2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Original Paper

Arabian Journal of Geosciences

February 2015, Volume 8, Issue 2, pp 659­675

First online: 28 December 2013

Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate, Egypt

Abdulaziz M. Abdulaziz , Abdalla M. Faid

10.1007/s12517­013­1199­4

Copyright information

Abstract

Siwa represents the last virgin oasis in the western desert of Egypt. Recently, serious environmental changes pertaining to the invaluable groundwater resources, such soil salinity and expansion in surface lakes have developed due to excessive uncontrolled groundwater discharge associating land development for agriculture. The present work tackles these problems through monitoring the configuration of pressure head in carbonate and Nubian Sandstone aquifers using multilayer groundwater model. Several scenarios for pumping stress are tested, and the results indicated that the optimum pumping should be close to 520,000 m3/day with important disturbances in the pressure head encountered between Bahei El­Din Lake and Zeitoun Lake. This aquifer stress is capable of lowering the pressure head to stop artesian flow and inconsequence saves large water quantities draining daily to the lakes through natural flow and mitigates the waterlogging problems. In addition, minimal changes are observed in the eastern part of the modeled area suggesting additional production wells to tap the aquifer system at this barren area and initiate new development projects. Such results demonstrate the potential of groundwater flow modeling in water resources management to define the optimum pumping scenarios capable to mitigate environmental problems.

Keywords

3D Groundwater modeling Groundwater management Arid regions Siwa Oasis Egypt

Introduction

Over the past century, the total population of Egypt increased from 11 million in 1907 to 73.4 million in 2004, while the area of cultivated land has only increased from 2.25 million to around 3.5 million ha (Abdulaziz et al. 2009). As a consequence, the area of land per capita has fallen from 0.2 to 0.05 ha during the same period (FAO 2005). To accommodate the increasing demands for food, attentions are usually paid to reclaiming the desert, but the success of such agricultural projects is entirely restricted to the availability of sustainable water resources. This is probably the reason behind the development programs of the “New Valley Project” and the “South Egypt Development Project,” great projects of land reclamation around the oases of the Western Desert and south Egypt respectively.

While the agricultural land reclamation of the New Valley Project initiated in 1960 is based exclusively on groundwater from the Nubian Sandstone aquifer, the south Egypt development project commenced in 1995 is based on the conjugated use of groundwater from the same aquifer and surface water pumped from Nasser Lake (Ebraheem et al. 2004). Nubian Sandstone is Cambrian to Late Cretaceous and is predominantly continental sediments extending over 2,000,000 km2 underneath the eastern Sahara with confinement condition prevailing to the north of latitude 25° N due to the development of the post­Cretaceous dense limestone, shale, and clay. Groundwater in the Nubian Sandstone aquifer,

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 1/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

hydrologically unsteady aquifer over the past thousands of years, is known to be fossil water and its abstraction is unrenewable (Hellstrom 1940; Pallas 1980; Heinl and Thorweihe 1993; Ebraheem et al. 2002; Gossel et al. 2004). The calculated in­place groundwater of the Nubian Sandstone aquifer is arguable which generally tends to increase over time: 3,000 km3 (Ambroggi 1966), 20,000 km3 (Gischler 1976), 50,000 km3 (Heinl and Thorweihe 1993), and 28,000 km3 (Ebraheem et al. 2002). But generally only a small portion of these volumes can be suitable for exploitation. Isotopic studies on groundwater supported these implications and asserted the deficit of recent recharge to the Nubian Sandstone aquifer in Egypt. The estimated age of groundwater from the Nubian Sandstone aquifer has broadly varied with the different techniques reporting 10,000–33,000 years (Abdelghafour 1993) and 25,000–40,000 years (Heinl and Brinkmann 1989) using Carbon­14, but much older values (200,000 years–1 Ma; Sturchio et al. 2004) are estimated using Krypton­81 and Chlorine­36 techniques. The recharge to the fractured carbonate aquifer underneath Siwa Oasis is largely disputable between local recharge from the deep Nubian Sandstone aquifer and surface recharge at a faraway catchment area near Gebel Akhdar, Libya (El­Shazly and Abdel­Mogheeth 1991). Accordingly, several pre­development studies are devoted mainly for evaluating the groundwater potential and provide detailed characterization of groundwater quality and quantity in these regions (Ball 1927; Diab 1972; Ezzat et al. 1962; Ezzat 1974, 1976; Hesse et al. 1987; Nour 1996; MPWWR 1998).

Large­scale land development usually associates serious ecological and socioeconomic problems, such as waterlogging phenomenon (Masoud and Koike 2004), rapid falling water level and groundwater depletion (Nour 1996; Konikow and Kendy 2005; Venot and Molle 2008), salt water intrusion (Kashef 1983; Ebraheem et al. 1997; Werner and Simmons 2009), and disturbance to the overall groundwater system (Vrba and Pêkný 1991; Abdulaziz 2007; Ahmed et al. 2012). The situation becomes worse if high natural discharge of poor­quality groundwater through natural springs and/or uncontrolled dug wells is involved, which is recognized in Siwa Oasis. Water management is a decisive action with numerous goals that may be partly conflicting to maintain and improve the state of water resources (Pahl­Wostl 2007). The principle objectives of water resources management and plans are to grant the increasing water demands for different uses in a most environmentally effective, socially acceptable, and economically efficient manner (Biswas 2004).

Recently, groundwater flow models became the most influential and effective tools that help water resource planners making water plans and depicting management policies through time series with different scenarios using transient simulations. However, managing such transient simulations involves large transient data sets including well pumping data, recharge data, and observation data (Abdulaziz 2007). In addition, Anderson and Woessner (1992) indicated that specifying storage characteristics of the hydrostratigraphic units, influences of the initial conditions and model boundaries, propagation of hydraulic stresses to reach the model boundaries, and discretization in spatial and temporal domains add inherently more complications to transient simulations. Generally, the key factors that influence the model validity involve the selection of boundary conditions that properly simulate the common steady­state calibration and assigning the appropriate time steps with minimal influences on the numerical outputs. Finally, the use of model­ generated head values ensures the initial head data and consistency of input hydrologic parameters (Franke et al. 1987).

Siwa represents the smallest oasis located in the Egyptian part of the extensive Libyan Desert and depends exclusively on groundwater resources and drainage water reuse. Recently, farmers started to experience the challenging rising water level in the soil zone together with the groundwater salinity and the escorted waterlogging and soil salinization, especially in the topographically low lands at the proximity of the drainage lakes. To evaluate the adverse environmental impact of such excessive groundwater, the Desert Research Center has initiated a program to periodically monitor the water levels and electric conductivity of groundwater in 60 piezometers tapping the shallow aquifer and soil zone (DRC 1988). A second program sponsored by Research Institute for Groundwater (RIGW) in 2001 carried out extensive hydrogeologic and hydrochemical investigations and surveyed 1,265 piezometers throughout Siwa Oasis to determine wells with excessive discharge and evaluate the variations in chemical composition (Sakr et al. 1999). These studies indicated that more than 50 % of the naturally flowing groundwater through natural springs or wells tapping the confined aquifers is dispensed into salt ponds through the poor drainage system. Through an action plan over the past few years, 80,000 m3/day discharged through 180 critical well was managed and terminated to save approximately 40 % of the disposed groundwater (El­ Hossary 1999). In the present study, the main objective is to develop a multilayer three­dimensional groundwater flow model to Siwa Oasis based on the thorough conceptual understanding of the aquifer system, the established monitoring network, and the available well data. Such a model enables the prediction of future aquifer repercussions to different pumping stresses and the potential of the groundwater system over a relatively long period of time, 50 years. This information is extremely valuable in evaluating the available groundwater resources and demonstrates the significances groundwater modeling applications.

