Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Modelling Flow in Aquifer

Download as pdf or txt
Download as pdf or txt
You are on page 1of 20

Jordan Journal of Civil Engineering, Volume 1, No.

2, 2007

Using Modflow and MT3D Groundwater Flow and Transport Models As a


Management Tool for the Azraq Groundwater System
Wail Y. Abu-El-Shar1) and Randa I. Hatamleh2)
1)

Associate Professor, Department of Civil Engineering, Jordan University of Science and Technology,
Irbid, Jordan, E-mail: wail@just.edu.jo
2)
PhD Student, Department of Civil Engineering, The Hashemite University, Zarqa, Jordan,
E-mail: rhatamleh@gmail.com

ABSTRACT
Numerical modeling has emerged as an effective tool for managing groundwater resources and predicting future
responses, especially when dealing with complex aquifers systems and heterogeneous formations. Among these
models, MODFLOW and MT3D are the most commonly used simulators for groundwater flow and solute transport
in subsurface systems, respectively. These models have been used herein as a management tool for the Azraq basin,
one of the most important groundwater resources for domestic and agricultural sectors in Jordan. Groundwater
extraction from this basin already exceeded the safe yield of the aquifer, and a sharp drop in the water table, a dry
out of the springs at the center of the basin and the problem of increased salinity in many parts of the aquifers have
been reported. Currently, more than 600 wells including governmental, private and unauthorized wells are operating
within the basin boundary. In its attempts to restore and sustain the aquifer, the Ministry of Water and Irrigation
(MWI), the official entity in charge of water resources in Jordan, is considering several scenarios of controlled
pumping. In this study, five suggested scenarios of pumping with different abstraction rates for years 2005 through
2020 have been explored by using the three dimensional finite difference flow model [MODFLOW (PM5)] to
simulate the flow system, and the solute transport model (MT3D) to predict the transport of total dissolved solids
given in terms of Electric Conductivity (EC). These scenarios include: first, maintaining the current pumping rate of
57 MCM for the study period; second, reducing the current pumping rates by half; third, increasing the pumping
rate by half; fourth, reducing the pumping rates in public wells by half and maintaining the current rates for other
wells; and finally reducing the pumping rates by half for the farm wells (private) and maintaining the rates at the
other wells. Results indicate that the first and fourth scenarios have similar effect on the drawdown. Also, the
second and fifth scenarios have similar effects and provide the lowest drawdown values. The third scenario gives
the worst drawdown. The transport of Total Dissolved Solids (TDS) given in terms of Electric Conductivity (EC),
has also been explored. Different parameters including EC, recharge, model boundary and advection parameters
were adjusted to run the model. Simulation results indicated that the effect of the different scenarios on the values of
EC is less profound than the effects on the drawdown values. The third scenario caused a slight increase of EC
values over the values simulated by other scenarios.
Keywords: Numerical Models, Modflow, MT3D, Groundwater Flow, Transport, Management
Scenarios, Azraq Basin.

Received on 1/1/2007 and Accepted for Publication on


1/4/2007.

