University of Nevada, Reno

Groundwater Resource Sustainability in Wadi Watir Watershed, Sinai, Egypt

A dissertation submitted in partial fulfillment of the

requirements for the degree of Doctor of Philosophy in

Hydrogeology

by

Mustafa Eissa

Dr. James Thomas/Dissertation Advisor

August, 2012

THE GRADUATE SCHOOL

We recommend that the dissertation prepared under our supervision by

MUSTAFA EISSA

entitled

Groundwater Resource Sustainability in Wadi Watir Watershed, Sinai, Egypt

be accepted in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

James Thomas, Ph.D., Advisor

Greg Pohll, Ph.D., Committee Member

Ronald Hershey, Ph.D., Committee Member

Scott Bassett, Ph.D., Committee Member

Mohamed Gomaa, Ph.D., Committee Member

Birant Ramazan, Ph.D., Graduate School Representative

Marsha H. Read, Ph. D., Associate Dean, Graduate School

August, 2012 i

Abstract

The Wadi Watir delta is a tourist area in the arid southeastern part of the Sinai

Peninsula, Egypt, where development and growth of the community is constrained by the amount of groundwater that can be withdrawn sustainably. To effectively manage groundwater resources in the Wadi Watir delta, the origin of groundwater recharge, groundwater age as related to the timing of groundwater recharge, groundwater chemistry, upwelling of a deep saline groundwater and seawater intrusion along the coast of the delta need to be understood. Mineral identification, rock chemistry, water chemistry, and isotopes (δ2H, δ18O, 14C, δ13C, 87Sr/86Sr, δ37Cl and δ81Br) were used to identify recharge sources, mixing of different waters (including a saline groundwater and seawater), and groundwater age. The chemical evolution of groundwater as it flows from the upland areas in the watershed to the Wadi Watir delta was evaluated using inverse geochemical modeling. The geochemical model NETPATH was used to identify possible geochemical processes that account for the chemical and isotopic changes in groundwater throughout the watershed and in alluvial aquifers of the delta. Geochemical models are presented that identify and quantify the amount of water-rock reactions, mixing of different waters, and evaporative concentration of shallow groundwater. The geochemical and isotopic data in combination with the modeling show that groundwater in the Wadi

Watir watershed is primarily recent recharge within the watershed and that groundwater salinity is controlled by mixing of chemically different waters and dissolution of minerals and salts within the aquifers. The main channel area of Wadi Watir, up gradient from the delta, receives groundwater flow primarily from the El Shiekh Attia area. Groundwater from the El Shiekh Attia and main channel areas is the primary source of groundwater ii supplying alluvial aquifers of the Wadi Watir delta. The Wadi El Ain area, which is also up gradient of the main channel area, does not supply much groundwater to the main channel area based on isotopic data. The same storms that recharge the El Sheikh Attia area also recharge the Wadi El Ain area, but the isotopic signature of the groundwater in the Wadi El Ain area is more depleted in δ18O and δ2H than the El Sheikh Attia area groundwater because of the rainout effect and altitude effect. This isotopically depleted groundwater is not evident in the main channel area or the alluvial aquifers of the Wadi

Watir delta.

Groundwater in alluvial aquifers of the Wadi Watir delta can be divided into two groups, based on their isotopic and chemical compositions. Group I is characterized by relatively depleted δ18O values ranging from -3.82 to -2.58 ‰ as compared to the other groundwater (Group II). Most of the Group I groundwaters are located away from the coast with many of the sites along the mountain front, particularly where the Wadi Watir drainage enters the delta. This groundwater is isotopically similar to groundwater in the upper watershed recharge areas of El Sheikh Attia and the main channel. Most Group II groundwaters are located in shallow wells along the coast and near wetted Sabkha areas.

The δ18O values of these groundwaters range from –0.66 to +6.86‰. The majority of

Group II groundwater has undergone some evaporation as indicated by an isotopic evaporation trend from winter rain and Group I groundwaters to Group II groundwaters.

Two wells located near the coast have elevated ion concentrations and isotopic values that indicate seawater intrusion. Group I groundwaters in the community well field have a component of deep saline water as indicated by δ37Cl and δ81Br. The deep saline water is depleted in δ81Br and enriched in δ37Cl suggesting a marine origin. Over pumping of the iii well field has resulted in upwelling of this deep saline groundwater. Geochemical modeling indicates that 6 to 12% of deep saline groundwater mixes with the well field water and 8 to 12.5% of seawater mixes with groundwater along the coast.

Groundwater level and salinity data were used to develop a three-dimensional, variable-density, groundwater flow and salinity transport model using the SEAWAT modeling environment. The model was developed to estimate average annual groundwater recharge to the alluvial aquifers of the Wadi Watir delta and to simulate annual groundwater pumping, evaluate upwelling of saline water from beneath the well field, and quantify the extent of seawater intrusion along the coast for different pumping scenarios. The model was calibrated using groundwater level and salinity changes.

Modeling results for the period 1982 to 2009 showed that: the extent of seawater invasion along the coast was about 200 m; the main factors controlling groundwater salinity are pumping stresses and the availability of recharge; the daily extraction rate from individual deep drilled wells of the main well field ranged from 200 to 1400 m3/day with an annual average rate of 3100 m3/day for all the wells; and, the estimated annual average recharge to the delta using different modeling scenarios ranges from 4200 to 6000 m3/day.

iv

Acknowledgements

I would like to offer my sincerest gratitude to my advisor Dr. James Thomas who supported me through my dissertation with his patience; this research could not be possible without his encouragement and support.

I am also thanking my Committee:

Thanks to Dr. Greg Pohll for his expert guidance with groundwater modeling. Dr.

Ron Hershey has given valuable advice and insight to my work with geochemical modeling and isotopes. I really appreciate Dr. Scott Bassett and Dr. Birant Ramazan for helping and guiding my research proposal.

This research was sponsored by the Egyptian Cultural and Educational Bureau at

Washington DC. Thanks go to both the George Burke Maxey Fellowship and Dr. James

Thomas for funding my studies.

The Desert Research Institute in Reno, Nevada, USA and the Desert Research

Center in Cairo, Egypt, have provided the support and equipment I needed to produce and complete my dissertation.

I would like to thank Dr. Orfan Shouakar-Stash at the University of Waterloo and

Dr. Simon Poulson at the University of Nevada Reno for conducting isotope analyses.

I really appreciate the help from Drs. Maher Dawoud, Kamal Dahab, Mohammed

Gomaa, Ahsraf Shabana and Abdelfattah El Shiekh for helping with the field work.

Moreover, I will never forget Dr. Clay Cooper, Dr. Alexandra Lutz, Ms. Laura Craig,

Sean Thomas, and Mrs. Debi Noack for helping me with my dissertation.

Finally, I thank God, my parents, as well as my wife and my kids.

v

CONTENTS Abstract ...... i Acknowledgements ...... iv List of Figures ...... vii List of Tables ...... x

CHAPTER 1 ...... 1 1. Introduction ...... 1 2. Hypothesis ...... 3 3. Background ...... 3 3.1 Climatic Conditions...... 4 3.2 Geomorphic and Hydrogeologic Setting ...... 6 4. Groundwater Recharge...... 14 5. Groundwater Chemistry ...... 15 6. Conceptual Model of Groundwater Recharge, Flow, and Salinity Sources ...... 16 7. Objectives ...... 17 8. Approach and Methods ...... 18 9. References ...... 20

CHAPTER 2: Geochemical and Isotopic Evolution of Groundwater in the Wadi Watir Watershed, Sinai Peninsula, Egypt ...... 25 Abstract ...... 26 1. Introduction ...... 27 2. Background ...... 29 2.1 Study Area ...... 29 2.2 Geology ...... 29 3. Methods ...... 32 3.1 Field and Laboratory Methods ...... 32 3.2 Water-Rock Reaction Modeling...... 35 4. Results and Discussion ...... 39 4.1 Groundwater Flow System ...... 39 4.2 Environmental Isotopes ...... 43 vi

4.3 Water-Rock Reaction Models using NETPATH ...... 50 4.4 Groundwater Ages...... 53 5. Summary and Conclusions ...... 56 6. Acknowledgements ...... 57 7. References ...... 57

CHAPTER 3: Groundwater Resource Sustainability in the Wadi Watir Delta, Gulf of Aqaba, Sinai, Egypt ...... 63 Abstract ...... 64 1. Introduction ...... 66 2. Background ...... 68 2.1 Study Area ...... 68 2.2 Geology and Hydrogeology ...... 68 3. Methods ...... 69 3.1 Field and Laboratory Methods ...... 69 3.2 Water-Rock Reaction Modeling...... 74 3.3 Groundwater Flow Modeling ...... 75 4. Results and Discussion ...... 76 4.1 Groundwater Chemistry ...... 76 4.2 Environmental Isotopes ...... 79 4.3 Water-Rock Reaction Modeling...... 86 4.4 Groundwater Ages...... 90 5. Groundwater Flow Model ...... 91 5.1 Model Discretization and Hydrogeologic Parameters ...... 91 5.2 Initial Conditions of the Groundwater Flow Model ...... 94 5.3 Groundwater Withdrawal ...... 94 5.4 Boundary Conditions...... 97 5.5 Model Calibration ...... 97 5.6 Recharge ...... 100 5.7 SEAWAT Model Results ...... 104 6. Summary and Conclusions ...... 109 7. References Cited ...... 110

vii

CHAPTER 4: Groundwater Recharge and Seawater Intrusion of the Quaternary Coastal Plain Aquifer in the Wadi Watir Delta, Sinai, Egypt ...... 117 Abstract ...... 118 1. Background ...... 120 1.2 Geology and Hydrogeology ...... 120 2. Methods ...... 123 2.1 Field and Laboratory Methods ...... 123 2.2 Water-Rock Reaction Modeling...... 125 2.3 Groundwater Flow Modeling ...... 126 3. Results and Discussions ...... 126 3.1 Chemical Characteristics of the Delta and the Deep Saline Groundwaters ..... 126 3.2 Isotopic Characteristics of the Delta and Deep Saline Groundwaters ...... 129 3.2.1 Environmental Isotopes ...... 129 3.2.2 Mechanisms of Water-Rock Interaction using Strontium Isotopes ...... 129 3.2.3 Mechanism of Salinization and Chloride Isotope ...... 135 3.2.4 Bromine Isotope Systematics ...... 137 3.3 Geochemical Modeling of Groundwater in the Wadi Watir Delta Area ...... 140 3.4 Solute-transport Modeling...... 143 3.4.1 Model Parameters ...... 148 3.4.2 Boundary Conditions and internal sinks ...... 148 3.4.3 Initial Conditions ...... 149 3.4.4 Model Calibration ...... 151 4. Summary and Conclusions ...... 163 5. References ...... 166 CHAPTER V: Conclusion ...... 172

viii

List of Figures

1-1. Location map of the study area...... 2 1-2. Topography and rain gauge locations in the Wadi Watir watershed and in the nearby vicinity (Cools et al., 2012)...... 7 1-3. Geomorphological map of the study area, ...... 8 1-4. a) Basement Rocks in the Wadi Watir watershed; b) Fractures and joints in granitic rocks in the Wadi Watir watershed; c)Basic dykes intruded into the granitic rocks at Furtaga Spring; d) Light colored limestone of the El Tih Plateau in the watershed; e) Alluvial fan deposits of the Wadi Watir delta; f) Quaternary deposits of Wadi Watir delta ...... 9 1-5. Generalized hydrogeological map of the study area...... 12 1-6. Wells tapping different aquifers in the Wadi Watir watershed and its delta...... 13 1-7. Conceptual model of the hydrogeology and groundwater flow in the Wadi Watir watershed and for the Wadi Watir delta...... 14 1-8. Field trip pictures from 2009 showing (a) groundwater sampling, (b) measuring depth to water, and (c) conducting infiltration tests using a double ring infiltrometer...... 19 2-1. Location of groundwater samples in the Wadi Watir watershed...... 28 2-2. Geological map of the study (modified from CONOCO, 1987)...... 30 2-3. Hydrogeological cross section A-B, C-D and E-F...... 31 2-4. Water level contours (meters) in the El Shiekh Attia area...... 41 2-5. General Hydrogeological map of the study area...... 42 2-6. Amount of Preciptation (mm) versus δ18O (‰) in rainwater for some selected stations in Sinai (El Arish and Rafah); Israel (Soreq, Sadoth, Ramond and bet dagan) and Jourdan (Shoubak)...... 45 2-7. δ18O (‰) versus δ2H (‰) plot for the rainwater during the rainy seasons in the study area (October to Aprill)...... 46 2-8. Relationship between δ18O and δ2H for groundwater in the El Shiekh Attia, Wadi El Ain, and the Main Channel areas...... 47 2-9. Relationship between δ18O and Cl for groundwater in the El Shiekh Attia, Wadi El Ain, and the Main Channel areas...... 47 3-1. Location of the study area and groundwater samples...... 67 3-2. Stratigraphic cross-section of Wadi Watir delta constructed from well log information and geophysical interpretations (modified from Abbas et al., 2004)...... 70 3-3. Major-ion diagram (Piper Diagram) of groundwater in the Wadi Watir delta. Groundwater is predominantly Na + Ca and Cl + SO4...... 77 3-4. Total dissolved solids and chloride concentrations in groundwater for 1999 (After Said, 2004) and 2007...... 78 ix

3-5. Isotopically distinct groundwaters shown as Group I (-3.82 to -2.58 ‰ δ18O; - 19.5 to -13.4 ‰ δ2H) and Group II (-0.66 to +6.86 ‰ δ18O; -5.9 to +22.5 ‰ δ2H)...... 80 3-6. δ18O versus δ2H for groundwater in the Wadi Watir watershed and delta areas..... 81 3-7. δ18O versus chloride for groundwater in the Wadi Watir delta...... 83 3-8. δ18O versus bromide for groundwater, seawater, and sabkha in the Wadi Watir delta. Group I (-3.82 to -2.58 ‰ δ18O; -19.5 to -13.4 ‰ δ2H) and Group II (-0.66 to +6.86 ‰ δ18O; -5.9 to +22.5 ‰ δ2H)...... 84 3-9. The finite difference grid cells used for the model in the x, y (A), and z (B) directions...... 92 3-10. Measured versus model calculated water level cells in the groundwater flow model that contained wells...... 98 3-11. Estimated recharge for the Wadi Watir delta between the periods from 1982 to 2009...... 103 3-12. Simulated groundwater levels from the calibrated MODFLOW model...... 105 3-13. SEAWAT model output for a vertical west to east cross section through well No. 35, showing increasing salinity through time...... 106 3-14. Salinity with depth after El-Refeai (1992)...... 107 3-15. Total dissolved solids (mg/l) recorded at Well No. 35 and Well No. 42 from 1996 to 2009...... 108 4-1. a) Location map of Wadi Watir basin, Sinai Peninsula, Egypt; b) Location map of Wadi Watir delta and groundwater wells...... 121 4-2. Sulin's Diagram for groundwater in the Wadi Watir delta (2007 and 2009)...... 122 4-3. Positive correlation of strontium versus calcium...... 131 4-4. 87Sr/86Sr versus Sr in the Wadi Watir upper watershed and delta groundwaters.. 133 4-5. a) δ 37Cl ‰ (SMOC) versus Cl (ppm) for groundwater in the Wadi Watir delta. Chloride concentration of seawater is divided by 10 to fit on the plot. b) δ 37Cl ‰ (SMOC) in different rock types in the Wadi Watir watershed...... 136 4-6. a- δ 81Br ‰ SMOB versus Br and b- δ 37Cl ‰ SMOC versus δ 81Br ‰ SMOB of groundwater in the Wadi Watir delta...... 139 4-7. The finite difference grid cells used for the model a) in the x and y direction and b) in the z direction...... 147 4-8. Salinity (mg/L) with depth (meter) in groundwater of the Quaternary aquifer in the Wadi Watir delta (modified from El-Refeai, 1992)...... 150 4-9. Calcutated versus oberved water levels (head)...... 152 4-10. Well location map of the salinity observations...... 155 4-11. Calcutated versus oberved salinity...... 157 4-12. Breakthrough curves for selected wells in the Wadi Watir delta...... 160 4-13. Estimated recharge and pumping (m3/day) using head observations and couples head and salinity observations through the model time (1982-2009)...... 162 x

List of Tables

1-1. Flash flood occurrence and volume (Himida, 1997; JICA, 1999 and Cools et al., 2012)...... 5 2-1. Results of petrographic analyses of rock samples...... 36 2-2. Results of X-ray fluorescence of granite samples...... 36 2-3. Constraints, phases, and parameters used in NETPATH models...... 37 2-4. Chemical (meq/L) and isotopic (‰) data for groundwater samples collected in March 2007...... 38 2-5. NETPATH modeling results (mmol/L) for the Wadi Watir watershed...... 51 2-6. Mineral saturation indices for phases in NETPATH geochemical models...... 51 2-7. Calculated groundwater ages using NETPATH...... 55 3-1. Well information and chemical and isotopic data for groundwater samples collected in March 2007 and August 2009...... 71 3-2. Constraints, phases, and parameters used in NETPATH models...... 74 3-3. Water chemical analyses for some wells in the Wadi Watir delta in 1999 (After Said, 2004) and 2007...... 79 3-4. NETPATH modeling results for water-rock interactions and mass transfer (mmol/L) for the Wadi Watir basin...... 85 3-5. Mineral saturation indices for phases in geochemical models calculated using NETPATH...... 88 3-6. Groundwater ages for Wadi Watir delta corrected for water-rock interactions using NETPATH...... 90 3-7. The amount of groundwater pumping from water supply wells on the delta...... 96 3-8. Flash flood occurrence and volume (Himida, 1994; JICA, 1999 and Cools et al., 2012)...... 96 3-9. Modeled and simulated groundwater levels in different years through the modeled time...... 99 3-10. Simulated groundwater recharge for injection wells in the groundwater flow model...... 102 4-1. 87Sr/86Sr, 37Cl, and 81Br in groundwater and rock sample in the Wadi Watir Basin...... 132 4-2. NETPATH modeling results for water-rock interactions and mass transfer (mmol/L) for the Wadi Watir delta...... 141 4-3. Mineral saturation indices (SI) for phases in geochemical models calculated using NETPATH...... 142 4-4. Parameters for the groundwater model in the Wadi Watir delta...... 146 4-5. Modeled and simulated groundwater levels in different years through the modeled time...... 153 xi

4-6. Total dissolved solids for groundwater in the Wadi Watir delta in 1996, 1998, 2007 and 2009...... 156 4-7. The amount of groundwater pumping from water supply wells (drilled wells) in the delta...... 158 4-8. Data for pumping rates obtained from the model after calibrating with salinity and water level changes over time...... 158 4-9. Simulated groundwater recharge for injection wells in the groundwater flow model...... 161

1

CHAPTER 1

1. Introduction

In the past decade, rapid development in the Wadi Watir delta, Gulf of Aqaba,

Sinai Peninsula, Egypt (Figure 1-1), primarily related to tourism, has resulted in increased demand on groundwater resources. Groundwater is being developed for various needs in coastal and arid areas of the Sinai where there are no other sources of fresh water. Groundwater beneath the Wadi Watir delta is the main source of potable water for this area. Groundwater occurs as a thin lens of fresh water and is very sensitive to pumping induced stresses (Shalaby, 1997). Thus, groundwater withdrawals have to be carefully managed to avoid deterioration of this valuable resource by up welling of underlying saline groundwater and intrusion of seawater along the coast. To understand historic and current groundwater conditions in this area, and to predict future changes in groundwater availability and salinity, water chemistry and isotopic data were used to: identify mixing of different waters; characterize the evolution of groundwater chemistry along flow paths; determine the age of the groundwater; and constrain a groundwater flow and salinity transport model. The chemical and isotopic methods and groundwater flow and transport models used in this study, to delineate groundwater resources in the

Sinai Peninsula area of Egypt and understand the role of upwelling saline groundwater and sea water intrusion along the coast, are applicable to other similar hydrogeological settings throughout the world in arid to semi-arid environments. The research presented in this dissertation combined mineral identification, rock chemistry, water chemistry, and

2

MEDITERRANEAN

Cairo SINAI

R E 34 00 34 30 D S E EGYPT A 29 Wadi Watir Watershed

30

f o El Shiekh a Mediterranean Bet Dagan Attia Area b Sea a Soreq f l q u A Sadoth

31 G

Port Said El Arish Israel

Rafah r

Main i Ramond Jordan 00 t a Sinai Shoubak a

29 Wadi Channel a

t W

Egypt l

i

Nekhel El Ain e

d

D

Ras 00 a

El-Naqab

29 W Wadi Watir Boundary Nuweiba 00 Location Saint Saudi Arabia Cathrine Sites El-Tor 0 20 40 Km Ras b Country Nasari Bourders Red Scale Sea 34 00 35 00 Figure 1-1. Location map of the study area.

3 isotopic signatures of groundwater with groundwater flow and transport modeling to determine the physical and geochemical processes that produce the observed water chemistry in the study area. Recent increases in groundwater salinity limits the amount of available potable groundwater.

2. Hypothesis

In the Wadi Watir delta, groundwater quality has been degraded during the last decade. This research tested the hypothesis that groundwater quality degradation is caused by high pumping rates, which results in seawater intrusion along the coast and upwelling of deep saline groundwater beneath the main well field. There are discrepancies among previous studies concerning the reasons for increased groundwater salinization. Some studies attributed the increasing groundwater salinization to leaching of aquifer matrix salts and seawater intrusion caused by over pumping (Ismail, 1998; Abd

El Hafez, 2001; Said, 2004; El-Sayed, 2006). Other studies attribute the increasing salinity to upwelling of a deep saline groundwater beneath the delta (El-Kiki et al., 1992;

El-Refeai, 1992; Shalaby, 1997).

3. Background

The study area is located between longitude 34○ 38 َ and 34○ 41 َ E and latitude

and 29○ 03 َ N in the southeastern part of the Sinai Peninsula, Egypt َ 57 ○28

(Figure 1-1). The Wadi Watir watershed drains east toward the Gulf of Aqaba. The Wadi

Watir delta is located at the downstream portion of Wadi Watir watershed. The Wadi

Watir watershed is considered to be the most important wadi in this region because the

4 city of Nuweiba, a tourist destination, and Nuweiba Harbor, are located on the delta.

Ships sailing from Nuweiba Harbor link Egypt with Saudi Arabia and Jordan.

3.1 Climatic Conditions

Groundwater originates as infiltrating precipitation in the study area, thus it is important to understand the climate of the area. The study area is arid to hyper arid

(UNESCO, 1963; Issar and Gilad, 1982; El-Refeai, 1984; El-Sammany, 2011; and Cools et al., 2012). Mean temperature in the Wadi Watir ranges between 22 and 24 ○C in July and 2 and 12 ○C in January (Greenwood, 1997). Relative humidity is higher in the north

Sinai at the El Arish and Port Said stations (70-80%), and is lower at the south Sinai station of Saint Cathrine (15-40% ) (Greenwood, 1997). Meteorological station locations are shown on Figure 1-1a. The average annual rate of potential evapotranspiration in the

Sinai Peninsula (1750 mm/yr) is 30 times greater than the average annual precipitation

(Parsons and Abrahams, 1994; Tolba and Gaafer 2003; and Cools et al., 2012).

Rainfall is rare and unpredictable in the study area (El-Baz et al., 1998). Airflow from the Mediterranean Sea is the main source of precipitation, which generally occurs in the winter (October-May) (Dames and Moore, 1983; Greenwood, 1997). Rainfall is higher in the northern Sinai at Rafah and varies in the mountainous regions with increasing precipitation with elevation (Issar and Gilad, 1982). At El Tih Plateau, within the Wadi Watir watershed, rainfall ranges from 25 to 50 mm/yr, whereas, at the Gulf of

Aqaba coast it ranges from 15-20 mm/yr (Greenwood, 1997). Cools et al. (2012) estimated the average annual rainfall for the study area from 1979 to 2006 to be 35 mm/yr. Precipitation in the Sinia is strongly restricted to individual storms. The Wadi

5

Watir experiences flash floods about every 2-3 years (El-Sammany, 2011). These severe storm events which produce flash floods are the main recharge events for aquifers of the

Wadi Watir watershed. Flash floods in Wadi Watir from 1987 to 2008 were reported by

Himida (1997), JICA (1999), and Cools et al. (2012) (Table 1-1).

Table 1-1. Flash flood occurrence and volume (Himida, 1997; JICA, 1999 and Cools et al., 2012). Flood Data Volume Flash flood Remarks (106 m3) Oct-16-1987 45 Very High Disaster Dec-20-1987 -- Low -- Apr-20-1988 5 Moderate Catchment outlet Oct-16-1988 15 High Catchment outlet Mar-12-1990 -- Low -- Oct-20-1990 35 High Catchment outlet Mar-22-91 -- Moderate -- Mar-1993 -- High Catchment outlet Oct-1993 -- High Catchment outlet Jan-1-1994 -- Moderate Catchment outlet Nov-2-1994 -- Moderate -- Nov-17-1996 -- Moderate -- Jan-14-1997 -- Moderate -- May-16, 17-1997 4.4 Moderate Catchment outlet May-28-1997 0.27 Low Catchment outlet Oct-18-1997* -- Very High Disaster Jan-15-2000 -- Low -- Dec-9-2000 -- Low -- Oct-(27-31)-2002 -- Moderate -- Nov-3-2002 -- Low -- Dec-15-2003 -- Low -- Feb-5-2004 -- Low -- Oct-29-2004 -- Low -- Oct-24-2008 -- Low -- *This flood continued for six days (Oct. 19-24); -- No data

6

3.2 Geomorphic and Hydrogeologic Setting

The geomorphological and geological setting of the study area has been investigated by many authors including Said (1962), El-Refaei (1992), El Shamy (1992),

Shalaby (1997), Shabana (1998) and Aggour et al. (2000). The Wadi Watir attains maximum elevation in the upper watershed at the southern portion of the study area where the ground elevation exceeds 1250 m above sea level (Figure 1-2). Ground elevation at the delta ranges between 0 (at the coast) to 42 m at the mountain front.

The study area can be divided into the general geomorphic units of: (1) high mountainous terrain; plateau; coastal plain; and inland plain (Aggour et al., 2000)

(Figure 1-3). The high mountainous terrain is characterized by sharp peaks and steep slopes (Figure 1-4a) and slopes steeply toward the east. This terrain is composed mainly of the Arabo-Nubian massive (Said, 1962), which forms the southern Mountain Region consisting of mainly rugged igneous and metamorphic rocks (Figure 1-3). The mountainous terrains are generally good aquifers because groundwater percolates easily through fractures and joints (Figure 1-4b), however, dykes often cut the granitic rocks and act as barriers to groundwater flow (Figure 1-4c).

The plateau terrain is the El Tih Plateau in the study area, which occupies the western part of the study area (Figures 1-3b and 1-4d) and is dominated by a Cretaceous limestone interbedded with shale (El-Baz et al., 1998). The El Tih Plateau is the principal recharge area for the lower Cretaceous sandstone outcrops at the base of the El Tih escarpment (Issar, 1985; Mill and Shata, 1989). Inland plains occur along the drainages of the upper watershed, whereas the coastal plain is along the Gulf of Aqaba extending

7 inland to the mountain block-alluvial contact (Aggour et al., 2000) (Figure 1-3b). The inland plain is primarily eroded limestone deposits and the coastal plain consists of poorly sorted sediments derived from the weathering of igneous and sedimentary rocks from the watershed (Figures 1-4e and 1-4f).

Figure 1-2. Topography and rain gauge locations in the Wadi Watir watershed and in the nearby vicinity (Cools et al., 2012).

8

The Mediterranean 31

00 Mediterranean Coastal Plain

a Central Table Land 34 00 34 30 30 1- High Land 00 High Mountaneous Terrain

29 El Tih Plataeu S El Egma u d Plateau 2- Low Land r 30 P la Inland Plain i n El Tih Coastal Plain

Plateau a

z 29 b

e a

u q

S

00 A f

a f

o E

b o

f l l a f

Q l u Southeastern q

a u G a A 29 Mountain G P f la o Region 00 i f Major n l Fault u G 28 b Wadi Watir boundary 00

33 00 34 00 Figure 1-3. Geomorphological map of the study area,

9

a b

c d

f

e

Figure 1-4. a) Basement Rocks in the Wadi Watir watershed. b) Fractures and joints in granitic rocks in the Wadi Watir watershed. c) Basic dykes intruded into the granitic rocks at Furtaga Spring. (Dykes act as groundwater flow barriers). d) Light colored limestone of the El Tih Plateau in the watershed. e) Alluvial fan deposits of the Wadi Watir delta. f) Quaternary deposits of Wadi Watir delta (Alternating beds of fine and coarse alluvial material associated with the different flooding cycles).

10

The oldest aquifers in the study area are granitic rocks that mainly occur in the western part of the study area and are recharged through fractures in the rock

(Figure 1-4b) (Shabana, 1998). The occurrence of groundwater in this aquifer is mainly controlled by the presence of basic dykes, which act as barriers to groundwater flow

(Figure 1-4c). The Furtaga spring area is just up gradient of one of these dykes in the main channel area. Groundwater discharging from the springs and in nearby wells in this area have salinity that ranges from 1140 and 3400 mg/L. At Furtaga springs, groundwater discharge is about 8 m3/hr in 2009 and was 78 to120 m3/hr in 1982. Groundwater in nearby wells has depth ranging from 4.23 to 5.83 m below land surface.

The Lower Cretaceous aquifer occurs locally in the upper reaches of the main channel

(Figures 1-5 and 1-6a) and extends north of the study area. This aquifer is composed mainly of sandstone with shale layers (Yehia, 1994; Yehia, 1998). The Upper Cretaceous aquifer is composed of primarily carbonate rocks and outcrops in the Wadi El Ain area

(Figure 1-1) (Shabana, 1998). This aquifer is under unconfined conditions and depth to water ranges from 4.73 to 9.73 m in the Wadi El Ain area. Groundwater salinity ranges from 1650 to 2250 mg/L.

Quarternary aquifers in the El Sheikh Attia area and along the main channel

(Figure 1-1) are composed mainly of sand, gravel and boulders of carbonate rock embedded in a silty matrix (Shabana, 1998) (Figures 1-5 and 1-6a). Groundwater is present in the Quaternary aquifer under unconfined conditions with depth ranging from

10.2 to 12.2 m and water levels ranging from 553 to 563 m above sea level in the El

Sheikh Attia area. Groundwater flows from north to south following the main slope of the

11 ground elevation (Eissa et al., 2012, in review). Groundwater salinity ranges from 720 to

1190 mg/L. The Quaternary aquifer in Wadi El-Ain area is composed mainly of of sand, pebbles, cobbles and boulders and detrital fragments from the weathering of granitic and sedimentary rocks. Depth to water ranges from 2.8 to 4.0 m and salinity ranges from

2890 to 3490 mg/l.

The Quaternary aquifers of the Wadi Watir delta are composed of alluvial deposits of sand, gravel and boulders of carbonate, sandstone, and granitic rocks embeded in a silty and clayey matrix (Shalaby, 1997). Groundwater in this aquifer occurs under unconfined conditions (El-Kiki et al., 1992). The Quaternary aquifers of the Wadi Watir delta can be divided into five layers (Abbas et al., 2004) (Figure 1-7). The uppermost two layers are generally <10 m thick and are comprised of heterogeneous alluvial deposits. The third layer is a sandy clay layer and is 30 to 45 m thick. The fourth layer is comprised of sand and gravel and is 20 to 40 m thick. The fifth layer is comprised of sand interlayered with shale and is 20 to 50 m thick. These five layers are underlain by bedrock which is composed mainly of very low permeability granitic rocks. The Wadi Watir delta aquifers are the main groundwater source in the study area. The aquifers are tapped by thirty nine wells (Figure 1-6b) and have a low storage capacity with high transmisivity ranging between 8000 and 2800 m2/day (El-Refeai, 1992).

12

29 (19) 419 30

29

a 20 b (18) 430 a q (2-7 & 9) [548] A 29 (22-24) f [469] o 10 f l (21) (25&27) [643] 660 (29-31) (16) u [347] 237 G (13&14) ( 17, a & b) 29 [641] (15) [0.9] 257 00 0 10 (20) 688 Km

34 30 34 40 34 50 water wells tapping the Quaternary Quaternary aquifer massive carbonate rocks aquifer forming the watershed water wells tapping the upper Cretaceous upper Cretaceous aquifer aquifer (Site No. ) lower Cretaceous water wells tapping the Water Level (m) aquifer lower Cretaceous aquifer [Average water level] water wells tapping Precambrian the Precambrian aquifer aquifer Figure 1-5. Generalized hydrogeological map of the study area.