Site description

Siwa Oasis occupies an EW elongated depression in the northwest part of the western, 300 km from Mersa Matruh, between latitudes 29°05′00″ N and 29°25′00″ N and longitudes 25°05′00″ E and 26°06′00″ E, with an area of approximately 1,200 km2 (Fig. 1). Topographically, five local depressions are easily recognized in Siwa Oasis, the area lying below the zero elevation contours, and vary in altitude between ∼1 and −18 m above the mean sea level (a.m.s.l.) (Fig. 2). These depressions host the important lakes recognized from west to east as Maraqi, Siwa, Aghourmy, and Zeitoun lakes. The climate is typical arid to semi­arid with a negligible rainfall, high evaporation rate, and moderate to high humidity (Parsons 1963). Siwa was populated in historic times and has increased from 3,000 in 1840 to approximately 23,000 in 2009.

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 2/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 1

Location map of Siwa area

Fig. 2

Geomorphological map of the important features developed in Siwa Oasis and the surrounding area.

As a part of the Western Desert, the climate in Siwa Oasis is very hot in summer and mild in winter with average maximum–minimum temperature between 20 and −5 °C in January and 38 and 21 °C in July (EMA 2012). Rainfall is scarce with an average annual rainfall of 13 mm, but humidity is relatively high ranges from 22 % in May to 45 % in December depending on the associating daily evaporation rate (average 17 mm in June–5.2 mm in December). The people of Siwa involve a mixture of Berber, Bedouin, and Sudanese races and have their own local language, a Berber dialect that is unrelated to Arabic language. The important activity in Siwa is farming that showed a surplus increase from 2,000 acres in 1962 to 12,000 acres in 2012 and the main crops are dates and olives.

At the environs of Siwa Oasis, the Middle Eocene chalky limestone (75 m) is exposed and overlain by quartzitic gravel and silisified wood west of Timeira (Fig. 2). In subsurface (at Qattara, well located in the west of Qattara depression close to Siwa Oasis, Fig. 1), the Moghra Formation, a clastic fluviomarine delta­front sequence of Early Miocene that grades

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 3/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

laterally to marine facies (Said 1990), unconformably overlies the Upper Eocene (Fig. 3). At the depression, a 94­m­thick Marmarica Formation of the Middle Miocene forms the greater part of Siwa Oasis and comprises mainly limestone, dolomite, and shale (Fig. 2). It mainly forms the northern scarp (78 m height) and many of the hills at Gebel El­Dakrour, Mortazak, Zomag, El­Mawta, and Khameisa (Fig. 2) (Gindy and El Askary 1969). Tertiary rocks are covered by the Quaternary alluvium and eolian deposits that constitute 2–3 m soil zone that is replaced by salt or sabkha at the proximity of the lakes. Siwa Oasis occupies a regional NNW–SSE synclinal fold (El Shazly et al. 1978) and is characterized by well­developed NW–SE and ENE–WSW structural lineaments. The tectonic evolution in Siwa area indicates a complicated geologic history of uplift and subsidence with developed folds, horsts, and grabens (Fig. 4) (Said 1960; Rizkalla 1975; Rizkalla and Awad 1990). This results in remarkable variations in formation thickness.

Fig. 3

Composite stratigraphic section in Siwa basin (modified from EGPC 1992)

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 4/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 4

Composite geological cross section from west to east across Siwa basin; scale is approximated

Water wells and springs are the only water supply available in Siwa Oasis with the majority tapping the shallow carbonate aquifer at depths between less than 10 and 120 m and few deep wells (±1,000 m) that pump a total of 400,000 m3/day to meet the water demands for agriculture and municipal uses. Most of these wells, especially the shallow and the hand­dug wells, are poorly designed and lack to casing and discharge valves to control the artesian flow. Given that irrigation wells and springs are usually supplemented by water storage tanks to store the irrigation water of salinity between 1,200 and 7,000 ppm, it is common to encounter a localized waterlogging nearby these tanks at each farm.

Material and methods

In the present study, a three­dimensional finite difference model was developed to simulate groundwater flow in the Nubian Sandstone and carbonate aquifers underneath Siwa Oasis. MODFLOW 2000 (Harbaugh et al. 2000) was selected to numerically solve the governing flow equation based on water balance with a fully implicit finite different approximation (Wang and Anderson 1982). All calculations and modeling processes were accomplished using GMS 6.5 software (Aquaveo 2007). To develop a calibrated groundwater flow model that simulates the flow through the present complex stratigraphy of the study area, the methodology presented in Fig. 5 was applied.

Fig. 5

The steps applied in groundwater model construction in Siwa Oasis

Building the three­dimensional stratigraphic model

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 5/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

The deep wells (e.g., Siwa 1, Zeitoun, Desouki, and Kohla) disclosed successions of Paleozoic, Mesozoic, and Cenozoic sediments (Fig. 3). The Paleozoic sediments involve alternation of sandstone, shale, and dolomite developed by epicontinental and marine sedimentation. The Hercynian unconformity inhabits Paleozoic–Mesozoic contact (El­Sweify 1975) that is easily identified in most deep wells. The Mesozoic rocks (Alam EI­Bueib and Kharita formations) comprise sandstone with siltstone and shale with carbonate intercalations deposited in a shallow marine environment (Fig. 3). In Egypt, the clastic fluviomarine Lower Cretaceous sediments between the crystalline basement and the Upper Cenomanian carbonates are known as the Nubian Sandstone (GPC (General Petroleum Company) 1991). Beneath Siwa, this sandstone is 2,500–3,000 m thick and acts as a single groundwater system that is usually encountered 500–700 m below ground surface (Fig. 4). The upper section of this system is characterized as Bahariya Formation (Norton 1967) and represents the main freshwater provider in the eastern part of Siwa Oasis as it is completely eroded in the western part (Figs. 3 and 4). A sequence of 500–600 m thick of limestone, shales, and marls, with some abundant salt and gypsum of Upper Cretaceous (Abu Roash and Khoman formations) and Tertiary (Apollonia formation), superimposes the Nubian Sandstone (Figs. 3 and 4). These rocks are covered by 2 to 3 m of clay, providing soils for agriculture, and eolian sand of the Great Sand Sea located to the south of the study area.