-153-

2007 JUST. All Rights Reserved.

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

INTRODUCTION
With nearly half the population of Jordan drawing its
drinking water from groundwater resources, issues of
groundwater flow and contaminant transport in
subsurface systems have elicited considerable interest
from both public and private sectors. Among the twelve
groundwater basins in Jordan, Azraq basin is one of the
most important since it provides the three main cities of
Jordan, namely: Amman, Irbid and Zarqa with significant
portions of its domestic water supplies. It consists of a
three aquifer system, the upper (unconfined), middle and
lower aquifers, separated by two confining formations.
The water is mainly pumped from the upper aquifer and
only few wells penetrated the middle aquifer and the deep
aquifer, respectively.
In the beginning, water was pumped from the basin
to supply the city of Irbid only, and the increased water
demand due mainly to the explosive population growth in
the cities of Amman and Zarqa, led to drill new well
fields and draw groundwater at high rates. Furthermore,
private wells in the basin increased dramatically in the
past two decades; some of these wells are not even
authorized. The total pumping from the basin was
estimated to reach 57.7 MCM/yr in the year 2001 (AlHadidi and Subah, 2001), which is far beyond the safe
yield of the basin (30 MCM/yr.). Absence of full control
on private wells and its pumping rates in addition to
pumping from the water authority well fields resulted in
overexploitation of water storage and deterioration of
groundwater quality in some parts of the aquifer.
The Azraq basin received great attention, and
numerous studies were conducted for estimating the
potential safe yield and for assessing current and potential
future scenarios (Agrar and Hydrotechnick, 1977;
Humphry and Sons, 1978 and 1982; Rimawi, 1985; AlMomani, 1993; Al-Kharabsheh, 1995; Ayed, 1996;
UNDP- Azraq Oasis Conservation Project, 1996;
Jordanian Consulting Engineering, 1997; Al-Khatib,
1999; Hatamleh, 2001; Al-Hadidi and Subah, 2001;
Rihani, 2003; Moqbel, 2004). All computer modeling
studies of Azraq basin (except for Rihani, 2003 and

- 154 -

Moqbel, 2004) used the processing ModFlow and MT3D


to come up with future scenarios for water management
of the basin. For instance, the water budget obtained from
the steady state simulation organized by the United
Nations Development Program (UNDP) and the Azraq
Oasis Conservation Project (1996) employed Processing
ModFlow (PM3) to predict the behavior of the shallow
aquifer for the years 2005, 2015 and 2025. These results
formed the basis of the solute transport modeling study
(via MT3D) organized by Jordanian Consulting
Engineers (1997) which also included a hydrochemical
evaluation of the basin.
Concerns about increasing salinity in the basin began
to rise since the mid 1980s, and numerous studies were
conducted to assess the severity of the situation and to
determine the salinity characteristics of the groundwater
at Azraq (Worzyk and Huser, 1987; Al-Waheidi, 1990;
Al-kharabsheh, 1991; Rimawi, 1985; Ayed, 1996;
Jordanian Consulting Engineering, 1997; Moqbel, 2004).
Among the first of these was a study conducted by
Worzyk and Huser (1987), which aimed at locating the
fresh water/ salt water interface, especially at the Qaa
area and the well fields. The main conclusion of this
study was that water quality deterioration in the upper
aquifer is due to the full utilization and overpumping of
the aquifer and would most likely originate at the center
of the basin and then extend towards the north, northwest
and northeast parts of the basin. Al-Waheidi (1990)
showed that the saline zones extended to a maximum
depth of 65m within the upper aquifer. However, two
types of salinity were characterized in the upper aquifer
(Al-Kharabsheh, 1991), namely; Na-HCO3 in the north
and NaCl in the south and southwest.
Rimawi (1985) conducted an isotope hydrology study
and investigated the hydrochemical properties of the water
at Azraq. Some of the groundwater stored in the basin was
dated back to 20000 years ago. Recent waters (from the
present) were also found to exist in the basin. Ayed (1996)
studied
the
hydrological,
hydrogeological
and
hydrogeochemical characteristics of the basins three
aquifer system. This study indicated that water from the
upper aquifer system is potable in the Qaa area (around

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

the middle of the aquifer) while water in the south flowing


toward the north and west of Azraq has high TDS values.
All the above mentioned studies reported a general
lack of data about the hydrogeologic system that exists at
Azraq basin. For example, only four observation wells do
exist for the upper aquifer and no observation is available
for the middle and deep aquifers. Nevertheless, important
decisions need to be made by the MWI in order to restore
and sustain the aquifer systems. The study presented
herein used Processing Modflow PM5 and the solute
transport code (MT3D) to construct a groundwater flow
model and to model contaminant transport in the basin.
These models were then used as a management tool for
assessing the current situation and forecasting future
responses to assumed coming events. Five different
future pumping scenarios have been considered.
DESCRIPTION OF THE STUDY AREA
Location
Azraq Basin is one of the largest basins in the northern
parts of Jordan. According to the Palestine grid, the basin
lies from 250 to 400 east and from 55 to 230 north as
shown in Figure (1). It covers an area of approximately
12710 km, of which 94 percent lies within Jordan and the
reminder in Syria and Saudi Arabia. The area is bounded
by Syria to the north, Hammad Basin to the east, Sirhan
Basin to southeast, Mujib Basin to the south and Zarqa
Dhuleil Basin to the west.
Topography and Drainage
Azraq Basin is a depression surrounded by hilly relief.
All the water courses within its boundary drain in the
center of the basin, Qaa Al-Azraq, which is the lowest
point in the basin. The elevation of Qaa Al-Azraq is
about 503 to 512 m above mean sea level (a.m.s.l.). The
elevation rises to 900 m (a.m.s.l.) in the eastern, southern
and western parts of the basin. It increases sharply to
1550 m (a.m.s.l) in the north.
Geology
The outcropping formations in the Azraq Basin are