13

34 00 34 30 34 39 34 41 Wadi Watir Watershed 29 Legend a 30 19 Hand Dug Wells 43 18 Drilled Wells 41 a b 42 a q Basement Rocks A 40 0 1 2

39

29 Km. Sabkha Deposits

f o 00

G 29

Alluvial deposits f l u u G 02 lf 0 20 40 Km

Scale 38 o 37 f Legend 23 1 El Shiekh 33 Attia 30 Sampling sites in different aquifers 2 21A22 36 A 5 Quaternary 24 29 35 q 3 20 28 a Upper Cretaceous 25 b 7 19 27 32 26 a 4 Lower Cretaceous 8 18 31 34 9 6 0 1500 m Precambrian 29 17 44 00 45 50 53 46 Wadi El Ain 27 Main Channel 14 23 24 47 48 28 49 12 Sabkha 1326 22 51 21 1125 10 29 B 30 31 Wadi Watir 16 Delta 52 15 28 b c 54 a 55 20 17 58 0 1.5 Km 0 10 Km ba b ei w N

Figure 1-6. Wells tapping different aquifers in the Wadi Watir watershed and its delta.

14

Precip tation from Storms Evaporation of Runoff water medite rranian before infiltration

Wadi Watir watershed

R un Seepage of f W a ter G roun dwat Evaporation of Sabkha er flo w Wadi Watir delta Water and through Gulf of Aqaba shallow groundwater

A Higher Pumping B 30 Sabkha C

10

)

l e

v Recharge Alluvial deposits

e -10 Cone of l

Depression (Thin layer of Fresh water) a

e S -30 and & gravel S s (Brakish ( ground

water)

r

n e

o -50

t

i t

Upwelling of a

a w v Saline Water

e -70

a

l

e E

Sand intercalated with clay (Saline groundwater) S -90 Bedrock Seawater Intrusion -110 Zone

Figure 1-7. Conceptual model of the hydrogeology and groundwater flow in the Wadi Watir watershed and for the Wadi Watir delta. Well "A" represents drilled wells in the well field affected by upwelling of deep saline groundwater due to over pumping. Well "B" represents wells affected primarily by water-rock interactions. Well "C" represents shallow hand-dug wells affected by leaching of sabkha salts, evaporation and seawater intrusion. The Wadi Watir delta cross section is modified after (Abbas et al., 2004).

4. Groundwater Recharge

Average annual recharge, between 1998 and 2007, for the Wadi Watir watershed was estimated by Milewski et al. (2009) to be 193 x 106 m3/yr. This value is similar to the average recharge value of 165 x 106 m3/yr estimated by Masoud (2009) between 1960

15 and 1990. Elewa et al. (2011) ranked the Nubian Sandstone aquifer (Lower Cretaceous) in eastern Sinai and the Furtaga Springs areas in Wadi Watir as promising areas with very high annual groundwater recharge.

El Ghazawi (1999) and El Sayed (2002) suggested that the Wadi Watir delta aquifers are recharged directly from the outlet of Wadi Watir drainage and by discharge from the Furtaga springs. Discharge from Furtaga springs varies considerably according to the amount of rainfall and storm frequency during the year. Issar and Gilad (1982) showed that discharge from Furtaga Springs ranged from 115 to 3600 m3/day with an average rate of 1800 m3/day from 1969 to 1979. The Research Institute for Water

Resources (1989) estimated the discharge of Furtaga Springs to be 1644 m3/day, and Idris

(1995) estimated the spring flow to range from 192 to 2500 m3/day. The springs stopped flowing in August 2009 due to a prolonged drought. Ismail (1998) estimated the subsurface recharge to the Wadi Watir delta aquifers in April 1995 to be 2200 m3/day.

5. Groundwater Chemistry

According to Shalaby (1997), groundwater in the shallow aquifers of the delta form a lens-like shape of slightly brackish water overlaying a deeper saline groundwater.

This thin lens of relatively fresh water is very sensitive to groundwater pumping. El-

Refeai (1992) measured groundwater salinity with depth in a well in the well field and showed that the salinity increased with depth ranging from 900 mg/L at 2 m above sea level to 18,000 mg/L at 35 m below sea level.

According to Said (2004), groundwater in the Wadi Watir delta near the coast is characterized by high salinity because of seawater intrusion. Abd El Hafez (2001)

16 reported that groundwater in the Wadi Watir delta is from a mixed origin of meteoric water and seawater. Ismail (1998) reported salt water-intrusion in the northern part of the delta because of upwelling of saline water.

6. Conceptual Model of Groundwater Recharge, Flow, and Salinity Sources

Groundwater in the Wadi Watir watershed is recharged by infiltration of precipitation and surface runoff when the area receives heavy storms, which occur primarily in the winter. Water infiltrates through the soil and percolates through the joints, fissures and fractures or directly infiltrates fractures in exposed rock and flows through the unsaturated zone to the water table (Figure 1-7). The groundwater generally flows from the upper watershed to the Wadi Watir delta. Wadi Watir delta aquifers receive most of their recharge from the mountain blocks.

Groundwater chemistry and salinity in the Wadi Watir delta is derived from five main sources: (1) water rock interactions; (2) upwelling of deep saline groundwater due to over pumping in the well field; (3) seawater intrusion along the coast; (4) leaching of salts from sabkha deposits near the coast; and (5) evaporation of shallow groundwater in the delta area. Groundwater recharge, groundwater flow and salinity sources are shown in conceptual cross sections of the study area in Figure 1-7. Well "A" represents drilled wells in the well field affected by upwelling of deep saline groundwater due to over pumping. Well "B" represents wells affected primarily by water-rock interactions. Well

"C" represents shallow hand-dug wells affected by leaching of sabkha salts, evaporation and seawater intrusion.

17

7. Objectives

The main objectives of this research were to estimate the average annual groundwater recharge for the Wadi Watir delta aquifers and to evaluate the different geochemical processes affecting groundwater chemistry in the Wadi Wadir watershed and its delta. Particular focus was placed on determining the recharge sources for the watershed and the delta aquifers and salinity increases in the groundwater related to management of groundwater withdrawal rates in the delta. Specific objectives include:

1. Evaluate the amount of annual recharge from the mountain blocks to the Wadi

Watir delta aquifers.

2. Estimate the annual pumping rates from the Wadi Watir delta aquifers.

3. Simulate the upwelling of higher salinity groundwater into fresher shallow

groundwater in the well field of the Wadi Watir delta.

4. Determine the extent of seawater intrusion along the coast.

5. Determine the main source(s) of groundwater recharge in the upper watershed

and Wadi Watir delta aquifers.

6. Determine geochemical reactions and physical processes that produce the

observed water chemistry in the study area. These physical processes include

mixing of chemically and isotopically different groundwaters, evaporation of

shallow groundwater, and mixing with seawater along the coast.

7. Determine if most of the groundwater that is currently being used is from recent

recharge or is paleo-groundwater recharged thousands of years ago under wetter

climatic conditions.

18

8. Approach and Methods

Data for amounts and isotopic composition (δ18O and δ2H) of rainfall were downloaded from the Global Network of Isotopes in Precipitation (GNIP: IAEA and

WISER, 2008). These data were used to calculate the monthly and annual weighted average isotopic signature for rainfall in the study area. Six stations were used including

El Arish and Rafah Stations from the Sinai, Shoubak from Jordan, and Sadoth, Bet

Dagan, Ramond, and Soreq from Israel (Figure 1-1a).

Water-sampling was conducted in March 2007. Twenty-nine groundwater samples were collected from the upper Wadi Watir watershed; nine from hand-dug wells in the El Shiekh Attia area, nine from hand-dug wells and one from a drilled well in the main channel area, and ten from hand-dug wells in the Wadi El Ain area. Thirty-nine groundwater samples were collected from the Wadi Watir delta aquifers; 32 samples from hand-dug wells and seven samples from drilled wells. Six additional samples were collected in September 2009 (Figure 1-8a). Five pumping tests were performed to determine hydraulic conductivity for different water-bearing formations in the delta

(Figure 1-8b). Five infiltration tests (Figure 1-8c) were conducted to determine vertical hydraulic conductivity of the surficial layer in the Wadi Watir delta using the double ring infiltrometer method (Philip, 1957; Hanks and Aschroft, 1980).

19

a b

c

Figure 1-8. Field trip pictures from 2009 showing (a) groundwater sampling, (b) measuring depth to water, and (c) conducting infiltration tests using a double ring infiltrometer.

Rock samples representing different aquifers were collected throughout the Wadi

Watir watershed. Mineralogy of these rock samples was identified by examination of thin-sections using polarized light microscopy. Rock chemical composition was determined by x-ray fluorescence.

Water stable isotope analysis was conducted for 13C, 14C, 37Cl, 81Br and 87Sr/86Sr for interpretation of geochemical and physical reactions and mixing of different waters.

Twenty rock samples were analyzed for 87Sr/86Sr, 37Cl and 81Br. Isotopic analyses were

20 performed using standard methods (McNutt, 1990; Eggnkamp, 1994 ; and Shouakar-

Stash et al. 2005).

A groundwater flow model (MODFLOW) that includes seawater intrusion

(SEAWAT) was developed to; (1) evaluate the amount of annual recharge from the mountain blocks to the Wadi Watir aquifers, (2) estimate annual pumping rates,

(3) simulate upwelling of higher salinity groundwater into fresher shallow groundwater, and (4) evaluate seawater intrusion along the coast. This modeling effort included determining the safe yield of the delta alluvial aquifers to avoid seawater intrusion along the coast and upwelling of deep saline groundwater into the well field.

Geochemical modeling that used groundwater chemistry, δ18O and δ2H groundwater data, and rock mineralogy and chemistry was conducted to determine geochemical reactions and physical processes that produce the observed water chemistry.

87Sr/86Sr, 37Cl and 81Br data were used to determine water-rock interactions, mixing of different waters, upwelling of saline water, and seawater intrusion. Geochemical models included carbon isotope data to determine groundwater age in different aquifers.

9. References

Abbas, A.M., Atya, A.A, Al-Sayed, E.A, and Kamei, H., 2004. Assessment of groundwater resources of the Nuweiba area at Sinai Peninsula, Egypt by using geoelectric data corrected for the influence of near surface inhomogeneities, Journal of Applied Geophysics 56(2):107-122. Abd El Hafez, A.A., 2001. Chemical evaluation and possible treatment sea water intrusion in the groundwater in some coastal areas, South Sinai. M. Sc. Thesis, Fac. Sci., Al Azhar Univ. (girls), 221 p.

21

Aggour, T.A., Shabana, A.R., Shided, A.G., and Yehia, M.M., 2000. Hydrogeological conditions of the water bearing formations in Wadi Watir basin with emphasis on the deep ones. 2nd International Conference on Basic Science and Advanced Technology. Fac. Sci., Assiut Univ., 13 p. Cools, J., Vanderkimpen, P., El Afandi, G., Abdelkhalek, A., Fockedey, S., El Sammany, M., Abdallah, G., El Bihery, M., Bauwens, W., and Huygens M., 2012. An early warning system for flash floods in hyper-arid Egypt Nat. Hazards Earth Syst. Sci., 12, 443–457. Dames and Moore 1983. Sinai development study, final report. Submitted to the advisory committee for reconstructions, ministry of development, Arab Republic of Egypt. Eggenkamp, H.G.M., Middleburg, J.J., Kreulen, R., 1994. Preferential diffusion of 35Cl relative to 37Cl in sediments of Kau Bay, Halmahera, Indonesia. Chem. Geol. 116, 317–325. Eissa, M.A., Thomas, J.M., Hershey, R.L., Dawoud, M.I., Pohll, G., Gomaa, M.A., and Kamal A.D. (in review, Environmental Earth Sciences). Geochemical and Isotopic Evolution of Groundwater in the Wadi Watir Watershed, Sinai Peninsula, Egypt. El-Baz, F., Kusky, T., Himida, I., and Abdl-Mogheeth, S., 1998. Groundwater potential of the Siani Peninsula, Egypt. Sponsored by United States Agency for International Development (AID) in Cairo, Egypt. project conducted jointly between the Boston University Center for Remote Sensing (BU/CRS) and the Desert Research Center (DRC) in Cairo, Egypt. Elewa, H., Qaddah, A., 2011. Groundwater potentiality mapping in the Sinai Peninsula, Egypt, using remote sensing and GIS-watershed-based modeling. Hydrogeology Journal 19:613-628. doi 10.1007/s10040-011-0703-8. El Ghazawi, M.M., 1999. Reconsideration of hydrogeologic setting in the delta of Wadi Watir, southern Sinai. Bulletin of Science, Mansoura Univ., Egypt, vol. 26 (1).

22

El-Kiki, M.F., Eweida, E.A, El-Refeai, A.A., 1992. Hydrogeology of the Aqaba rift border province. Proc 3rd Conf Geol Sinai Develop, Ismailia, Egypt, 91-100. El-Refeai, A.A., 1984. Geomorphological and hydrogeological studies on El-Qaa plain, Gulf of Suez, Sinai, Egypt. M. Sc. Thesis, Fac. Sci., Cairo Univ., Egypt,298p. El-Refeai, A.A., 1992. Water resources of southern Sinai, Egypt geomorphological and hydrogeological studies. Ph.D. Thesis, Fac. Sci., Cairo Univ., Egypt, 357p. El-Sammany, M., 2011. Forecasting of Flash Floods Over Wadi Watier – Sinai Peninsula Using the Weather Research and Forecasting (WRF) Model. Nile Basin Water Science& Engineering Journal, Vol.3, Issue2, 2010. El Sayed, M.H., Abd El-Samie, S.G., Sallouma, M.K., and Said, M.M., 2002. Hydrochemical parameters to detect salt water intrusion in the groundwater aquifer of delta wadi Watir, Gulf of Aqaba, Southeast Sinai, Egypt‖. Bulletin of Egypt. J. Appl. Sci; 17(2) pp. 350-389. El-Sayed, M.H., 2006. Comparative study of water quality of the Quaternary aquifers in Wadi Watir basin and its delta, southeast Sinai, Egypt. Egyptian J. Desert Res. 56, No. 1, 17-46. El Shamy, I.Z., 1992. Towards the water management in Sinai Peninsula", Proc. 3rd conf. Geol. Sinai Develop., Ismailia, pp. 63-70. Greenwood, N.H., 1997. The Sinai: A Physical Geography. University of Texas Press. Hanks, J.R., and Aschroft, G.L., 1980. Applied soil physics, Springer Verlag Berlin/Heidelberg/New York, 1980. 159 pp. (cloth.). Himida, I.H., 1997. Water resources of Wadi Watir. Internal report, Desert Research Center (In Arabic). IAEA and WISER, 2008. Water Isotope System for data analysis, visualization, and Electronic Retrieval: The WISER Version 0.7. Idris, H., 1995. Springs in Egypt, Environmental Geology (Vol. 27) 99-104.

23

Ismail Y.L., 1998. Hydrogeological and hydrochemical studies on Wadi Watir area, South Sinai, Egypt. PhD Thesis, Fac Sci, Suez Canal Univ, Egypt. Issar, A., and Gilad, D. 1982. Groundwater flow system in the arid crystalline province of Southern Sinai‖. Jour.Desert Sci. Hydro., Vol.27, pp. 309-325. Issar, A., 1985. Fossil water under the Sinai-Negev Peninsula: Scientific American, V. 253, pp. 104-110. Japan International Cooperation Agency (JICA) and Water Resources Research Institute (WRRI) 1999. South Sinai groundwater resources study in the Arab Republic of Egypt, Main Report. Pacific consultants international, Tokyo, 290p. Masoud, A.A., 2009. Runoff modeling of the wadi systems for estimating flashflood and groundwater recharge potential in Southern Sinai, Egypt, Arab Journal of Geoscience, doi:10.1007/s12517-009-0090-9. McNutt, R.H., Frape, S.K., Fritz, P., Jones, M.G., and MacDonald, I.M., 1990. The 87Sr/86Sr values of Canadian Shield brines and fracture minerals with applications to groundwater mixing, fracture history, and geochronology. Geochim. Cosmochim. Acta 54, 202–215. Milewski, A., Sultan, M., Yan, E., Becker, R., Abdeldayem, A., Soliman, F., and Gelili, K., 2009. A remote sensing solution for estimating runoff recharge in arid environments. Journal of Hydrology 373:1-14. Mill, A.C., Shata, A., 1989. Groundwater assessment of Sinai, Egypt. Groundwater, Vol. 27, P. 793-801. Parsons, A.J., and Abrahams, A.D., 1994. Geomorphology of Desert environments Parsons A.J, and Abrahams, A.D, (eds), Chapman and Hall, London, P. 3-12. Philip, J. R., 1957. The theory of infiltration. The infiltration equation and its solution Soil Sci. 83: 345-357. Research Institute for Water Resources., (RIWR) 1989. Hydrogeology of deep aquifers in the western desert and Sinai. Research Institute for Water Resources (RIWR), ― Sinai water resources study (phase II)‖. Internal Report, 191p, 1989.

24

Said, M.M., 2004. Geochemistry of groundwater in coastal areas, South Sinai, Egypt. Ph.D. Thesis, Fac. Sci., Ain Shams Univ., Egypt. Said, R., 1962. The Geology of Egypt. Elsevier, Amsterdam. Shabana, A.R., 1998. Geology of water resources in some catchment areas draining in the Gulf of Aqaba, Sinai-Egypt. PhD Thesis, Fac Sci, Ain Shams Univ, Egypt 246. Shalaby, A.I., 1997. Geomorphology and hydrogeology of Wadi Watir basin, S. E. Sinai, Egypt‖. M.Sc. Thesis, Fac. Sci., Mansoura Univ., Egypt, 185p. Shouakar-Stash, O., Frape, S.K., Drimmie, R.J., 2005. Determination of bromine stable isotopes using continuous-flow isotope ratio mass spectrometry. Anal. Chem. 77, 4027–4033. Tolba, A. F. and Gaafer, K. (2003) On estimation of potential evapotranspiration in Egypt Egyptian Meteorological Authority-Meteorological Research Bulletin, ISSN 1687-1014, 2003. UNESCO, 1963. Bioclimatic map of the mediterranean region (East Sheet). Paris. Yehia, M.M., 1994. Hydrological and hydrochemical studies on deep aquifer in some localities in Sinai Peninsula. Ph.D. Thesis, Faculty of Science, Menoufia University. Yehia, M.M., 1998. Potentiality of lower cretaceous aquifer system at south eastern Sinai, Egypt. Sci. J. Fac. Sci., Menufiya Uni., Vol. XII, 21–43.

25

CHAPTER 2

Geochemical and Isotopic Evolution of Groundwater in the Wadi Watir Watershed,

Sinai Peninsula, Egypt

Mustafa A. Eissa1,2,*, James M. Thomas2, Ronald L. Hershey2, Maher I. Dawoud3, Greg

Pohll2, Kamal A. Dahab3, Mohamed A. Gomaa1, and Ashraf R. Shabana1

1Desert Research Center, Division of Water Resource, Matariya, Cairo, Egypt

2Desert Research Institute, Division of Hydrologic Sciences, Reno, Nevada, USA

3Facility of Science, Geology Department, Menoufiya University, Egypt

*[email protected], telephone 775.673-7627, fax 775.673.7363

26

Abstract

The Wadi Watir delta in the Wadi Watir watershed is a tourist area in the arid southeastern part of the Sinai Peninsula, Egypt, where development and growth of the community on the delta is constrained by the amount of groundwater that can be withdrawn sustainably. To effectively manage groundwater resources in the Wadi Watir delta, the origin of groundwater recharge, groundwater age, and changes in groundwater chemistry in the watershed needs to be understood. Mineral identification, rock chemistry, water chemistry, and the isotopes of hydrogen, oxygen, and carbon in groundwater were used to identify the sources, mixing, and ages of groundwater in the watershed and the chemical evolution of groundwater as it flows from the upland areas in the watershed to the developed areas at the Wadi Watir delta. Groundwater in the Wadi

Watir watershed is primarily from recent recharge while groundwater salinity is controlled by mixing of chemically different waters and dissolution of minerals and salts in the aquifers. The same storms that recharge the El Sheikh Attia area also recharge the

Wadi El Ain area, but the isotopic signature of the groundwater in the Wadi El Ain area is more depleted in δ18O and δ2H than the El Sheikh Attia area because of the rainout effect. The down-gradient Main Channel area receives groundwater flow primarily from the El Shiekh Attia area. Groundwater in the Main Channel area is the primary source of groundwater supplying the aquifers of the Wadi Watir delta.

Keywords: hydrochemical modeling, isotopes, groundwater recharge, Wadi Watir, Sinai

Peninsula, Egypt.

27

1. Introduction

In the past decade, rapid development in the Wadi Watir delta, in the Wadi Watir watershed, Gulf of Aqaba, Sinai Peninsula, Egypt (Figure 2-1), primarily related to tourism, has resulted in increased demand on groundwater resources. Groundwater is being developed for various needs in coastal and arid areas of the Sinai where there are no other sources of fresh water. Groundwater beneath the Wadi Watir delta is the main source of potable water for this area. Groundwater occurs as a thin lens of fresh water and is very sensitive to pumping induced stresses. Thus, groundwater withdrawals have to be carefully managed to avoid deterioration of this valuable resource by up welling of underlying saline groundwater and intrusion of seawater along the coast. To understand historic and current groundwater conditions in this area, and to predict future changes in groundwater availability and salinity, water chemistry and isotopic data are currently being used to: (1) identify mixing of different waters; (2) characterize the evolution of groundwater chemistry along flow paths; (3) determine the age of the groundwater; and

(4) constrain a groundwater flow and solute transport model. The chemical and isotopic methods used in this study to delineate groundwater resources in the Sinai Peninsula area of Egypt are applicable to other similar hydrogeological settings throughout the world in arid to semi-arid environments. This study combines mineral identification, rock chemistry, water chemistry, and isotopic signatures of groundwater to determine the physical and geochemical processes that produce the observed water chemistry, which because of salinity increases, limits the amount of potable groundwater that is available for use in the Wadi Watir delta.

28

34 00 34 30 29 Wadi Watir Watershed 30 19

18 a b a q A MEDITERRANEAN

29 Cairo SINAI

f o 00

STUDY AREA R E f D l S E u A G EGYPT 0 20 40 Km

Scale

a 1 El Shiekh Attia Legend 2 5 Sampling Sites in different aquifers 3 Quaternary (Alluvial) 7 Cenomanian Turonian (Wata-Raha Fm.) 4 8 Lower Cretaceous 9 6 (Sandstone) 0 1500 m Precambrian Basement (Granitic)

Wadi El Ain 27 Main Channel 14 28 23 24 12 1326 22 21 1125 10 29 30 31 Wadi Watir 16 Delta 15 b c a 20 17 0 1.5 Km 0 10 Km ba ei w N

Figure 2-1. Location of groundwater samples in the Wadi Watir watershed.

29

2. Background

2.1 Study Area

The Wadi Watir watershed is located in the southeastern part of the Sinai

Peninsula, Egypt (Figure 2-1). The watershed drains east toward the Gulf of Aqaba and is considered to be the most important wadi in this area because the city of Nuweiba, a tourist destination, and Nuweiba Harbor, are located on the Wadi Watir delta. Nuweiba

Harbor links Egypt with Saudi Arabia and Jordan.

2.2 Geology

The Wadi Watir watershed is underlain by rocks ranging in age from Precambrian to Quaternary (Figures 2 and 3). The Precambrian basement rocks are mainly composed of metamorphic and igneous rocks, including gneisses, metagabbro, metasediments, older gabbros and younger granites that are cut by basic dykes, joints, fractures and fault zones

(EGSMA, 1981). Cambrian rocks include the Araba and Naqous formations (Hassan,

1967; Said, 1971). The Araba Formation (early Cambrian) is formed mainly of thick bedded, grayish and reddish, fine to medium-grained ferruginous sandstone. The Naqous formation (late Cambrian) is composed mainly of fluvial deposits of friable, white to light grey, fine grained, and pebbly quartz with authigenic clay minerals (Issawi and Jax, 1982;

Shabana, 1998; and Abdel-Rahman 2002). The Naqous formation uncomformably overlies the Araba formation and is overlain by the Malha Formation (Abdel-Rahman

2002). Cretaceous rocks are also present throughout the Wadi Watir watershed. They consist of three formations; the Malha Formation (early Cretaceous), which represent a

30

y

r Qw Wadi Deposits a

n Sb Sbkha Deposits g r Qw 29 e ms t Q Undifferentiated

D a (Wadi deposits, sand and gravel)

30 u e

Q Thebes Group Egma F. S gnm n tete e (Chalky limestone with chert bands)

gni c Mt g o tpe Esna F. E (Marine shale) Sudr F. tpe gnm S (Marine chalk with thin

r shale intercalations)

e s

a p

F-1 p Matulla F.

u Mt U

o (Marine sandstone, marl, and shale) tete E b e B c Raha F.

F g a g a t kuc (Fossiliferous limestine with shale

F-2 e

Q intercallations and oyster beds) r g r C e q C Malha F. mgi w M (Fluviatile white sandstone,

g o s S L locally conglomerate) A u o Abu Durba F. g r

e Cd (Dark shale, silt and sandstone, marl

g f kuc f 29 i and limestone intercallation)

gni n gnm o 00 o vd Tertiary Volcanics (Basaltic dikes) Mt b Qw A r

a rc mgi f Ring Complex, Syenitic to alkali-feldspars l C g g Alkalifeldspar granitic rocks "Younger Granite" g Qw Mt g Calc-alkaline granitic rocks "Younger Granite"

u n a

i g Calc-alkaline foliated quartzdioriteic Cd r to granodioritic rocks "Older Granite"

gni M b M gnm G gb Gabbroic rocks Q m Qw rc gni a ha Molasse-type conglomerate

c to siltstone sequence e

0 20 Km. r ms

mva gni Metasediments P ha g mgi mgi Intrausive metagabbro to metadiorite g gni Leucocratic medium to high grade metamorphic rocks 34 10 34 30 34 50 gnm Migmatite, granite gneisses, gneiss, schist, and amphibolite Drainage Cross Sections Major Faults Figure 2-2. Geological map of the study (modified from CONOCO, 1987).

31

North a South South North A B b C D Elevation (m) Wadi Watir Confined 1200 Elevation (m) 1200 Furtaga Well Well 800 No. 19 Springs G.Abu 800 No. 18 Rutha 400 F-2 400 F-1

0.00 0.00 M Dy F2 -400 -400

-800

y

r

a n

West East c i r Qw

Wadi Deposits s Rajabiah Fm.

e

a

t r

a Sandstone with gravel u c u E F J bands and shale beds

Elevation Q

Thebes Group Egma F. e Naqus Fm. e

(meter) tete t

n a

n (Chalky limestone with chert bands) N (Sandstones mainly fluviatile origin

a

e

L

i

c

1200

r

interbeded with shallow marine sand)

o

b E

Esna F. y l

tpe m Araba Fm. r F-1 a

(Marine shale) a Variclolored Sandstone

C

E

with ferruginous bands

600 y

Sudr F. r

a i

S (Marine chalk with thin t r Basic Dykes

shale intercalations) e T Matulla F.

Mt n

a r

0.0 (Marine sandstone, marl, and shale) i

e r

p Metamorphic, metasediments, ignous b p Raha F.

m granites, gabbros and diorites

U s

kuc (Fossiliferous limestine with shale a

c

u

e o

intercallations and oyster beds) r

e c -600 P

a Galala F. t G e (Limestone and dolomite)

r Major Faults

C r

e Malha F. w

o M (Fluviatile white sandstone,

-1200 L locally conglomerate) Water Level Figure 2-3. Hydrogeological cross section A-B, C-D and E-F. Locations Fig. of3 these cross sections are shown on Figure 2 - 2. Cross sections are modified from Yehia, 1998.

32 clastic facies deposit of continental fluviatile environment of white sandstone locally conglomerated (Salah et al., 2010 and CONOCO, 1987), and the Wata and Raha formations (late Cretaceous). The Wata Formation consists mainly of alternative beds of limestone and marl at the base; clay marl sandstone and bands of limestone in the middle; and dolomitic limestone and limestone at the top. The Raha Formation is primarily composed of fossiliferous alternating beds of limestone, shale, and sandy limestone at the base; shale and marl in the middle; and marly limestone and dolomite at the top

(Shabana, 1998). These two formations were deposited in shallow to deep marine open to restricted environments (Gertsch et al., 2008). The Quaternary deposits consist of erosional products of different sedimentary and rocks (El-Shazly et al., 1974; Eyal et al.,

1980) and they are composed mainly of fine-to-course sands, gravels, and boulders of carbonate and rocks embedded in a silty and clayey matrix (El Kiki et al., 1992).

In Figure 2-2, there are two sets of faults that trend northwest-southeast and north- south. Beadnell (1927) and Said (1962) studied these two sets in the El Shiekh Attia area and concluded that the northwest-southeast faults are older than the north-south trending faults.

3. Methods

3.1 Field and Laboratory Methods

Water samples were collected in March 2007. Thirty-four samples were collected from 25 hand-dug wells, four drilled wells, and three springs. Water samples were filtered in the field using a 0.45 μm cellulose acetate filter; samples for major-ion and

33 isotopic analysis were collected in 1-L polyethylene bottles. Depth to water, total depth, pH and electrical conductivity (EC) were measured in the field at the sampling location.

Electrical conductivity was measured with an YSI model 35 conductivity meter. The pH was measured with a WTW model LF 538 pH meter. The EC and pH meters were calibrated once daily. Major-ion water chemistry analyses were conducted at the

Egyptian Desert Research Center Water Central Laboratory using the methods of

Rainwater and Thatcher (1960) and Fishman and Friedman (1985). Calcium and magnesium were determined by titration using Na2EDTA. Sodium and potassium were determined by flame photometry using a standard curve. Carbonate and bicarbonate were determined by titration using sulfuric acid. Sulfate was determined by spectrophotometry.

Chloride was determined by volumetric titration using silver nitrate. Silica, as SiO2, was determined by colorimetry using molybdate [(NH4)6Mo7O24.4H2O].

For quality assurance/quality control, each sample was analyzed in duplicate, and if the difference between the sum of cations and anions is more than 5%, the sample analysis was repeated until an acceptable percent difference was obtained. For flame photometry and the spectrophotometry analysis, a set of standard solutions were measured for every set of 10 samples. If the standard was not verified, then the standards were measured again until verification was met and the samples were re-analyzed.

Stable isotopic analyses were conducted at the University of Nevada, Reno.

Stable isotopic analyses were performed using a Micromass Iso Prime stable isotope ratio mass spectrometer. δ2H analyses were performed using the method of Morrison et al.

(2001). δ2H results are reported in units of ‰ VSMOW with an uncertainty of ±1‰

34

18 (1 standard deviation). δ O analyses were performed using the CO2-H2O equilibration method (Epstein and Mayeda, 1953). δ18O results are reported in units of ‰ VSMOW with an uncertainty of ±0.2‰ (1 standard deviation). Carbon-14 analyses were conducted at the University of Arizona, Accelerator Mass Spectrometry Laboratory and results are reported as percent modern carbon (pmc) with an uncertainty of ±0.4 pmc. Mineralogy was identified by examination of rock thin-sections using a polarizing light microscope.

Rock chemical composition was determined by x-ray fluorescence at the University of

Nevada, Reno.

Ground-surface elevations were surveyed for seven wells in the El Shiekh Attia area and three wells in the delta area using a global positioning system (GPS) LEICA

TCRA1103 PLUS ROBOTIC 3 instrument with a laser reflector total station. These ground-surface elevations were used with depth-to-water measurements to draw water- level contours and determine groundwater flow direction in the El Shiekh Attia area. The digital elevation model DEM (USGS, 2004), map of Shuttle Radar Topographic Mission

(STRM-90) was used to determine the ground-surface elevation for other wells and springs since GPS surveys were not conducted in these areas. These data, along with depth-to-water measurements, were used to determine the general direction of groundwater flow in the main channel aquifers. Because there is a large absolute average vertical elevation error (16 m) for the DEM data (Gorokhovich et al., 2006), water-level measurement points close to each other and having similar water-level elevations less than the DEM elevation error were assigned an average water-level elevation.