Mapping stratigraphy and allocating the hydraulic parameters through the aquifer system are achieved using the available borehole data. The relationship between the hydrogeological units of 80 shallow and 13 deep boreholes selected in the study area (Fig. 6) was first checked to confirm a relative spatial continuation of the layers in the spatial domain. The hydraulic conductivity and other hydraulic parameters were determined from either borehole samples or pumping tests published in literature (Anderson and Woessner 1992; Morris and Johnson 1967; Domenico 1972) and unpublished reports (RIGW 1997; Sakr et al. 1999). The resulting solids of the study area were divided vertically into 18 layers, with discretizing each layer to 3,600 cells of square mesh 1,000 × 1,000 m2 each and occupying a space of x 337,450∼417,450 m, y 3,216,270∼3,261,270 m, z 56∼−2,000 m representing the aquifer system together with an E–W cross section through these solids as shown in Fig. 7. As shown in Fig. 7, three main aquifers are identified in the constructed hydrostratigraphic model, called solid. The upper shallow clastic aquifer comprises 0∼70 m thick of loose sand and gravel sediments that locally cover the middle­southern part during the Quaternary and generally possess little hydrogeological influences/importance on the aquifer system of the area. The middle fissure carbonate aquifer is made of hard limestone with intercalations of shale and clay of the Tertiary age and average thickness 500 m that provide the major share of natural surface groundwater discharge. The lower aquifer is made of Nubian Sandstone that comprises a great thickness (500–700 m thick) of highly permeable sandstone of pre­Tertiary in the modeled area. The carbonate aquifer is partially covered by a relatively thin clay layer, layer 1, and is simulated by four layers, named 2, 5, 7, and 9. Several discontinuous marl (layer 3) and anhydrite (layers 4 and 8) strata together with a relatively continuous lower clay layer (layer 6) are mapped in three dimensions through the carbonate aquifer. The Nubian Sandstone aquifer comes in contact with the carbonate aquifer and is simulated in nine layers that extend from layer 10 to layer 18 (Table 1) to minimize the effect of the lower boundary of the model on the simulation results through the multilayer infinite thickness.

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 6/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 6

3D view of the 13 deep wells together with their lithological succession (upper) used in building solids of Siwa model and map view (lower) showing their spatial distribution (black circles). The blue hexagons in the lower map represent the location of the six monitoring stations

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 7/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 7

3D view of the gridded solid (lower) and east–west cross section (upper) showing the complex distribution of the different units in Siwa model. The y­axis follows the north direction; figures are not to scale

Table 1

The hydraulic parameters for the different rock units obtained during the final optimizations of the static model in Siwa area

Hydraulic conductivity Level (m) (m/day) Model Specific Specific Lithology layers yield storage (m−1)

Max Min Horizontal Vertical

1 Upper clay 42 −113 0.00015 0.00006 0.02 0.0125

2 Limestone 50 −282 0.8 to 7.2 0.73 0.24 0.000036

3 Marl −55 −398 0.25 0.0001 0.04 0.00016

4 Anhydrite −136 −449 0.00000001 0.0000000001 0 Negligible

5 Limestone −19 −610 0.8 to 8 0.73 0.23 0.000042

6 Lower clay −57 −728 0.00015 0.00006 0.02 Negligible

7 Limestone −74 −933 0.8 to 9 0.73 0.21 0.000047

8 Anhydrite −230 −1,031 0.00000001 0.0000000001 0 Negligible

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 8/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

8 Anhydrite −230 −1,031 0.00000001 0.0000000001 0 Negligible

9 Limestone −530 −1,235 0.8 to 7.8 0.73 0.21 0.000047

10:18 Sandstone −531 −2,100 0.41 to 6 0.22 0.19 0.000069

Leakance (in meters) 0.016; average evaporation (in meters per day) 0.012; average precipitation (in meters per day) 0.00004

Building the conceptual model

In conceptual models, all model parameters are designed as sources and sinks, boundary conditions, or observation points in the map module using the GIS tools. A basic assumption in Siwa model considers insignificant water level changes at the model boundary, particularly the deeper Nubian Sandstone aquifer system, during the study period. This can be simply acknowledged if the local flow system was considered as a part of the regional flow system (∼2 million km2 areal extension and more than 400 m thick) extending throughout the western desert of Egypt and southeast Libya to the north and northeast Chad and part of north Sudan to the south. In addition, the high flux springs and the gushing artesian flow of drilled shallow and/or deep wells (400,000 m3/day) with approximately no change in the pressure head indicate steady influx through the aquifer system in Siwa area. Further, the deepest parts of freshwater bearing Nubian Sandstone are encountered underneath Siwa Oasis while one of the highest levels in the subsurface is found in Kharga Oasis (Fig. 1), indicating that the groundwater head in study area is located down­gradient to locations of the potential groundwater abstraction located south and southwest. These conditions together with salt–freshwater contact on Nubian Sandstone aquifer located in lower levels to the north of Siwa Oasis (Fig. 1) provide a shield protecting the groundwater head in Siwa Oasis from significant changes. Therefore, all model boundaries were assigned as specified head boundary in which unlimited volume of water inters or leaves the system based only on hydraulic properties and aquifer stress. It should be clear that the results of this model depend strongly on the validity of the specified head assumption, and head perturbations would be encountered if this assumption is violated with degree of perturbations correlatable to the deviation from the specified head. Published research (Brinkmann et al. 1987; Ball 1927; Said 1990; Ebraheem et al. 2002, 2003, 2004), unpublished reports (Ezzat et al. 1962, 1974; Pallas 1991; RIGW 1997; Sakr et al. 1999), and field measurements were utilized to determine groundwater level at more than 75 points at model boundaries in the years of 2000 and 2010.

Faults located in the Nubian Sandstone and carbonate aquifers are represented by wall boundaries (vertical flow conduits). The important lakes in Siwa such as Aghourmy, Bahei El­Din, and Zeitoun lakes were mapped as a general head with lake bed conductance (0.0001 and 0.003 m2/day) to swiftly manipulate the variations in hydraulic head. Due to high evaporation and scarce precipitation, recharge to shallow aquifer is considered negligible. Infiltration from agricultural fields was simulated as recharge zones with recharge rate of 0.0000058 to 0.00005 m/day to the shallow aquifer (values are approximated using Abdulaziz 2007). Due to the presence of the surface sticky clay layer covering most of the study area, evaporation from wetlands and sabkha is considered as a part of the aquifer system discharge through wells as most of these features are developed as effluent discharge sites for the artesian springs and wells.

Aquifer discharge from natural springs and wells is simulated by 192 production wells of average daily discharge 400– 500 m3 and 356 production wells with 850–1,000 m3/day. Aquifer discharge locations and quantities were determined based on the inventory datasets delivered by Desert Research Center (DRC 1988) and Research Institute for Groundwater (RIGW 2001). Additional discharge sites and quantities that simulate future abstractions were allocated based on field observations on 2010 and remotes sensing data by Masoud and Koike (2006). The majority of these wells are screened within the upper and middle carbonate aquifer with the few wells targeting the Nubian Sandstone aquifer (RIGW 2001). More than two thirds of the total water production was simulated to abstract the shallow zone of the carbonate aquifer, i.e., wells and springs less than 120 m depth form the surface (RIGW 2001). Additional 100 wells mostly targeting the deeper layer of the carbonate aquifer and the upper layer of the Nubian Sandstone aquifer are placed in the model together with increasing the pumping rate to simulate aquifer stress through the scenarios of the transient model. For model calibration, ten observation wells distributed throughout the study area were constructed with two, five, and three wells tapping the shallow, middle, and deep aquifers, respectively. Groundwater head in the ten observation wells reported in Sakr et al. 1999 and head data recorded in the field measurements were utilized for the 2000 static and 2010 transient model, respectively. An additional 30 monitoring wells located at six monitoring stations were set up in study area to monitor the groundwater head with six wells in the upper shallow aquifer occupying layer 2, 18 wells for the middle carbonate aquifer distributed equally to monitor layers 5, 7, and 9 of the model, and six wells for the deeper Nubian Sandstone aquifer tapping layer 10.