the Rijam and Wadi Shallala in the central and eastern


part of the basin. The northern part of the Azraq Basin is
dominated by Miocene to Pleistocene basalt, whereas to
the west and south by the Rijam and Muwaqqar
Formations of the Late Cretaceous (Early Tertiary) age.
Cretaceous rocks in Jordan are subdivided into two main
sequences; the Early and Late Cretaceous. The Early
Cretaceous rocks are locally known as Kurnub, and the
Late Cretaceous rocks are further subdivided into the
Ajlun and Belqa Groups. There are two major fault
systems in the Azraq Basin, one trending east to west and
a second trending northwest to southeast.
Hydrology
The climate of the Azraq Basin is characterized by
two well-defined seasons, a hot and dry summer, and a
wet and cold winter. Most of the study area is arid, while
a small portion can be considered as semi-arid. The
average annual minimum and maximum temperatures are
11.6C and 26.6C. The average relative humidity varies
from 49.9% to 61% in summer and from 56% to 82% in
winter. The wind direction is northwest in summer,
shifting to southeast in the winter. The annual average
wind speed is 11.6 km/hour, which ranges between 10
and 18 km/hour in the winter and from 7 to 12 km/hour in
the summer. The average daily evaporation observed
from a class A pan is 10.4 mm/day and varies from 5 to
19 mm/day in summer and from 3 to 12 mm/ day in
winter. The mean annual rainfall in the Azraq basin
varies between 75 and 400 mm in a wet water year, and
varies between 50 and 300 mm in a normal water year,
whereas it ranges from 10 to 175 mm in dry conditions. It
is clear that the mean annual rainfall decreases across the
basin from northwest to southeast. A number of major
wadi's drain into the central mud flat of Qaa Azraq.
These wadis are characterized by wide, shallow flowbeds with relatively low slopes.
HYDROGEOLOGY
Aquifer Systems
Groundwater aquifers in Jordan can be divided into

- 155 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

three main hydraulic complexes (Agrar and


Hydrotechnick, 1977): (1) the Shallow Aquifer System
(Upper Aquifer, B4/B5); (2) the Upper Cretaceous
(Middle Aquifer System, B2/A7) Amman-Wadi Sir
Hydraulic Complex; and (3) the Deep Sandstone (Lower
Aquifer System), Kurnub and Disi Hydraulic Complex.

The Upper (Shallow) Aquifer and the Middle Aquifer


Systems are separated by the Muwaqqar (aquiclude)
formation (B3). Two major groups of springs existed in
the central part of the Azraq Basin comprising the main
discharge outlets of the Shallow (Upper) Phreatic
Aquifer.

Figure 1: Map of Jordan showing the location of Azraq basin (Al-Hadidi and Subah, 2002).
The Upper Aquifer System is an unconfined aquifer
except at the middle of the basin (Qaa Azraq). The Qaa
Azraq deposits form a lens up to 15 meters thick
overlying the aquifer and give rise to leaky artesian

- 156 -

conditions. Before the publicly owned well-field


extraction to Amman began in 1982, this aquifer
discharged as springs and seeps near Azraq Al-Shishan
and Azraq Al- Doruz and as evaporation from the shallow

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

water table in the vicinity of the Qaa Azraq.