35

3.2 Water-Rock Reaction Modeling

Inverse geochemical modeling (Plummer, 1992) has been widely used in interpreting geochemical processes that account for the hydrochemical and isotopic changes in groundwater. The inverse geochemical model NETPATH 2.0 (Plummer et al.,

1994) was used because of its ability to compute the mixing proportions of two to five waters and the net geochemical reactions that account for the observed chemical composition changes in groundwater along a flow path. The geochemical reactions and physical processes that produced the observed water chemistry were evaluated and several possible models are presented that identify water-rock interactions, mixing, and evaporation. Inverse geochemical modeling does not produce unique results, but this type of modeling does produce results that account for the changes in observed water chemistry along a flow path for phases (minerals, salts, or gases) that are known to occur in the aquifers and that are constrained by mineral and gas saturation data. A mineral or gas can only enter the water if the water is under saturated with respect to the mineral or gas, and it can only leave the water (precipitate from, or degas) if the water is super saturated with respect to the mineral or gas.

First, potential groundwater flow paths were identified using water-level information, δ2H and δ18O data, and major-ion chemistry. To determine water-rock interactions that are producing the observed water chemistry, the mineral phases of the aquifers were identified. Mineral phases were obtained from analysis of 12 rock samples from aquifers in the study area (Tables 2-1 and 2-2). Granitic rocks form the backbone of

36 the mountain block recharge area and are the main source of Quaternary alluvial deposits in the study area.

Table 2-1. Results of petrographic analyses of rock samples. Mineral Composition No Area Major Minor and Accessories Alterations Texture Rock Type Main plag, biot, mica magnetite, medium grained 1 argillaceous granite granite Channel ortho, qz chlr plutonic rocks strongly argillized plag with very fine Main 2 plag, qz chlr, silica more chlr and iron oxides, red grained, basalt Channel stained hybabysal intensively argillized felds, biot medium to Main 3 plag, qz, ortho biot altered to iron oxides show coarse grained granite Channel hematization perthetic texture Main argillized granite with 4 plag, ortho biot, apatite granite Channel hematization argillized iron oxides and Main Plag pyrox, augite, oliv, serp pyrox replaced by serp, chlr olivine basalt 5 Channel and serp strongly iron stained

Main iron oxides, oliv, 6 pyrox, plag medium grained gabbros Channel magnetite, illmnite Main 7 plag, qz, ortho biot, muscovite, zircon less argillized felds medium grained biotite granite Channel Wadi ortho is strongly weathered, 8 ortho, plag, qz biot biotite granite El-Ain plag with oxidized biot Wadi plag replaced by clay and medium to 9 plag, pyrox, qz chlr, magnetite, clay gabbros El-Ain pyrox by chlr coarse grained Wadi clay minerals, chlr, pyrox altered to chlr and plag medium to 10 plag, pyrox gabbros El-Ain zircon, opaqoues to clay coarse grained

Wadi qz, ortho, plag, coarse grained 11 opaqoues weak to moderate argillization granite El-Ain felds graphic texture Wadi no opaqoues no apatite oxidation of biot with iron medium grained 12 plag, ortho, qz granite El-Ain no zircon stained texture Plag-plagioclase; ortho-orthoclase; felds -feldspars; qz-quartz; biot-biotite; chlr-chlorite; serp-serpentine; pyrox-pyroxene; oliv- olivine.

Table 2-2. Results of X-ray fluorescence of granite samples. Rock Loss on Basin No. SiO TiO Al O Fe O MnO MgO CaO Na O K O P O Type 2 2 2 3 2 3 2 2 2 5 Ignition Main Granite 1 73.41 0.15 15.91 1.11 0.04 0.02 1.28 3.53 3.65 0.06 0.83 Channel Main Granite 3 75.03 0.16 14.23 1.77 0.03 0.00 0.63 3.04 4.70 0.01 0.41 Channel Main Granite Channel 4 74.77 0.14 13.90 1.63 0.04 0.01 1.02 2.81 4.66 0.02 1.00 Main Granite Channel 7 74.26 0.16 14.94 1.38 0.04 0.26 1.39 2.95 3.92 0.05 0.65 Wadi El Granite Ain 8 76.48 0.08 12.32 1.15 0.02 0.00 0.49 2.28 4.69 0.01 2.48 Wadi El Granite Ain 11 77.85 0.08 12.67 1.30 0.01 0.04 0.20 1.98 5.13 0.02 0.72 Wadi El Granite 12 78.73 0.08 12.12 1.23 0.02 0.00 0.45 2.26 4.67 0.00 0.43 Ain

37

The water-rock reaction models developed for this study were constrained by the major-ion concentrations of the groundwater and the mineral phases in the aquifers of the study area (Tables 2-3 and 2-4). Halite was included as a phase in NETPATH models because of its occurrence in carbonate rocks of marine origin (Raha and Wata

Formations) and because it is embedded in terrestrial sediments. Calcite and dolomite were included as phases in the model because of the presence of carbonate rocks in the study area. Clay sheets are present in the sediments so the clay minerals montmorillinite and illite were also included in the model. X-ray fluorescence results for granitic samples are presented in Table 2-2. These data were used to determine the average chemical composition for a composite granite for the NETPATH models that represents felsic igneous rocks found throughout the study area (Ca0.33Mg0.01Na1.02K1.72Si21Al3.74Fe0.74).

Table 2-3. Constraints, phases, and parameters used in NETPATH models. Constraints Phases Parameters Calcite, Composite Granite Calcium, Carbon, (Ca Mg Na K Si Al 4Fe ), 0.33 0.01 1.02 1.72 21 3.7 0.74 Evaporation, Mixing, Magnesium, Dolomite, Cation Exchange, Gypsum, Carbon Isotopic Potassium, Sodium, Halite, Illite, Exchange Sulfur Mafic Montmorillonite, Na-Montmorillonite, Silica,

Samples with major-ion analysis electroneutrality error of greater than 5% were not used in water-rock reaction models. Major-ion analyses were not corrected to 0% electroneutrality so any electroneutrality error was carried through each model. An acceptance criterion for a valid water-rock reaction model was the precipitation or dissolution of no more than 10 mmol/L of a phase, unless one of the major ions making

38

Table 2-4. Chemical (meq/L) and isotopic (‰) data for groundwater samples collected in March 2007.

TD- Total Depth (meter); DTW-Depth To Water (meter); W.L-Water Level Above Sea Level (meter); AQ- Aquifer Type; Qt-Quaternary; Bs-Basement; LC-Lower Cretaceous; UC-Upper Cretaceous; epm- equivalents per million; -- no data; Fwg-Flowing; * - Local rain water chemistry, fter El-Sayed, 2006; ** Local rain isotopic data from 1960-1987 after (Abdel Samie 1995 and Abd El Samie and Sadek 2001); *** water chemistry data after Shalaby, 1997; hydrologic parameters after Yehia, 1998; and isotopic data after JICA, 1999.

39 up the phase in question had a dissolved concentration greater than 10 mmol/L in the initial and final waters. In this case, more than 10 mmol/L of a phase could precipitate or dissolve, but the amount of the phase precipitating or dissolving could not be greater than twice the dissolved concentration of the major ion making up the phase in question.

4. Results and Discussion

4.1 Groundwater Flow System

The study area contains four main water bearing formations that are recharged by flash floods when the area receives heavy winter storms (October to April) and by infiltration of precipitation on the mountain block (RIWR, 1989, Himida, 1997 and JICA,

1999). For the Wadi Watir basin catchment area, the annual rainfall recharge between

1998 and 2007 showed an average value of 192.7x106 m3/yr (Milewski et al., 2009). This is close to the average annual value of 164.9x106 m3/yr estimated by Masoud (2009) between 1960 and 1990. Sultan et al. (2011) estimated the total average annual recharge for the Nubian Sandstone aquifer in Sinai between the periods 1998 and 2007 at approximately 13x106 m3/year, while the local average annual recharge for the Nubian outcrops in Wadi Watir watershed is 1.83x106 m3/year. Elewa et al. (2011) ranked the

Nubian Sandstone aquifer (Lower Cretaceous) in eastern Sinai and the Furtaga Springs areas in Wadi Watir as promising areas with very high annual groundwater recharge.

Wadi Watir comprises many aquifers ranging from land surface to 800 m deeper under the ground surface. These aquifers include the Quaternary alluvial aquifer, the upper

Cretaceous (Wata and Raha formations) carbonate aquifer, the lower Cretaceous

40 sandstone aquifer (Malha Formation), and the Precambrian granitic aquifer. The

Quaternary, upper Cretaceous, and Precambrian aquifers are generally unconfined, water- table aquifers, while the lower Cretaceous is a confined aquifer. The Precambrian aquifer underlies most of the study area (Figure 2-3 a-c). The Quaternary aquifer is present in the eastern Wadi Watir alluvial fan, the main channel of Wadi Watir, the Wadi El Ain area, and in the El Shiekh Attia area (Figures 2-1 and 2-5). Water-level contours in the

Quaternary alluvial aquifer of the El Shiekh Attia area show that groundwater flows from northeast to southwest, generally following the surface-water drainage (Figure 2-4). The average groundwater elevation for wells tapping the Quaternary aquifer in the El Shiekh

Attia and Wadi El Ain areas is 559 m and 641 m, respectively; further downstream in the alluvial aquifer, the water level is 237 m at site 16 and 0.9 m (average) in the delta at sites

17, a, and b (Figure 2-5 and Table 2-4) . These data indicate that groundwater in the

Quaternary alluvial aquifer flows down gradient in the drainages to the delta. The groundwater in Wadi El Ain occurs in three different aquifers, the Quaternary, the upper

Cretaceous, and the Precambrian. Water-level elevations are highest in the upper

Cretaceous at sites 20 (688 m) and 21 (660 m) (Figures 2-1 and 2-5, Table 2-4). Water- levels are lower in the Precambrian (643 m, average of sites 25 and 27) and in the

Quaternary alluvial aquifer (641 m, average of sites 13 and 14) (Figures 2-1 and 2-5,

Table 2-4).

41

a El Shiekh Attia 1

23 5 7

8 9 6 4 0 1500 m

Scale

Figure 2-4. Water level contours (meters) in the El Shiekh Attia area.

The lower Cretaceous aquifer occurs locally in the upper reach of the Main

Channel (sites 22, 23, and 24; Figure 2-1) and extends northward in the central Sinai

Peninsula (Figure 2-5). Water in the confined, lower Cretaceous aquifer (Yehia, 1998) flows northward from sites 22, 23, and 24 (average water-level elevation 469 m) toward sites 18 (430 m) and 19 (419 m) (Figure 2-5). The Precambrian aquifer outcrops in the mountains and underlies most of the study area (Figures 2-3 and 2-5). This aquifer outcrops at Wadi El Ain (sites 25, 26, 27, and 28; Figure 2-1) and in the Main Channel at

Furtaga Springs (sites 29, 30, 31; Figure 2-1, Table 2-4). The average water–level elevation for wells tapping the Precambrian aquifer at Wadi El Ain is 643 m (sites 25 and

42

29 (19) 419 30

29

a 20 b (18) El-Hazim 430 a q Plateu (2-7 & 9) [559] A 29 (22-24) f [469] o 10 f l (21) (25&27) [643] 660 (29-31) (16) u [347] 237 G (13&14) ( 17, a & b) 29 [641] (15) [0.9] 257 00

(20) 0 20 Km. 688

34 30 34 40 34 50 Water Wells tapping Quaternary Carbonate massive The Quaternary Aquifer Aquifer forming the watershed Water Wells tapping The Cenomanian-Turonian Upper Cretaceous (Upper Cretaceous Aquifer Aquifer (Site No. ) Water Wells tapping Lower Cretaceous Water Level (m) The Lower Cretaceous Aquifer Aquifer [Average water level] Water Wells tapping Precambrian The Precambrian Aquifer Aquifer

Figure 2-5. General Hydrogeological map of the study area.

43

27) while it is much lower in the Main Channel at an average of 347 m at Furtaga Springs

(sites 29-31) (Figure 2-5). At Furtaga Springs, low permeability dykes create barriers to groundwater flow in this area of the Main Channel (Figure 2-3a).

4.2 Environmental Isotopes

Hydrogen (δ2H) and oxygen (δ18O) isotopic ratios of water are ideal tracers that can be used to determine the source(s) and mixing of groundwater because they are part of the water molecule, are not involved in geochemical reactions, and are sensitive to physical processes such as groundwater mixing and evaporation (Dansgaard, 1964; Clark and Fritz, 1997). In the study area, the isotopic composition of groundwater varies according to the location and the source(s) of groundwater recharge. The stable isotope ratios of hydrogen and oxygen of groundwater, along with water chemistry data, are shown in Table 2-4.

The Wadi Watir watershed contains four main water bearing formations that are recharged by flash floods when the area receives large winter storms (October to April) and by infiltration of precipitation on the mountain block (RIWR, 1989, Himida, 1997 and JICA, 1999). These storms mainly originate from the eastern Mediterranean (Dames and Moore, 1983). Airflow from the Mediterranean Sea is the main source of precipitation that falls during winter (Greenwood, 1997). Rainfall amounts and the isotopic composition (δ18O and δ2H) were obtained from the Global Network of Isotopes in Precipitation (GNIP). The International Atomic Energy Agency (IAEA), in collaboration with the World Meteorological Organization (WMO), established a worldwide network of meteorological stations to monitor the δ18O and δ2H composition

44 of precipitation (IAEA and WISER, 2008)

(http://www.univie.ac.at/cartography/project/wiser/ index.php). These data were used to calculate the monthly and annual precipitation amount weighted isotopic average of rainfall for the Wadi Watir watershed. Seven stations were used including El Arish and

Rafah Stations in Sinai, Shoubak in Jordan, and Sadoth, Bet Dagan, Ramond and Soreq in Israel (see the meteorological location sites in Figure 2-1, Chapter I). These stations were selected according to the availability and continuity of data and their location sites.

Figure 2-6 shows the monthly average δ18O and the monthly average amount of precipitation for each station. From these data, most precipitation occurs during the winter from October to April and has a more depleted isotopic signature than precipitation in September and May. The precipitation amount weighted isotopic average for each month for all stations is presented on Figure 2-7. This plot shows that amount weighted isotopic values for September and May are isotopically heavier than precipitation during the winter. The precipitation amount weighted average isotopic value for all precipitation between October and April was -4.99 and -21.37 for δ18O and δ2H, respectively, which plotted close to the Mediterranean Meteoric Water Line (Gat et al.,

1969; IAEA 1981) and represented the source of groundwater recharge for the Wadi

Watir watershed. The Sinai Peninsula is located in an arid to hyper arid region

(UNESCO, 1963; Cools et al., 2012) so precipitations that fell in May and September were relatively small amounts and were isotopically enriched Precipitation events during these months likely did not contribute to recharging the aquifers located in the Wadi

Watir watershed.

45

Figure 2-6. Amount of Preciptation (mm) versus δ18O (‰) in rainwater for some selected stations in Sinai (El Arish and Rafah); Israel (Soreq, Sadoth, Ramond and bet dagan) and Jourdan (Shoubak). The location sites for these stations are indicated in (Figure 2-1a, Chapter 2).

46

40 30 20 10 ) 0

‰ Sept -10 Oct

H ( H Mar Nov 2 -20 Feb May δ Jan Apr -30 Dec -40 Weighted average (Winter Rain) -50 -60 -9 -7 -5 -3 -1 1 δ18O (‰)

Figure 2-7. δ18O (‰) versus δ2H (‰) plot for the rainwater during the rainy seasons in the study area (October to Aprill). Weighted average of Winter Rain is estimated by - 4.99 and -21.37 for δ18O and δ2H, respectively. Global Meteoric Water Line after (GMWL; Craig, 1961) and Mediterranean Meteoric Water Line after (MMWL; Gat et al. (1969) and IAEA (1981).

Groundwater from the Quaternary alluvial aquifer in the El Shiekh Attia area

(Figure 2-8) plots close to the Global Meteoric Water Line (GMWL) (Craig, 1961), and is plotted on the evaporation trend line to verify the equation δ2H=2.17 * δ18O - 10.59 and issues from the average of weighted winter rain (IAEA and WISER, 2008), and is low in chloride (Figure 2-9). These data indicate that the main source of groundwater in the Quaternary alluvial aquifer in the El Shiekh Attia area is recent precipitation, which comes from the eastern Mediterranean (Dames and Moore, 1983 and Greenwood, 1997) as heavy winter storms from October to April (RIWR, 1989, Himida, 1997 and JICA,

1999).

47

Figure 2-8. Relationship between δ18O and δ2H for groundwater in the El Shiekh Attia, Wadi El Ain, and the Main Channel areas. The samples outside of the El Shiekh Attia samples circle and Wadi El Ain samples ellipse are the Main Channel area samples. Global Meteoric Water Line (GMWL; Craig, 1961) and Mediterranean Meteoric Water Line (MMWL; Gat et al. (1969) and IAEA (1981).

Figure 2-9. Relationship between δ18O and Cl for groundwater in the El Shiekh Attia, Wadi El Ain, and the Main Channel areas. The samples outside of the El Shiekh Attia samples circle and Wadi El Ain samples ellipse are the Main Channel area samples.

48

The isotopic signatures of groundwater in the Quaternary alluvial aquifer, upper

Cretaceous carbonate aquifer, and Precambrian granitic aquifer in the Wadi El Ain area

(Figures 2-1 and 2-8) are isotopically depleted relative to weighted average of winter rain and groundwater from the El Shiekh Attia area, and have much higher chloride (Figure 2-

9). These data suggest that groundwater in Wadi El Ain is derived from the east

Mediterranean storms, but is more depleted than the isotopic signature of average winter rain because of increased rain out effect as storms continue moving southward.

According to Aggour et al. ( 2000), the Wadi Watir watershed is subdivided into two main sub basins; Wadi Watir (Trend N-S) and Wadi El Ain (trends NE-SW). The Wadi

Watir trend is structurally and topographically low and controlled by El Shiekh Attia fault (Said, 1962). According to Cools et al., 2012; The Wadi El Ain watershed area is relatively higher (exceed 1250 m above sea level) than the Wadi Watir watershed (750-

1000 above sea level). The elevation effect fractionates the isotopic signatures due to rainout of δ18O and δ2H , where the depletion varies between -0.15 to -0.5 ‰ per 100 meter (Clark and Fritz, 1997).

Groundwater in the lower Cretaceous aquifer flows northward from the Main

Channel area; site 19 in the very northern part of the Wadi Watir watershed is the isotopically lightest groundwater found in the watershed and the winter rain weighted average (Figures 2-1 and 2-8) and has considerable chloride (Figure 2-9). Site 18, which is located between sites 22, 23, and 24 in the Main Channel area and site 19 (Figure 2-1), is isotopically very similar to recent precipitation and groundwater in the alluvial aquifer in El Shiekh Attia, and has low chloride. Groundwater in the lower Cretaceous aquifer in

49 the Main Channel area at site 23 appears to be a mixture of winter rain and isotopically light groundwater from the lower Cretaceous aquifer similar to that found at site 19; site

23 also lies close to a mixing line between winter rain and site 19 on Figure 2-8.

Similarly, site 23 lies near a mixing line between winter rain (and site 8) and site 19 in

Figure 2-9 comparing isotopic signatures and chloride concentrations. These interpretations are consistent with those found by Yehiya (1998) and Abd El Samie and

Sadek (2001).

Isotopic data for groundwater in the Precambrian aquifer in the Main Channel area (sites 29 and 31; Figures 1 and 8) fall on an evaporation line originating from

Quaternary alluvial aquifer groundwater in the El Shiekh Attia area. These data suggest that groundwater in the Precambrian aquifer in the Main Channel area is derived either from the up-gradient El Shiekh Attia area or local recharge from winter rain. Note that the chloride concentrations in sites 29 and 31 have not increased because of evaporation and are similar to chloride concentrations in El Shiekh Attia (Figure 2-9). Chloride data suggest that local recharge of recent precipitation may be the more likely source of water at sites 29 and 31.

Two groundwater samples in the study area, sites 15 and 27, have undergone substantial evaporation and lie along an evaporation lines from winter rain and groundwater in the Wadi El Ain area, respectively (Figure 2-8). Site 27 is located within the Wadi El Ain area in the Precambrian aquifer (Figure 2-1). Site 15 is located on a tributary to the Main Channel (Figure 2-1) in the Quaternary alluvial aquifer; although site 15 has a similar evaporated isotopic signature to that of site 27, it is geographically

50

(and hydrologically) unrelated to groundwater in the Wadi El Ain area. Using the average

δ18O value for un-evaporated groundwater from the Wadi El Ain area (Figure 2-8; Table

2-4; average of sites 13, 14, 20, 21, 25, 26, and 28) and the Rayleigh equation (Clark and

Fritz 1997), an evaporation factor for site 27 is 1.2. Moreover some of the evaporation observed for shallow groundwaters is likely directly from hand-dug wells; because hand- dug wells are generally about one meter in diameter so that the large water surface area in the open well allows for evaporation of water in a hand dug well especially if they are not withdrawing much water from the well.

Wells 17, a, b, and c are located in the Quaternary alluvial aquifer in the Wadi

Watir delta, down gradient from the Main Channel area (Figure 2-1). Groundwater from these wells lies on an evaporation line originating from the up gradient El Shiekh Attia area (Figure 2-8). Isotopic data suggest that groundwater in the Quaternary alluvial aquifer originates either from the El Shiekh Attia area or local recharge of winter rain.

Chloride concentrations in the Wadi Watir delta are much higher than in the El Shiekh

Attia area or up-gradient groundwater in the Main Channel area (Figure 2-9).

Groundwater in the delta is likely influenced by underlying saline groundwater reported by (El-Refeai, 1992; El Kiki, 1992 and El Sayed, 2006) which may be upwelling because of increased groundwater pumping.

4.3 Water-Rock Reaction Models using NETPATH

In the El Shiekh Attia area, groundwater in the Quaternary alluvial aquifer flows down gradient from northeast to southwest, increases in dissolved solids (Table 2-4), and then flows toward the Main Channel area. Groundwater in this area is not evaporated

51

(Figures 2-8 and 2-9). Geochemical modeling results suggest that gypsum, dolomite, halite, and granite dissolve as groundwater flows down gradient (e.g. site 2 to site 4 or site 9) while calcite and clays are precipitated and some cation exchange occurs (Table 2-

5). Calculated mineral saturation indices (SI) are consistent with changes in mineral phases in NETPATH models (Table 2-6).

Table 2-5. NETPATH modeling results (mmol/L) for the Wadi Watir watershed.

Positive values mean the phase is going into solution while negative values mean the phase is being removed from the solution. Cal=Calcite; Gyp=Gypsum; Dol=Dolomite; Ilt=Illite; Mont-Maf=Mafic Montmorillonite; Na-Mont=Sodium Montmorillonite; Gr=Composite Granite; Biot=Biotite; Hal=Halite; Ex-Cation Exchange; -- No Data

Table 2-6. Mineral saturation indices for phases in NETPATH geochemical models. Basin No. Calcite Gypsum Dolomite Illite Ca-Mont Albite Anorth Kspar Chalcedy Halite El Shiekh Attia 2 0.23 -1.13 0.50 0.88 1.12 -1.32 -3.58 0.03 0.11 -4.59 El Shiekh Attia 4 -0.29 -0.96 -0.65 1.56 2.32 -1.13 -3.56 0.01 0.13 -4.22 El Shiekh Attia 9 0.14 -0.90 0.27 0.89 1.44 -1.62 -3.81 -0.42 -0.03 -4.20 Wadi El Ain 14 0.26 -0.64 0.49 1.03 1.04 -1.64 -3.91 0.12 -0.12 -3.12 Delta Watir 17 0.33 -0.54 0.39 0.17 0.41 -1.39 -3.92 -0.60 -0.14 -3.69 El Shiekh Attia 18 0.65 -1.04 1.44 -2.23 -2.09 -2.95 -4.33 -1.90 -0.21 -4.18 Wadi El Ain 20 0.22 -0.54 0.37 0.15 0.37 -1.98 -4.08 -0.77 -0.23 -3.35 Main Channel 23 0.24 -0.57 0.41 1.29 1.50 -1.50 -3.71 0.15 -0.07 -3.73 Wadi El Ain 25 0.13 -0.27 0.12 1.82 2.04 -1.14 -3.66 0.53 -0.04 -3.05 Main Channel 31 0.35 -0.78 0.48 1.54 1.84 -0.86 -3.27 0.55 0.22 -4.27

Positive values indicate over saturation; negative values indicate under saturation.

In the Wadi El Ain area, groundwater occurs in three different aquifers, yet all are similar isotopically indicating that all are derived from the same source. Up gradient groundwater (site 20) in the upper Cretaceous aquifer flows down gradient to the

Precambrian (e.g. sites 25 and 27) and Quaternary (e.g. site 14) aquifers. Water-rock

52 reaction models for flow from the upper Cretaceous to the Precambrian suggest that gypsum, dolomite, granite, and halite dissolve while calcite and clays are formed with some cation exchange (Table 2-5).

Models suggest that Quaternary alluvial aquifer groundwater is a mixture of upper

Cretaceous (66%) groundwater and recent precipitation (3%) with dissolution of gypsum, dolomite, granite, and halite while calcite and clays are formed (Table 2-5). Petrographic analyses of rocks from the Wadi El Ain area (Table 2-1) are consistent with the modeled dissolution of igneous rocks and formation of clays. Calculated mineral SIs are consistent with dissolution and precipitation of mineral phases in NETPATH models (Table 2-6) including de-dolomitization driving the dissolution of dolomite despite over-saturated SIs

(Back et al., 1983).

Water-rock reaction modeling of groundwater flow in the lower Cretaceous aquifer suggests that groundwater at site 18 is a mixture of up-gradient groundwater from site 23 (36%) and recent precipitation (64%). Models suggest dissolution of gypsum, dolomite, and halite while calcite and clays are formed (Table 2-6).

Groundwater in the Precambrian aquifer in the central Main Channel area emanates from Furtaga springs (sites 29, 30, 31; Figures 1 and 5). Water-rock reaction models of this groundwater can be made by mixing up-gradient groundwater from El

Shiekh Attia area (e.g. site 2; 94%) and recent precipitation (6%; Table 2-6) or by simply reacting El Shiekh Attia area groundwater with minerals. These models suggest dissolution of gypsum, dolomite, composite granite, and halite while calcite and clays are formed with some cation exchange (Table 2-6).

53

From the Main Channel, groundwater flows down gradient into the Wadi Watir delta (e.g. site 17). Water-rock reaction models of groundwater in the Quaternary alluvial aquifer in the delta can be made by with up-gradient groundwater from the Precambrian aquifer at Furtaga Springs (e.g. site 31) dissolving gypsum, biotite, and halite while precipitating calcite and clays with cation exchange (Table 2-6). Wadi Watir delta groundwater can also be derived from up-gradient Quaternary alluvial aquifer water (e.g. site 2) by dissolving gypsum, dolomite, biotite, and halite while precipitating calcite and clays with cation exchange. Successful water-rock reaction models could also be produced that had up-gradient alluvial aquifer water (70% site 2) mixing with recent precipitation (30%) to produce Wadi Watir delta groundwater by dissolving gypsum, dolomite, biotite, and halite while precipitating calcite and with cation exchange.

4.4 Groundwater Ages

Carbon-13 (δ13C) and carbon-14 (14C) were analyzed in groundwater samples from one well in the El Shiekh Attia area (site 1); three wells (sites 18, 20, and 23) and two springs (sites 29 and 31) in the Main Channel area; and three wells (site a, b and c) in the Wadi Watir delta area (Table 2-4). Water-rock reaction models were developed to explain the changes in water chemistry along flow paths, including mixing of different waters (Table 2-5), and to correct groundwater ages as needed for interaction of groundwater carbon isotopes with aquifer materials. The carbon isotope age correction models adjusted groundwater 14C activity for various geochemical reactions including inorganic carbon reactions in the soil zone, carbonate mineral dissolution, and isotopic exchange with carbonate minerals in the aquifer matrix. Carbon-14 ages were corrected

54 using the Original Data Method in NETPATH; this method has been used to highlight the influence of isotopic exchange and carbonate dissolution through the flow path (Plummer et al, 1994).

13 14 The δ C value and C activity of CaCO3 in the aquifer matrix were assumed to be 0 ‰ and 0 pmc, respectively. When correcting groundwater ages, the 14C activity of

14 the initial water (A0) at, or near, the recharge area and the C activity of groundwater along a flow path must be defined (Wigely et al., 1978). The initial water may be a groundwater sample from up gradient, or an assumption that represents the water chemistry at the recharge area (Van der Kemp et al. 2000). For modeling, carbon isotopic data for site 1 in the El Shiekh Attia area was used as the initial water for estimating down-gradient groundwater ages because: (1) it was the most up-gradient water in the El

Shiekh Attia area, which supplies most of the water to the down-gradient Main Channel area; (2) it was the most dilute sample in the project area (expect for one spring); (3) it contained the most negative δ13C value indicative of recent recharge through the soil zone; and (4) it had a 14C activity of 74 percent modern carbon (pmc), which combined with a δ13C value of -10 ‰ indicates that this groundwater is modern in age (<1000 years old).

All model corrected 14C groundwater ages were modern, except for groundwater at site 18, which had a corrected 14C of 4,600 years (Table 2-7).

55

Table 2-7. Calculated groundwater ages using NETPATH. Initial Water Carbon 13C 14C Final Isotopic Age Basin Site Site Computed Observed Computed Observed Ev Water Exchange (years) 1 2 (‰) (‰) (pmc) (pmc) (mmol/L) Wadi El 1 20 1.7 -5.45 -5.70 34.88 80.5 Modern Ain 1 23 18 0.1 -8.11 -8.10 57.05 32.6 -- 4,626 Main 1 29 0.1 -7.48 -7.9 53.65 73.5 -- Modern Channel 1 31 0.1 -9.79 -9.70 71.49 73.5 1.14 Modern 1 a c 0.4 -4.17 -4.20 25.70 30.5 -- Modern Site 1 represents recent groundwater recharge. Ev-Evaporation Factor

The drilled well at site 18 taps into the lower Cretaceous aquifer. Shiftan (1961) reported that the water in the Nubian Sandstone aquifer in the study area is of Pleistocene age. Issar (1972) used 14C to date groundwater in this aquifer and reported ages between

13,000 and 30,000 years. Abd El Samie and Sadek (2001) estimated the age for groundwater at site 19 in the lower Cretaceous aquifer to be 22,000 years. The well at site

18, although tapping the lower Cretaceous aquifer, is clearly a mixture of recent precipitation and older groundwater in the lower Cretaceous aquifer as indicated by isotopic data, chemistry data, and water-rock reaction modeling. Sultan et al. (2011) reported that the Nubian sandstone aquifer in Sinai was primarily recharged by previous wet climates, and currently receives modern meteoric recharge under dry climatic conditions. Rosenthal et al. (2007) confirm that the lower Cretaceous aquifer in the central eastern Sinai, and at the Negev comprise paleo water mostly recharged and replenished during the Pleistocene age.

Groundwater in the Precambrian aquifer and down gradient in the Quaternary alluvial aquifer in the Main Channel and in the Wadi Watir delta are modern in age

(Table 2-7) indicating that groundwater is from recent precipitation and not groundwater recharged under a different climatic regime.

56

5. Summary and Conclusions

The study area has four primary aquifer units: Quaternary alluvium, upper

Cretaceous carbonate, lower Cretaceous sandstone, and Precambrian granite.

Groundwater in the El Shiekh Attia area has isotopic signatures that are not evaporated and are similar to the winter rain indicating that recent precipitation is the main source of groundwater recharge to the Quaternary alluvial aquifer in this area. Groundwater in the

Wadi El Ain area is isotopically depleted relative to groundwater in the El Shiekh Attia area and the winter rain, due to rainout and depletion of isotopic signatures as a result of latitude and altitude effect. Groundwater in the Main Channel area is primarily derived from the up-gradient El Shiekh Attia area and recent precipitation as indicated by isotopic and water chemistry data and water-rock reaction modeling. Groundwater chemistry in the study area evolves as it flows from up-gradient, dilute water in the El Shiekh Attia area (site 1), down gradient through the El Sheikh Attia area, and then down the Main

Channel area to the Wadi Watir delta (sites 17, a, b, and c). Dissolution of gypsum, halite, and igneous rocks are the primary contributors to increases in major-ion concentrations, while precipitation of calcite and formation of clays remove ions from solution.

Groundwater chemistry in the down gradient part of the Wadi El Ain area also results from mixing groundwater with different chemistries in addition to the water-rock reactions.