Finally, the conceptual model is exported to the MODFLOW grid, so that all hydraulic data are automatically arranged in arrays coinciding with the grid cells and solvable by MODFLOW algorithm. Accepting two orders of magnitude contrast in hydraulic conductivity of the successive aquifer layers to cause refraction in flow lines (Anderson and Woessner 1992) and sufficient to justify the placement of impermeable boundary (Freeze and Witherspoon 1967), the confinement conditions were automatically assigned, based on variations in hydraulic conductivity of the hydrostratigraphic units.

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 9/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Results

Steady­state model

To ensure a precise model conversion and truthful solution, the simulation input parameters were checked for errors and warnings before model execution. For solving flow equation (Eq. 1), head change and residual criteria for convergence were set to 0.01 m and 0.01 m3/day, respectively, while the maximum iteration was set to 25. ∂ ∂h ∂ ∂h ∂ ∂h ∂h (K ) + (K ) + (K ) − ω = S ∂x xx ∂x ∂y yy ∂y ∂z zz ∂z s ∂t (1) where K xx , K yy , K zz are hydraulic conductivity values (in meters per day) along x­, y­, and z­axis; h is the hydraulic head (in meters); ω is a source and sink term (volumetric flux per unit volume per day); S s is the specific storage (per meter); and t is time (in days). After the first successful run in the unstressed steady state, the resulting wet and dry cells in addition to the erroneous head distribution are fixed during model calibration by adjusting the input parameters until the calculated and observed head in observation points reach acceptable match. Sensible parameters that effectively used in model calibration include hydraulic conductivity, recharge rate, and conductance values. The sets of these parameters were systematically adjusted starting with K values, recharge rates, and finally conductance values, and the model was repeatedly saved and run. Table 1 shows the final input parameters that were used in the calibrated unstressed steady­state model. Figure 8 shows the simulated head distribution in the Nubian Sandstone aquifer as calculated from the calibrated unstressed steady state that represents the conditions of aquifer system in January, 2000. In addition, the calibration indicator bars of 11 observation wells are also presented in Fig. 8 that shows good calibration (a green bar indicates less than 0.5 m difference between calculated and observed groundwater head) throughout the area, but the northern part of the modeled area had intermediate calibration (a yellow bar indicates 0.5–1.5 m difference in groundwater head). The poor calibration indicator at the southwestern boundary (a red bar indicates poor calibration that reports more than 1.5 m difference in head) is encountered due to the skeptic head value reported to this well. Figure 9 presents the plot of the calculated and observed head values for ten observation points; skeptic value was eliminated, used in model calibration.

Fig. 8

Plane view to the distribution of groundwater head in meters (a.m.s.l.) calculated in the steady state for the upper layer of Nubian sandstone, layer 10 of Siwa model. The error bar corresponds to target ±0.5 to a meter interval. The bar is green when lies entirely within the target, yellow when error is less than 200 %, and red when error is greater than 200 %

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 10/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 9

Statistical plot to the observed and calculated head for the observation wells shown in Fig. 8 (symbols indicate individual wells)

Transient state model

Typically starts with a calibrated steady­state model, transient simulations are usually recommended to analyze time­ dependent hydrogeological problems. In the present transient simulation, most of the input parameters of the calibrated unstressed steady model were utilized, and the model was set to convertible confinement conditions, confined or unconfined, based on K value contrast of the alternating model layers (Freeze and Witherspoon 1967). The storage characteristics of hydrostratigraphic units were fine­tuned using the data in Table 1, considering the possible ranges of specific yield values published in literatures (Morris and Johnson 1967; Domenico 1972). The 36 head data of 2010 (13 deep and 23 shallow wells) with 95 % confidence and 10 % head interval were imported as observed transient head. Started with the 2000 steady state and using try and error technique, the transient model was calibrated through fine tuning the hydraulic parameters. The transient model was considered calibrated and represents the hydrogeological setting in Siwa area on 2010 when the difference between the observed and calculated heads was minimal and the statistical parameters of calibration (mean error, standard error, and RMS) approached the values in Table 2. Figure 10 shows a plot of calculated and observed head value for ten observation points at the same sites shown in Fig. 8. Generally, a good match between observed and calculated values except for intermediate match at the proximity of Aghormy Lake is a poor match for an observation well located to the East of Siwa Lake. The head distribution of the calibrated interpretive transient model follows the general trends of the steady state (Fig. 8) with insignificant changes to be presented, and the groundwater flow budgets calculated for the transient state are presented in Table 3. To set up the model for future stress predictions, ten stress periods of 5 years each were assigned throughout the total simulation period with five time steps assigned to each stress period. All model parameters were assumed relatively unchanged during the stress period. Table 2

The statistical parameters for assessment the calibration of transient model

Statistical parameters Value (m)

Mean error −0.185

Mean absolute error 0.42

Root mean square 0.461

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 11/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 10

Plot of the observed and calculated head for the calibrated transient model obtained in the observation points shown in Fig. 8 (symbols indicate individual wells)

Table 3

In­ and out­flow budgets calculated from the 2010 transient model

Budget item In (m3/day) Out (m3/day)

Constant heads 5,996,035 1,059,843

General heads 28 4,533,683

Wells 0 402,537

Total 5,996,063 5,996,063

In − out % discrepancy 2.14E−07

Model prediction and management scenarios

Theoretically, several strategies can be adapted to groundwater development in Siwa region, but the selected strategy should remain a practical methodology (Farid and Tuinhof 1991). Three main groundwater development strategies were developed for evaluation using drawdown in head against groundwater abstraction over 50­year period. These strategies include low (400,000 m3/day), intermediate (500,000 and 600,000 m3/day), and relatively high (800,000 m3/day) pumping stresses. In all scenarios, the additional groundwater withdrawal was extracted from the present wells in addition to 100 supplementary sites distributed throughout the modeled area with the majority placed in the eastern part, especially in the reclamation lands and around the main populated areas.

The first strategy simulates the current situation and adopts the present aquifer discharge, 400,000 m3/day, from both the carbonate and Nubian Sandstone aquifers with the bulk volume abstracted from the carbonate aquifer. The second strategy considers two scenarios: The first involves increasing water extraction gradually to 500,000 m3/day through the modeled time span with some additional wells discharging the Nubian Sandstone aquifer. The second scenario resembles the previous scenario but adopts increasing the total abstraction to 600,000 m3/day. The third strategy predicts the aquifer situation with increasing the pumping rate gradually to 800,000 m3/day and placing the additional wells to tap both the carbonate and Nubian Sandstone aquifers. The decline in the pressure head of the Nubian Sandstone aquifer and limestone aquifer is monitored during the proposed stress periods of the different scenarios at six observation stations distributed throughout the modeled area and the results are presented in Figs. 11 and 12.