The Middle Aquifer System is a confined aquifer due
to the aquiclude, bituminous marl of the (B3) formation.
This aquifer system underlies the Upper Aquifer System
and outcrops in the western part of the basin. More than
15 water wells penetrating the B2/A7 Aquifer System
were drilled in the Azraq Basin.
The Middle and Lower Aquifer Systems are
separated by marl and marly limestone of the lowpermeability Ajlun Group (Al/6). The Lower Aquifer
System is a confined aquifer. There is only one well in
the Azraq Basin completely penetrating the B2/A7
Aquifer System. It is only 40 m through the underlying
Kurnub Sandstone (Lower Aquifer), with a total depth
of about 1299 m. It was found that the Kurnub Group
seems to be a promising ground source for future
planning. The Middle and Lower Aquifer Systems occur
at depths varying from 400 to 3000 m with brackish to
saline water (Gibbs, 1993). The ground water exploited
from the Middle and Lower Aquifer Systems have not
been utilized to date.
The saturated thickness of the Upper Aquifer System
in the Azraq basin varies from one place to another. The
depth of the water table varies from a few meters in the
center of the basin to more than 300 m in the north and
northeast parts of the basin at the Basalt Plateau.
More than 1100 water wells have been drilled in the
basin, most of which are located north and east of the
Azraq Oasis. About 600 wells are in operation. The
depths of these wells range from a few meters in the
center of the basin to more than 400 m to the north and
northeast of the basin (BGR/WAJ, 1996).
The yields of the wells are variable and range between
7 and 107 m3/hour in the B4/B5 aquifers, between 13 and
200 m3/hour in the Basalt aquifer and between 14 and
304 m3/hour in the Basalt/Rijam (B4) aquifer.
Groundwater Level Fluctuation
The groundwater of the Upper Aquifer System
generally occurs under unconfined conditions in the
Azraq Basin. The water table is not stable and fluctuates
according to wet and dry seasons and also according to

the withdrawal rate of groundwater. Four observation


wells (F1022, F1043, F1060 and F1280) are used to
monitor the water level fluctuations of the Upper Aquifer.
These wells have total depths of 88, 255, 116 and 195 m,
respectively. The F1060 borehole terminates in the basalt;
the others penetrate both basalt and B4 limestone (BGR,
WAJ, 1996).
Discharge
Prior to the sixties, the Upper Aquifer System in the
Azraq Basin was relatively undeveloped. Drilling
activities started in the early sixties and the extensive
drilling program accelerated in the 1980s. Until 1990, the
discharge from the Upper Aquifer System occurred at
springs in the center of the basin and water pumped from
boreholes. Unfortunately, spring flow declined to zero by
the end of 1990. Now, the total number of boreholes
reached 1100 and the total annual groundwater extraction
has continually increased.
Storage
The evaluation of storage in the Upper Aquifer
System of the Azraq Basin is estimated by the amount of
water that can be released before the decline of the water
table depth becomes the controlling factor of groundwater
development in the area. The saturated area in the Azraq
Basin is about 9900 km2, the average saturated thickness
of the Upper Aquifer System is 150 m and the average
specific yield is 0.015 (Ayed, 1996).
Recharge
The average annual recharge to the upper aquifer of
the Azraq Basin was estimated to be about 34 MCM (AlHadidi and Subah, 2001).
Hydrogeochemistry
The historical hydrochemical data for the wells in
Azraq Basin were collected from more than 960 samples
collected from wells scattered all over the basin (UNDP
and Azraq Oasis Conservation Project (1996)). These
samples were subjected to correlation analysis in order to
study the interrelation between the different parameters.