Corrected 14C groundwater ages in the study area are all modern, except for one sample from the lower Cretaceous aquifer that has a corrected age of 4,600 years. These groundwater ages have very important implications for groundwater management in the

57 area. Groundwater being pumped in the area is older groundwater recharged under a different climatic regime, but is being replenished by recent recharge. Thus, this resource can be managed in a sustainable way by not pumping more than is being recharged on an average annual basis. If this groundwater resource was much older, then it could only be managed as a one-time use and not as a sustainable resource.

6. Acknowledgements

Acknowledgement goes to the Egyptian Cultural and Educational Bureau at

Washington DC for funding this work and to Dr. Simon Poulson, Department of

Geological Sciences and Engineering, University of Nevada, Reno, for conducting the stable isotopic analyses.

7. References

Abdel-Rahman IF (2002) Physical and chemical characteristics of silica sand deposits of Wadi Watir region, Sinai, Acta Mineralogica-Petrographica, Szeged 2002, 43:79- 83. Abd El-Samei SG (1995) Isotopic and hydrochemical studies on the groundwater of Sinai Peninsula Ph.D Thesis, chemistry dept. Faculty of science, Ain Shams Univ. Table 22, p. 125. Abd El Samei SG and Sadek MA (2001) Groundwater recharge and flow in the Lower Cretaceous Nubian Sandstone aquifer in the Sinai Peninsula, using isotopic techniques and hydrochemistry, Hydrogeology Journal 9:378-389. Aggour, T.A., Shabana, A.R., Shided, A.G., and Yehia, M.M., 2000. Hydrogeological conditions of the water bearing formations in Wadi Watir basin with emphasis on the deep ones. 2nd International Conference on Basic Science and Advanced Technology. Fac. Sci., Assiut Univ., 13 p.

58

Back W, Hanshaw BB, Plummer LN, Rahn PH., Rightmire CT, Rubin M (1983) Process and rate of dedolomitization—Mass transfer and 14C dating in a regional carbonate aquifer. Geol Soc Am Bull 94(12):1415-1429. Beadnell HJL (1927) The wilderness of Sinai. Arnold, London, 180 pp. Clark I, Fritz P (1997) Environmental Isotopes in Hydrogeology. Lewis Publishers, New York. CONOCO (1987) Geologic map of Egypt, scale 1:500,000. Continental Oil Company, Egypt. Cools. J., Vanderkimpen P., El Afandi, Abdelkhalek A., Fockedey S., El Sammany M., Abdallah G., El Bihery M., Bauwens W., and Huygens M., 2012. An early warning system for flash floods in hyper-arid Egypt. Nat. Hazards Earth Syst. Sci., 12, 443–457, 2012. Craig H (1961) Isotopic variations in meteoric waters. Science 133:1702-1703. Dames and Moore 1983. Sinai development study, final report. Submitted to the advisory committee for reconstructions, ministry of development, Arab Republic of Egypt. Dansgaard W (1964) Stable isotopes in precipitation. Tellus 16:436–468. EGSMA (1981) Geological Map of Egypt, Scale 1:2000000. El Beialy YS, Head JM, and El Atfy SH (2010) Palynology of the Mid-Cretaceous Malha and Galala Formations, Gebel El Minshera, North Sinai, Egypt, PALAIOS, 25:517-526. doi: 0.2110/palo.2009.p09-128r. Elewa H, Qaddah A (2011) Groundwater potentiality mapping in the Sinai Peninsula, Egypt, using remote sensing and GIS-watershed-based modeling. Hydrogeology Journal 19:613-628. doi 10.1007/s10040-011-0703-8. El Kiki M.F., Eweida EA, El-Refeai AA (1992) Hydrogeology of the Aqaba rift border province. Proc 3rd Conf Geol Sinai Develop, Ismailia, Egypt, 91-100. El-Refeai AA (1992) Water resources of southern Sinai, Egypt geomorphological and hydrogeological studies. Ph.D. Thesis, Fac. Sci., Cairo Univ., Egypt, 357p.

59

El Shayeb H (1994) Geological and geophysical exploration for groundwater in central Sinai, Egypt. Ph.D thesis, Fac. of Sc. El Menoufiya Univ., 172 P. El-Sayed M.H (2006) Comparative study of water quality of the Quaternary aquifer in Wadi Watir Basin and its delta, Southeast Sinai, Egypt. Egyptian J Desert Res 56(1):17-46. El-Shazly EM, Abddel-Hady MA, El-Ghawaby MA, El-Kassas IA, El-Shazly MM (1974) Geology of Sinai Peninsula from ERTS-1 satellite images. Acad Sci Res Tech Remote Sensing Project . Epstein S, Mayeda T (1953) Variation of 18O content of waters from natural sources. Geochim Cosmochim Acta 4:213-224. Eyal M, Bartov Y, Shimron AE, Bentor YK (1980) Sinai geologic map, scale 1:500,000. Ministry of Energy and Infrastructure, Res Earth Sci Israel. Farid MS, Yehia M, El Deftar T (1995) Annual Conf of the National Water Research Center, Ministry of Publication Works and Water Resources, 26–28 Dec, Cairo. Fishman, MJ and Friedman LC (1985) Methods for determination of inorganic substances in water and fluvial sediments. U.S. Geol. Surv. Book 5,Chapter A1.Open File Report 85-495, Denver, Colorado, U.S.A. Gat, J.R., Mazor, E. and Tzur, Y., 1969. The stable isotope composition of mineral waters inthe Jordan Rift Valley, Israel. J. Hydrology, 76: 334-352. Gertsch B, Keller G, Adatte T, Berner Z, Kassab AS, Tantawy AAA, El-Sabagh AM, and Stueben D (2008) Cenomanian–Turonian transition in a shallow water sequence of the Sinai, Egypt, International Journal of Earth Science (Geol Rundsch) doi:10.1007/s00531-008-0374-4. Gorokhovich Y, Voustianiouk A (2006) Accuracy assessment of the processed STRM- based elevation data by CGIAR using field data from USA and Thailand and its relation to the terrain characteristics. Greenwood, N.H., 1997. The Sinai: A Physical Geography. University of Texas Press.

60

Hassan AA (1967): A new Carboniferous occurrence in the Abu Durba, Sinai, Egypt. 6th Arab. Petroleum Conference, Baghdad, 2, 8p. I.A.E.A., 1981. Stable isotope hydrology. Deuterium and oxygen-18 in water cycle. In: J.R. Gat,and R. Gonfiantini, (ed.), International Atomic Energy Agency Technical Report No. 210,Vienna, 339 p. IAEA and WISER, 2008. Water Isotope System for data analysis, visualization, and Electronic Retrieval: The WISER Version 0.7. Issawi B, Jux U (1982): Contribution on the stratigraphy of the Paleozoic rocks in Egypt. Geological Survey of Egypt, 64, 28. Himida IH (1997) Water resources of Wadi Watir. Internal report, Desert Research Center, Hydrology Dept. (In Arabic). IAEA and WISER, 2008, Water Isotope System for data analysis, visualization, and Electronic Retrieval: The WISER Version 0.7. Issar AAB, and Michaeli A (1972) On the Ancient Water of the Upper Nubian Sandstone Aquifer in Central Sinai and Southern Israel. Journal of Hydrology 17. Ismail Y.L (1998) Hydrogeological and hydrochemical studies on Wadi Watir area, South Sinai, Egypt. PhD Thesis, Fac Sci, Suez Canal Univ, Egypt. JICA - Japan International Cooperation Agency (1999) South Sinai Groundwater Resources Study in the Arab Republic of Egypt Report, Pacific Consultants International and Sanyu Consultants Inc., Tokyo, Japan. Masoud AA (2009) Runoff modeling of the wadi systems for estimating flashflood and groundwater recharge potential in Southern Sinai, Egypt, Arab Journal of Geoscience, doi:10.1007/s12517-009-0090-9. Milewski A, Sultan M, Yan E, Becker R, Abdeldayem A, Soliman F, and Gelili K (2009) A remote sensing solution for estimating runoff recharge in arid environments. Journal of Hydrology 373:1-14.

61

Morrison J, Brockwell T, Merren T, Fourel F, Phillips AM (2001) On-line high-precision stable hydrogen isotopic analyses on nanoliter water samples. Anal Chem 73:3570-3575. Plummer LN (1992) Geochemical modeling of water-rock interaction: Past, present, future. In: Kharaka YK Maest AS (eds) Water-Rock Interaction, Balkema, Rotterdam, Netherlands: 23-33. Plummer LN, Prestemon EC, Parkhurst DL (1994) An interactive code (NETPATH) for modeling net geochemical reactions along a flow path, version 2.0. US Geol Surv Water-Resour Invest Rep 94-4169:130. Rainwater FH, Thatcher LI (1960) Methods for collection and analysis of water samples. U.S. Geol. Survey, Water Supply 1454, 301p. Research Institute for Water Research (RIWR) (1989) Sinai Water Resources Study (Phase II). Internal Report. Rosenthal E, Zilberbrand M and Livshitz Y (2007) The hydrochemical evolution of brackish groundwater in central and northern Sinai (Egypt) and in the western Negev (Israel). Journal of Hydrology 337:294-314. Said R (1962) The Geology of Egypt. Elsevier, Amsterdam. Shabana AR (1998) Geology of water resources in some catchment areas draining in the Gulf of Aqaba, Sinai-Egypt. PhD Thesis, Fac Sci, Ain Shams Univ, Egypt:246. Shalaby AI (1997) Geomorphology and hydrogeology of Wadi Watir basin, S. E. Sinai, Egypt. M.Sc. Thesis, Fac. Sci., Mansoura Univ., Egypt, 185p. Sultan M, Metwally S, Milewski A, Becker D, Ahmed M, Sauck W, Soliman F, Sturchio N, Yan E, Rashed M, Wagdy A, Becker R, and Welton B (2011) Modern recharge to fossil aquifers: Geochemical, geophysical, and modeling constraints Journal of Hydrology 403:14-24. UNESCO, 1963. Bioclimatic map of the mediterranean region (East Sheet). Paris.

62

United States Geological Survey (USGS), 2004. Reprocessing by the Global Land Cover Facility, 2004, (30) Arc Second SRTM Elevation. College Park, Maryland: the GLCF, Retrieved from World Wide Web. http://www.landcover.org. Van der Kemp WJM, Appelo CAJ, Walraevens K (2000) Inverse chemical modeling and radioactive dating of paleogroundwaters: The Tertiary Ledo-Paniselian aquifer in Flanders, Belgium. Water Resources Research 36(5):1277-1287. Wigely TML, Muller AB (1981) Fractionation corrections in radiocarbon dating: Radio carbon 23(2):173-190. Yehia MM (1998) Potentiality of the Lower Cretaceous aquifer system at Southeastern Sinai, Egypt. Sci. Journal. Fac. Sci., Menoufiya Univ. Vol. XII (1998), 21-43.

63

CHAPTER 3

Groundwater Resource Sustainability in the Wadi Watir Delta, Gulf of Aqaba,

Sinai, Egypt

Mustafa A. Eissaa,b,*, James M. Thomasb, Greg Pohll2, Ronald L. Hersheyb, Kamal A.

Dahabc, Maher I. Dawoudc, Mohamed A. Gomaaa, and Abdelfattah ElShiekha

aDesert Research Center, Division of Water Resource, Matariya, Cairo, Egypt

bDesert Research Institute, Division of Hydrologic Sciences, Reno, Nevada, USA

cFaculty of Science, Geology Dept., Menoufiya University, Egypt

64

Abstract

To manage groundwater in a sustainable way in the Wadi Watir delta, the amount of average annual groundwater recharge to the area needs to be quantified and potential seawater intrusion needs to be evaluated. The delta is a tourist area in the arid southeastern part of the Sinai Peninsula, Egypt, where development is constrained by the amount of groundwater that can be regenerated in a sustainable manner. The study area contains an alluvial aquifer, which is underlain by impermeable Precambrian basement rock. The scarcity of rainfall during the last decade, combined with high pumping rates, has resulted in degradation of water quality in the main supply wells along the mountain front. Increasing salinity in the water supply wells has resulted in reduced groundwater pumping. Additionally, seawater intrusion along the coast has resulted in increased salinity in some wells. A three-dimensional groundwater flow model

(MODFLOW/SEAWAT) was calibrated using groundwater level changes and pumping rates from 1982 to 2009. Model simulations estimated that the annual average groundwater recharge rate to the study area by 1.58 x 106 m3/year (about 4300 m3/day), through the model duration time. A variable-density flow and seawater intrusion model

(MODFLOW/SEAWAT) was used to evaluate seawater intrusion for different pumping rates and well field locations. Model simulation results showed some seawater intrusion along the coast. Water chemistry and stable isotope data were used to calculate seawater mixing with groundwater along the coast. These data were also used in geochemical modeling (NETPATH) to determine the sources and mixing of different groundwaters from the mountainous recharge areas and within the delta aquifers. Geochemical

65 modeling showed that the salinity of the groundwater is controlled by dissolution of minerals and salts in the aquifers along flow paths and mixing of chemically different waters, including upwelling of saline groundwater and seawater intrusion. Future groundwater pumping needs to be closely monitored to limit upwelling of saline groundwater into the well field near the mountain front and along the coast to limit seawater intrusion.

Keywords: groundwater sustainability; groundwater modeling; seawater intrusion; water chemistry; isotopes; Wadi Watir, Egypt.

66

1. Introduction

In the past decade, rapid development in the Wadi Watir delta (Figure 3-1), a tourist area on the Gulf of Aqaba, has resulted in a growing demand on groundwater resources. Groundwater is being utilized to meet various needs in coastal and arid areas of the Sinai where there are no other sources of fresh water. Groundwater beneath the

Wadi Watir delta, which includes the Nuweiba Harbor area, is the main source of potable water for this area. Groundwater occurs as a thin lens of fresh water and is very sensitive to pumping induced stresses and showed high salinity at the coast due to upwelling of seawater (Shalaby, 1997 and Ismail, 1998). Thus, groundwater withdrawals have to be carefully planned and monitored to avoid deterioration of this valuable resource by upwelling of underlying saline groundwater and intrusion of seawater. To understand historic and current groundwater conditions in this area and predict future changes in groundwater availability and potential salinity increases caused by pumping, we employed a model to evaluate the response of this coastal aquifer under different withdrawal rates and at different pumping locations. Water chemistry and isotopic data were used to (1) evaluate potential groundwater flow paths and mixing of different water sources (including seawater intrusion); (2) determine the evolution of groundwater chemistry (including evaporative concentration); (3) estimate the age of the groundwater; and (4) constrain groundwater flow and solute transport models.

67

34 39 34 41 Hand Dug Wells

Drilled Wells 43 Infiltration Test sites 41 42 Cross Section 40 0 1 2 39 Basement Rocks Km. 29 G Sabkha Deposits u l 02 Alluvial deposits f

38 eld o Fi ell 37 f W 23

30 33 21 22 A 24 29 36 35 q 20 A 28 a 25 b 19 27 32 26 a 18 31 34 29 17 44 00 45 50 53 46 47 48 49 51 Sabkha B 34 00 34 30 29 52 30

28 54 58 55 MEDITERRANEAN 29 Cairo SINAI 00 Wadi Watir

0 10 20 Study Area EGYPT Scale

Figure 3-1. Location of the study area and groundwater samples.

68

2. Background

2.1 Study Area

The Wadi Watir delta area is located in the downstream portion of Wadi Watir on the southeastern part of the Sinai Peninsula, Egypt, between longitude 34○ 38` and 34○

41` E and latitude 28○ 57` and 29○ 03` N (Figure 3-1). The Wadi Watir watershed drains toward the Gulf of Aqaba and is considered the most important valley in this region because the city of Nuweiba, a tourist destination, is located on its delta and Nuweiba

Harbor is located on the delta coast. Ships sailing from Nuweiba Harbor link Egypt with

Saudi Arabia and Jordan.

2.2 Geology and Hydrogeology

The watershed area above the Wadi Watir delta is the source of surface water and groundwater recharge for the delta. This area consists of rocks ranging in age from

Precambrian to Quaternary (CONOCO, 1987). The deepest (Basement) rocks are mainly composed of acidic granites that are cut by basic dykes. The Araba Formation (early

Cambrian) is formed mainly of thick-bedded, grayish and reddish, fine to medium- grained sandstone. The Naqous Formation (late Cambrian) is composed mainly of fractured and jointed layers of white, coarse, and pebbly sandstone with authigenic clay minerals (Shabana, 1998). The Wata Formation (lower Cretaceous) consists mainly of alternating beds of limestone and marl at the base, clay marl sandstone with bands of limestone in the middle, and dolomitic limestone and limestone at the top. The Raha

Formation (lower Cretaceous) is primarily composed of alternating beds of limestone,

69 shale, and sandy limestone at the base; shale and marl in the middle; and marly limestone and dolomite at the top (Shabana, 1998).

The Wadi Watir delta is composed mainly of Quaternary deposits and constitutes the main water-bearing formation in the Nuweiba District. The Quaternary aquifer of the

Wadi Watir Delta is composed mainly of fine-to-coarse sands, gravels, and boulders of carbonate, sandstone, and granitic rocks embedded in a silt and clay matrix (El-Shazly et al., 1974; Eyal et al., 1980; El Kiki et al., 1992a). The Quaternary aquifer of the Wadi

Watir delta can be divided into five layers (Figure 3-2; Abbas et al., 2004). The uppermost two layers are generally <10 m thick and grouped as surface layers. They are comprised of heterogeneous alluvial deposits. The third layer is a sandy clay layer and is between 30 and 45 m thick. The fourth layer is comprised of sand and gravel and is between 20 and 40 m thick. The fifth layer is comprised of sand interlayered with shale and is 20 to 50 m thick. These five layers are underlain by bedrock which is composed mainly of very low permeability granitic rocks.

The Wadi Watir delta Quaternary aquifer is an unconfined water-table aquifer.

Depth to water in this aquifer ranges from 2.33 to 40.8 m. The water level in the Wadi

Watir delta Quaternary aquifer varies from <1 m to about 2 m above sea level.

3. Methods

3.1 Field and Laboratory Methods

Water samples were collected from seven drilled wells and 32 hand-dug wells in the Wadi Watir delta in March, 2007. In September, 2009, four drilled wells (17, 19, 20,

21; Table 3-1) and two hand-dug wells (35, 37; Table 3-1) were resampled. Samples were

70 filtered through a 0.45 µm filter and collected in polyethylene bottles for both chemical and isotopic analyses. pH and electrical conductivity (EC) were measured in the field.

Electrical conductivity was measured with an YSI model 35 conductivity meter; pH was measured with a WTW model LF 538 pH meter. Meters were calibrated once daily during the field campaign. For determination of groundwater flow direction, depth to water, total depth, and ground-surface elevation were measured in 22 wells using a

LEICA TCRA1103 PLUS ROBOTIC 3 instrument.

W E A B Surface layers 50 (Alluvial deposits) Scale

30 0 1000 (meters)

10 )

m Sandy clay (

-10

h

t p e Sand & gravels D -30 (aquifer layer) -50

-70

-90 Sand intercalated with clay

-110 Bed rock Figure 3-2. Stratigraphic cross-section of Wadi Watir delta constructed from well log information and geophysical interpretations (modified from Abbas et al., 2004). The cross-section location is shown on Figure 3-1 and is in a south-north direction across the delta (A-B).

71

Table 3-1. Well information and chemical and isotopic data for groundwater samples collected in March 2007 and August 2009. Well depths and water levels are in meters; chemical data are in meq/L; stable isotopic data are in ‰; and carbon-14 data are in percent modern carbon. Well T.D D.T.W G.E W.L pH TDS epm SiO Br δ18O δ2H δ13C 14 No. 2 C (pmc) Type (m) (m) (m) (m) mg/l Ca Mg Na K CO3 HCO3 SO4 Cl (ppm) (ppm) (‰) (‰) (‰) WS Hand Dug ------7.43 726 6.14 3.93 4.12 0.31 0.00 2.67 8.08 3.29 10.83 2.0 -- -- -10 74 17 52 33.3 -- -- 6.86 3540 18.06 11.16 29.12 0.41 0.6 2.00 16.21 42.82 15.35 17.7 -3.23 -15 -- -- 18 55 28.9 -- -- 7.19 3609 18.06 13.39 29.12 0.41 0.00 2.00 16.55 43.52 15.96 22.1 -3.15 -15.7 -6.4 56.7

19 Drilled 60.5 36.7 37.3 0.6 7.73 2600 12.64 6.25 24.15 0.34 0.00 2.00 19.59 19.76 9.97 11.9 -3.67 -18.4 -- -- 20 59.5 39.7 41.2 1.5 7.26 1925 13.55 4.82 12.62 0.34 0.00 2.00 10.87 20.00 17.44 11.5 -3.61 -17.9 -6.1 76.02 21 2007 58.5 40.8 41.5 0.7 7.25 2538 18.06 8.93 16.84 0.37 0.27 1.73 10.05 31.76 17.61 30 -3.71 -19.4 -- -- 22 -- 37 -- -- 7.2 2941 13.55 15.62 22.90 0.40 0.00 2.00 11.46 37.41 17.09 17.5 ------23 44 35.7 37.2 1.5 7.75 2821 7.23 21.42 21.71 0.40 0.00 1.47 13.30 34.35 17.44 15.5 -3.29 -15.1 -- -- 24 15.8 14.9 15.8 0.9 7.48 2317 10.80 9.26 18.59 0.41 0.00 1.90 15.26 22.29 10.1 17 -3.57 -18.7 -4.2 30.59 25 18 15.6 16.3 0.7 7.5 2358 12.64 11.43 17.07 0.37 0.00 2.13 11.14 27.05 10.05 15.2 -3.82 -19.5 -- --

26 13 10.9 -- -- 8.01 2069 8.31 11.43 23.52 0.37 0.80 1.87 8.11 21.17 15.87 16.8 ------27 ------7.68 2691 9.03 10.71 26.17 0.38 0.00 2.67 12.87 30.00 19.78 15.9 ------28 ------7.76 3005 9.03 17.85 25.48 0.42 0.27 2.27 16.34 32.35 20.57 17.6 -3.77 -20.1 -- -- 29 11.8 10.2 1.1 0.9 7.2 4725 23.40 19.14 37.47 0.82 0.00 3.45 29.74 50.50 8.92 27 ------30 11.5 9.83 1.1 1.27 7.68 1830 9.00 8.90 11.69 0.41 0.00 2.07 8.88 18.20 6.32 17 ------31 ------7.8 3239 7.23 19.64 27.60 0.40 0.53 1.73 20.78 32.35 13.01 17.3 ------Hand 32 Dug 10.5 9.2 10.4 1.2 7.32 3280 10.84 16.07 29.12 0.46 0.00 2.67 16.77 36.47 15.35 19.9 ------33 8.2 6.4 7.50 1.1 7.66 3246 8.55 9.35 21.67 0.43 0.00 3.62 13.49 22.29 14.74 17 -0.33 -3.4 -- -- 34 2007 5.5 4.8 -- -- 7.33 6447 19.87 33.92 61.43 1.17 0.00 4.13 18.38 89.40 11.70 36.5 -0.27 -6.6 -- -- 35 7 4.1 5 0.9 7.31 11447 49.68 53.56 86.92 1.52 0.00 3.47 49.57 148.21 26.56 76.1 -2.69 -15 -- -- 36 ------7.75 2606 10.84 8.93 24.15 0.37 0.00 2.00 16.99 23.53 19.35 13.3 -2.9 -15.3 -- -- 37 3.5 2.4 3.1 0.9 7.82 2902 13.50 11.57 21.67 1.64 0.29 2.41 19.34 25.02 15.96 19 -2.72 -14 -- -- 38 3.5 2.4 3.1 0.7 7.58 2716 11.25 12.02 23.28 0.51 0.00 3.45 20.23 23.20 7.62 16 ------39 7 4.1 5.2 1.1 7.2 4400 12.64 36.60 27.60 0.62 0.00 1.60 31.93 41.17 24.21 18.7 0.72 3.8 -- -- 40 8 4.4 5.5 1.1 7.59 1772 7.95 4.28 16.05 0.37 0.00 1.07 10.09 18.62 3.45 9.57 ------41 6 4.9 6.2 1.1 7.18 2886 7.23 15.18 27.60 0.43 0.00 1.60 13.30 34.35 16.31 14.4 -0.66 -5.9 -- -- 42 7.5 6 7.5 1.4 7.25 5881 24.84 31.24 49.62 0.74 0.00 2.40 15.78 84.69 15.44 40.7 -2.09 -13 -- --

72

Table 3-1. Well information and chemical and isotopic data for groundwater samples collected in March 2007 and August 2009. Well depths and water levels are in meters; chemical data are in meq/L; stable isotopic data are in ‰; and carbon-14 data are in percent modern carbon (continued). Well T.D D.T.W G.E W.L pH TDS epm SiO Br δ18O δ2H δ13C 14 No. 2 C (pmc) Type (m) (m) (m) (m) mg/l Ca Mg Na K CO3 HCO3 SO4 Cl (ppm) (ppm) (‰) (‰) (‰) 43 -- 16.7 -- -- 8.2 10466 9.03 18.75 55.26 2.26 0.00 0.00 105.37 102.57 ------44 23.5 22 -- -- 7.42 1750 11.92 2.50 12.62 0.34 0.00 0.93 12.82 15.06 11.44 8.66 ------45 17 14.7 16.1 1.4 7.13 1785 10.84 9.28 11.04 0.34 0.00 1.87 7.23 22.11 16.39 13.1 ------46 18 15.8 17.3 1.5 7.13 1850 9.39 7.14 14.04 0.37 0.00 2.67 9.65 19.53 14.48 10.7 -2.58 -13.4 -- -- 47 ------7.49 2991 9.03 14.28 26.88 0.42 0.00 2.40 18.29 29.64 18.91 14.5 ------48 15 12.7 14.1 1.1 7.42 2740 9.03 16.96 21.71 0.40 0.00 2.40 15.90 28.23 21.96 13.2 ------49 12 10.4 -- -- 6.56 7503 25.29 37.49 68.35 1.64 0.00 2.00 33.11 91.75 3.28 46.9 6.86 22.5 -7.9 73.49 50 14 11.5 12.4 0.9 7.36 4088 10.84 29.46 32.40 0.67 0.00 4.00 20.67 45.88 17.00 27.5 2.58 7.4 -- -- 51 ------8.21 942 2.26 0.89 12.62 0.30 0.27 2.13 1.42 12.61 5.36 5.63 1.73 10.1 -7.6 72.97 52 7 5.8 -- -- 7.85 2947 9.03 8.93 31.55 1.37 0.80 4.53 12.11 32.35 31.86 14.8 1.43 7.1 -- -- 53 13 11.3 12.4 1.1 6.84 10716 40.64 41.96 91.68 2.38 0.00 2.53 63.00 115.28 1.02 60.8 5.94 21.5 -- -- 54 ------7.7 3226 9.03 14.28 32.40 0.73 0.80 2.67 15.69 35.29 19.00 16.9 ------55 ------6.99 2960 9.90 9.80 32.40 0.5 0.00 2.7 6.8 41.2 16.13 18.9 ------17 ------7.43 4062 25 10.5 32.78 0.5 0 1.69 11.48 58.86 20.2 -- -3.3 -- -- 19 Drilled ------7.69 2190 16.9 5.49 13.83 0.36 0 1.72 11.73 23.97 20.8 -- -3.9 -- -- 20 2009 ------7.66 2104 15.65 5.55 13.57 0.35 0 1.67 11.50 22.74 19.8 -- -3.8 -- -- 21 ------7.71 1928 13.45 5.43 13.78 0.36 0 1.74 10.67 19.94 19 -- -3.8 -- -- 35 ------7.18 13853 63.50 60.5 114.78 2.11 0 3.39 53.96 184 31 -- -3.0 -- -- Hand Dug 2009 37 ------7.77 2884 13.45 12.5 21.83 0.60 0 3.02 16.83 29.71 26.9 -- -2.70 -- -- Shed ------7.43 720 6.14 3.93 4.12 0.31 0 2.67 8.08 3.29 10.83 2.0 -- -- -10 74 Rain ------7.00 7.1 0.6 23 -- -- 14.5 6.4 0.15 --15.6 -- -4.3 -17.1 -- -- Sea ------8.08 41415 544 1480 12300 464 -- 154 3280 23300 -- 81.2 1.64 9.8 -- -- Sbakha -- 7.6 624 1883 15877 547 -- 117 3688 28500 -- 150 14.4 ------T.D-Total Depth; D.T.W-Depth To Water; TDS-Total Dissolved Solid; G.E-Ground Elevation; W.L-Water Level; epm-equivalents per million; -- no data; * -(mg/l) El-Sayed, M. H. 2006; ** Local rain data from 1960-1987 after Abdel Samei 1995; WS- Sample represent the Wadi Watir Watershed

73

Major-ion analyses were conducted at the Desert Research Center, Water Central

Laboratory, Cairo, Egypt, using the analytical methods described in Rainwater and

Thatcher (1960) and Fishman and Friedman (1989). For analysis quality assurance and quality control, each sample was run in duplicate. If the difference between the sum of cations and anions was more than 5 percent, the sample was reanalyzed until an acceptable percentage difference was obtained.

Stable isotopic analyses were conducted at the Nevada Stable Isotope Lab,

University of Nevada, Reno following the methods of Morrison et al. (2001) for δ2H and

Epstein and Mayeda (1953) for δ18O. δ2H results were reported in units of per mil (‰)

VSMOW with an uncertainty of ±1‰ (1 standard deviation) and δ18O in per mil

VSMOW with an uncertainty of ±0.2‰ (1 standard deviation). Carbon-14 analyses were conducted at the University of Arizona Accelerator Mass Spectrometry Laboratory.

Results were reported in percent modern carbon (pmc) with an uncertainty of ± 0.4 pmc.

Infiltration tests were conducted to determine vertical hydraulic conductivity of surficial layers using the double-ring infiltrometer method (Philip 1957a; Hanks and

Aschroft 1980). Aquifer tests were performed to determine hydraulic conductivity for different water-bearing units.

Mineralogy was identified by examination of rock thin-sections using polarized light microscopy. Rock chemical composition was determined by x-ray florescence

(XRF) at the Nevada Bureau of Mines and Geology, Reno, Nevada (Eissa et al., 2012, in review).

74

3.2 Water-Rock Reaction Modeling

Inverse geochemical modeling (Plummer, 1992) has been widely used for interpreting geochemical processes that account for hydrochemical and isotopic changes in groundwater. The inverse geochemical software NETPATH (Plummer et al., 1994) was used because it can calculate mixing proportions of two to five waters, net geochemical reactions that account for observed chemical composition changes in groundwater along a flow path, and changes in carbon isotopes. The geochemical reactions and physical processes that produced observed water chemistries in the Wadi

Watir delta were evaluated and several possible results are presented that identify water- rock interactions, mixing, and evaporation. Inverse geochemical modeling does not produce unique results; however, this type of modeling does provide insight into the possible water-rock reactions that could produce the changes in observed water chemistry along a flow path. In these simulations, water-rock reactions are constrained by mineral and gas saturation information. A mineral or gas can only enter water if the water is under saturated with respect to the mineral or gas (dissolve) or can only leave the water

(precipitate or degas) if the water is super saturated with respect to the mineral or gas.

Table 3-2. Constraints, phases, and parameters used in NETPATH models. Constraints Phases Parameters Calcite, Composite Granite (Ca Mg Na K Si Al 4Fe ), Calcium, Carbon, 0.33 0.01 1.02 1.72 21 3.7 0.74 Dolomite, Biotite, Cation Exchange, Evaporation, Mixing, Magnesium, Gypsum, Halite, Illite, Carbon Isotopic Exchange Potassium, Sodium, Sulfur Mafic Montmorillonite, Na-Montmorillonite

75

Possible groundwater flow paths were identified using water-level data, the conservative isotopic tracers δ2H and δ18O, and major-ion chemistry (Table 3-1). The water-rock reaction model simulations were constrained by major-ion concentrations of the groundwater and mineral phases (Table 3-1) in the aquifers of the delta. Halite was included as a phase because it is often present in terrestrial deposits in arid lands. Calcite and dolomite were included because of the presence of carbonate boulders in the delta.

Clays are present in sediments and deeper rock units so the clay minerals montmorillinite and illite were included. Minerals of mafic and felsic rocks were identified in thin- sections of rock samples from aquifers located in the mountain block recharge area; the mountain block is also the main source of alluvial deposits in the delta (Eissa et al., 2012, in review). A composite granite (Ca0.33Mg0.01Na1.02K1.72Si21Al3.74Fe0.74), constructed from

XRF data of felsic igneous rocks found in the Wadi Watir watershed, was used as a phase in the water-rock reaction model simulations (Eissa et al., 2012, in review).