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 12/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 11

Plane view to the head distribution in upper carbonate aquifer (layer 5), lower carbonate aquifer (layer 7), and Nubian Sandstone aquifer at the end of the 50­year time span due to different pumping stresses: 400,000 m3/day (a, left), 500,000 m3/day (a, right), 600,000 m3/day (b, left), and 800,000 m3/day (b, right)

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 13/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Fig. 12

Monitoring the decline of pressure head associating the different pumping scenarios in the important layers of the aquifers using six monitoring stations (location is shown in Fig. 6)

Discussion

The calibrated groundwater flow model of Siwa Oasis provides a detailed description to the configuration of the hydrogeological system and spatial distribution of the groundwater level in the consecutive layers of the important aquifers, carbonate and Nubian Sandstone, associating the different proposed scenarios (Fig. 11a, b). The average reported pressure head in the artesian flowing deep wells in Siwa varies between 70 m a.m.s.l. in areas of great aquifer discharge, the area between Bahei El­Din Lake and Zeitoun Lake, and 120 m a.m.s.l. in the south eastern area. Alternatively, the groundwater head in the carbonate aquifer varies in average between 50 and 60 m a.m.s.l. that shows a considerable difference to the pressure head in the Nubian Sandstone aquifer. Such a hydrologic setting favors a dominant local recharge to the carbonate aquifer via the upward leakage from the Nubian Sandstone aquifer through fault plains and the developed fracture system. This may explain the higher TDS in water abstracted from the carbonated aquifer (>3,500 ppm) compared to that reported in the Nubian sandstone aquifer (∼2,000 ppm) through passing across the Eocene sediments where TDS dramatically increases. In addition, the high flux springs developed along major fault plains and tapping the carbonate aquifer (e.g., Abu Shrouf with a natural flow of 9,600 m3/day) introduce an obvious evident to the in situ recharge to carbonate aquifer. Other sources of recharge to the carbonate aquifer are assumed to occur through water influx from surface water runoff during Post­Middle Miocene time (El­Hossary 1999). In fact the contradicting age dating and wide range of ages reported to groundwater from the Nubian aquifer as discussed earlier lead to consider appreciable recharge or

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 14/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

at least mixture with water from fluvial environment dominated at the Post­Miocene time in Egypt. This explains the presence of good quality groundwater overlying a bulk poor quality at the shallow parts of the carbonate aquifer and notable signs of water erosions in carbonates and other surface units such as old channels, dry gorges, and cavities along Siwa­Matruh road (El­Hossary 1999).

The dominant groundwater flow in Siwa area follows WNW direction with two major local flow systems encountered at the vicinity of Maraqi and Siwa lakes. These local flow systems are associated to discontinuity in the hydrostratigraphic units that correlate with the extensional faulted bedrocks, and such setting strongly supports the theory that the area is a structurally controlled tectonic depression (El­Askary 1969; Mohammad et al. 1999). The converging groundwater flow toward the center of Maraqi and Siwa lakes (Fig. 11) depicts higher rates of groundwater effluent to the lakes and indicates zones of high vertical hydraulic conductivity resulting from the intersection of conjugate sets of faults as defined by Masoud and Koike (2006). This hydrogeological setting together with high groundwater discharge through natural springs and/or uncontrolled dug wells may ultimately lead to further extension in the surface area of the lakes; for example, the surface extension of the lake Siwa has increased from 5.82 in 1987 to 25.97 km2 in 2003 (Masoud and Koike 2004). Furthermore, the tectonic history not only controlled the groundwater flow toward the topographic depressions occupied by the lakes but also configured the orientation of the lakes (Knetsch and Yallouze 1955) toward the NW following the normal faults while the coupled slip movements may dissect large lakes into parts such as the NE­trending regional fault dislocating Massir Lake from Zeitoun Lake (Fig. 2) (Masoud and Koike 2006).

Monitoring the ground water levels in shallow and deep wells throughout Siwa Oasis indicated direct hydraulic interaction between the Nubian Sandstone aquifer and the overlying carbonate aquifer (El­Hossary 1999). This stands true for settings at the proximity of high vertical hydraulic conductivity and/or where the aquitards are thin or extinct; otherwise, such interaction seems unclear as observed in evaluating the present model output. The pressure head in the different layers of the steady­state and transient models (Fig. 12) indicated that the head in the upper two layers (layers 2 and 5) of the carbonate aquifer is almost identical, except for the middle part where the separating marl layer is well developed (Fig. 7). Such similarity indicates hydraulic continuity and uniform distribution of the confining pressure through the two layers and therefore can be denoted as the upper carbonate aquifer. Characterizing it as the lower carbonate aquifer, the pressure head distribution in the third carbonate layer (layer 7 of the multilayer model), bounded at the top by the lower clay aquitard and at the bottom by the anhydrite aquifuge, of the carbonate aquifer is distinctly different from that of the fourth carbonate layer (layer 9 of the multilayer model) that occurs in contact with Nubian Sandstone aquifer. Although the fourth layer in the carbonate aquifer represents the stratigraphic extent to the third layer, particularly the area located southeast of Siwa Lake, their head closely resembles that of the Nubian Sandstone aquifer due to their hydrostratigraphic position (Fig. 7). This represents a typical conjugate control of the developed geologic structure and hydrostratigraphy on groundwater head distribution in the multilayer aquifer system of Siwa Oasis.

During different transient model trials, monitoring of pressure head in carbonate aquifer, particularly the upper part, showed a distinct fluctuation associating pumping both the carbonate and Nubian Sandstone aquifers and the fluctuation rhythm appears dependent to the pumping scenarios and the influences of the neighboring hydrologic disturbances (Figs. 11 and 12). This obviously indicates the direct groundwater replenishment of both shallow clastic and carbonate aquifers through the vertical leakage from the deeper Nubian Sandstone that widely extends throughout Egypt and part of Libya. To evaluate the feasibility of various exploitation plans for the groundwater resources in Siwa Oasis, the calibrated transient model is utilized to evaluate the impact of a total discharge of 400,000, 500,000, 600,000, and 800,000 m3/day from both carbonate and Nubian Sandstone aquifers. In addition, the results may help to indicate the optimum pumping rate that controls the natural flow from springs and untapped shallow hand­dug wells. Over 50 years, the transient model run showed insignificant change in groundwater level of the alternating layers throughout Siwa area with ongoing pumping rate, 400,000 m3/day, except for the area at the environ of monitoring station 1 (Fig. 12). The pressure head distribution appears highly sensitive to the vertical hydraulic conductivity, presence and continuity of the aquitard, and the input of boundary conditions (tested only in this scenario to evaluate the effect of constant boundary hypothesis). Similarly, the other scenarios showed minimal effect of pumping on the model layers below the lower carbonate aquifer, and generally, the drawdown did not exceed a meter. However, both the upper and lower carbonate aquifer showed significantly different responses to various pumping stress from the west to the east of the modeled area (Figs. 11 and 12). Spatially, the western part, the area around Bahei El­Din Lake, and middle part, the area between Siwa Lake and Zeitoun Lake, of the model demonstrated the important drawdown in pressure head compared to the eastern part through the different pumping stresses with the significant changes resulting from 800,000 m3/day scenario. The insignificant changes in groundwater level of east Siwa, the area between Massir Lake and Zeitoun Lake, with different aquifer stresses are attributed to the minimal groundwater abstraction associating seldom human activities in this area.