- 157 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

The high significant relationships are indicated between


EC and Cl, Na, Mg, Ca and SO4 in descending order and
low significant relationships between EC and pH and
CO3. The values of EC were calculated by multiplying
the values of TDS (mg/L) by 0.64 (Jordanian Consulting
Engineering, 1997).
FLOW AND TRANSPORT MODEL
Conceptual Model
The conceptual model for Azraq Basin adapted by this
study consists of three hydrogeological layers, two aquifers
and a confining layer. The top soil formation (B4/B5) of
the unconfined aquifer is taken as the first layer, followed
by the confining layer (B3) as the second layer. The third is
the Middle aquifer (B2/A7). Interaction between the
various aquifer systems is represented by leakage terms.
These layers in addition to the surrounding boundaries are
distinguished from each other by the hydraulic
conductivity for each element in its field.
Sources and sinks of water in the basin (precipitation,
baseflow, springs and recharge) are assigned to the
corresponding grid blocks within the domain.
Flow and Transport Governing Equation
A general form of the governing equation which
describes the three dimensional movement of
groundwater flow of constant density through the porous
media is (Freeze and Cherry, 1979):
h h h
h
Kx + Ky + Kz w = Ss
x x y y z z
t

(1)

where: Kx, Ky, Kz are values of hydraulic


conductivity along the x, y and z coordinate axes (L/t); h:
is the potentiometric head (L); w: is the volumetric flux
per unit volume and represents sources and/or sinks of
water per unit time (t-1); Ss: is the specific storage of the
porous material (L-1); and t: is time (t).
The first part of this problem was run to get a steady
state solution that takes the form:

- 158 -

h
h h
Kx + Ky + Kz w = 0.0
x
x y
y z
z

(2)

From the steady state solution, the hydraulic


conductivity for model aquifers can be found. Then the
equation is solved for transient case in order to solve for
storage coefficient.
The partial differential equation for three-dimensional
transport of contaminants in groundwater is (Freeze and
Cherry, 1979):
C
C
(vi C ) + qs C s +
=
Dij

t xi xi xi

(3)

k =1

where: C: the concentration of contaminant dissolved


in groundwater; t: time (t); x i the distance along the
respective Cartesian co-ordinate axis; Dij: the
hydrodynamic dispersion coefficient; i: the seepage or
linear pore water velocity; qs: the volumetric flux of
water per unit volume of aquifer representing sources
(positive) and sinks (negative); Cs: the concentration of
the sources or the sinks; : the porosity of the porous
medium; and Rk: chemical reaction term.
FLOW MODEL
Model Input
The model domain and grids used in this study are the
same for both the flow and transport model. The model
covered 5431.5 km2 and the whole area is divided into 81
columns and 54 rows. The total number of cells is 4374
cells. The area of the largest cells is 8.69 km2 and the area
for the smallest cells is 0.14 km2. Similar grid sizes have
been used by a previous groundwater flow study (AlKhatib, 1999). The smallest cells are located in Qaa
Azraq where most of the wells exist. The boundary
conditions of the upper aquifer (B4/5 and alluvium) are
described in Figure (2). Constant head boundary is in the
north, east and south of the domain. The topographic
elevation of Jabel Al-Doruz in the north (a physical
boundary) defined the constant head where flow comes
from this side. No flow boundary conditions are in the

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

western, northwestern, southwestern, northeastern and


southeastern parts of the domain. And the variable head
boundary is in the middle.
For the middle aquifer (B2/A7 formation), the
boundary conditions in the north, south and some

locations at the east and at the west of the domain are


constant head as shown in Figure (3). No flow boundaries
are in the cells where the stream flow lines are not
perpendicular to water level contours.

Figure 2: Steady state flow calibration matching map between calculated and
measured water levels for the upper aquifer.
Steady State Calibration
Steady state calibration for the flow model was
achieved by comparing the hydraulic heads obtained from
available groundwater level contour maps of the first and
second model layer and the calculated hydraulic heads of
the MODFLOW simulation in order to simulate the flow

lines of these layers. During calibration, horizontal and


vertical hydraulic conductivities and recharge values
were adjusted in sequential model runs to match the
simulated heads and measured head. The comparison
between measured and simulated water levels for the
upper and middle aquifers is shown in Figures (2) and

- 159 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

(3), respectively.
As shown in Figure (4), the horizontal hydraulic
conductivity values for the upper aquifer ranged from
0.10-90.00 m/ day for the first model layer (B4/B5),

8.4E-5 m/ day for aquitard (B3) formation; and the


hydraulic conductivity values ranged from 0.001-30.00
m/ day for the middle aquifer (B2/A7) formations as
Figure (5) shows.