Only water-chemistry samples with an electro-neutrality of 5% or less were used in water-rock reaction modeling simulations; no attempt was made to correct water analyses to electro-neutrality. Correcting for this very small analytical error could bias model simulation results.

3.3 Groundwater Flow Modeling

We used the computer program SEAWAT (Guo and Langevin, 2002) for groundwater flow and transport modeling because it is capable of simulating three- dimensional, variable-density groundwater flow in porous media and seawater intrusion into coastal aquifers. This program combines a groundwater flow model, MODFLOW

76

(Harbaugh et al., 2000), with a solute transport model, MT3DMS (Zheng and Wang,

1999), into a single program that solves the density-dependent groundwater flow and solute-transport equations derived by Guo and Langevin (2002).

4. Results and Discussion

4.1 Groundwater Chemistry

Groundwater in the Quaternary alluvial aquifer in the Wadi Watir delta had total dissolved solids (TDS) ranging from 940 to 13,800 mg/l (Table 3-1). The highest TDS

(5,800 to 13,800 mg/L) generally occurred in wells near the coast (Figure 3-1 and Table

3-1; No. 34, 35, 42, and 43) or in hand-dug wells (Figure 3-1 and Table 3-1; No. 49 and

53). Groundwater was chemically homogenous consisting predominantly of the cations

Na and Ca; and the anions Cl and SO4 (Figure 3-3).

In the delta, TDS increased in all wells from 1999 to 2007 (Figure 3-4; Table 3-1 and Table 3-3). This increase in TDS was particularly apparent in the main well field

(wells 17, 18, 20, and 22), which supplies most of the potable water to the Wadi Watir delta. Total dissolved solids also increased along the coast in wells 35 and 42. Total dissolved solids in wells in the main well field increased even though groundwater withdrawal rates were reduced during drought conditions. Conservative ions (chloride, iodide, and bromide) also increased in most wells from 1999 to 2007 (Table 3-3).

77

Figure 3-3. Major-ion diagram (Piper Diagram) of groundwater in the Wadi Watir delta.

Groundwater is predominantly Na + Ca and Cl + SO4.

78

TDS 1999 TDS 2007

12000

10000

8000

6000

4000

Concentration(ppm) 2000

0

Well No.

Cl (1999) Cl (2007)

6000

5000

4000

3000

2000

Concentration(ppm) 1000

0

Well No.

Figure 3-4. Total dissolved solids and chloride concentrations in groundwater for 1999 (After Said, 2004) and 2007. The well numbers go from the mountain front to the coast.

79

Table 3-3. Water chemical analyses for some wells in the Wadi Watir delta in 1999 (After Said, 2004) and 2007. Well TDS Cl Br I I No. 1999 1999 1999 1999 2007 (mg/l) (mg/l) (mg/l) (mg/l) (mg/l) 17 2453 1086 13.9 0.028 0.052 18 2047 761 9.8 0.020 0.033 20 1476 516 13.9 0.014 0.030 22 2566 1528 18.6 0.037 0.046 24 1549 735 9 0.018 0.084 25 1861 621 9.2 0.018 0.041 28 2038 679 9.4 0.019 0.033 35 6038 2320 28 0.028 0.119 37 2460 637 6.0 0.018 0.192 42 2678 972 11.3 0.033 0.175 45 1736 641 7.5 0.015 0.029

4.2 Environmental Isotopes

In 2007, groundwater in the Wadi Watir delta had a wide range of δ18O (-3.82 to

+6.86 ‰) and δ2H (-20.1 to +22.5 ‰) values (Table 3-1). Groundwater in the delta could be divided into two groups. Group I was characterized by relatively depleted δ18O and

δ2H values ranging from -3.82 to -2.58 ‰ and -19.5 to -13.4 ‰, respectively. Most of

Group I was located away from the coast with many of the sites along the mountain front and where the Wadi Watir drainage enters the delta (Figure 3-5). The isotopic values for

Group I were similar to the δ18O and δ2H of groundwater in the upper watershed recharge areas (Eissa et al., 2012, in review), although within the isotopically heavier end of the range (Figure 3-6). Group II is located in shallow wells along the coast and near the wetted Sabkha areas; the δ18O and δ2H values ranged from -0.66 to +6.86 ‰ and -5.9 to +22.5 ‰, respectively. Moreover, some of the evaporation observed for shallow groundwaters is likely directly from hand-dug wells; because hand-dug wells are

80

34 39 34 41 Legend

Hand Dug Wells

Drilled Wells 41 Hand Dug Seawater Intrusion 42 0 1 2 Hand Dug 39 Km. Evaporation G 29 u Basement Rocks l f 02 Sabkha Deposits

o (Group I) f 37

23 33 A 21 q 24 36 35 a b 20 25 a 19 28 18 29 17 34 53 00 50 46 (Group II) 51 49 Sabkha

52 28

58

Figure 3-5. Isotopically distinct groundwaters shown as Group I (-3.82 to -2.58 ‰ δ18O; -19.5 to -13.4 ‰ δ2H) and Group II (-0.66 to +6.86 ‰ δ18O; -5.9 to +22.5 ‰ δ2H).

81

GMWL 40 MMWL

30

20

10 Watershed 0

Delta Water (Group I) -10

H (‰) H 2

-20 Delta Water (Group II) δ δ

-30 Winter Rain -40

-50 -8 -6 -4 -2 0 2 4 6 8 δ 18O (‰)

Figure 3-6. δ18O versus δ2H for groundwater in the Wadi Watir watershed and delta areas. Watir rain amount weighted average (from Eissa et al., in review) Global Meteoric Water Line (GMWL; Craig, 1961) and Mediterranean Meteoric Water Line (MMWL; Gat et al. (1969) and IAEA (1981).

82 generally about one meter in diameter so that the large water surface area in the open well allows for evaporation of water in a hand dug well especially if they are not withdrawing much water from the well. Group II locations were evaporated and plot along an evaporation line from winter rain (from Eissa et al., in review) and Group I groundwater

(δ2H = 4.7 x δ18O – 2.6; Figure 3-6).

Two wells (35 and 42) were not included in Group I or Group II because they were located along the coast in the area of other Group II samples (Figure 3-5) and had elevated ion concentrations, but were isotopically lighter (Table 3-1) like Group I.

Seawater mixing with groundwater occurred in these two wells as they contained higher

TDS, Cl, and Br then other wells in the delta (Tables 1 and 4). δ18O versus Cl and δ18O versus Br plots (Figures 3-7 and 3-8) showed that these samples plot on a mixing line between Group I and seawater.

Evaporative enrichment of shallow groundwater samples (evaporation factor) was calculated using the Rayleigh equation (Clark and Fritz, 1997) and the average isotopic composition of groundwater in the up-gradient watershed area. The annual average relative humidity and temperature were obtained from nearby meteorological stations between the periods of 1983 to 1994 (Saint Cathrine, Ras Nasari and Ras El-Naqab stations; Shabana, 1998). The annual average temperature was 20.54 o C and the average relative humidity was 39.45%. Calculated evaporation factors for Group II ranged from

1.1 (well 35) to 1.8 (well 49) (Table 3-4).

83

800

700

600

500

400 Delta Groundwater Cl (epm) Cl

300 Winter Rain

200 Seawater Intrusion Group II Seawater

100 Group I Sbakha Water 0 -6 -1 4 9 14 δ 18O (‰)

Figure 3-7. δ18O versus chloride for groundwater in the Wadi Watir delta. Group I (-3.82 to -2.58 ‰ δ18O; -19.5 to -13.4 ‰ δ2H) and Group II (-0.66 to +6.86 ‰ δ18O; -5.9 to +22.5 ‰ δ2H). Watir rain amount weighted average (from Eissa et al., in review). Sabkha water analyses for chloride (Average of the concentration of chloride in six samples) and δ18O are after (Gavish, 1974).

84

Figure 3-8. δ18O versus bromide for groundwater, seawater, and sabkha in the Wadi Watir delta. Group I (-3.82 to -2.58 ‰ δ18O; -19.5 to - 13.4 ‰ δ2H) and Group II (-0.66 to +6.86 ‰ δ18O; -5.9 to +22.5 ‰ δ2H). Watir rain amount weighted average (from Eissa et al., in review). Sabkha water analyses for bromide and δ18O is after (Gavish, 1974).

85

Table 3-4. NETPATH modeling results for water-rock interactions and mass transfer (mmol/L) for the Wadi Watir basin. Initial Initial Mixing Phases precipitated or dissolved Ev using (Site 1) (Site 2) Final Percent Ev using Rayleigh Numbers Water Na- Mont NETPATH Site 1 Site 2 Cal Biotite Gyp llt Gr Ex Hal Dol Eq in Fig. 1 Mont -Maf Water Shed -- 20 -- -- -5.07 2.12 1.40 ------16.48 - -- -- 20 -- 18 ------2.83 2.86 -- -20.47 -3.59 -- -- 23.84 0.012.76 -- -- 20 -- 25 -- -- -1.21 1.76 0.13 -- -7.64 ------6.99 0.62 -- -- 20 -- 27 -- -- -3.38 0.64 1.00 ------1.85 9.91 1.96 -- -- 20 -- 28 -- -- -9.44 1.12 2.75 ------0.33 12.25 4.77 -- -- GROUP I 20 -- 31 -- -- -13.02 0.70 4.98 ------1.40 12.26 6.29 -- -- 20 -- 33 -- -- -1.43 0.32 0.62 ------3.42 -- 1.29 1.1 1.1 20 -- 36 -- -- -4.19 -- 3.07 ------2.13 2.95 3.51 2.01 -- -- 20 -- 38 -- -- -5.55 0.06 4.70 ------3.76 3.19 3.47 -- -- 20 -- 17 -- -- 0.3 2.09 2.69 -- -18.35 -4.29 -- 22.64 ------20 -- 46 ------0.48 -0.6 -- -5.7 -- -- 1.89 -0.47 0.42 -- -- 17 -- 49 ------2.74 ------6.95 -- 1.14 2.09 - 2 1.8 GROUP II 17 -- 50 -- -- -14.65 -- 1.55 ------1.92 -0.41 -- 0.54-- 1.1 1.3 46 -- 52 -- -- -5.99 -- -- -13.46 -- -- 4.84 6.18 3.30 1.3 1.3 17 -- 53 -- -- -13.52 -- 9.99 ------0.23 23.11 6.39 1.7 1.7 Seawater 20 Sea 35 87.5 12.5 -- 3.64 4.94 -- -51.8 -- 2.69 ------1.6 1.1 Intrusion 20 Sea 42 92 8 -15.76 ------2.02 -12.11 9.42 8.07 1.1 1.1

Chl-Chlorite; Cal-Calcite; Gyp-Gypsum; Ilt-Illite; Biot-Biotite; Mont-Maf-Montmorillonite Mafic; Gr-Composite Granite;Ex-Cation Exchange; Hal-Halite; Dol-Dolomite; the initial water is groundwater originating in the watershed above the Wadi Watir delta; Ev-Evaporation Factor; -- No Data; *constraint ignored.

86

4.3 Water-Rock Reaction Modeling

Water chemistry data, in addition to isotopic data, can be used to evaluate potential groundwater flow paths (Clark and Fritz 1997). To be a valid groundwater flow path, the observed water chemistry at the end of a flow path has to have undergone reasonable geochemical reactions and/or changed by mixing with chemically different waters along the flow path (Plummer et al., 1994). If the change in water chemistry along a flow path cannot be explained by thermodynamically feasible reactions (minerals or gases that are under saturated in the groundwater need to dissolve and minerals or gases that are saturated in the groundwater need to precipitate or dissolve) or by mixing, then the flow path cannot be validated with groundwater chemistry. For a water-rock reaction model simulation to be considered valid, precipitation or dissolution of any phase along the proposed flow path could not be greater than 15 millimoles per liter (mmol/L). This approach was limited by the available mineral, gas, chemical, and isotopic data (Hershey et al., 2007).

Potential flow paths in the Wadi Watir delta were first delineated based on groundwater levels; groundwater flow is generally from the upper watershed areas in the west to the delta in the east and within the delta from west to east (El Kiki, 1992; El

Ghazawi, 1999; Ismail, 1998). Stable isotopic data also had to support a potential flow path for it to be included in geochemical modeling. Flow paths modeled were: (1) from the upper watershed to well 20, a water supply well in the well field along the mountain front that represented the most dilute groundwater in the delta; (2) and from well 20 to other Group I and Group II wells down gradient east of the well field; (3) from Group I to

87

Group II wells in the southern part of the delta; and, (4) seawater intrusion using up- gradient well 20 mixing with seawater to make down-gradient wells 35 and 42 near the coast.

Valid water-rock reaction model simulations for the watershed, Group I, and II included precipitation of calcite and/or clays (montmorillonite) with dissolution of biotite or composite granite, gypsum, halite, and dolomite and/or exchange of calcium and magnesium in groundwater with sodium in clay minerals. Evaporative enrichment was not important for Group I simulations, but was important for Group II. For seawater intrusion, calcite or clay precipitated, biotite and/or composite granite dissolved, and either gypsum or halite and dolomite dissolved plus calcium and magnesium exchanged with sodium. Evaporative enrichment was also important in seawater intrusion simulations.

Water-rock reaction model simulations from the main well field to other down- gradient Group I and Group II wells explained observed water chemistry changes without evaporative concentration except for well 33 (Table 3-4). Mass transfer of major ions by dissolution or precipitation of various minerals was consistent with saturation index (SI) values (Table 3-5). In two simulations, mass transfer of no more than 15 mmol/L was exceeded by halite dissolution because of the large difference in chloride concentrations between initial and final waters. However, these simulations were considered valid because the likely source of chloride along these flow paths was halite dissolution (not

88

Table 3-5. Mineral saturation indices for phases in geochemical models calculated using NETPATH. No. Calcite Dolomite Gypsum Albite Anorthite K-spar Halite Ca-Mont Illite 17 -0.39 -0.96 -0.54 -0.83 -3.54 -0.03 -2.71 2.68 1.81 18 -0.08 -0.24 -0.55 -0.73 -3.40 0.08 -2.70 2.21 1.61 20 -0.03 -0.48 -0.68 -0.95 -3.35 0.14 -3.40 2.24 1.57 24 0.02 0.01 -0.69 -2.45 -4.59 -1.45 -3.18 -0.25 -0.52 25 0.17 0.34 -0.76 -1.54 -3.86 -0.54 -3.14 0.88 0.56 27 0.28 0.68 -0.85 -0.47 -3.44 0.36 -2.91 1.58 1.91 28 0.26 0.85 -0.81 -0.44 -3.44 0.44 -2.89 1.47 1.40 31 0.06 0.59 -0.83 -1.00 -3.95 -0.19 -2.85 0.64 0.65 33 0.38 0.84 -0.83 -0.93 -3.71 0.03 -3.12 1.15 0.98 35 0.49 1.08 -0.03 0.43 -2.66 1.31 -1.69 2.88 2.72 36 0.29 0.53 -0.66 -0.53 -3.39 0.32 -3.05 1.42 1.23 38 0.35 0.77 -0.61 -1.77 -4.22 -0.76 -3.07 0.27 0.12 41 -0.54 -0.72 -0.96 -0.72 -3.75 0.14 -2.83 2.21 1.71 42 0.13 0.41 -0.59 -0.59 -3.33 0.29 -2.18 2.06 1.73 46 -0.19 -0.46 -0.87 -1.15 -3.68 -0.06 -3.36 2.16 1.49 48 -0.04 0.23 -0.80 -0.42 -3.37 0.50 -3.01 2.25 1.89 49 -0.64 -1.17 -0.34 -2.64 -5.17 -1.60 -2.00 0.47 -0.09 50 0.13 0.74 0.74 -0.59 -3.58 0.38 -2.63 1.95 1.73 52 0.67 1.38 -0.88 0.23 -3.03 1.53 -2.79 1.99 2.21 53 0.27 0.78 -0.82 -0.44 -3.51 0.57 -1.78 1.47 1.47 Positive values indicate that a mineral is supersaturated in the water and will precipitate from the water, and negative values indicate that a mineral is under saturated in the water and will dissolve if present.

89

evaporative concentration), which also added sodium, which then had to be removed by precipitation of a sodium bearing phase, in this case, Na-montmorillonite.

Water-rock reaction model simulations for Group II examined groundwater flow from the southern part of the main well field to wells to the southeast including wells 49,

50, 52, and 53 (Figure 3-5). Water-rock reactions to Well 51 were excluded because groundwater in this well was likely diluted by water from a car wash next to the well.

Initial water for these simulations were from either well 17 in the southern portion of the well field or from well 46, the most dilute groundwater in the southern part of the delta.

Mass transfer of major ions by dissolution or precipitation of various minerals was consistent with SI values (Table 3-5) except for dolomite for wells 50, 52, and 53 where de-dolomitization may be driving dissolution of dolomite despite SI values indicating saturation with respect to dolomite (Back et al., 1983). NETPATH calculated evaporative concentration ranged from 1.1 to 2, consistent with evaporation factors calculated with the isotopic data and the Rayleigh equation (1.1 to 1.8; Table 3-4).

Water-rock reaction model simulations also supported groundwater-seawater mixing at the coast in two wells as indicated by isotopic, chloride, and bromide data

(Figures 3-6 and 3-7). Mixing dilute groundwater water from well 20 in the well field with 8 to 12.5 percent seawater produced the observed water chemistry at wells 35 and

42. NETPATH calculated evaporative concentration was consistent with evaporation calculated with the isotopic data and the Rayleigh equation (Table 3-4).

90

4.4 Groundwater Ages

δ13C and carbon-14 were analyzed in groundwater samples from one well in the watershed above the delta and five wells in the delta (wells 18, 20, 24, and 49). Water- rock reactions modeled with NETPATH were used to correct groundwater ages. The watershed area and well 20 were used as the initial source for estimating down-gradient groundwater ages because (1) they were the most up-gradient water; (2) were the most dilute water in the delta; and (3) contained the most negative δ13C values and the highest carbon-14 pmc.

Two of the four wells had modern corrected carbon-14 groundwater ages while wells 18 and 24 had ages of 2370 and 3950 years, respectively (Table 3-6). These wells may be influenced by upwelling of older groundwater (Figure 3-2; sandstone and clay layer above the bedrock) as a result of over pumping in the well field (well 18) and from local community wells (well 24). These results indicated that groundwater in the delta are comprised of recently recharged water and not paleo-recharge water from a past wetter climate.

Table 3-6. Groundwater ages for Wadi Watir delta corrected for water-rock interactions using NETPATH. Carbon δ13C (‰) 14C Initial Initial Final Isotopic Age Basin Computed Observed Computed Observed Site 1 Site 2 Water Exchange (years) (‰) (‰) (pmc) (pmc) (mmol/L) Watershed* -- 20 0.00 -2.69 -6.10 9.17 76 Mode rn Wadi 20 -- 18 1.70 -6.04 -6.40 75.24 56.47 2372 Watir delta 20 -- 24 1.00 -4.31 -4.20 47.24 30.59 3593 18 49 1.00 -7.87 -7.90 56.11 73.49 Mode rn *Watershed sample is indicated by WS (Table 3-1).

91

5. Groundwater Flow Model

5.1 Model Discretization and Hydrogeologic Parameters

To determine the sustainable yield of the Wadi Watir delta and unconsolidated deposit aquifers, we employed a groundwater flow model. The model boundaries were chosen to be the Gulf of Aqaba to the east, the mountain block unconsolidated contact to the west, and where this contact meets the Gulf of Aqaba to the north and south

(Figure 3-1). A finite-difference grid oriented along the North-South and the East-West axes of the Wadi Watir Delta was used for the model. There were 60 columns and 120 rows with uniform grid spacing of 98 m (323 feet) in both directions (Figure 3-9A). In the vertical direction (Figure 3-9B), the model grid consisted of 27 layers representing the five hydrostratigraphic units of the delta aquifers (Abbas et al., 2004). Simulated layers from 1 to 12 correspond to the A1 layers group, with hydraulic conductivity of 4 m/day representing the upper most first three hydrostratigraphic units of the heterogeneous alluvial deposits (stratigraphic units in cross section A-B (Figure 3-2) and the modeled cross section C-D (Figure 3-9A and B). Layers 13 through 18 corresponded to A2 with hydraulic conductivity of 11 m/day (El-Refeai, 1992). The A2 layer was the fourth hydrostratigraphic unit, comprised of Pliocene to Pleistocene age; mainly sand and gravel with sand and clayey sand (Mabrouk and Nasr, 1997). Layers 19 to 27 corresponded to the A3 layer with hydraulic conductivity of 0.001 m/day. Layer A3 represented the fifth hydrostratigraphic unit (Figure 3-2), mainly composed of sand and clay intercalations.

The hydraulic conductivity values for the first unit (A1) were obtained from pumping tests in March 2007. The values of hydraulic conductivities for each layer (A1, A2 and

92

3216000 A Head Observations 1 Constant Head Coubled with Constant Concentration Boundary

3215000 Pumping Wells

1616 Inactive Cells

5 41a Artificial Recharge 45 Wells coupled with 3214000 46 42a a point source of 1000 mg/l 44 40a 43 39a

3213000 10 2121

33

3212000 15 38a

5542 4 4 37a 3131 33a 23a 1515 54 37 3211000 30a 10 10 5 58 21a 2 3232 5 M8 53 6 6 20a 3333 7 9 9 24a 362 29a 35a 51 50 36 38 11 C 11 41 D 19a 39 20 M1 34 34 25a1717 52 32a 403535 3210000 1919 18 20 18 1 1 20 M2 22 22 18a 3209000 13 13 45a 23 17a 53a 23 25 56 50a 49 24 2457 46a 47 3208000 2525 27 48a 29 27 26 29 48 8 28 26 28 12 14148 12

30 30 30 3207000

3206000 35

3205000 0 1 2 Km. 660000 661000 662000 663000 664000 665000

B

Figure 3-9. The finite difference grid cells used for the model in the x, y (A), and z (B) directions. The number of artificial recharge wells is 5 increments, and the sites denoted by letter "a" are by the author indicated in Table 3-1.

93

A3) are corresponded within the ranges of hydraulic conductivities for unconsolidated sediments (Fetter 2001, and Freeze and Cherry 1979). Land surface elevations used in the model were obtained from the Digital Elevation Map STRM-90 m (USGS, 2004) coupled with 23 ground elevation surveyed spots representing water points. The model layer thickness was determined from the geo-electrical cross section of Abbas et al. (2004)

(Figure 3-2).

Vertical hydraulic conductivity values were assumed to be one tenth of the values of the horizontal ones. Five infiltration tests have been conducted for the surficial layer

(A1) at different sites selected according to different soil types investigated in the field site, using the double-ring infiltometer methods (Philip 1957a; Hanks and Aschroft

1980). The values of vertical hydraulic conductivities at these sites (Sites 1 to 5 in Figure

3-1) are 0.01, 0.09, 0.13, 1.56, and 1.67 m/day, respectively with a geometric mean of

0.19. Two slug tests have been carried out at two shallow dug wells (represent Layer A1 in the model Figure 3-9B) (Wells 25 and 42 in Figure 3-1) and the hydraulic conductivity has been estimated using the method of Bower and Rice 1976. The hydraulic conductivity estimated at site 25 was 0.43 m/day while it was 2.83 m/day at site 42. Total porosity was assumed to be homogeneous and was set to 0.1, where aquifer materials composed mostly of poorly sorted silt, sand, gravel and clay (Fetter 2001 and Johnson,

1976). The storage coefficient value used for the model is 3.14x10-4 (Himida, 1997). The longitudinal dispersivity also was assumed to be homogeneous and set by 100 meters, and the ratio of horizontal transverse dispersivity to longitudinal dispersivity was assumed to be 0.1, while the ratio of vertical transverse dispersivity to longitudinal

94

dispersivity was assumed to be 0.01 (Gelhar et al., 1992). The diffusion coefficient was assumed to be 10-9 cm/sec.

5.2 Initial Conditions of the Groundwater Flow Model

Initial water levels were specified for the groundwater flow model by interpolating between the known values of the water level in February 1986 (El Kiki et al., 1992) and in November 1994 (Ismail, 198). The initial groundwater salinity concentrations were assumed to be 1000 mg/l for the entire study area according to the average of chemical analyses of three nearby samples collected from Furtaga Springs, located about 10 kilometer west in the upstream in the Wadi Watir watershed and represent the main source of groundwater recharge for the Wadi Watir delta (El Ghazawi,

1999). Results from early simulations suggested that better estimates of initial groundwater salinity would reduce the simulated time required to achieve a stable position of the saltwater interface from seawater intrusion into the Wadi Watir delta aquifers. Model simulations were conducted until the position of the interface between seawater and groundwater reached the dynamic equilibrium. Results from the steady- state simulation, including both water levels and groundwater salinity, were used as initial conditions for the transient simulation.

5.3 Groundwater Withdrawal

We simulated pumping wells using the well package in MODFLOW (Harbaugh et al., 2000). Pumping wells are treated in the package by specifying the location of each well and its pumping rate. Actual pumping started in the delta after 1982; before that time, pumping was restricted to 15 shallow hand-dug wells to meet domestic uses (El

95

Ghazawi 1999). Due to the extension of cultivated projects in Sinai between 1982 and

1996, twelve wells were drilled by Sinai Development Authority and Desert Research

Center (El Ghazawi 1999). Six of them were productive, whereas some were abounded due to high salinity or low productivity (Shalaby, 1997). There are nine water supply wells, (eight drilled; one hand dug) located along the west part of the study area where

Wadi Watir enters the delta (Table 3-7 and Figure 3-9A). The pumping rates from all hand dug wells assumed to be 1 m3/day except well sites 25 and 35 have 10 m3/day, according to the average daily withdrawal rates that obtained from well owners during our field visits. The pumping rates for the drilled wells obtained from the well site engineer, and it is varied from 1982 to 2009 (Table 3-7) depending on groundwater availability and flash flood intensity and/or frequency (Table 3-8). Within the model, each well has its individual withdrawal rate specified for each stress period, specified according to the variation of pumping within the time. The total average withdrawal rates for all eight wells were 3640 to 3900 m3/day from 1986 to 1994 and 2500 m3/day from

1994 to 1998. Because of the drought that started after 1998, discharge was decreased to

1600 m3/day starting from December 1998 to 2009 (Table 3-7).

96

Table 3-7. The amount of groundwater pumping from water supply wells on the delta. Dec Dec Dec Feb 1982 Feb 1986 June 1987 Dec 1994 1998 2001 2008 To To To To To To To Well No. Feb 1986 June 1987 Dec 1994 Dec1998 Dec Dec Apr 2001 2008 2009 Q Q Q Q Q m3/day Q m3/day Q m3/day m3/day m3/day m3/day m3/day M1 -870 -870 -700 -500 Stopped M2 -- -700 -700 -500 18a -1070 -1100 -500 -500 -500 -500 17a -1000 -1000 -500 -200 -200 Stopped M8 Not -400 -500 Stopped 21a Drilled yet Not Drilled -200 -200 -200 Not Drilled 20a Yet Not Drilled Yet -200 -200 -200 Yet 19a -500 -500 -500 Withdra -870 -3640 -3900 -2500 -1600 -1600 -1400 wal The well Sites are indicated in Figure 3-9A. Wells donated by letter "a" after the present authors and indicated in (Table 3-1).

Table 3-8. Flash flood occurrence and volume (Himida, 1994; JICA, 1999 and Cools et al., 2012). Flood Data Volume Flash flood Remarks (106 m3) Oct-16-1987 45 Very High Disaster Dec-20-1987 -- Low -- Apr-20-1988 5 Moderate Catchment outlet Oct-16-1988 15 High Catchment outlet Mar-12-1990 -- Low -- Oct-20-1990 35 High Catchment outlet Mar-22-91 -- Moderate -- Mar-1993 -- High Catchment outlet Oct-1993 -- High Catchment outlet Jan-1-1994 -- Moderate Catchment outlet Nov-2-1994 -- Moderate -- Nov-17-1996 -- Moderate -- Jan-14-1997 -- Moderate -- May-16, 17-1997 4.4 Moderate Catchment outlet May-28-1997 0.27 Low Catchment outlet Oct-18-1997* -- Very High Disaster Jan-15-2000 -- Low -- Dec-9-2000 -- Low -- Oct-(27-31)-2002 -- Moderate -- Nov-3-2002 -- Low -- Dec-15-2003 -- Low -- Feb-5-2004 -- Low -- Oct-29-2004 -- Low -- Oct-24-2008 -- Low -- *This flood continued for six days (Oct. 19-24); -- No data

97

5.4 Boundary Conditions

Since the model was set up to simulate seawater-groundwater mixing into coastal aquifers simultaneously involving both groundwater flow and solute transport processes, boundary conditions for these two processes were required. The model base was assumed to be a no-flow boundary, represented by the consolidated granitic aquitard bed rock at the base. For the flow model, a constant head boundary was set to sea level at the coast, representing the mean seawater level at the shore line of the Gulf of Aqaba. Whereas a constant concentration boundary was applied for the 27 model layers at the coast, this concentration was assumed to be 41,000 mg/l based on the analysis of Gulf water samples (Table 3-1). The Wadi Watir delta aquifers were primarily recharged from Ain

Furtage and from the subsurface flow from basement aquifer (El Ghazawi, 1999) along the western boundary of the groundwater flow model area. So the boundary between the basement and unconsolidated delta aquifers was simulated by a point-source boundary with a value of 1000 mg/l. This value was based on the average of salinity of three samples collected from nearby Furtaga Springs, located about 10 kilometers west in the upstream (Figure 3-9A) (Eissa et al., 2012, in review).

5.5 Model Calibration

Model calibration was achieved through trial and error by adjusting the values of recharge through the injection recharge wells at the boundary between mountain blocks and the alluvial aquifer, until there was general agreement between simulated and measured water levels. A total of 58 observations were used (Figure 3-9A). The relative error (RE = [(Abs (Head obs – Head calc) / (N x (Head max – Head min)] has been used to

98

calculate the error percent between observed and calculated head. The model calibration was obtained when a reasonable value of the relative error of the modeled water level became 12.6% (Figure 3-10; Table 3-9). The relatively high of error values obtained between measured and simulated heads mainly attributed to the highly transient state through long time periods which occurred due to extensive pumping through the drilled wells located very close, with distance ranges from 250 to 500 meters in between.

Moreover, the flat water level in the Wadi Watir delta, where the maximum water level relative to sea level is 2.5 meter and the minimum is 0.23 meter, with a mean average value of 1 meter, makes it harder to get all head observations matched with the calculated ones.

3

2.5

2

1.5 Feb-1986

Nov-1994 1

CalculatedHead (meter) Apr-1995

0.5 Oct-1995

Mar-2007 0 0 0.5 1 1.5 2 2.5 3 Observed Head (meter) Figure 3-10. Measured versus model calculated water level cells in the groundwater flow model that contained wells. (Observations in Feb-1986 by El-Refeai, 1992; Nov-1994; and Apr-1995 by Ismail, 1998; and Oct-1995 by El Ghazawi, 1999).