The monitoring programs conducted by RIGW (1996) and local community interview during field work indicated that the naturally flowing shallow wells and springs tapping the carbonate aquifer and located at the environs of active pumping deep wells that tap the Nubian Sandstone aquifer show a gradual reduction in artesian flow followed eventually by complete cease of natural flow. These observations may indicate local interaction between the Nubian Sandstone aquifer and the upper carbonate aquifer that could be facilitated by development of vertical fault systems. El­Hossary 1999 reported that the shallow wells stop natural flow when the drawdown in pressure head approaches 4 m, and the deep wells may behave in correspondence when the pressure head falls to 60 m a.m.s.l. Based on these criteria and according to the current transient model runs, most of the shallow wells in the western part of the study area are expected to discontinue artesian flow while the majority wells in the middle part encounter a considerable reduction in flow rate within the first 10 years if the pumping rate increased to 500,000 m3/day. Alternatively when the total pumping increases to 600,000

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 15/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

and 800,000 m3/day, the simulation results demonstrated drawdown in groundwater head between 4 and 18 m at the western and middle parts of Siwa area which would prevent the natural flow of both shallow wells and dramatically reduce artesian flow from deep wells after 4 months to 2 years, and ultimately, most pumping wells should be supplemented with water pumps. The groundwater level monitored in the eastern part of the Nubian Sandstone and carbonate aquifers did not show important variations, less than 2 m, and appeared stable to a great extent through most pumping scenarios. This is apparently pertinent to the few pumping wells tapping the system that are incompetent to cause important disequilibrium and reconfiguration of pressure head within the model layers, and in consequence, a trivial change in groundwater level has been observed through the tested time span. This indicates the capability of the Nubian Sandstone and carbonate aquifers to supply additional groundwater, despite the additional pumping stress added at this area in the different scenarios, for future land development plans.

Detailed water budget analysis for different layers (Table 4) through the different scenarios indicated that there is more than 320,000–570,000 m3/day of groundwater abstracted the carbonate aquifer, particularly layers 5 and 7 of the upper carbonate aquifer. While considerable volume of the additional pumping stress targeted the lower parts of the model, layers 9 and 10, 25­35% the total stress was assigned to the upper carbonate aquifer. In addition, the simulation results of various pumping stresses indicate that the recommended pumping scenario should fall between 500,000 and 600,000 m3/day and such increase in groundwater abstraction should be gradual. These values are defined by marked fall in groundwater head of the shallow aquifers that considerably reduce the natural flow of the hand­dug wells and springs tapping them (Fig. 12). Accordingly, several trial runs have been tested to define the optimum pumping rate that preserves the groundwater system balance under the current boundary conditions. The model run that best fits the above criteria has adopted a total abstraction of 520,000 m3/day from the carbonate and Nubian Sandstone aquifers. This total abstraction is expected to save large water quantities draining daily to the different lakes and mitigate the waterlogging problems in Siwa Oasis (Sakr et al. 2002). Furthermore, additional production wells should tap the area east of Zeitoun Lake that seems suitable to host new development projects. Finally, it is recommended to continue the monitoring programs of groundwater level and soil salinity to update the model input and apparently test the reliability of the model output. Table 4

Daily pumping rates from the different layers in Siwa aquifer system during the different scenarios

Layer Daily pumping rate scenarios (×1,000 m3/day)

2 40 40 40 40

5 145 150 160 170

7 170 190 220 300

9 20 40 70 100

10 25 80 110 190

Total 400 500 600 800

Conclusion

Simulation of impact of present and future groundwater extraction scenarios from the non­replenished carbonate and Nubian Sandstone Aquifers in Siwa Oasis indicated the presence of local flow systems at the vicinity of Bahei El­Din Lake and Siwa Lake, interrupting the regional WNW groundwater flow system. These local flow systems are tectonically dependent and generally associate zones of significant high vertical hydraulic conductivity that match the development of normal faults. The multilayer model simulations indicated that most shallow wells will cease artesian flow if the total abstraction increases to 500,000 m3/day while phenomenon of natural flow would disappear if the pumping rate increased to 800,000 m3/day. The optimum pumping rate that best fits the aquifer potentiality and controls the artesian flow was found close to 520,000 m3/day from the carbonate and Nubian Sandstone aquifers. The insignificant disturbances in pressure head with various pumping stress in the eastern part of the modeled area are attributed to the few production wells and lack of important human activities. The gleaned information represents the importance of multilayer

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 16/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

groundwater flow models in solving intervening groundwater problems and demarcating the broad portrait of water resources management policy.

Acknowledgments

The authors gratefully acknowledge the financial support of National Authority of Remote Sensing and Space Sciences during the field work. The innumerous RIGW and Desert Research Center employees who collected and processed the field data together with the internal reports are truly appreciated for providing well and pumping data that constitute the foundation of this work.