Figure 3: Steady state flow calibration matching map between calculated and
measured water levels for the middle aquifer.
Transient Calibration
For transient simulation, the time was divided into
fifteen-stress period. The first one stared from the
beginning of the 1970s to the end of 1984 with fifteen
time steps. This period represented the steady state
period where there was no significant drawdown in the
water level. The second starts from the beginning of
1985 till the end of that year with one time step. And the

- 160 -

same will be for the next thirteen-stress periods with


one time step for each. The data from 1970 to the end of
1994 were used for transient calibration and from 1995
to 1998 for verification. About 7 MCM was discharged
each year for 1994 until 1998 from the upper aquifer
through illegal wells. Abstraction from the two aquifers
are only available from 1981 as total abstract without
specifying the location of abstraction wells, so wells

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

discharge was randomly distributed. In the study area


there are only four monitoring wells in the (B4/B5)
formation. And there is no monitoring well for the
middle aquifer. The available data were used to
calibrate the transient state of the model. The first step
of the calibration was to assign an initial value for
storage coefficients and specific yield for each model

layer. These values were taken from previous studies.


Calibration is done using trial and error procedure by
changing the specific yield, storage coefficient, and with
very limited range in the hydraulic conductivity values.
Good performance of the model was observed through
the transient simulation to fit between simulated and
observed drawdown in the observation wells.

Figure 4: Calibrated hydraulic conductivity for the upper aquifer.


Model Verification
Verification of the code was carried out to establish
greater confidence in the model by using the set of
calibrated parameter values and stresses to reproduce a
second set of field data. Figures (6) to (9) present a
comparison between the observed and the calculated
drawdown in the observation wells F1022, F1043, F1060
and F1280, respectively. The results show good
agreement between the observed and calculated
drawdown for the government wells, while there is

noticeable difference between the observed and


calculated drawdown in the farms wells. This conforms to
the existence of illegal wells responsible for unrecorded
discharge in the farm area.

- 161 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

Figure 5: Calibrated hydraulic conductivity for the middle aquifer.


SOLUTE TANSPORT MODEL
Mass transport of total dissolved solids has been
investigated by using MT3D, which is a computer model
for simulation of advection, dispersion and chemical
reactions of soluble contaminants in three-dimensional
groundwater flow systems. Concentrations of the total
dissolved solids are presented herein in terms of EC.
Different parameters including EC, recharge, model
boundaries, particle tracking algorithm and advection
parameters and methods were adjusted to run the model
by using methods of characteristics (MOD). Figure (10)
shows the initial values of EC s/ cm at the beginning of
1985 for the upper aquifer system (JCE, 1997). The
highest EC concentrations were found in the well field

- 162 -

area in the center and the north of the domain.


Transient simulation as well as flow calibration begin
with the steady-state initial conditions with the same
boundaries, parameters and methods of advection and ends
before or when a new steady state is reached. Both the time
and the stress period were divided into several steps. The
first step started from the beginning of the 1970s till the
end of 1984 with fifteen time steps. This represents the
steady state period where there was no significant change
in EC concentrations. While the simulation of transient
conditions started in 1985 when a significant drawdown in
water level occurred and caused an increase in the values
of EC. Calibration was achieved by adjusting the advection
parameters, and the amounts of recharge were also
changed during the process of calibration.