99

Table 3-9. Modeled and simulated groundwater levels in different years through the modeled time. Observation Observe Calculated Calc.-Obs. (m) Time 1 1.4 1.0 -0.39 Feb-86 No. Head (m) Head (m) (month-Year) 2 1.9 1.2 -0.73 Feb-86 3 1.8 1.3 -0.47 Nov-94 4 1.3 1.0 -0.29 Nov-94 5 0.8 0.7 -0.05 Nov-94 6 0.2 0.5 0.24 Nov-94 7 1.2 0.8 -0.39 Nov-94 8 0.6 1.1 0.46 Nov-94 9 2.2 1.5 -0.69 Nov-94 10 1.0 0.6 -0.34 Apr-95 11 1.2 0.8 -0.38 Apr-95 12 0.3 0.7 0.40 Apr-95 13 0.8 0.6 -0.20 Apr-95 14 0.6 1.1 0.56 Apr-95 15 1.3 1.1 -0.20 Apr-95 16 1.0 1.1 0.09 Oct-95 17 1.0 0.7 -0.27 Oct-95 18 0.6 0.6 -0.03 Oct-95 19 1.0 0.7 -0.30 Oct-95 20 1.1 0.7 -0.39 Oct-95 21 2.5 2.0 -0.47 Oct-95 22 1.0 0.7 -0.34 Oct-95 23 0.5 0.6 0.11 Oct-95 24 0.3 0.7 0.38 Oct-95 25 0.7 1.3 0.52 Oct-95 26 1.0 0.7 -0.27 Oct-95 27 0.8 0.7 -0.03 Oct-95 28 1.0 0.7 -0.27 Oct-95 29 0.8 0.8 -0.05 Oct-95 30 0.9 0.8 -0.12 Oct-95 31 1.8 1.1 -0.66 Oct-95 32 1.1 0.8 -0.33 Oct-95 33 1.5 1.0 -0.52 Oct-95 34 1.1 0.7 -0.41 Oct-95 35 1.0 0.7 -0.33 Oct-95 36 0.8 0.7 -0.05 Oct-95 19a 0.6 1.1 0.37 Mar-07

100

Table 3-9. Modeled and simulated groundwater levels in different years through the modeled time (continued). Observation Observe Calculated Calc.-Obs. (m) Time 20a 1.5 1 -0.51 Mar-07 No. Head (m) Head (m) (month-Year) 21a 0.7 1.1 0.37 Mar-07 23a 1.5 1.6 0.1 Mar-07 24a 0.9 1 0.12 Mar-07 25a 0.7 1 0.3 Mar-07 29a 0.9 0.9 0.03 Mar-07 30a 1.2 0.9 -0.25 Mar-07 32a 1.2 0.8 -0.38 Mar-07 33a 1.1 0.8 -0.3 Mar-07 35a 0.9 0.5 -0.36 Mar-07 37a 0.9 0.5 -0.35 Mar-07 38a 0.7 0.6 -0.08 Mar-07 39a 1.1 1.3 0.2 Mar-07 40a 1.1 1.3 0.17 Mar-07 41a 1.1 1.2 0.14 Mar-07 42a 1.2 1.3 0.08 Mar-07 45a 1.4 1.7 0.35 Mar-07 46a 1.5 1.5 -0.04 Mar-07 48a 1.1 1.3 0.2 Mar-07 50a 0.9 1 0.1 Mar-07 53a 1.1 1 -0.14 Mar-07 Observations in Feb-1986 by El-Refeai, 1992; Nov-1994; and Apr-1995 by Ismail, 1998; and Oct-1995 by El Ghazawi, 1999. Observation sites are indicated in Figure 3-9A. Observations donated by letter "a" are by the author and represented in Table 3-1.

5.6 Recharge

Annual average rainfall for the Watir valley is 35 mm/year, which ranging from

10 mm/year at the coast to 50 mm/year in the high land areas (Cools et al., 2012). Since this is a small amount, the modeled recharge was assumed to be zero. The main recharge to the delta came mainly from subsurface groundwater flow from the mountain blocks to the delta (Ismail 1998 and El Ghazawi 1999). Flash floods that recharged the delta aquifers occurred on average about every six years (Himida, 1994); but from October

1997 to the end of 2009, there were no flash floods (Table 3-8).

101

Rainfall precipitation for the whole Wadi Watir watershed is estimated by

70.23 x 106 m3/year, while the groundwater recharge for the entire Watir Watir watershed is 38.63 x 106 m3/year (JICA, 1999). According to Milewski (2009) estimates the average annual recharge for the Wadi Watir watershed (1998–2007) by 31.30 x 106 m3/year.

Concerning the Wadi Watir delta, El Ghazawi (1999) and El Sayed (2002) mentioned that the aquifer is recharged from subsurface and from the outlet of Wadi

Watir through the surface overflow water from Ain Furtaga. Discharge from Ain Furtaga varies considerably according to the amount of rainfall and storm frequency during the year (Issar and Gilad 1982). The recorded data from 1969 to 1979 after (Issar and Gilad,

1982) showed the discharge from Furtaga Spring was ranged from 115 m3/day to 3600 m3/day with average 1800 m3/day during this time period. RIWR (1989) estimate the discharge of Furtaga Spring to be 1644 m3/day, (Idris, 1995) estimate it by 2500 m3/day and it was estimated during the two field trip to be 192 m3/day in March 2007, and the spring stopped flowing in our field visit in August 2009.

In this study, the main recharge to the study area was simulated by 39 injection wells located along the boundary between the consolidated rock of the mountain block and the unconsolidated deposits of the Wadi Watir delta aquifers (Figure 3-9A).

According to calibrated model results indicated in Table 3-10, total recharge was estimated to be 2950 m3/day as steady state (prior to 1982), 3100 m3/day between the period 1982 to 1987, 6230 m3/day between the period of 1987 to 2002, and 4060 m3/day through 2002 to 2009 (Figure 3-11). These estimated recharge values coincide with the total withdrawal amount reported by Himida (1994) of 5000 m3/day from the drilled

102

wells and the flood intensity and frequency, with a relatively higher from 1987 to 2002 and lower through the draught period from 2002 to 2008 (Table 3-8). The annual average for the groundwater recharge for the Wadi Watir delta is estimated by 1.58x106 m3/year which is equivalent to 4200 m3/day through the model time (1982-2009).

Table 3-10. Simulated groundwater recharge for injection wells in the groundwater flow model. Recharge for Recharge from Recharge from Recharge from Well No. the steady state 1982-1987 1987-2002 2002 -2009 in 1982 (m3/day) (m3/day) (m3/day) (m3/day) 1 80 80 80 80 2 80 80 80 80 3 80 80 80 80 4 80 80 80 80 5 80 80 80 80 6 80 150 150 150 7 80 80 80 80 8 80 150 150 150 9 80 150 150 150 10 80 80 80 80 11 80 80 80 80 12 80 80 80 80 13 100 110 110 110 14 100 110 400 200 15 100 110 400 200 16 100 110 400 200 17 100 110 400 200 18 100 110 400 200 19 100 110 400 200 20 100 110 400 200 21 100 110 400 200 22 100 110 300 150 23 100 110 300 150 24 100 110 300 150 25 30 40 150 80 26 30 40 100 80 27 30 40 150 80 28 30 40 100 40 29 60 40 40 40

103

Table 3-10. Simulated groundwater recharge for injection wells in the groundwater flow model (continued). Recharge for Recharge from Recharge from Recharge from Well No. the steady state 1982-1987 1987-2002 2002 -2009 in 1982 (m3/day) (m3/day) (m3/day) (m3/day) 30 60 40 40 40 31 60 40 40 40 32 60 100 400 100 33 80 30 30 30 34 100 40 40 40 35 60 30 30 30 36 30 40 40 40 37 60 30 30 30 38 60 30 30 30 39 60 30 30 30 Total 2950 3100 6230 4060 "Well No." donates to artificial recharge well sites used for the model and are shown in Figure 3-9A in a white box.

Figure 3-11. Estimated recharge for the Wadi Watir delta between the periods from 1982 to 2009.

104

5.7 SEAWAT Model Results

Variable-density flows (VDF) in porous media show worldwide phenomena in sea water intrusion problems (Langevin 2001). The SEAWAT-2000 code has been applied to simulate variable-density groundwater flow (Langevin et al., 2002).

SEAWAT-2000 contains all the processes of MODFLOW-2000 and also includes the variable-density flow process (as an alternative to the constant density groundwater flow process) and the integrated MT3DMS transport process (Guo and Langevin, 2002;

Langevin et al., 2003). SEAWAT-2000 used to simulate the transient system.

Our transient SEAWAT model results (1982–2009) showed a steepening of the hydraulic gradient due to groundwater extraction from the aquifer. The model results showed groundwater level change as a result of increased pumping and less recharge during the last decades, as previously described. We found that local pumping formed significant cones of depression around pumping sites that tended to increase in area through different times in 1988 to 2009. This indicated that the aquifer cannot sustain these large pumping rates (Figure 3-12 A-D). In November 1998, groundwater flow results, do not show a cone of depression, mainly due to higher recharge amount for the aquifer from many flood events through this time with a disastrous one in October 1997

(Table 3-8) when the floods lasts for six days.

From SEAWAT results shown in the vertical cross section C` - D` passed through well 35 (Figure 3-13 A-D), it is clear that, at the west (the left side of the cross section), the groundwater salinization observed at the group of drilled wells located at the outlet of

Wadi Watir at the front of the mountain block (Sites 17–23) is not a seawater intrusion

105

A B

C D

Figure 3-12. Simulated groundwater levels from the calibrated MODFLOW model. 1982 (Steady State) B) Nov-1988 (2465-day) C) Nov-1998 (6147-day) D) March-2009 (9890-day).

106

A

B

C

D

Figure 3-13. SEAWAT model output for a vertical west to east cross section through well No. 35, showing increasing salinity through time. Simulated salinity from the model for the years A) 1982 (Steady State) B) Nov-1988 (2465-day) C) Nov-1998 (6147-day) D) March- 2009 (9890-day).

107

problem. Modeled salinity values at these sites did not exceed 1000 mg/l, however, the observed salinities recorded at these wells ranged between 1925 mg/l (well

20) and 3609 mg/l (well 19) although the recharge water salinity averaged 1000 mg/l.

The isotopic signature of these wells didn’t show any evaporation or mixing with seawater. The high salinity was mainly due to upwelling of deep saline water due to over pumping through these wells. El-Refeai (1992) measured salinity with depth at deep drilled wells close to these sites and reported the salinity has increased up to 18,000 ppm at 75 m depth under from the ground surface (Figure 3-14). The deep saline layer is not incorporated in SEAWAT-2000.

Figure 3-14. Salinity with depth after El-Refeai (1992).

At the east, the right side of the C` - D` cross section, the SEAWAT model results showed that the solute contour lines concave toward the left, due to the presence of a

108

sand layer of relatively higher conductivity (K=11 m/day) which overlies by the alluvial deposits (k = 4 m/day) and is underlain by sand intercalated with shale and clay (k =

0.001 m/day) from the bottom. Seawater encroachment increased with time from 1982 to

2009 at the coast as a result of pumping stresses shown in (Figure 3-13 A-D). The

SEAWAT modeling results showed that seawater is invaded by the delta land about 200 m toward the west. The solute inland transgression was due to the transient state and occurred as a result of groundwater extraction through the modeled time (1986 to 2009).

The observed total dissolved solids at well 35 and well 42 increased with time through the period from 1996 to 2009 (Figure 3-15).

Figure 3-15. Total dissolved solids (mg/l) recorded at Well No. 35 and Well No. 42 from 1996 to 2009. The sallinity record in 1996 and 1998 after (Said, 2004)

109

6. Summary and Conclusions

The Wadi Watir delta is located at the southeastern part of the Sinai Peninsula, bordered along the north, south, and east by the Gulf of Aqaba. The climate in this area is arid and semi-arid. Deterioration of groundwater quality and seawater intrusion problems could become a crisis for the local economy.

Our chemistry and environmental isotope results for groundwater in the Wadi

Watir delta showed the main source for the groundwater recharge comes from the Wadi

Watir watershed. The groundwater isotopic signature in the Wadi Watir delta discriminated into two groups. Group I was characterized by relatively low groundwater salinity and depletion of isotopic signature and received its water mainly through subsurface flow coming from the mountain blocks and did not show any evaporation trend. Group II was characterized by relatively high saline groundwater and isotopic enrichment due to dissolution of wetted Sabkha and/or seawater intrusion. This group was restricted and parallel to the coastline or located in the vicinity of the Sabkha zone.

Groundwater samples Nos. 35 and 42 graphed and located at the trend line between seawater and recharge water due to mixing with Gulf of Aqaba water as a result of over pumping.

The geochemical NETPATH model simulations confirmed previous results

(Plummer et al., 1994). The NETPATH model discriminates groundwater into two groups. Group I is low saline groundwater which has evolved due to dissolution of salts and minerals through the aquifer matrix along the flow path. The NETPATH model converges with a low mass transfer without showing any evaporation or mixing with

110

seawater or Sabkha waters. Group II is characterized by relatively high salinity. The mixing percent calculated by the NETPATH model ranged between 12% (Well 35) to 5%

(well 42) seawater to 88% to 93% groundwater, respectively.

SEAWAT-2000 was used to simulate the transient groundwater system, cross section C-D passed through well No. 35 located on the eastern side at the coast and showed transgression of salinity toward the inland by about 200 meters due to the seawater encroachment as a result of the groundwater abstraction through the modeled time. The model simulations confirmed that seawater intrusion was recorded at the coast while salinization of the groundwater inland at the outlet of Wadi Watir mainly was due to over upwelling of the deep saline layer as a result of over pumping. Further investigation of upwelling of the saline deep layer is recommended and along with additional modeling in the future to determine the main source for solutes and salinization leading to groundwater deterioration. The main recharge to the delta comes from the outlet of Wadi Watir at the front of well field location sites, model simulations estimated that the annual average groundwater recharge rate to the study area is 1.58 x

106 m3/year (about 4300 m3/day), through the model duration time (1982-2009).

7. References Cited

Abbas, A.M., Atya, A.A., Al-Sayed E.A., Kamei H., 2004. Assessment of groundwater resources of the Nuweiba area at Sinai Peninsula, Egypt by using geoelectric data corrected for the influence of near surface inhomogeneities, Journal of Applied Geophysics 56(2):107-122.

111

Back W., Hanshaw B.B., Plummer L.N., Rahn P.H., Rightmire C.T., Rubin M., 1983. Process and rate of dedolomitization, mass transfer and 14C dating in a regional carbonate aquifer. Geol Soc Am Bull 94 (12): 1415-1429. Bakker, M., Oude Essink. Bouwer, H. and Rice, R.C., 1976. A slug test method for determining hydraulic conductivity of unconfined aquifers with completely or partially penetrating wells, Water Resources Research, vol. 12, no. 3, pp. 423-428. G.H.P., and Langevin, C.D., 2004. The rotating movement of three immiscible fluids, a benchmark problem. J Hydrol 287:270–278. Clark, I. and Fritz, P., 1997. Environmental isotopes in hydrogeology. Lewis Publishers, New York. CONOCO Continental Oil Company, 1987. Geologic map of Egypt (Scale 1:500,000). Cools. J., Vanderkimpen P., El Afandi, Abdelkhalek A., Fockedey S., El Sammany M., Abdallah G., El Bihery M., Bauwens W., and Huygens M., 2012. An early warning system for flash floods in hyper-arid Egypt. Nat. Hazards Earth Syst. Sci., 12, 443–457, 2012. Eissa, M.A., Thomas, J.M., Hershey, R.L., Dawoud, M.I., Pohll, G., Gomaa, M.A., and Kamal A.D., (in review, Environmental Earth Sciences). Geochemical and Isotopic Evolution of Groundwater in the Wadi Watir Watershed, Sinai Peninsula, Egypt. El Ghazawi, M.M., 1999. Reconsideration of hydrogeologic setting in the delta of Wadi Watir, southern Sinai. Bulletin of Science, Mansoura Univ., Egypt, vol. 26 (1). El Kiki, M.F., Eweida, E.A., and El-Refeai, A.A., 1992. Hydrogeology of the Aqaba rift border province. Proc. of the 3rd Conf. Geol. Sinai Develop., Ismailia, Egypt. pp. 91-100. El-Refeai, A.A., 1992. Water resources of southern Sinai, Egypt geomorphological and hydrogeological studies. Ph.D. Thesis, Fac. Sci., Cairo Univ., Egypt, 357p. File Report 00-92, 121P.

112

El Sayed, M.H, Abd El-Samie, G., Sallouma, M.K., and Said, M.M., 2002. Hydrochemical parameters to detect saltwater intrusion in the groundwater aquifer of delta Wadi Watir, Gulf of Aqaba, Southeast Sinai, Egypt. Bulletin of Egypt. J. Appl. Sci. 17(2):350-389. El-Shazly, E.M., Abddel-Hady, M.A., El-Ghawaby, M.A., El-Kassas, I.A., El-Shazly, M.M., 1974. Geology of Sinai Peninsula from ERTS-1 Satellite Images, Acad. Sci. Res. Tech., Remote Sence. Project. Epstein, S. and Mayeda, T., 1953. Variation of O18 content of waters from natural sources. Geochimice et Cosmochimica Acta 4:213-224. Eyal, M., Bartov, Y., Shimron, A.E., Bentor, Y. K., 1980. Sinai geologic map, Scale 1:500,000. Ministry of Energy and Infrastructure. Admin. For Res. In Earth Sci., Israel. Fetter C.W., 2001. Applied Hydrogeology 4th Edition. Prentice-Hall, Inc. Fishman, M. J. and Friedman, L.C., 1989. Methods for the determination of inorganic substances in water and fluvial sediments. Techniques of Water-Resources Investigations of the United States Geological Survey, Book 5, Chapter A1, USGS, Alexandria, VA. Freeze, R.A., Cherry, J.A., 1979. Groundwater. Prentice-Hall, Englewood Cliffs: 604 pp. Gat, J.R., Mazor, E. and Tzur, Y., 1969. The stable isotope composition of mineral waters inthe Jordan Rift Valley, Israel. J. Hydrology, 76: 334-352. Gavish, E., 1974. Geochemistry and mineralogy of a recent sabkha along the coast of Sinai,. Gulf of Suez. - Sedimentology 21 .-397-414. Gelhar, L.W., Welty, C., and Rehfeldt, K.W., 1992, A critical review of data on field- scale dispersion in aquifers, water resource. Research, 28(7), 1955-1974. Guo, W. and Langevin, C.D., 2002. User’s guide to SEAWAT. A computer program for the simulation of three-dimensional, variable-density groundwater flow. USGS Techniques of Water Resources Investigations, Book 6, Chapter A7.

113

Hanks, R.J., and Ashcroft, G.L., 1980. Applied soil physics. Advanced Studies Agricultural Science. Springer, Berlin, Heidelberg, New York, 159 pp. Hassan, A.A., 1967. A new carboniferous occurrence in the Abu Durba, Sinai, Egypt: 6th Arab. Petrol Congr., Baghdad, 2, 8p. Harbaugh, A.W., Banta, E.R., Hill, M.C., and McDonald, M.G., 2000. MODFLOW- 2000, the U.S. Geological Survey Modular Ground- Water Model-User Guide to Modularization Concepts and the Groundwater Flow Processes. U.S. Geological Survey Open-File. Hershey R. L., Heilweil V. M., Gardner P., Lyles B., Earman S., Thomas J., and Lundmark K.W., 2007. Ground-water Chemistry Interpretations Supporting the Basin and Range Regional Carbonate-rock Aquifer System (BARCAS) Study, Eastern Nevada and Western Utah. Prepared by Desert Research Institute, Nevada System of Higher Education and U.S. Geological Survey. DHS Publication No. 41230 Himida, I.H., 1994. Water resources in delta Wadi Watir (Nuweiba area) and improvement of the hydrologic setting in the area (in Arabic). Internal Technical Report, Desert Research Center, 39 p. Himida, I.H., 1997. Water resources of Wadi Watir. Internal report, Desert Research Center (In Arabic). I.A.E.A., 1981. Stable isotope hydrology. Deuterium and oxygen-18 in water cycle. In: J.R. Gat,and R. Gonfiantini, (ed.), International Atomic Energy Agency Technical Report No. 210,Vienna, 339 p. Ismail, Y.L., 1998. Hydrogeological and hydrochemical studies on Wadi Watir area, South Sinai, Egypt. Ph.D. Thesis, Fac. Sci., Suez Canal Univ., Egypt, 239 p. Issar, A., and Gilad, D., 1982. Groundwater flow system in the arid crystalline province of Southern Sinai. Jour. Desert Sci. Hydro. 27:309-325.

114

J.I.C.A., 1999. Water Resources Research Institute, JICA (1999). South Sinai groundwater resources study in the Arab Republic of Egypt. Final report, El Qanater El Khayria, NWRC, Cairo, Egypt. Johnson, A.I., 1976. Specific yield-Compilation of specific yield for various materials. US. Geological Survey Water-Supply Paper 1662-D. Langevin, C.D., 2001. Simulation of groundwater discharge to Biscayne Bay, southeastern Florida: US. Geological Survey Water-Resource Investigation Report 00-4251, 127p. Langevin D., Christian, and Guo W., 2002. User's Guide to SEAWAT: A Computer Program for Simulation of Three-dimensional Variable-Density Ground-Water Flow U.S. Geological Survey Techniques of water resource investigations 6-A7. Langevin C.D, Shoemaker, W.B., and Guo W., 2003. MODFLOW-2000, the U.S. Geological Survey Modular Groundwater Model-Documentation of the SEAWAT-2000 Version with Variable-Density Flow Process (VDF) and the Integrated MT3DMS Transport Processes (IMT). U.S. Geological Survey, Tallahassee, Florida. Langevin, C.D., 2003. Simulation of submarine groundwater discharge to a marine estuary: Biscayne Bay, Florida. Ground Water 41(6). Langevin, C.D., Shoemaker, W.B., and Guo, W., 2003. MODFLOW-2000, the U.S. Geological Survey Modular Groundwater Model-Documentation of the SEAWAT-2000 version with the variable density flow process (VDF) and the integrated MT3DMS Transport Process (IMT). USGS Open-File Report 03-426. Mabrouk M.A., and Nasr I.M., 1997. The Hydrogeologic Setting In The Delta Of Wadi Watir, Gulf of Aqaba, A Geoelectrical Resistivity Sounding Approach. Desert Inst. Bull., Egypt. 47 No. 1, 149-162. Milewski, A., Sultan, M., Eugene, Y., Abdeldayem, A., and Abdel Gelil, K., 2009. A remote sensing solution for estimating runoff and recharge in arid environments. J Hydrol 373:1–14.

115

Morrison, J., Brockwell, T., Merren, T., Fourel, F. and Phillips, A.M., 2001. On-line high-precision stable hydrogen isotopic analyses on nanoliter water samples. Analytical Chemistry, 73: 3570–3575. Philip, J.R., 1957a. The Theory of infiltration, the infiltration equations and its solution. Soil Sci., 83: 345-357. Plummer L.N., 1992. Geochemical modeling of water-rock interaction: Past, present, future. In: Kharaka YK Maest AS (Eds) Water-Rock Interaction, Balkema, Rotterdam, Netherlands: 23-33. Plummer, L.N., Prestemon E.C. and D.L. Parkhurst, 1994. An interactive code (NETPATH) for modeling net geochemical reactions along a flow path, version 2.0. U.S. Geological Survey Water Resource Investigations Report 94-4169, 130 p. Rainwater, F. H. and Thatcher, L. I., 1960. Methods for collection and analysis of water samples. U.S.Geol. Survey, Water Supply, 1454, 301p. Research Institute for Water Resources (RIWR) 1989. Sinai water resources study (phase II). Internal Report, 191p. Said, M.M., 2004. Geochemistry of groundwater in coastal areas, South Sinai, Egypt. Ph.D. Thesis, Fac. Sci., Ain Shams Univ., Egypt. Said, R., 1971. Explanatory notes to accompany the geological map of Egypt: Egypt Geol. Surv., No. 56, 123 p. Shabana, A.R., 1998. Geology of water resources in some catchments areas draining in the Gulf of Aqaba, Sinai-Egypt. Ph.D. Thesis, Fac. Sci., Ain Shams Univ., Egypt, 246 p. Shalaby, A.I., 1997. Geomorphology and hydrogeology of Wadi Watir basin, S. E. Sinai, Egypt‖. M.Sc. Thesis, Fac. Sci., Mansoura Univ., Egypt, 185 p. U. S. Environmental Protection Agency, 1973. Water quality criteria, 1972. Washington, D.C., U.S. Government Printing Office, 594 p.

116

United States Geological Survey (USGS), 2004. Reprocessing by the Global Land Cover Facility, 2004, (30) Arc Second SRTM Elevation. College Park, Maryland: the GLCF, Retrieved from World Wide Web. http://www.landcover.org. Zheng, C., Wang, P.P., 1999. MT3DMS, a modular three-dimensional multi-species transport model for simulation of advection, dispersion and chemical reactions of contaminants in groundwater systems; documentation and user's guide. U.S. AERDCC Report SERDP-99-1, Vicksburg, MS.

117

CHAPTER 4

Groundwater Recharge and Seawater Intrusion of the Quaternary Coastal Plain

Aquifer in the Wadi Watir Delta, Sinai, Egypt

Mustafa A. Eissa1,2*, James M. Thomas2, Greg Pohll2 Orfan Shouakar-Stash3, Ronald L.

Hershey2, Kamal A. Dahab4, Maher I. Dawoud4, and Mohamed A. Gomaa1

1Desert Research Center, Hydrogeochemistry Dept. Matariya, Cairo, Egypt

2Desert Research Institute, Division of Hydrologic Sciences, Reno, Nevada, USA,

[email protected]

3University of Waterloo, Department of Earth Sciences, Waterloo, Ont., Canada

4Menoufiya University, Faculty of Science, Geology Dept., Shebein El Kom, El

Menoufiya Egypt

118

Abstract

The hydrogeology of the Wadi Watir delta located in Sinai, Egypt consists of alluvial aquifers underlain by impermeable Precambrian basement. The scarcity of rainfall during the last decade, combined with high pumping rates, has resulted in the degradation of water quality in the main supply wells along the mountain front and in wells along the coast. This deterioration in the water quality has required a reduction in groundwater production. The main factors controlling groundwater salinity in the delta are water-rock interactions, seawater intrusion near coastal areas, mixing with deep saline water due to over pumping, and evaporation. The chemistry and stable isotopes of groundwater, including 2/1H, 18/16O, 87Sr/86Sr, δ37Cl and δ81Br, were utilized to determine the groundwater chemical evolution and the sources of groundwater in the delta aquifers.

This in turn, provided a better understanding of the mixing scenarios between the different water types and especially the extent of seawater intrusion along the coast. Also during this investigation, 87Sr/86Sr, δ37Cl and δ81Br of different rock types from the watershed were determined. This information was important for improving our understanding of the impact of water-rock interactions on groundwater chemistry and recharge sources. Inverse geochemical modeling has been used for determining geochemical processes that account for the hydrochemical and isotopic changes in the groundwater and several possible models that identify water-rock interactions, mixing, and evaporation are presented.

The isotopic data were also utilized to develop a three-dimensional, variable- density, flow and transport model using the SEAWAT modeling environment. The

119

models were developed to estimate average annual groundwater recharge and to simulate annual groundwater pumping, upwelling of saline water from beneath the well field, and seawater intrusion along the coast for different pumping scenarios. The models were calibrated using groundwater level and salinity changes.

Geochemical and SEAWAT modeling results show that: (1) groundwater in the main well field contains 6 to 12% of deep saline groundwater; (2) groundwater in some wells along the coast contain 8 to 12.5% of seawater; (3) the main factors controlling groundwater salinity are the pumping stresses and the availability of recharge; (4) For the time period (1982-2009), the daily extraction rate from the individual deep drilled wells of the main well field ranged between 200 and 1400 m3/day with an annual average rate of 3100 m3/day for all the wells; and (5) the annual average recharge to the delta is 6000 m3/day for 1982-2009.

Keywords: groundwater sustainability, groundwater modeling, seawater intrusion, water chemistry, isotopes, Wadi Watir Egypt.

120

1. Background

1.1 Study Area

The Wadi Watir delta area is located at the downstream portion of Wadi Watir on the southeastern part of the Sinai Peninsula, Egypt, between longitude 34○ 38` and 34○

41` E and latitude 28○ 57` and 29○ 03` N (Figure 4-1 a and b). The Wadi Watir watershed drains toward the Gulf of Aqaba and is considered to be the most important watershed in this region because the city of Nuweiba, a tourist destination, is located on its delta and

Nuweiba Harbor is located on the coast of this delta. Ships sailing from Nuweiba Harbor link Egypt with Saudi Arabia and Jordan.

1.2 Geology and Hydrogeology

The Wadi Watir delta consists of alluvial aquifers underlain by impermeable

Precambrian basement. The delta Quaternary deposits are composed mainly of fine-to- course sands, gravels, and boulders, derived from primarily carbonate and basement rocks, often within a silty and clayey matrix (El-Shazly et al., 1974; Eyal et al., 1980; El

Kiki et al., 1992).

The Quaternary deposits of the Wadi Watir delta can be divided into five layers

(see Figure 3-2 in Chapter 3; Abbas et al., 2004). The uppermost two layers are generally

<10 m thick, are grouped as surface layers and are comprised of heterogeneous alluvial deposits. The third layer is a sandy clay layer and is between 30 and 45 m thick. The fourth layer is sand and gravel and is between 20 and 40 m thick. The fifth layer is sand with interlayered clay and is 20 to 50 m thick. These five layers are underlain by bedrock

121

34 00 34 30 29 a Wadi Watir Watershed 30

Well Site

a El Shiekh Attia b [At1 & At2] a q Main A MEDITERRANEAN Channel

Wadi El Ain

[An1-An3] Furtaga 29 Cairo SINAI Springs f o 00

STUDY AREA R E f D l S E u A G EGYPT 0 20 40 Km

Scale

34 39 34 41 Legend

Hand Dug Wells 43 Drilled Wells 41 Cross Section 42 40 0 1 2 Basement Rocks 39 Km. 29 Sabkha Deposits G u l 02 Alluvial deposits f

38 eld o Fi ell 37 f W 23

30 33 21 22 A 24 29 36 35 q 20 A 28 a 25 b 19 27 32 26 a 18 31 34 29 17 44 00 45 50 53 46 47 48 49 51 Sabkha B

52

28 54 58 55

Figure 4-1. a) Location map of Wadi Watir basin, Sinai Peninsula, Egypt; b) Location map of Wadi Watir delta and groundwater wells. Dashed line A-B indicates location of cross section shown in Figure 4-2.

122

50

) 1 Legend > 1 % [Ca:Cl] ) = 1

m K < [Ca:Cl ] Water Type + p a N e 40 ( ( r - l Deep Drilled Wells 42 C r 17b 21b 45

Hand Dug Wells 22 [Mg:Cl] 30 35b 35 34 23 17 (2007) 19b 25 20b 30

) 18 1 17b (2009) K

20 17 49 <

+ 20 50

] a l 21b 55 39

N 48 C 46

( 41

r 37b

/ r

29

a 28

-

N 53 l 26

r 10 [

C 27 31

54

r SO4 r 40 24 (epm%) 44 37 47 50 40 30 20 10 52 33

38 36 10 20 30 40 50 19 51 r Mg (epm%)

10 l

C

1

>

r

]

l

-

)

C

r K

20 /

+ a

a

N

r

N

[

(

[Na:SO4] r 30

1 > 1 l = C 1 r < ) - 40 ) K % + 4 a O N S m ( r r p e [Na:HCO3] (

50

Figure 4-2. Sulin's Diagram for groundwater in the Wadi Watir delta (2007 and 2009).

which is composed mainly of very low permeability granitic rocks. The Wadi Watir delta alluvial aquifers are unconfined water-table aquifers. Depth to water in these aquifers ranges from 2.3 to 40.8 m below land surface. The water level in the Wadi Watir delta aquifers varies from < 1 m to about 2 m above sea level (see Table 3-1 in Chapter 3).

123

2. Methods

2.1 Field and Laboratory Methods

Thirty-nine water samples were collected from 32 hand-dug wells and seven drilled wells in March 2007. In September 2009, an additional six samples were collected. Depth to water, total well depth, pH, and electrical conductivity (EC) were measured in the field. For the determination of groundwater flow direction, ground surface elevation surveying for 22 selected wells at the delta area were measured using the LEICA TCRA1103 PLUS ROBOTIC 3 instrument, a reflector less total station. EC was measured with a YSI model 35 conductivity meter. The pH was measured with a

WTW model LF 538 pH meter. The EC and pH meters were calibrated daily. Water samples were filtered in the field using a 0.45µm cellulose acetate filter paper; and samples for major-ion and isotope analyses were collected in polyethylene bottles. Major- ion water chemistry analyses were conducted in Cairo, Egypt at the Desert Research

Center, Water Central Laboratory using the methods of Rainwater and Thatcher (1960) and Fishman and Friedman (1985). Calcium and magnesium were determined by titration using Na2EDTA. Sodium and potassium were determined by flame photometer using a standard curve. Carbonate and bicarbonate, were determined by titration using sulfuric acid. Sulfate was determined by spectrophotometry. Chloride was determined by volumetric titration using silver nitrate. Silica as (SiO2) was determined by colorimetry using molybdate.

Each sample was analyzed in duplicate and if the difference between the sum of cations and anions was more than 5%, the sample was reanalyzed until an acceptable

124

percent difference was obtained. For the flame photometer and the UV-visible spectrophotometer analyses, a set of standard solutions was measured for every set of 10 samples. If the standard was not verified within 5% the set was measured again until it was within an acceptable range.

To determine the isotopic ratio signatures in different rocks within different aquifers located upstream of Wadi Watir, twenty rock samples were crushed and ground to a powder mortar. A volume ratio of 1:1 rock and water was used and then the sample was shaken for three weeks to reach the equilibrium between water and rock. The samples were centrifuged using the ICE Model HN-SII and filtered through 0.45µm filter paper to obtain the water extract for each sample, which should have the same isotopic signature as the rock within the aquifer.