References

Abdelghafour OH (1993) A three dimensional groundwater flow model for Wadi Howar Area, Darfur Province, Western Sudan. M.S. thesis, Ohio University (unpublished), 127p Abdulaziz AM (2007) Applications of remote sensing, GIS, and groundwater flow modeling in evaluating groundwater resources; two case studies: East Nile Delta, Egypt and Gold Valley, California, USA. Ph.D. dissertation, University of Texas at El Paso Abdulaziz MA, Hurtado JM, Aldouri R (2009) Application of multitemporal Landsat data to monitor land cover changes in the Eastern Nile Delta region, Egypt. Int J Remote Sens 30(11):2977–2996 CrossRef (http://dx.doi.org/10.1080/01431160802558675) Ahmed MA, Abdel Samie SG, Badawy HA (2012) Factors controlling mechanisms of groundwater salinization and hydrogeochemical processes in the Quaternary aquifer of the Eastern Nile Delta, Egypt. Environ Earth Sci 68:369–394. doi:10.1​ 007/s​ 12665­012­1744­6­012­1744­6 (http://dx.doi.org/10.1007/s12665­012­1744­6­012­1744­6) Ambroggi RP (1966) Water under Sahara. Sci Am 214:21–29 CrossRef (http://dx.doi.org/10.1038/scientificamerican0566­21) Anderson MP, Woessner WW (1992) Groundwater modeling simulation of flow and advective transport. Academic, New York, 381 pp Aquaveo (2007) Groundwater modeling system, GMS 6.5­Version: 6.5.3. Aquaveo, Provo. http://w​ ww.a​ quaveo.c​ om (http://www.aquaveo.com/) . Ball J (1927) Problems of the Libyan Desert. Geogr J 70:21–38,105–128, 209–224 Biswas AK (2004) Integrated water resources management: a reassessment. Water Int 29(2):248–256 CrossRef (http://dx.doi.org/10.1080/02508060408691775) Brinkmann PJ, Heinl M, Hollander R, Reich G (1987) Retrospective simulation of groundwater flow and transport in the Nubian Sandstone Aquifer System. Berl Geowiss­Abh (A) 75(2):465–516 GPC (General Petroleum Company) (1991) Salinity study on Pre­Upper Cenomanian (PUC) sediments in Siwa Area. Internal technical report in the General Petroleum Company, Nasr City, Cairo Diab MS (1972) Hydrogeological and hydrochemical studies of the Nubian Sandstone water­bearing complex in some localities in United Arab. Republic. Ph.D. thesis, Geol Dept, Faculty of Science, Assiut University, Egypt Domenico PA (1972) Concepts and models in groundwater hydrology. McGraw­Hill, New York, 405 p Ebraheem AM, Senosy MM, Dahab KA (1997) Geoelectrical and hydrogeochemical studies for delineating ground­water contamination due to salt–water intrusion in the northern part of the Nile Delta, Egypt. Ground Water 35(2):216–222. doi:10.1​ 111/j​ .1​ 745­6584.1​ 997.t​ b03728.x​ (http://dx.doi.org/10.1111/j.1745­6584.1997.tb03728.x) Ebraheem AM, Riad S, Wycisk P, Seif El­Nasr AM (2002) Simulation of impact of present and future groundwater extraction from the non­ replenished Nubian Sandstone Aquifer in southwest Egypt. Environ Geol 43:188–196. doi:10.1​ 007/s​ 00254­002­0643­7 (http://dx.doi.org/10.1007/s00254­002­0643­7) Ebraheem AM, Garamoon HK, Riad S, Wycisk P, Seif El Nasr AM (2003) Numerical modeling of groundwater resource management options in the East Oweinat area, SW Egypt. Environ Geol (2003) 44:433–447. doi:10.1​ 007/s​ 00254­003­0778­1 (http://dx.doi.org/10.1007/s00254­ 003­0778­1) Ebraheem AM, Riad S, Wycisk P, Sefelnasr AM (2004) A local­scale groundwater flow model for groundwater resources management in Dakhla Oasis, SW Egypt. Hydrogeol J 12:714–722 CrossRef (http://dx.doi.org/10.1007/s10040­004­0359­8) EI­Sweify A (1975) Subsurface Paleozoic stratigraphy of Siwa­Faghur area, Western Desert, Egypt_9th Arab petrol. Congo Dubai, paper no. 119 (B­3), 44 p El­Askary MA (1969) Some geological studies on Siwa Depression, Western Desert, UAR. Ph.D. thesis, Faculty of Science, Alexandria University, p 127 El­Hossary MF (1999) Evaluation and management of groundwater resources in Siwa area with emphasis on the Nubia sandstone aquifer. Ph.D. thesis, Faculty of Science, Ain Shams University, p 143 El­Shazly MM, Abdel­Mogheeth SM (1991) Remarks on the water resources and the potential of horizontal agricultural expansion in Siwa Oasis. 1st workshop on rehabilitation and development of Siwa Oasis, 26–29 November 1991, Mersa Matruh, Egypt El­Shazly EM, Abdel­Hady MA, El­Shazly MM (1978) Application of Landsat imagery in the geological and soil investigations in the central Western Desert, Egypt. Int Symp Remote Sens Environ Proc 12(2):857–866 EMA, Egyptian Meteorological Authority (2012) http://w​ ww.e​ ma.g​ ov.e​ g/m​ ap?m​ enu=3​ &l​ ang=e​ n (http://www.ema.gov.eg/map? menu=3&lang=en) Ezzat MA (1974) Groundwater series in the Arab republic of Egypt, exploitation of groundwater in the El­Wadi E1­Gedid project area, parts I– IV. General Desert Development Authority, Ministry of Irrigation, Cairo Ezzat M (1976) Exploration of groundwater, Kharga Oasis. Desert irrigation. Ministry of Irrigation, Cairo Ezzat MA, E1­Badry HM, Ibrahim MM (1962) Preliminary report on the hydrology of the New Valley, Western Desert, Egypt, with special emphasis on Kharga Oasis. Prepared for the General Desert Development Organisation, in co­operation with the U.S. Geol. Surv. And under the auspices of the U.S. Agency of the International Development, (USAID), 181 pp, 40 figs FAO (2005) Fertilizer use by crop in Egypt. Food and Agriculture Organization of the United Nations, Land and Plant Nutrition Management Service, Land and Water Development Division, Rome