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

Figure 6: Comparison of observed and simulated draw for well F1022


(Farm well area).
MODEL PREDICTIONS
Model predictions have been conducted in order to
evaluate the response of the model for five future
scenarios. These scenarios vary in terms of pumpage rates
for the different operating well groups in the basin and
were developed jointly by the staff of the Ministry of
Water and Irrigations as follows: the first scenario
assumes that the current withdrawal rate of 57 MCM/
year continues to be the same during the simulation
period that starts in 2005 till 2020; the second scenario
explores the effect of reducing the pumping rate by 50%
(annual withdrawal rate 28.5 MCM) on the drawdown
and EC values; in the third scenario, the effects of
increasing all the pumping rates by 50% (annual
withdrawal rate 85.5 MCM) during the period of 2005 to
2020 are examined. This situation might happen if the
Azraq Basin is considered as an alternative for other
water resources; the fourth scenario assumes that all the
pumping rates of public wells have been reduced by 50%
and pumping rates of private wells have been kept the
same as in the year 2000 during the period of 2005 to

2020; the fifth scenario investigates the effects of


reducing all the pumping rates of private wells by 50%
while maintaining the public wells pumping rates at those
of the year 2000 for the period of 2005 to 2020.
Simulation results of the first scenario, in which a
total pumping rate of 57MCM is maintained during the
period from 2005 till 2020, are summarized in Table (1).
As shown in this table, the maximum drawdown
increased from 25.7m in year 2005 to 32.4m in year 2020
while the values of EC slightly varied for the same period
(from 1882 to 1905 s). The maximum drawdown values
for all the scenarios considered are given in Table (2).
These indicate that the first and fourth scenarios have
similar effect on drawdown. Also, the second and fifth
scenarios have similar effects and provide the lowest
drawdown values. Among the five scenarios considered,
the third scenario has the worst effect on drawdown
values and especially for the year 2020. Figure (11)
shows the simulated drawdown in the year 2020 for the
upper aquifer (scenario one). Figure (12) shows the
simulated values of EC in the year 2020 for the upper
aquifer (scenario three).

- 163 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

Table 1: Prediction results for the flow and transport model with the same current
discharge 57 MCM/year (The first scenario).
Year
Observation well
Max.
F1022
F1043
F1060
F1280
Draw down (m)
25.7
5.3
13.6
11.1
11.7
2005
EC (s/cm)
1882
1069
853
1776
820
Draw down (m)
28.0
6.2
14.7
12.9
12.8
2010
EC (s/cm)
1886
1099
867
1781
924
Draw down (m)
32.4
7.8
16.8
15.9
15
2020
EC (s/cm)
1905
1122
873
1804
929

Year
2005

Table 2: Maximum simulated drawdown values for the five scenarios.


Max.
Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5
Drown down (m)
25.7
20.3
30.6
25.8
20.2

2010

Drown down (m)

28.0

20.5

35.6

28.0

20.5

2020

Drown down (m)

32.4

20.8

43.7

32.4

20.8

The EC values for the year 2020 are shown for the first
and third scenarios in Figures (13) and (14), respectively.
There is no significant change in the values of EC through
the different scenarios for the same year.

shows high sensitivity to specific yield at lower values


rather than at higher ones. On the other hand, the transport
model showed no response for the change of any parameter
except for very long period of discharge.

SENSITIVITY ANALYSIS

CONCLUSION

Sensitivity analysis was conducted herein to quantify


the uncertainty in the calibrated model caused by
uncertainty in the estimates of hydraulic conductivity,
specific storage and the typical values of layer thickness,
initial head and recharge. These parameters were
systematically changed between -25% to +25% with
increments of 5% from its base value to determine what
effect these changes have on the predicted drawdown
values and EC. Results indicated that for steady state
conditions, the model is slightly more sensitive to changes
in horizontal hydraulic conductivity than to changes in
recharge values and not sensitive to specific yield. For
transient conditions, it is not sensitive to horizontal
hydraulic conductivity, slightly sensitive to recharge and

The models used in this study, Modflow (PM5) and


MT3D, provided an effective tool for managing Azraq
Basin by evaluating the effect of the different alternatives
under consideration by the Ministry of Water and
Irrigation in Jordan. Simulation results indicate that
increasing the current pumping rates by 50% caused the
maximum drawdown and should be avoided. The first
and fourth scenarios have similar effect on drawdown.
Also, the second and fifth scenarios have similar effects
and provide the lowest drawdown values. The effect of
the different scenarios on the values of EC is less
profound than the effects on the drawdown values. The
third scenario caused a slight increase in EC values over
the values simulated by other scenarios.