The δ37Cl, δ81Br and 87Sr/86Sr ratios were analyzed at the University of Waterloo

Environmental Isotopes Laboratory (Table 4-1). The method and procedures described by

McNutt et al., (1990) was used for 87Sr/86Sr, via thermal ionization mass spectroscopy with an analytical precision of 0.00001‰. The chloride stable isotopes were determined using the method described in Eggnkamp (1994), and bromide stable isotopes were determined using the method described in Shouakar-Stash et al., (2005b). Analyses were

37 81 performed on CH3Cl for Cl and CH3Br for δ Br with a precision of 0.1 ‰ for both isotopes. The isotopic ratio was reported in per mil (‰) using isotopic ratio mass spectroscopy (IRMS).

125

2.2 Water-Rock Reaction Modeling

Inverse geochemical modeling (Plummer, 1992) has been widely used for interpreting geochemical processes that account for the hydrochemical and isotopic changes in groundwater along flow paths. The inverse geochemical modeling program used in this study is NETPATH (Plummer et al., 1994). NETPATH calculates net geochemical reactions that account for the observed changes in chemical composition along a flow path between the initial and final water. The geochemical reactions and physical processes that produced observed water chemistry in the study area were determined. This inverse mass-balance model approach does not produce a unique solution, so several possible models are presented for water-rock interactions, mixing, and evaporation in the geochemical modeling section of this paper.

The water-rock reaction models developed for this study were constrained by the major-ion concentrations of the groundwater and the mineral phases in the aquifers of the study area. Halite was included as a phase because it is generally dominant in most of terrestrial deposits of marine origin sediments (Raha and Wata Formations in the Wadi

Watir delta). Also, calcite and dolomite were included because of the presence of carbonate rocks as boulders in the delta. Biotite and albite are dominant in basement rocks, which are embedded as boulders and cobbles in the alluvial deposits. Clay sheets are also present in the sediments and deep layers, so the clay minerals montmorillinite and illite were included in the model as well (see Table 3-2 in Chapter 3). Mineral phases for the mafic and felsic rocks were obtained from thin section analysis of rock samples from aquifers located in the mountain block recharge area, as they are the main source of

126

Quaternary alluvial deposits in the study area. A composite granite that represents felsic igneous rocks found throughout the study area (Ca0.33Mg0.01Na1.02K1.72Si21Al3.74Fe0.74) is also included as a phase in the model and constructed according to the average chemical composition of the X-ray fluorescence of granitic rocks (Eissa et al., 2012 in review).

Water chemistry electroneutrality of 5% or less was the criterion for using a water chemistry analysis in the geochemical models. Samples not meeting this criterion were not used in the geochemical models.

2.3 Groundwater Flow Modeling

The computer program SEAWAT (Guo and Langevin, 2002) was used for groundwater flow and transport modeling for this study because it is capable of simulating three-dimensional variable-density groundwater flow in porous media and seawater intrusion into coastal aquifers. This program combines a groundwater flow model, MODFLOW (Harbaugh et al., 2000), with a solute transport model, MT3DMS

(Zheng and Wang, 1999), into a single program that solves the density-dependent groundwater flow and solute-transport equations derived by Guo and Langevin (2002).

3. Results and Discussions

3.1 Chemical Characteristics of the Delta and the Deep Saline Groundwaters

The primary potable water supply for the Wadi Watir delta is from a well field located next to the mountain block where the Wadi Watir drainage enters the delta

(Figure 4-1; sites 17-23). Pumping of these wells began in 1982, and since that time the pumping rate has varied depending on groundwater availability and recharge (El

Ghazawi, 1999). The major source of recharge is infiltration of infrequent heavy rains

127

and floods in the mountain block in the Wadi Watir watershed and its delta (El-Refeai,

1992). Several studies were conducted over the years to investigate and determine the history of groundwater salinization on the front of Wadi Watir delta and at the coast.

Groundwater in shallow aquifers of the Wadi Watir delta contain total dissolved solids (TDS) ranging from 940 to 13,853 mg/l (see Table 3-1 in Chapter 3). The highest

TDS groundwater generally occurs in wells near the coast (Figure 4-1 and see Table 3-1 in Chapter 3; wells 35, 42, and 43). According to Shalaby (1997), groundwater in the shallow aquifers of the delta form a lens-like shape of slightly brackish water overlaying a deeper saline groundwater. Groundwater in the Wadi Watir delta near the coast is characterized by high salinity due to seawater intrusion (Said, 2004). Abd El Hafez

(2001) reported that groundwater in the Wadi Watir delta is from a mixture of meteoric water and seawater. Ismail (1998) reported salt water intrusion in the northern part of the delta due to upwelling of saline water. The deep saline layer in this aquifer has been investigated by (Mabrouk and Nasr, 1997). They studied the hydrogeological setting in the Wadi Watir delta and reported three different clay layers which reflect three different sequential phases of transgression and regression of the gulf water due to rise of sea elevation through the Pleistocene. The thin lens of relatively fresh water is less than two meters above sea level in the study area, so it is very sensitive to the groundwater pumping rate. El-Refeai (1992) and El Sayed et al. (2006) subdivided groundwater in the delta into two zones according to salinity; a shallow brackish water and a deeper extremely saline water. In summary, the previous studies indicated that groundwater salinity in the study area is due to seawater intrusion along the coast, upwelling of deep

128

saline groundwater, water-rock interaction with aquifer matrix, and evaporation. In addition, the recent drought lasting more than 10 years has led to increased groundwater salinization.

According to Sulin (1946), groundwater can be classified into two main groups.

Groundwater which has a rNa/rCl ratio >1 or <1 (rNa, is the ratio of Na to total cations in meq/l and rCl is the ratio of Cl to total anions in meq/l). The former case, according to

Sulin (1946) and Schoeller (1962) is subdivided into two fundamental water types. A ratio of [rNa - rCl]/[rSO4]>1 indicates a [Na:SO4] water type, and a ratio of [r(Na+K) - rCl]/[rSO4]<1 indicates a [Na:HCO3] water type. In the latter case, a ratio of [rCl - r(Na+K))/[rMg]<1 indicates a [Mg:Cl] water type, and a ratio of [rCl – r(Na+K)/[rMg]>1 indicates a [Ca:Cl] water type (Figure 4-2).

From Figure 4-2 it is clear that four samples have [rNa/rCl] >1, which plot on the lower left side of the diagram and they represent a [Na:SO4] water type (Sites 19, 38, 36 and 51). This indicates relatively low saline groundwater and a dominance of terrestrial salt leaching or rock-water interaction. The majority of groundwater samples have a

[rNa/rCl] <1, which plot on the upper right side of Sulin’s diagram and are characterized by [Ca:Cl] or [Mg:Cl] water types. Deep drilled wells (Sites 17-22 inclusive), a dug well located close to these wells (Site 45), and two dug wells located at the coast (Sites 35 and

42) are a [Ca:Cl] water type. The deep saline groundwater layer, which contains intercalated clay, represents three sequential phases of regression and transgression of seawater (Mabbrouk and Nasr, 1997; Abbas et al., 2004). The water type [Ca:Cl] dominates the deep drilled wells with high pumping rates (screed interval ranges from 1

129

to 15 meters below sea level). This is due to the upwelling of deep saline groundwater of old marine genesis from deeper zones of this aquifer because of over pumping. The water type [Mg:Cl] dominates the shallow dug wells with moderate or low pumping rates

(screened interval is 0.5 above to 1 meter below sea level). This water type results from leaching and dissolution of carbonate rocks including marl, limestone and dolomite in the aquifer matrix.

3.2 Isotopic Characteristics of the Delta and Deep Saline Groundwaters

3.2.1 Environmental Isotopes

Environmental isotopes (δ18O and δ2H) of groundwater samples show that groundwater in the Wadi Watir delta has been affected by seawater intrusion along the coast (Eissa et al., 2012 in review). Saline water also mixes with groundwater in the main well field near the mountain block several kilometers from the coast. However, the δ18O and δ2H values of this mixed groundwater are similar to the isotopic composition of other groundwater in the alluvial aquifers of the Wadi Watir delta, so potential mixing of a deep saline groundwater with the shallower groundwater cannot be determined using

δ18O and δ2H. Thus, in this paper 87Sr/86Sr, δ37Cl, and δ81Br isotopes are integrated with the chemistry of deep saline groundwater to address the origin of water and solutes, and to better constrain the mixing end members in the Wadi Watir delta aquifers.

3.2.2 Mechanisms of Water-Rock Interaction using Strontium Isotopes

87Sr/86Sr groundwater values are controlled by water rock interactions and they reflect the isotopic signature of strontium containing minerals in aquifers the water has flowed through (Layons et al., 1995). Generally, higher 87Sr/86Sr values are recorded in

130

felsic (granites) rather than mafic (basalt) rocks. Sr2+ behaves like Ca2+ and has higher concentrations in Ca bearing minerals (Layons et al., 1995). Calcium shows a positive correlation with Sr and they behave conservatively in groundwaters of the Wadi Watir delta (Figure 4-3). Groundwater 87Sr/86Sr values can be used to determine the geochemical history of groundwater (rock-water interactions), groundwater mixing, groundwater recharge source(s) and the source(s) of salinity (McNutt et al., 1984; Layons et al., 1995 and Clark and Fritz, 1997). Water-rock interactions and groundwater mixing, with seawater along the coast or from deep saline groundwater, are the most prevalent conditions affecting groundwater chemistry in the Wadi Watir delta.

Groundwater in the watershed above the delta was sampled from three different locations: Wadi El Ain, Furtaga Spring and El Shiekh Attia (Table 4-1 and Figure 4-1-a).

Groundwater from the El Shiekh Attia (Sites At1-2) and Wadi El Ain (Sites An1-3) areas in the watershed are more enriched with 87Sr/86Sr than the groundwater of Furtaga spring (Site

Fu) and the Wadi Watir delta groundwaters (Figure 4-4). The 87Sr/86Sr values of these watershed groundwaters are within the average range of 87Sr/86Sr of the Upper Cretaceous limestone and the Lower Cretaceous sandstone rocks which underlie the main area of Wadi

Watir watershed (Eissa et al., 2012 in review).

131

Figure 4-3. Positive correlation of strontium versus calcium.

132

Table 4-1. 87Sr/86Sr, 37Cl, and 81Br in groundwater and rock sample in the Wadi Watir Basin. 87Sr/86Sr δ37Cl δ81Br Location Map Site Sample Type SMOC SMOB average 2 σ average stdv average stdv

At1 Water 0.70794 0.00001 -0.06 0.07 -- -- Water At2 0.70794 0.00003 -0.32 0.04 1.05 0.25 Fu Water 0.70723 0.00002 -0.42 0.08 0.94 0.17 Wadi Watir Mc Water 0.70800 0.00002 -0.19 0.10 1.28 0.14 watershed Water An1 0.70806 0.00002 -0.52 0.10 0.94 0.05 Water An2 0.70809 0.00003 -0.48 0.05 -- -- Water An3 0.70823 0.00003 -0.26 0.13 -- -- 17 Water 0.70698 0.00003 0.22 0.10 0.31 0.07 19 Water 0.70711 0.00002 0.04 0.08 0.22 0.03 21 Water 0.70705 0.00002 -0.01 0.06 0.41 0.10 23 Water 0.70714 0.00003 0.09 0.13 0.37 0.03 25 Water 0.70717 0.00003 -0.19 0.09 0.38 0.07 28 Water 0.70729 0.00002 -0.09 0.03 0.51 0.04 33 Water 0.70743 0.00002 -0.50 0.08 0.32 0.08 34 Water 0.70741 0.00001 -0.38 0.08 0.33 0.03 35 Water 0.70758 0.00002 0.13 0.05 0.32 0.11 Wadi Watir delta 36 Water 0.70742 0.00003 -0.08 0.11 0.49 0.05 37 Water 0.70744 0.00002 -0.06 0.04 0.55 0.05 39 Water 0.70820 0.00003 -0.26 0.04 0.37 0.06 41 Water 0.70735 0.00003 -0.37 0.04 0.41 0.08 42 Water 0.70740 0.00002 -0.05 0.11 0.34 0.04 46 Water -- -- -0.03 0.08 0.38 0.10 50 Water 0.70753 0.00002 -0.29 0.05 0.54 0.09 52 Water 0.70733 0.00002 -0.85 0.05 0.29 0.09 Gulf Water 0.70925 0.00005 -0.01 0.03 0.57 0.11

Y.GrAn1 Rock (Younger Granite) 0.71557 0.00004 -0.20 0.02 -- -- Wadi El Ain Y.GrAn2 Rock (Younger Granite) 0.72182 0.00003 -0.58 0.10 -- -- Up.Cr Rock (Upper Cretaceous) 0.70835 0.00004 1.41 0.12 -- --

L.Cr Rock ( Lower Cretaceous) 0.70788 0.00005 -0.22 0.28 -- --

Main Channel O.Gr Rock (Older Granite) 0.7029 ------

Y.GrMc1 Rock (Younger Granite) 0.71330 0.00004 -0.62 0.10 -- --

Y.CrMc2 Rock (Younger Granite) 0.70854 0.00004 0.15 0.09 -- --

Y.Gr.F1 Rock (Younger Granite) 0.71206 0.00002 -0.12 0.14 -- --

Ain Furtaga Y.GrF2 Rock (Younger Granite) 0.71780 0.00003 -0.57 0.11 -- -- Y.GrF3 Rock (Younger Granite) 0.71228 0.00004 -0.69 0.17 -- -- Location sites for water and rock samples are indicated in (Figure 4-1).

133

0.7150 Avg Y.Gr (Younger Granite) 0.7140 Wadi Watir watershed

Hand Dug wells (Wadi Watir delta) 0.7100 Gulf Water Drilled wells (Wadi Watir delta) 0.7090

0.7088

0.7086 Up.Cr (Limestone) 0.7084 39 Wadi Watir watershed 0.7082 An3 An2 An1 At2 0.7080 Mc At1 L.Cr (Sandstone) 0.7078 35 Mixing two waters 0.7076 50 33 37 36 0.7074 41 52 42 34 28 Fu 0.7072 25 23 19 0.7070 17 21

0.7068 Mixing with deep saline groundwater

0.7030 O. Gr (Older Granite) 0.7020 0 0.2 0.4 0.6 0.8 1 Figure 4-4. 87Sr/86Sr versus Sr in the Wadi Watir upper watershed and delta groundwaters. An, At and Mc are samples collected from the Wadi El Ain, El Shiekh Attia, and main channel areas of the uppper Wadi Watir Watershed. Avg Y. Gr (Average Sr in 9 samples of Younger Granite Rocks), L. Cr (Lowe Cretaceous Sandstone), L.S (Lime stone of Upper Cretaceous) and O.Gr (Older granite Rocks).

134

The 87Sr/86Sr isotopic signature of shallow groundwater (hand dug wells) in the

Wadi Watir delta is similar to the isotopic signature of Furtaga Spring (Site Fu), although the shallow groundwater has higher strontium concentration than the up gradient Furtaga

Spring. 87Sr/86Sr values of shallow groundwater in the Wadi Watir delta plot between the drilled wells that contain deep saline groundwater and the Wadi Watir watershed groundwaters (Figure 4-4). Thus, the shallow groundwater 87Sr/86Sr isotopic signature is either from dissolution of minerals similar to that the Furtaga spring water flows through and/or is a mixture of groundwater recharge from the Wadi Watir watershed and deep saline groundwater beneath the main well field. These data also support that most of the recharge to the Wadi Watir delta aquifers comes from the Furtaga spring and El Shiekh

Attia areas of the Wadi Watir watershed (El Ghazawi 1999; El Sayed 2006 and Eissa et al., 2012 in review). Most Wadi Watir groundwater samples plot along a straight line between the recharge water from the mountain blocks coming from Furtaga springs (Site

Fu) and site 50. Furtaga springs and site 50 represent the end members of groundwater evolution from leaching and dissolution of aquifer rock matrix.

Groundwater from the deep drilled wells (Sites 17, 19, 21 and 23) is depleted in

87Sr/86Sr as compared to other samples in the study area (Figure 4-4), which is most likely due to mixing with a deep saline groundwater of unknown 87Sr/86Sr signature, but which likely obtains its 87Sr/86Sr composition from interaction with old granitic rock (depleted

87Sr/86Sr = 0.7029). Old granitic rocks underlie the Wadi Watir delta aquifers (see Figure

3-2 in Chapter 3).

135

3.2.3 Mechanism of Salinization and Chloride Isotope

In general, salt deposits are enriched with δ37Cl isotopes and saline groundwater becomes more enriched with depth (Kaufman et al., 1984). The main factors affecting

δ37Cl are rock-water interactions, mixing of waters with different δ37Cl compositions, and fluid phase separation which includes fractionation due to both dissolved elements and isotopes (Shmulovich et al., 1999; Liebscher et al., 2006a and Liebscher et al., 2006b).

Chloride is a conservative element, so δ37Cl can be used to help determine groundwater evolution, recharge source(s), and mixing of different waters. The Standard Mean Ocean

Chloride (SMOC) isotopic ratio of δ37Cl in seawater is set as 0.00 ‰ (Kaufman et al.,

1984). The δ37Cl shows more enrichment in felsic-host waters than in mafic-host waters

(Frap et al., 2004).

δ37Cl shows variation in the Wadi Watir delta groundwaters (Figure 4-5a). δ37Cl ranges between -0.85 (Site 52) and +0.22 ‰ SMOC (Site 17), with a value of -0.01 ‰ for the Gulf of Aqaba water. The majority of shallow groundwaters in the Wadi Watir delta have higher chloride concentration, but similar δ37Cl values as the recharge water

37 from the watershed (Sites An1-3, Mc and At1-2) and within the δ Cl ranges in the granitic and Lower Cretaceous rocks at Wadi El Ain, Furtaga Spring and the Main Channel areas

(Y.GrAn1, Y.GrAn2, Y.GrMc1, Y.Gr.F1, Y.GrF2, Y.GrF3 and L.Cr) (Table 4-1 and

Figure 4-5b). This increase in chloride concentration reflects leaching and dissolution of chloride from the aquifer matrix rich (i.e. felsic and marine carbonate rocks). Most of the shallow groundwater is categorized as a [Mg:Cl] water type (Figures 4-2 and 4-5b). Most groundwater samples in the delta have negative values, which indicate that δ37Cl is

136 mainly from rock weathering (Eggekamp, 1994). All rock samples have a negative value of δ37Cl except the Upper Cretaceous limestone (Up.Cr) and the younger granite samples in the Main Channel (Y.CrMc2) (Figure 4-5b). a

An1 At2 MC Groundwater in Wadi Watir watershed An2 At1 Fu An3 b

Deep saline groundwater Gulf

Shallow groundwater in delta Watir

Y. Gr An2 Y. Gr F1 Y. Gr Mc2 Granitic rocks Y. Gr F3 Y. Gr F2 Y. Gr Mc1 Y. Gr An1

Limestone and sandstone L.Cr Up.Cr

-1.00 -0.50 0.00 0.50 1.00 1.50 2.00 δ37Cl

Figure 4-5. a) δ 37Cl ‰ (SMOC) versus Cl (ppm) for groundwater in the Wadi Watir delta. Chloride concentration of seawater is divided by 10 to fit on the plot. b) δ 37Cl ‰ (SMOC) in different rock types in the Wadi Watir watershed. The location sites, rock types and the map sites are indicated in (Figure 4-1 and Table 4-1).

137

The deep drilled wells in the well field (Sites 17, 19, 21 and 23), one shallow well near the well field (Site 46), and two shallow dug wells located at the coast (Sites 35 and

42) are characterized by enrichment of δ37Cl relative to all groundwater located up gradient in the watershed or other groundwater in the delta. Some groundwater samples are even enriched relative to the gulf water sample. This enrichment is mainly due to: (1) over pumping of the deeper drilled wells in the well field (Sites 17, 19, 21 and 23) or as a result of the cone of depression near the well field (site 46); (2) the withdrawal of deep saline groundwater of marine origin; (3) possibly the electrostatic effect of subsurface clay sheets in the Wadi Watir delta (Mabbrouk and Nasr 1997 and Abbas et al., 2004); and/or (4) mixing with seawater along the coast (Sites 35 and 42). All these enriched waters have a [Ca:Cl] water type which characterizes the deep saline water. Kaufman et al. (1984), suggests that enrichment of δ37Cl may be due to fractionation through a brine aquifer, where the salt deposits are enriched in δ37Cl with respect to seawater. The majority of shallow groundwater in the Wadi Watir delta plots between two end member waters; the recharge water which comes from the Wadi Watir watershed (modern water) and water in the well field which contains some deep saline groundwater (old water)

(Figure 4-5a). The different location of samples on the 87Sr/86Sr versus Sr plot and the

δ37Cl versus Cl plot are likely due to the dissolution of sea salt or salt spray (Shouakar-

Stash et al., 2005b) from the Gulf of Aqaba.

3.2.4 Bromine Isotope Systematics

138

Natural distribution of δ81Br has only been reported for sedimentary formation waters, with the fractionation processes poorly documented (Stotler et al., 2010).

Eggenkamp and Coleman (2000) were the first scientists since 1920 to report the δ81Br signature of natural samples with good precision. Shouakar-Stash et al. (2005a) subsequently presented a method for determining δ81Br with excellent precision and accuracy, using continuous-flow mass spectrometry, with seawater as the standard. In the study area, groundwaters show a large variability with δ81Br ranging from +0.22 (Site 19) to +1.28 ‰ (Site Mc) and including a sample from the Aqaba Gulf (Gulf) (Figure 4-6a).

All samples show enrichment of bromide isotopes as compared to the standard; however samples from the deep drilled wells are depleted compared to most other samples in the study area. This depletion in δ81Br results from mixing with upwelling saline water which has an old marine genesis (Figure 4-6a). Samples from wells along the coast (Sites 35 and

42) plot between the shallow groundwaters and the Aqaba Gulf indicating that these waters have been influenced by seawater intrusion (Figure 4-6a).

Said (2004) postulated that the [Ca:Cl] water type in the Wadi Watir delta is mainly a result of subjecting a deep marine water to a reducing condition under continuous marine sedimentation of rich organic matter for long periods of geologic time.

Under these conditions, bicarbonate is completely removed by precipitation and sulfate is reduced, leaving high concentrations of chloride remaining in solution. Figure 4-6a shows depletion of δ81Br in the deep drilled wells (Sites 17, 19, 21 and 23). This depletion, which is characteristic of the deep saline water, is mainly due to fractionation upon rock

139

a

1.50 Wadi Watir Watershed

‰ Hand Dug Wells (Wadi Watir delta)

Br Br Drilled Wells (Wadi Watir delta) Mc

81 1.20 b Gulf Water δ At2 An1 Fu 0.90

0.60

0.30

0.00 -1.10 -0.90 -0.70 -0.50 -0.30 -0.10 0.10 0.30 δ37Cl ‰

Figure 4-6. a- δ 81Br ‰ SMOB versus Br and b- δ 37Cl ‰ SMOC versus δ 81Br ‰ SMOB of groundwater in the Wadi Watir delta.

formation accompanied by reducing conditions and an altered pH (Stotler et al., 2010).

Figure 4-6b shows that the enrichment of δ37Cl of the deep drilled well waters is associated with a depletion of δ81Br. The majority of shallow groundwater sites (hand

140 dug wells) plot between the enriched δ81Br recharge water (Wadi Watir watershed) and depleted δ81Br deep drilled wells indicating a mixed Br source or increased Br concentration from atmospheric deposition of Aqaba Gulf salt (Figure 4-6a).

3.3 Geochemical Modeling of Groundwater in the Wadi Watir Delta Area

Groundwater chemistry, in addition to isotopic data, was used to evaluate mixing of deep saline groundwater with well field groundwater and mixing of seawater with

Wadi Watir delta groundwater along the coast. This approach is limited by the available mineral, gas, chemical and isotopic data (Hershey et al., 2007). The mixing of deep saline groundwater with groundwater in the well field was first delineated based on isotopes and water chemistry (see previous sections of this report). Geochemical modeling was used to determine the percentages of mixing waters and to identify geochemical reactions that produce brackish water in the well field from upwelling of saline water and seawater intrusion along the coast. For the well field, the two end member waters that mix and undergo geochemical reactions to produce the observed water chemistry in the well field are recharge to the Wadi Watir delta, represented by site 20 water chemistry, and the deep saline groundwater represented by the most saline groundwater samples in the well field (Appendix 1, Table 4 Site 31 in Said, 2004) Well 20 was chosen to represent recharge to the delta because it is the most dilute groundwater (TDS of 1925 mg/l) in the well field. Said (2004) site 31 was chosen to represent the saline water end member because it is the most saline sample (TDS of 16,500 mg/l) collected from the well field wells. For seawater intrusion along the coast, one end member was the Gulf of Aqaba

(Table 4-2) and the other was site 20.

141

Table 4-2. NETPATH modeling results for water-rock interactions and mass transfer (mmol/L) for the Wadi Watir delta. Initial Initial Mixing Phases precipitated or dissolved (Site 1) (Site 2) Percent Final Ev using Ev using Numbers Water Site Na- Mont Site 2 Sea Cal Biot Gyp llt Gr Abite Ex Hal Dol NETPATH Rayleigh in Fig. 1 1 Mont -Maf Water Ev Eq 20 Deep 17 49 6 -- 4.9- -- 6.6 -- -- ...6- -- -- ..0- -- 6.9 -- -- 20 Deep 18 88 64 -- 9.4- -- 6.4 ------4.4- 4.0 4.2 -- -- 20* -- 35 87.5 -- 12.5 -- 3.64 4.49 -- -51.8 -- 2.69 ------1.6 1.1 20* -- 42 92 -- 8 -15.76 ------2.02 -- -12.11 9.42 8.07 1.1 1.1 Cal-Calcite; Biot-Biotite; Gyp-Gypsum; Ilt-Illite; Mont- Montmorillonite; Gr-Composite Granite;Ex-Cation Exchange; Hal-Halite; Dol-Dolomite; Ev-Evporation Factor; -- No Data *(After Eissa et al., in review)

142

Geochemical modeling shows that the observed water chemistry for wells in the well field, as represented by sites 17 and 18, is produced by: (1) mixing of 6 to 12% saline water with recharge water; (2) precipitation of calcite; (3) formation of mafic montmorillonite or incongruent mineral dissolution; (4) dissolution of gypsum, halite, and dolomite; and (5) exchange of sodium ions in the water with calcium ions of the aquifer matrix (Table 4-3). The mass transfer of major ions, as shown by the dissolution or precipitation of a mineral, is supported by saturation index (SI) values (Table 4-3).

Minerals that are under saturated in the groundwater (negative SI values) dissolve and minerals that are over saturated in the groundwater (positive SI values) precipitate.

Calcite SI values of the recharge water (site 20) and site 18 are negative, but within the range of +/- 0.10 for calcite saturation so calcite precipitation is acceptable for this model.

The calcite SI value site 17 is less than -0.10 so calcite precipitation in this model may be incorrect or occurring early along the flow path from the recharge area.

Table 4-3. Mineral saturation indices (SI) for phases in geochemical models calculated using NETPATH. No. Calcite Dolomite Gypsum Albite Anorthite K-spar Halite Ca- Illite Mont 17 -0.39 -0.96 -0.54 -0.83 -3.54 -0.03 -2.71 2.68 1.81

18 -0.08 -0.24 -0.55 -0.73 -3.40 0.08 -2.70 2.21 1.61

20 -0.03 -0.48 -0.68 -0.95 -3.35 0.14 -3.40 2.24 1.57

35 0.49 1.08 -0.03 0.43 -2.66 1.31 -1.69 2.88 2.72

42 0.13 0.41 -0.59 -0.59 -3.33 0.29 -2.18 2.06 1.73

Positive values indicate that a mineral is above saturation in the water and will precipitate from the water and negative values indicate that a mineral is under saturated in the water and will dissolve if present.

143

Geochemical modeling supports seawater intrusion along the coast as indicated by the δ37Cl and δ81Br isotopic data (Figures 4-5 and 4-6). Geochemical models show that the observed water chemistry for wells with seawater intrusion along the coast, as represented by sites 35 and 42, is produced by: (1) mixing of 8 to 12.5% seawater with

Wadi Watir groundwater; (2) precipitation of calcite; (3) formation of mafic montmorillonite or incongruent mineral dissolution; (4) dissolution of biotite, dolomite, gypsum, halite, and a composite granite; (5) exchange of sodium ions in the water with calcium ions of the aquifer matrix; and evaporative concentration of the groundwaters

(Table 4-3). The dissolution or precipitation of these phases is supported by SI calculations (Table 4-3). Although dolomite SI values indicate that dolomite should precipitate (positive SI values) rather than dissolve, dolomite dissolution occurs because of de-dolomitization (Back et al., 1983). Evaporative concentration of the groundwater calculated by the NETPATH models is very close to the evaporation factors calculated from the isotopic data using the Rayleigh equation (Table 4-2).

3.4 Solute-transport Modeling

Solute transport modeling was performed to (1) simulate the potential sea water intrusion at the coast, (2) simulate the vertical movement and migration of deep saline groundwater upward due to the region over pumping in the Wadi Watir delta, and (3) calculate the recharge from the basement mountain blocks and pumping rates for the delta using observed chemistry and observed head. SEAWAT (Guo and Langevin, 2002 and Langevin and Guo, 2006), a program for simulating water flow with variable density was used simulate the migration of solute from deeper saline layers and seawater

144 groundwater intrusion at the coast. Solute transport in porous media includes advection, molecular diffusion, and mechanical dispersion as described by Zheng and Bennett

(1995) and by the following partial differential equation:

( ) ( ) ( ) ∑ (1)

where:

D is the hydrodynamic dispersion coefficient [L2T-1]

Cs solute concentration of recharge water [mg/L]

ν is the water velocity (LT-1)

Rk is the rate of solute production or decay

In this project the R factor is not considered as this model is considered to be a solute transport non-reactive model. Baxter and Wallance (1916) developed the empirical relation between the densities and concentration of fresh, deep saline and sea waters concentrations as described in the following equation:

(2) where, E is a dimensionless constant factor, which is a function of C or the salt

-3 concentration [ML ] and and f are the densities of seawater and fresh water respectively [ML-3]. The slope E ( ) has a value of 0.7143 for the range of fresh or recharge water and sea water.

145

The SEAWAT program depends on the concept of fresh water equivalent head in a saline groundwater or seawater environment (Geo and Langevin, 2002) according to the following equation:

(3)

where, h is the saline or seawater head, hf is the equivalent fresh water head [L], ρf is the density of fresh water, ρ is the density of seawater [ML-3], and Z is elevation [L].

If we consider the sea level as the standard, Z will equal zero making the second

term of the equation equal to zero and simplified to .

The parameters and data used for simulations are summarized in Table 4-4. The model boundaries were chosen to be (1) the Gulf of Aqaba to the east, (2) the mountain block unconsolidated contact to the west, and (3) where the contact to the west meets the

Gulf of Aqaba to the north and south. A finite-difference grid oriented along the north- south and the east-west axes of the Wadi Watir Delta was used for the model. There are

60 columns and 120 rows with a uniform grid spacing of 98 m (323 feet) in both directions (Figure 4-7a and b). In the vertical direction, the model grid consists of 27 layers representing the five hydrostratigraphic units (see Figures 3-2 in Chapter 3, and

Figure 4-7a and 4-7b) of the delta aquifers (Abbas et al., 2004). The modeled layers from

1 to 12 correspond to the layer (A1) with a hydraulic conductivity of 4 m/day representing the first three hydrostratigraphic units of the alluvial deposits (Eissa et al., in review). Layers 13 through 18 correspond to Layer (A2) with a hydraulic conductivity of

11 m/day (El-Refeai, 1992). The layer (A2) is the fourth hydrostratigraphic unit, which

146 comprises Pliocene to Pleistocene age sand. Layers 19 to 27 correspond to the (A3) which formed mainly of sandy clay (Mabrouk and Nasr, 1997) with a hydraulic conductivity of 0.001 m/day (Fetter, 2001). Layer (A3) represents the fifth hydrostratigraphic unit, which is mainly composed of clay sand and clay intercalations representing three sequential phases of regression and transgression of Gulf water and is filled with high saline water (Mabrouk and Nasr, 1997). The land surface elevations used in the model were from the Digital Elevation Model STRM-90 m (USGS, 2004) coupled with 22 ground elevation surveyed spots representing water points. The model layer thickness was taken from the geo-electrical cross section of Abbas et al. (2004).