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 17/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Farid S, Tuinhof A (1991) Groundwater development planning in the desert fringes of the Nile Delta. Planning for Groundwater Development in Arid & Semi­arid Regions, Round Table Meeting (RTM­91), October 5–9, 1991, Cairo, Egypt: Edited by RIWG/IWACO Cairo–Rotterdam. Franke OL, Reilly TE, Bennett GD (1987) Definition of boundary and initial conditions in the analysis of saturated groundwater flow systems, an introduction. USGS, Techniques of Water Resources Investigations 03­B5, 15 pp Freeze RA, Witherspoon PA (1967) Theoretical analysis of regional groundwater flow: effect of water­table configuration and subsurface permeability variation. Water Resour Res 3:623–634 CrossRef (http://dx.doi.org/10.1029/WR003i002p00623) EGPC (Egyptian General Petroleum Corporation) (1992) Western Desert, oil and gas fields (a comprehensive overview). EGPC, Cairo, 431 pp Gindy AR, El Askary MA (1969) Stratigraphy, structure and origin of the Siwa depression, western desert of Egypt. Bull Am Assoc Petrol Geol 53:603–625 Gischler CE (1976) Present and future trend in water resources development in Arab countries. UNESCO report Gossel W, Ebraheem AM, Wycisk P (2004) A very large scale GIS based groundwater flow model for the Nubian Sandstone aquifer in eastern Sahara Egypt, northern Sudan, and eastern Libya. Hydrol J 12:698–713 Harbaugh AW, Banta ER, Hill MC, McDonald MG (2000) MODFLOW­2000, the US Geological Survey modular groundwater model­user guide to modularization concepts and the ground­water flow process. US Geological Survey Open­File Report 00­92 Heinl M, Brinkmann PJ (1989) A groundwater model of the Nubian aquifer system. Hydrol Sci 34:425–447 CrossRef (http://dx.doi.org/10.1080/02626668909491350) Heinl M, Thorweihe U (1993) Groundwater resources management in SW Egypt. In: Wycisk P, Meissner B (eds) Geopotential ecology, vol 6. Catena, Cremlingen­Destedt, pp 99–121 Hellstrom B (1940) The subterranean water in the Libyan desert. Geografiska Annaler Series A Phys Geogr 21:206–239 CrossRef (http://dx.doi.org/10.2307/519984) Hesse K, Hissne A, Kheir O, Schnäcker E, Schneider M, Thorweihe U (1987) Hydrogeological investigations in the Nubian sandstone aquifer system, eastern Sahara. Berl Geowiss­Abh (A) 75(2):397–464 Kashef AI (1983) Salt–water intrusion in the Nile Delta. Ground Water 21(2):160–167. doi:10.1​ 111/j​ .1​ 745­6584.1​ 983.t​ b00713.x​ (http://dx.doi.org/10.1111/j.1745­6584.1983.tb00713.x) CrossRef (http://dx.doi.org/10.1111/j.1745­6584.1983.tb00713.x) Knetsch G, Yallouze M (1955) Remarks on the origin of the Egyptian oasis­depressions. Bull Soc Geogr Egypt 28:21–33 Konikow LF, Kendy E (2005) Groundwater depletion: a global problem. Hydrogeol J 13(1):317–320. doi:10.1​ 007/s​ 10040­004­0411­8 (http://dx.doi.org/10.1007/s10040­004­0411­8) CrossRef (http://dx.doi.org/10.1007/s10040­004­0411­8) Masoud A, Koike K (2004) Soil salinity induced land cover change detection and analysis in Siwa region, NW Egypt. In: International Symposium on Geoinformatics for spatial Infrastructure Development in Earth and Allied Sciences (GISIDEAS04), Hanoi, Vietnam, 16–18 September 2004 Masoud A, Koike K (2006) Tectonic architecture through Landsat­7 ETM+/SRTM DEM­derived lineaments and relationship to the hydrogeologic setting in Siwa region, NW Egypt. J Afr Earth Sci 45(2006):467–477 CrossRef (http://dx.doi.org/10.1016/j.jafrearsci.2006.04.005) Mohammad MR, Sediek K, El­Sobky H, El­Raey M (1999) Structural analysis and groundwater potentialities using TM Image and field investigations, case study: Siwa region, Western Desert, Egypt. In: Second International Symposium on Operationalization of Remote Sensing, ITC, Enschede, the Netherlands, 16–20 August 1999 Morris DA, Johnson AI (1967) Summary of hydrologic and physical properties of rock and soil materials as analyzed by the Hydrologic Laboratory of the U.S. Geological Survey 1948–1960, USGS, Water Supply Paper 1839­D MPWWR (1998) Southern Egypt Development Project. Internal report, Ministry of Public Work and Water Resources, Cairo Norton P (1967) Rock stratigraphic nomenclature of the Western Desert. Internal report, Pan­American Oil Co., Cairo Nour S (1996) Groundwater potential for irrigation in the east Oweinat area, Western Desert, Egypt. Environ Geol 27:143–154 Pahl­Wostl C (2007) Transitions towards adaptive management of water facing climate and global change. Water Resour Manag 21(1):49–62. doi:10.1​ 007/s​ 11269­006­9040­4­4 (http://dx.doi.org/10.1007/s11269­006­9040­4­4) CrossRef (http://dx.doi.org/10.1007/s11269­006­9040­4) Pallas P (1980) Water resources of the Socialist People’s Libyan Arab Jamahiriya. In: Salem MJ, Busrewil MT (eds) Second Symposium on the Geology of Libya, Vol. 2. The Geology of Libya II, Academic, London, pp 539–594 Pallas P (1991) Deep groundwater resources of Siwa Oasis. FAO report, Rome, October 1991 Parsons RM (1963) Siwa Oasis area, final report, New Valley Project, Western Desert of Egypt, report to Egyptian Desert Development Organisation by The Ralf Parsons Engineering Company, Los Angeles, CA, USA DRC (Desert Research Center) (1988) Monitoring program to the water level of the shallow aquifer and soil zone in Siwa Oasis. Internal report RIGW (Research Institute for Groundwater) (1996) Well inventory in Siwa Oasis. Internal technical report, National Water Research Center, Cairo, Egypt RIGW (Research Institute for Groundwater) (1997) Development Project of the Siwa Oasis. Internal report, National Water Research Center, Cairo, Egypt RIGW (Research Institute for Groundwater) (2001) Southern Egypt development project. Research Institute for Ground Water, internal report, National Water Research Center, Cairo, Egypt. Rizkalla IR (1975) Interpretation of geophysical anomalies in Siwa Oasis, Western Desert, Egypt, for petroleum prospection. Ph.D. thesis, Fac. Sci., Cairo Univ., Egypt Rizkalla IR, Awad BM (1990) Modelling of the basement surface of Siwa­Qattara Depression area, northern western desert of Egypt. Qatar Univ Sci Bull 10:339–362 Said R (1960) New light on the origin of Qattara Depression. Bull Soc Geogr Egypt 33:37–44 Said R (1990) The geology of Egypt. Elsevier, Amsterdam Sakr SA, EI­Hossary MF, Attia FA (1999) Simulation of regional flow in a multilayer aquifer system, Siwa Oasis. Research Institute of Groundwater (RIGW), unpublished report Sakr SA, Gad M, Millette JA (2002) Mitigation options for water logging problems in closed basins: a case study from Siwa Oasis, Egypt. The Third International Conference Groundwater level control in urbanized areas. Mansoura University, Faculty of Engineering, December 17–19 2002.

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 18/19 2/24/2016 Evaluation of the groundwater resources potential of Siwa Oasis using three­dimensional multilayer groundwater flow model, Mersa Matruh Governorate…

Sturchio NC, Du X, Purtschert R, Lehmann BE, Sultan M, Patterson LJ, Lu ZT, Mueller P, Bigler T, Bailey K, O’Connor TP, Young L, Lorenzo R, Becker R, El Alfy Z, El Kaliouby B, Dawood Y, Abdallah AMA (2004) One million year old groundwater in the Sahara revealed by Krypton­ 81 Chlorine­36. Geophys Res Lett 31:1–4 Venot JP, Molle F (2008) Groundwater depletion in the Jordan Highlands: can pricing policies regulate irrigation water use? Water Resour Manag 22(12):1925–1941. doi:10.1​ 007/s​ 11269­008­9260­x (http://dx.doi.org/10.1007/s11269­008­9260­x) CrossRef (http://dx.doi.org/10.1007/s11269­008­9260­x) Vrba J, Pêkný V (1991) Groundwater­quality monitoring—effective method of hydrogeological system pollution prevention. Environ Geol 17(1):9–16. doi:10.1​ 007/B​ F01716070BF01716​ 070 (http://dx.doi.org/10.1007/BF01716070BF01716070) Wang HF, Anderson MP (1982) Introduction to groundwater modeling: finite difference and finite element methods. Freeman and Co, San Francisco, 237 pp (reprinted in 1995 by Academic Press). Werner AD, Simmons CT (2009) Impact of sea­level rise on sea water intrusion in coastal aquifers. Ground Water 47(2):197–204. doi:10.1​ 111/j​ .​ 1745­6584.2​ 008.0​ 0535.x​ (http://dx.doi.org/10.1111/j.1745­6584.2008.00535.x) CrossRef (http://dx.doi.org/10.1111/j.1745­6584.2008.00535.x)

Copyright information

© Saudi Society for Geosciences 2013

http://link.springer.com/article/10.1007/s12517­013­1199­4/fulltext.html 19/19