- 164 -

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

Figure 7: Comparison of observed and simulated drawdown for well F1043 (Governmental well area).

Figure 8: Comparison of observed and simulated drawdown for well F1060 (Farm well area).

- 165 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

Figure 9: Comparison of observed and simulated drawdown for well F1280


(Governmental well area).

- 166 -

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

Figure 10: Measured iso-EC (S/cm) for the upper aquifer system
(JCE, 1997).

- 167 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

Figure 11: Simulated drawdown in the year 2020


for the upper aquifer (scenario one).

- 168 -

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

Figure 12: Simulated values of EC in the year 2020


for the upper aquifer (Third scenario).

- 169 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

Figure 13: EC Values in the year 2020 for scenario one.

- 170 -

Jordan Journal of Civil Engineering, Volume 1, No. 2, 2007

Figure 14: EC Values in the year 2020 for scenario three.

REFERENCES
Agrar, Hydrotechnik and GTZ. 1977. National Water Master
Plan of Jordan. 8 Volume, Essen-Hanover.
Al-Hadidi K. and Subah, A. 2001. Jordan Badia Research
and Development Program: Integrated Studies of Azraq
Basin for Optimum Utilization of the Natural

Resources, Water Group, Vol. 4. Unpublished Report,


the Higher Council for Science and Technology,
Amman, Jordan.
Al-Kharabsheh. 1991. Hydrogeological and Hydrochemical
Study of the Upper Aquifer System in Azraq Basin,
Master Thesis, University of Yarmouk, Irbid, Jordan.
Al-Momani, M. 1993. Environmental Isotope and

- 171 -

Using Modflow and MT3D

Wail Abu-El-Shar and Randa I. Hatamleh

Hydrochemical Study of the Shallow and Deep


Groundwater in the Azraq Basin, Jordan, IAEA.
Al-Waheidi, M. 1990. Geoelectrical Resistivity of the
Central Part of Azraq Basin. M.Sc. Thesis, University of
Jordan, Amman, Jordan.
Ayed, R. 1996. Hydrology and Hydrogeology Study of the
Azraq Basin, Jordan, Ph.D. Thesis, Department of
Geology, University of Baghdad, Iraq.
Hatamleh, R. 2002. Using Modflow and MT3D
Groundwater Flow and Transport Model as a
Management Tool for the Azraq Groundwater System,
Master Thesis, Civil Engineering Department, Jordan
University of Science and Technology, Irbid, Jordan.
Humphry and Sons. 1978. Water Use Strategy, North
Jordan, Water Resources, London, Amman, 2.
Humphrys and Sons. 1982. Azraq Wellfield Evaluation:
Hydrochemistry and Monitoring, London-Amman.
Jordanian Consulting Engineers 1997. Water Quality and
Transport Model in the Azraq Basin. Final Report,
Prepared for the Water Authority of Jordan, Amman,
Jordan.

- 172 -

Moqbel, S. 2004. Modeling Groundwater Flow and Solute


Transport at Azraq Basin Using ParFlow and Slim-Fast,
Master Thesis, Civil Engineering Department, Jordan
University of Science and Technology, Irbid, Jordan.
Rihani, J. 2003. Application of the High Performance
Computing Techniques of ParFlow Simulator to Model
Groundwater Flow at Azarq Basin, Master Thesis, Civil
Engineering Department, Jordan University of Science
and Technology, Irbid, Jordan.
Rimawi, O. 1985. Hydrochemistry and Isotope Hydrology of
the Ground and Surface Water in North Jordan NE of
Mafraq, Dhuleil, Hallabat, Azraq Basin. Ph.D.
Dissertation, Tech. University of Munich, Munich,
Germany.
UNDP-Azraq Oasis Conservation Project .1996. Simulation
of Groundwater Flow in the Azraq Basin. Studies on the
Water Resources Sub-Project, Amman, Jordan.
Worzyk, P. H. and M. Huser. 1987. Geoelectrical Survey in
the Azraq Area of Northeast Jordan. Federal Institute for
Geosciences and Natural Resources, Federal Republic of
Germany, Hannover.

You might also like