Table 4-4. Parameters for the groundwater model in the Wadi Watir delta. Parameters

4 (A1); 11 Layer (A2) and Hydraulic Conductivity (m/day) -3 10 Layer (A3); Kx=Ky=(0.1)Kz

Total Porosity (%) 10

Longitudinal Dispersivity (m) 10

αL/ αT and αT/ αV 0.1

Molecular Diffusion Coefficient (m2/day) 10-9

Storage Coefficient (Himida 1997) 3.14x10-4

Density/Concentration Slope 0.7143

147

3216000 Head Observations 1 Inactive Cells

3215000 Constant Head Coupled

1616 with Constant Concentration Artificial Recharge Wells 5 41a 45 Boundary 3214000 46 42a coupled with a specifief 44 40a 43 39a Pumping Wells concentration of 1000 mg/l

3213000 2121 10 C D

33 3212000 15 Salinity 38a 4 5542 4 37a 20 Inactive

e (mg/l) 3131 33a 23a 1515 s 54 37

a Grid Cells 3211000 30a 10 5 10 900 58 21a 2 3232 5

M8 53 6 6 B 20a 3333 7 9 9 24a 362 29a 35a 51 50 36 38 1111 41 l

C D e 19a 39 34 34 0 20 M1 1717 1500

25a v 52 32a D 403535 Layers (A1) 19 3210000 19 18 e 20 18 1 1 20 M2 L 22 22 3000

18a a -20 e

13 13

3209000 S

23 45a 53a 23 ) 6000 56 17a 25 r Layer (A2) 50a 49

24 2457 e 46a t

47 e 3208000 2525 27 48a 29 -40 27 26 29 48 M 28 26

8 ( 18000 28 12 14148 12

30 h

30 30 t

3207000 p

e -60

D Layer (A3) 3206000 35

-90 3205000 0 1 2 Km. 669286 660000 660800 661600 662400 663200 664000 665125 660000 661000 662000 663000 664000 665000 Distance (Meter) Figure 4-7. The finite difference grid cells used for the model a) in the x and y direction and b) in the z direction.

148

3.4.1 Model Parameters

The model parameters are summarized in Table 4-4. The values of the vertical hydraulic conductivity are assumed to be one tenth of the values of the horizontal ones

(Eissa et al., in review). The total porosity is assumed to be homogeneous and is set at

30% (Khalil, 2010) in the model as the initial value. The longitudinal dispersivity is also assumed to be homogeneous in the flow system and is initially set at 100 m and is based on the overall model scale of 4000 m Gelhar et al. (1992). The ratio of the horizontal transverse dispersivity to the longitudinal dispersivity is assumed to be 0.1. The ratio of the vertical transverse dispersivity to the longitudinal dispersivity is assumed to be 0.01.

The diffusion coefficient is assumed to be 10-9 m2/day. The storage coefficient of this aquifer, as estimated by Himida (1997) is 3.14x10-4.

3.4.2 Boundary Conditions and internal sinks

The boundary conditions for both flow and transport are shown in Figure 4-7-a and 4-7b. The model base is assumed to be a no flow boundary and is represented by the consolidated granitic basement aquitard bed rock at the base. For the flow model, a constant head boundary is set to sea level at the coast, representing the mean sea water level at the shore line of the Gulf of Aqaba (Figure 4-7a). Whereas, a constant concentration boundary are applied for the 27 model layers at the coast with these concentration assumed to be 41,000 mg/L based on the analysis of the Gulf water sample

(Eissa et al., in review). The Wadi Watir delta aquifers are primarily recharged from the basement aquifer along the western boundary of the groundwater flow model area, so the boundary between the basement and the unconsolidated delta aquifers is simulated by a

149 point source boundary with value of 1000 mg/l. This value is based on the average of chemical analyses of three samples collected from nearby Furtaga Springs (Site Fu in

Figure 4-1-a and Table 4-1), located 7 kilometers west in the upstream (Eissa et al., 2012 in review). The local recharge from the rain fall is small (35 mm/year) so it is ignored in the model.

3.4.3 Initial Conditions

Initial water levels for the steady were specified for the groundwater flow model by kriging interpolating between the known values of the water level in February 1986

(El Kiki et al., 1992) and in November 1994 (Ismail, 198). Steady state conditions of groundwater flow and solute transport models have been identified from transient simulations under arbitrary initial conditions until reaching system stabilization and equilibrium for salt and physical flow within the models. The salinity obtained from the steady state has been modified with the salinity data of deep saline groundwater from El-

Refeai (1992) (Figure 4-7a and 4-7b and Figure 4-8). This deep saline groundwater has also been reported by (Shalaby, 1997; Mabrouk and Nasr, 1997; and Abbas et al., 2004 and El Sayed et al., 2006). This deep layers mainly composed of clayey sand and clay intercalations represents three sequential phases of regression and transgression of Gulf water and is filled with high saline water (Mabrouk and Nasr, 1997). El Refaee measured the salinity with depth in a drilled well located adjacent to the drilled well location sites.

The salinity of these drilled wells reaches 18000 mg/L at depth of 34 m below sea level

(Figure 4-8). According to the data availability, simulations were carried out for periods from 1982 to 2009. The steady state outputs for water level and salinity have been used

150 as the initial condition for the transient simulation of stressing the aquifer by pumping during the simulation periods. To simulate the pumping stresses, the pumping rates for the drilled wells were initially obtained from the well site engineer (see Table 3-8 in

Chapter 3). The pumping rates of these wells varied through the period 1982 to 2007 depending on the groundwater availability and flash flood intensity and/or frequency.

Within the model, each well has its individual withdrawal rate, specified for each stress period. The total withdrawal rate from the group of drilled wells located at the outlet of

Wadi Watir (well field area in Figure 4-1b). The pumping rate from the hand dug well is unknown and assumed to be 5 m3/day as the initial value based on field data collected from the well owners during field trips. The pumping rates for pumping through the drilled wells and hand dug wells used as initial values then it has been changed to calibrate the model using head and salinity observations.

20000 18000 16000 14000 12000 10000 8000

Salinity Salinity (mg/L) 6000 4000 2000 0 5 0 -5 -10 -15 -20 -25 -30 -35 Depth below Sea Level (meter)

Figure 4-8. Salinity (mg/L) with depth (meter) in groundwater of the Quaternary aquifer in the Wadi Watir delta (modified from El-Refeai, 1992).

151

3.4.4 Model Calibration

The model was calibrated using both salinity and head observations within different time periods. The recharge is the main factor controlling the head while the recharge rates, pumping, the value of dispersivity and the effective porosity are the main factors controlling the salinity. A total of 58 head observations were used through 1982 to

2007 (Figure 4-7a). Model calibration was achieved through a trial and error approach by adjusting the values of recharge through the injection recharge wells at the boundary between mountain blocks and alluvial aquifer, until reaching general agreement between simulated and measured water levels (Figure 4-9 and Table 4-5). The model calibration was obtained when a reasonable value of the relative error between the modeled and the observed water level was less than 12.6%. The salinity observations (Figure 4-10 and

Table 4-6) were selected according to the continuity of data and the availability of δ18O and δ2H data (Eissa et al., in review and Said, 2004) to ensure representation of the actual water salinity that is not affected by evaporation (Figure 4-11). According to the pumping stresses, the calibrated model simulates the transient changes in groundwater levels and salinity migration. Using both head and salinity observations gives more confidence for the estimated recharge and a better understanding of solute migration from the deep layers toward the fresh water lens of the shallow aquifer. The main water supply for the

Nuweiba town is the eight drilled wells located at the well filed area (Figure 4-7a), the town wells (Site 24 and 25), and the hand dug wells owned by the Bedouins across the delta and the hotels at the coast. To calibrate the model using the head and the salinity observations, the pumping rates through the drilled wells have been changed within (7 to

152

20 %) for each stress period through the model time (1982 to 2009) from the original data that we obtained it from the well site engineering (see Table 3-8 in Chapter 3). The estimated pumping rates are consistent with the historical pumping rates and the well type (Table 4-7 and Table 4-8). The pumping rates for the drilled wells ranged from 200 m3/day (Site 20) to 1400 m3/day (Site M2). However, for the hand dug wells, the pumping rates varies considerably from the initial value (5 m3/day) and ranged between

0.5 m3/day to 7 m3/day except for well 35 (owned by Amon Yaro resort hotel), where the pumping has increased with time from 2 m3/day in 1995 to 30 m3/day in 2009. The increased pumping of well 35 is linked to development activities in the area where hotels and resorts have been constructed over the last 15 years.

3.0

2.5

2.0

1.5

1.0

CalculatedHead (Meter) 0.5

0.0 0.00 0.50 1.00 1.50 2.00 2.50 3.00 Observed Head (Meter)

Figure 4-9. Calcutated versus oberved water levels (head).

153

Table 4-5. Modeled and simulated groundwater levels in different years through the modeled time. Observation Observe Calculated Calc.-Obs. (m) Time No.1 Head1.4 (m) Head1.6 (m) 0.19 (monthFeb-86Year) 2 1.9 1.8 -0.08 Feb-86 3 1.8 0.3 -0.89 Nov-94 4 1.3 1 -0.32 Nov-94 5 0.8 0.7 -0.06 Nov-94 6 0.2 0.5 0.26 Nov-94 7 1.2 0.8 -0.36 Nov-94 8 0.6 1.3 0.65 Nov-94 9 2.2 1.8 -0.42 Nov-94 10 1.0 0.6 -0.34 Apr-95 11 1.2 0.8 -0.31 Apr-95 12 0.3 0.8 0.47 Apr-95 13 0.8 0.9 0.02 Apr-95 14 0.6 1.3 0.74 Apr-95 15 1.3 1.1 -0.19 Apr-95 16 1.0 1.1 0.13 Oct-95 17 1.0 0.8 -0.18 Oct-95 18 0.6 0.5 -0.08 Oct-95 19 1.0 0.8 -0.20 Oct-95 20 1.1 0.9 -0.23 Oct-95 21 2.5 2.2 -0.32 Oct-95 22 1.0 0.7 -0.25 Oct-95 23 0.5 0.6 0.11 Oct-95 24 0.3 0.9 0.58 Oct-95 25 0.7 1.5 0.76 Oct-95 26 1.0 0.7 -0.15 Oct-95 27 0.8 0.9 0.07 Oct-95 28 1.0 0.9 -0.13 Oct-95 29 0.8 0.9 0.12 Oct-95 30 0.9 0.9 -0.01 Oct-95 31 1.8 1.1 -0.66 Oct-95 32 1.1 0.8 -0.32 Oct-95 33 1.5 1.1 -0.40 Oct-95 34 1.1 0.7 -0.37 Oct-95 35 1.0 0.7 -0.28 Oct-95 36 0.8 1.2 0.42 Oct-95 19a 0.6 1 -0.08 Mar-07

154

Table 4-5. Modeled and simulated groundwater levels in different years through the modeled time (continued). Observation Observe Calculated Calc.-Obs. (m) Time No.20a Head1.5 (m) Head0.8 (m) -0.01 (monthMar-Year)07 21a 0.7 0.9 0.17 Mar-07 23a 1.5 1.4 -0.51 Mar-07 24a 0.9 0.9 -0.52 Mar-07 25a 0.7 0.9 -0.35 Mar-07 29a 0.9 0.8 0.26 Mar-07 30a 1.2 0.8 0.20 Mar-07 32a 1.2 0.7 0.27 Mar-07 33a 1.1 0.7 0.20 Mar-07 35a 0.9 0.4 -0.21 Mar-07 37a 0.9 0.6 0.05 Mar-07 38a 0.7 0.6 -0.28 Mar-07 39a 1.1 1.4 0.16 Mar-07 40a 1.1 1.4 -0.58 Mar-07 41a 1.1 1.4 0.35 Mar-07 42a 1.2 1.4 -0.09 Mar-07 45a 1.4 1.6 -0.40 Mar-07 46a 1.5 1.3 -0.10 Mar-07 48a 1.1 1.2 0.17 Mar-07 50a 0.9 0.8 -0.06 Mar-07 53a 1.1 0.8 -0.44 Mar-07 Observations in Feb-1986 by El-Refeai, 1992; Nov-1994; and Apr-1995 by Ismail, 1998; and Oct-1995 by El Ghazawi, 1999. Observation sites are indicated in Figure 4-9A. Observations donated by letter "a" are by the author and represented in Table 4-1.

155

3216000 1

3215000

1616

5 41a 45 3214000 46 42a 44 40a 43 39a

3213000 10 2121

33

3212000 15 38a

5542 4 4 37a 3131 33a 23a 1515 54 37 3211000 30a 10 10 5 58 21a 2 3232 5 M8 53 6 6 20a 3333 7 9 9 24a 362 29a 35a 51 50 36 38 11 C 11 41 D 19a 39 20 M1 34 34 25a1717 52 32a 403535 3210000 1919 18 20 18 1 1 20 M2

2222 18a 3209000 13 13 45a 53a 2323 25 56 17a 50a 49 24 2457 46a 47 3208000 2525 27 48a 29 27 26 29 48 8 28 26 28 12 14148 12

30 30 30 3207000 Sbakha Deposites (Evaporites)

3206000 35

3205000 0 1 2 Km. 660000 661000 662000 663000 664000 665000 Figure 4-10. Well location map of the salinity observations. All observation plotted is tabulated in (Table 4-6). Filled circular pink color is observations for water salinity collected in 2007 after (Eissa et al., in review) with equivalent observation collected in 1996 and 1998 after (Said, 2004); diamoned red color sympoles is salinity obervations after (Siad, 2004).

156

Table 4-6. Total dissolved solids for groundwater in the Wadi Watir delta in 1996, 1998, 2007 and 2009. No Y X TDS-1996 TDS-1998 TDS-2007 TDS-2009 No. (Said, 2004) 17 3209371 660051 -- 3540 4062 --

18 3209780 660083 2047 2374 3609 -- MS30 19 3210339 660133 2799 1895 2600 2190 MS29 20 3210555 660206 2453 -- 1925 2104 MS33 21 3210731 660432 2566 -- 2538 1928 MS35 22 3210712 660623 -- -- 2941 -- -- 23 3211259 661283 -- -- 2821 -- -- 24 3210757 662281 -- -- 2317 -- -- 25 3210563 662284 -- -- 2358 -- -- 26 3210052 662615 -- -- 2069 -- -- 27 3210204 662728 -- -- 2691 -- -- 28 3210440 662667 -- -- 3005 -- -- 29 3210868 662547 1901 3420 4725 -- MS14 30 3211019 662602 2059 2360 1830 - MS13 31 3209948 662865 -- -- 3239 -- -- 32 3210358 662954 -- -- 3280 -- -- 33 3211345 662826 1908 1712 3246 -- MS12 34 3209653 663288 -- -- 6447 -- -- 35 3210709 663445 6038 7759 11447 13853 MS20 36 3211065 662830 2111 -- 2606 -- MS18 37 3211733 662877 2460 2093 2902 2884 MS10 38 3211735 663011 -- -- 2716 -- -- 39 3213781 661743 3569 3534 4400 -- MS9 40 3213953 661741 2468 2240 1772 -- MS8 41 3214405 661772 2827 2948 2886 -- MS1 42 3214254 661698 2678 -- 5881 -- MS2 43 3215209 662314 -- -- 10466 -- -- 44 3208987 660323 -- -- 1750 -- -- 45 3208746 660002 -- -- 1785 -- -- 46 3208322 660428 -- -- 1850 -- -- 47 3207958 660566 -- -- 2991 -- -- 48 3207918 660758 -- -- 2740 -- -- 49 3207724 660703 2924 2889 7503 -- MS37 50 3208578 661816 -- -- 4088 -- -- 51 3207393 661680 -- -- 942 -- -- 52 3205833 660806 3094 -- 2947 -- MS25 53 3208775 662023 -- 10716 -- --

54 3205510 660772 -- -- 3226 -- -- 55 3205185 660605 5623 -- 2960 -- MS24

MS3 3214129 661755 2526 5853 ------MS5 3214056 661821 3493 3403 ------MS6 3213981 661790 2557 3655 ------MS7 3213888 661694 6147 2463 ------MS11 3210970 663099 2691 2710 ------MS15 3210759 662534 1549 1945 ------MS16 3210626 662243 1861 ------MS17 3210539 662683 2038 ------

MS19 3210236 663434 9612 10515 ------MS22 3205004 660503 4367 ------MS23 3205115 660534 4559 4829 ------MS27 3206121 661137 10525 ------MS34 3210342 660445 1476 1832 ------MS31 3209105 660507 -- 16539 MS36 3208637 660062 1736 1864 ------MS denotes the equivalent numbers after the author (Said, 2004); -- No data.

157

16000 Normalized TDS with Evaporation factor

14000 No Isotopes data

12000 Linear (1:1 Trend Line)

10000

8000

6000

Calculated Salinity (mg/L) 4000

2000

0 0 2000 4000 6000 8000 10000 12000 14000 16000 Observed Salinity (mg/L)

Figure 4-11. Calcutated versus oberved salinity. The model is valid for the majority of model domains except for the area close to the coast and the area located in Sabkha deposits. These samples are denoted by an x symbol (black for no isotope data and red color for data normalized with an evaporation factor calculated using isotopic data). The isotopes data obtained from (Eissa et al., in review and Said, 2004). 158

Table 4-7. The amount of groundwater pumping from water supply wells (drilled wells) in the delta. Feb 1982 Feb 1986 June 1987 Dec 1994 Dec 1998 Dec 2001 Dec 2008 To To To To To To To Well Feb 1986 June 1987 Dec 1994 Dec1998 Dec 2001 Dec 2008 Apr 2009 No. Q Q Q Q Q m3/day Q m3/day Q m3/day m3/day m3/day m3/day m3/day M1 -700 -700 -700 -500 Stopped M2 -- -1300 -1400 -500 18a -1070 -1100 -400 -400 -400 -400 17a -708 -1000 -1000 -250 -250 Stopped M8 -400 -500 Stopped Not Drilled yet 21a -400 -400 -400 Not Drilled Yet 20a Not Drilled Yet Not Drilled Yet -200 -200 -200 19a -700 -700 -700 Withdra -700 -3778 -4600 -2900 -1950 -1950 -1700 wal The well Sites are indicated in Figure 4-9A. Wells donated by letter "a" after the present authors and indicated in (Table 4-1).

Table 4-8. Data for pumping rates obtained from the model after calibrating with salinity and water level changes over time. Well Time Pumping Rate Time Pumping Rate Well No. No. (Day) (m3/day) (Day) (m3/day) MS4, 5, 6, 11, 15 1982-2009 -0.5 34 1982-2009 -7 1982-1995 -2 1995-1998 -5 MS22, 23 1982-2009 -2 35 1998-2008 -20 2008-2009 -30 MS36 1982-2009 -3 36, 37, 38 1982-2009 0 MS3, 19, 26, 27 1982-2009 -5 39, 40, 41, 43, 44 1982-2009 -1 1982-1998 -1 30 1982-2009 -0.5 42 1998-2009 -2 1982-1993 -2 24, 25, 26 1982-2009 -1 45 1993-2009 -4 27, 28, 31, 46 1982-2009 -1.5 47 1982-2009 -3 48, 51, 52 1982-2009 -2 49, 50, 53 1982-2009 -5 1982-1998 0 1982-1995 0 32 29 1998-2009 -1.5 1995-2009 -3 1982-1987 0 33 1987-2009 -1 Time (day) starts from February 1982 (Start of model simulation time)

The dispersivity values of 10 m show a good match between calculated and observed salinity in Figure 4-13. The longitudinal dispersivity shows a direct relation with the scale observation function (Gelhar et al., 1992). The 10 meter longitudinal dispersivity according to Gelhar (1992) is located within the category of high confidence for the small scale filed site (1-1000 meter) and a low confidence for the scale of Wadi

159

Watir delta (for large scale 103–106 meter) using the contaminant events tracer test. The longitudinal dispersivity estimated by the theory of Gelhar et al., 1992 is over estimated according to (Chin and Wang 1992). The 10 meter dispersivity is 0.1 of the model cell size, where the fickian longitudinal dispersivity is 10% of the modeled cell size (100 m).

The pumping rates were also estimated from this model by changing the pumping rates until the break through curve of the calculated salinity matched the observed one in hand dug and drilled wells (Figure 4-12a-c). The total estimated average pumping rate from drilled wells is 3,100 m3/day, whereas it estimated by 60 m3/day from the shallow hand dug wells through the model time (1982 to 2009). The break through curves, have a different shapes and slopes due to different pumping stresses and rates through the modeled time. From these curves it is clear that the groundwater salinity is mainly controlled by the pumping rates due to upwelling of deep saline water which mixed with the thin fresh water lens (Shalaby, 1997 and El-Refeai, 1992) that exist at the upper most layer of the aquifer and assigned as initial conditions (Figure 4-7b).

The average recharge from the mountain blocks to the delta is ranged between

3900 m3/day (1982 to 1987) to 7770 m3/day (1987 to 2002) with an annual daily average of 6000 m3/day through the model time from 1982 to 2009 (Table 4-9 and Figure 4-13).

160

16000 35 a Calc. Salinity (Site 35) 14000 30

12000 25 Obs. Salinity (Site 35)

10000 /day) 20 3 Calc. Salinity (Site 34) 8000

15 Salinity (mg/L)Salinity 6000 Obs. Salinity (Site 34) 10

4000 (m rates Pumping Pumping rates (m3/day) 2000 5 Site 35

0 0 Pumping rates (m3/day) 0 2000 4000 6000 8000 10000 Site 34 Modeled Time (day)

12000 1600 b Calc. Salinity (Site 18) 1400 10000

1200 Obs. Salinity (Site 18)

) 8000

1000 /day) 3 Calc. Salinity (Site 19)

6000 800 Salinity (mg/L Salinity 600 4000 Obs. Salinity (Site 19)

400 Pumpingrates(m 2000 200 Pumping rates (m3/day) Site 18 0 0 0 2000 4000 6000 8000 10000 Modeled Time (day)

4500 6 c Calc Salinity (Site 44) 4000 Obs. Salinity (Site 44) 5 3500 Calc. Salinity (Site 48) 3000 4

) Obs. Salinity (Site 48)

/day) 3 2500 3 Calc. Saliniyt (Site 50) 2000

Obs. Salinity (Site 50) Salinity (mg/L Salinity 1500 2

Pumping rates (m3/day) Site Pumping rates (m rates Pumping 1000 44 1 Pumping rates (m3/day) Site 500 48 Pumping rates (m3/day) Site 0 0 50 0 2000 4000 6000 8000 10000 Modeled Time (day)

Figure 4-12. Breakthrough curves for selected wells in the Wadi Watir delta.

161

Table 4-9. Simulated groundwater recharge for injection wells in the groundwater flow model. Recharge for Recharge from Recharge from Recharge from 3 3 3 Well1 No. the steady80 state 1982-1987120 (m /day) 1987-2002120 (m /day) 2002 -2009120 (m /day) 3 2 in 198280 (m /day) 120 120 120 3 80 120 120 120 4 80 150 150 150 5 80 150 150 150 6 80 150 150 150 7 80 150 150 150 8 80 150 150 150 9 80 150 150 150 10 80 80 80 80 11 80 80 80 80 12 80 80 80 80 13 100 110 110 110 14 100 110 440 220 15 100 110 440 220 16 100 110 440 220 17 100 110 440 220 18 100 110 440 220 19 100 110 440 220 20 100 110 440 220 21 100 110 440 220 22 100 110 440 220 23 100 110 440 220 24 100 110 440 220 25 30 40 100 40 26 30 40 100 40 27 30 40 100 40 28 30 40 100 40 29 60 40 40 40 30 60 40 40 40 31 60 40 40 40 32 60 100 100 100 33 80 100 100 100 34 100 100 100 100 35 60 100 100 100 36 30 100 100 100 37 60 100 100 100 38 60 100 100 100 39 60 100 100 100 Total 2950 3900 7770 5110 "Well No." donates to artificial recharge well sites used for the model and are shown in Figure 4-9A in a white box. "Well No." donates to artificial recharge well sites used for the model and are shown in Figure 4-9A in a white box. "Well No." donates to artificial recharge well sites used for the model and are shown in Figure 4-9A in a white box. "Well No." donates to artificial recharge well sites used for the model and are shown in Figure 4-9A in a white box. "Well No." donates to artificial recharge well sites used for the model and are shown in Figure 4-9A in a white box.

162

Estimated Recharge (m3/day) Using Head Observation Pumping (m3/day) Data from The Well Site Engneering Estimated Recharge (m3/day) Using Head and salinity Observations Estimated Pumping (m3/day) Using Head and Salinity Observations

9000

High High

8000

Moderate Very

7000

/day)

3 Low 6000

5000

4000

3000

2000 Rechareg orpumpingRechareg amount (m 1000

0

Time (Year)

Figure 4-13. Estimated recharge and pumping (m3/day) using head observations and couples head and salinity observations through the model time (1982-2009). Estimated recharge and estimated pumping rates using head observation is after (Eissa et al., in review). 163

The model is valid for most regions within the model except for the area close to the coast and the area located in Sabkha deposits. The poor fit in the Sabkha is due to the fact that the model does not explicitly simulate rock-water interactions. In the coastal region the model performances poorly in areas of large horizontal concentration gradients. This can be attributed to a relatively large grid resolution (100m) and imprecise well locations. Two drilled wells located in the well field does not calibrated well as the salinity of these wells exceed or close to the maximum concentration assigned for the initial conditions after (El-Refaeai, 1992).

4. Summary and Conclusions

Groundwater chemistry in the Wadi Watir delta evolves from recharge water entering the Wadi Watir delta aquifers from the mountain blocks by water-rock reactions, upwelling of a deep saline water in the well field due to over pumping, seawater intrusion along the coast, and evaporation of some of the shallow groundwater. A Sulin diagram

(Sulin, 1946) for water samples collected in 2007 and 2009 supports these sources and processes that produce the observed water salinity. The relatively low salinity [Na:SO4] water type represents groundwater recharge to the delta aquifers. Groundwater in the deep drilled wells in the water supply well field and some wells along the coast are a

[Ca:Cl] water type. This water type reflects upwelling of a deep saline water beneath the well field and seawater intrusion along the coast. The majority of shallow groundwater samples in the Wadi Watir alluvial aquifers are a [Mg:Cl] water type reflecting water- rock interactions with carbonate rocks of marine origin.

164

The environmental isotopes 87Sr/86Sr, δ81Br, and δ37Cl support that groundwater in the well field contains water from upwelling of a deep saline water. Samples from the well field are depleted in 87Sr/86Sr and δ81Br indicating a deep saline groundwater mixing with the well field water. The δ81Br data are depleted because of their old marine genesis.

δ37Cl shows enrichment, because similar to δ81Br, chloride in the groundwater is derived from old marine deposits.

The groundwater flow and solute transport model was calibrated by obtaining a general agreement between the observed and modeled salinity and water level values.

The model is most sensitive to porosity, hydraulic conductivity, and dispersivity values.

Considerable water level decline was observed in the well field due to over pumping.

Increasing groundwater salinity along the coast resulted from the encroachment of seawater at the coast due to groundwater pumping. The extent of seawater intrusion varies in the different aquifers due to variability of hydraulic conductivity through the multiple layer aquifers. The layers most affected by salt water intrusion are the surficial layer (A1) and the sand layer (A2) due to pumping stresses from these layers. In this area, most of the drilled wells are screened across both the surficial layer and the high hydraulic conductivity sand layer.

The total average pumping rate for all the wells in the well field from 1982 to

2009 was estimated to be 3100 m3/day. In comparison, the total average pumping rate for all the hand dug wells is 60 m3/day. These data are very close to the values obtained from the well site engineers responsible for the operating the well field and owners of the hand dug wells. The model estimated pumping rates for wells in the well field ranges from 200

165 to 1400 m3/day, whereas, the model estimated pumping rates for the hand dug wells ranges from 0.5 to 5 m3/day, except for the hand dug wells located inside the hotels and resorts near the coast (Sites. 34 and 35) where pumping rates range from 7 to 30 m3/day through the last decade. The pumping rates are linked to development activities along the coast, where some rates have doubled over the last ten years.

The model calculated average annual recharge rate is 6000 m3/year. This value is about 1800 m3/year greater than the value calculated using only water level data (Eissa et al. 2012, in review). Calibrating a groundwater flow and solute transport model using both water level and salinity data should give a better estimate of average annual recharge than using water level data alone.

There are five primary sources of salinity that affect groundwater in the Wadi

Watir delta aquifers: (1) dissolution of minerals and salts from the aquifer matrix; (2) upwelling of deep saline groundwater due to over pumping; (3) seawater intrusion along the coast; (4) leaching of salts from sabkha deposits near the coast; and (5) evaporation of shallow groundwater. The problem of upwelling saline groundwater could be reduced by decreasing pumping rates and/or adjusting well screen intervals to be shallower in future drilled wells. Seawater intrusion along the coast could be lessened by drilling new wells at least 300 m from the shore line and the Sabkha area, and reducing pumping rates or keeping them low. In the Wadi Watir delta, groundwater located between the well field area and Sabkha area near the coast is characterized by a relatively low salinity groundwater as compared with the other groundwaters in the delta area. Thus, constructing more shallow wells in this area may provide more potable water. However,

166 even groundwater in this area cannot be extensively developed on a sustainable basis because of the small rate of average annual recharge to the Wadi Watir delta aquifers.

5. References

Abbas, A.M., Atya, A.A, Al-Sayed, E.A, and Kamei, H., 2004. Assessment of groundwater resources of the Nuweiba area at Sinai Peninsula, Egypt by using geoelectric data corrected for the influence of near surface inhomogeneities, Journal of Applied Geophysics 56(2):107-122. Abd El Hafez, A.A., 2001. Chemical evaluation and possible treatment sea water intrusion in the groundwater in some coastal areas, South Sinai. M. Sc. Thesis, Fac. Sci., Al Azhar Univ. (girls), 221 p. Back W., Hanshaw B.B., Plummer L.N., Rahn P.H., Rightmire C.T., Rubin M., 1983. Process and rate of dedolomitization, mass transfer and 14C dating in a regional carbonate aquifer. Geol Soc Am Bull 94 (12): 1415-1429. Bakker, M., Oude Essink. Baxter, G.P., and Wallance, C.C., 1916, Changes in volume upon solution in water of halogen salts of alkali metals: IX American Chemical Society Journal, no. 38, p. 70-104. Chin A.D., and Wang T., 1992. An Investigation of the Validity of First-Order Stochastic Dispersion Theories in Isotropic Porous Media. WATER RESOURCES RESEARCH, VOL. 28, NO. 6, PAGES 1531-1542. Clark, I., and Fritz, P., 1997. Environmental isotopes in hydrogeology. Lewis Publishers, New York. Eggkamp H.G.M., 1994. The geochemistry of chlorine isotopes. Doctoral thesis, Faculteit Aardwetenschappen, Universiteit Utrecht, The Netherlands, 150P. Eggenkamp, H.G.M., and Coleman, M.L. 2000. Rediscovery of classical methods and their application to the measurement of stable bromineisotopes in natural samples. Chemical Geology Volume 167, Issues 3–4, 15 June 2000, Pages 393–402.

167

Eissa, M.A., Thomas, J.M., Pohll, G., Hershey, R.L., Kamal A.D., Dawoud, M.I., Gomaa, M.A., and El-Shiekh A., (in review, Journal of Hydrology). Groundwater resources sustainability in the Wadi Watir Delta, Gulf of Aqaba, Sinai, Egypt. Eissa, M.A., Thomas, J.M., Hershey, R.L., Dawoud, M.I., Pohll, G., Gomaa, M.A., and Kamal A.D. (in review, Environmental Earth Sciences). Geochemical and Isotopic Evolution of Groundwater in the Wadi Watir Watershed, Sinai Peninsula, Egypt. El Ghazawi, M.M., 1999. Reconsideration of hydrogeologic setting in the delta of Wadi Watir, southern Sinai. Bulletin of Science, Mansoura Univ., Egypt, vol. 26 (1). El Kiki, M.F., Eweida, E.A, El-Refeai, A.A., 1992. Hydrogeology of the Aqaba rift border province. Proc 3rd Conf Geol Sinai Develop, Ismailia, Egypt, 91-100. El-Refaei, A.A., 1992. Water resources of southern Sinai, Egypt geomorphological and hydrogeological studies. Ph.D. Thesis, Fac. Sci., Cairo Univ., Egypt, 357p. El Sayed, M. H., 2006. Comparative study of water quality pf the Quaternary aquifer in Wadi Watir basin and its delta. Southeast Sinia, Egypt. Egyptian J. