer by deep
wells.
Leakage control Leakage control measures must
be implemented to minimize water losses from
water supply networks.
Acknowledgement The authors acknowledge highly useful support from Prof. Dr Zulfiqar Ahmad, Department
of Earth Sciences, Quaid-e-Azam University, Islamabad,
Pakistan. The support of Municipal Committee Buraydah
Al Qassim is also acknowledged for providing useful data.
References
Abdel-Aal, S. I., Sabrah, R. E., Rabie, R. K., & AbdelMagid, H. M. (1997). Evaluation of groundwater quality of irrigation in central Saudi Arabia, Arab Gulf.
Journal of Scientif ic Research, 15(2), 361–377.
Abulohom, M. S., Shah, S. M. S., & Ghumman, A. R.
(2002). Development of a rainfall–runoff model, its
calibration and validation. Journal of Water Resources
Management, 15(3), 149–163.
Environ Monit Assess
Anderson, M. P., & Woessner, W. W. (1992). Applied
groundwater modeling-simulation of f low and advective transport. San Diego: Academic.
AQUASTAT (2008). Irrigation in the Middle East Region
in figures. Water Report No 34, FAO, Rome, Italy.
Ayars, J. E., Christen, E. W., & Hornbuckle, J. W. (2006).
Controlled drainage for improved water management
in arid regions irrigated agriculture. Journal of Agricultural Water Management, 86(1–2), 128–139.
Banoeng-Yakubo, B., Yidana, S. M., Akabzaa, T., &
Asiedu, D. (2008). Groundwater flow modeling in the
Akyem area, southeastern Ghana. Journal of Environmental Hydrology, 16, 13.
Bureau de Recherches Geologiques et Minieres (BRGM),
& Abunayyan Trading Corporation (2006). Investigations of updating groundwater mathematical model(s)
for the Saq and overlying aquifers (Vol. 12). Report,
Ministry of Water and Electricity, Saudi Arabia.
Gary, S. J., Donna, M. C., & David, B. F. (2005). A numerical model and spread sheet interface for pumping test
analysis. Journal of Ground Water, 39(4), 582–592.
Hirekhan, M., Gupta, S. K., & Mishra, K. L. (2007). Application of WaSim to assess performance of a subsurface drainage system under semiarid monsoon climate.
Journal of Agricultural Water Management, 88(1–3),
224–234.
Hornbuckle, J. W., Christen, E. W., & Faulkner,
R. D. (2007). Evaluation of a multi-level subsurface drainage system in New South Wales, Australia.
Journal of Agricultural Water Management, 89(3),
208–216.
Hussein, M. T., Bazuhair, A. G., & Ageeb, A. E. (1992).
Hydrogeology of the Saq Formation east of Hail,
northern Saudi Arabia. Quarterly Journal of Engineering Geology, 25, 57–64.
Al-Salamah, I. S. (2000). To develop a useful management
tool for water resources allocation in the Saq aquifer
in Al-Qassim Region, Saudi Arabia. Ph.D. thesis, Iowa
State University, Ames, IA.
IPR Strategic Business Information Database. (2005).
Saudi Arabia: Results of survey on water needs, 21
August 2005.
Lloyd, J. W. (2007). The difficulties of regional groundwater resources assessments in arid areas. The Arabian
Journal for Science and Engineering, 32(1), 35–37.
McCuen, R. H. (2008). Calibration and reliability in
groundwater modeling: credibility of modeling. Journal of the American Water Resources Association, 44,
6 (ProQuest Science Journals, p. 1605).
Ministry of Agriculture and Water. (1984). Water atlas
of Saudi Arabia. Riyadh: Saudi Arabian Printing
Company Ltd.
Mujtaba, G., Ahmad, Z., & Ophori, D. (2008). Management of groundwater resources in Punjab, Pakistan
using a groundwater flow model. Journal of Environmental Hydrology, 16, 1.
Saibi, H., & Ehara, S. (2008). Hydrogeology and vulnerability assessment of groundwater resources in the
Mostaganem Plateau, Northwestern Algeria. Journal
of Environmental Hydrology, 16(1).
Sen, Z., & Al-Somayien, M. S. (1991). Some simple management criteria for confined Saq aquifer in Tabuk
Region, Saudi Arabia. Water Resources Management,
5, 161–179.
Sharaf, M. A., & Hussein, M. T. (1996). Groundwater quality in the Saq aquifer, Saudi Arabia. Journal of Hydrological Sciences, 4(5), 683–696.
Singh, R., Helmers, M. J., & Zhiming, Q. I. (2006). Calibration and validation of DRAINMOD to design subsurface drainage systems for Iowa’s tile landscapes.
Journal of Agricultural Water Management, 85(2006),
221–232.
Walpole, R. E., & Myers, R. H. (1985). Probability and
statistics for engineers and scientists (pp. 339–374).
London: Macmillan. ISBN-0-02-424170-9.
Wen-Hsing, C., & Wolfgang, K. (1999). Manual of processing MODFLOW, numerical model for groundwater
f low.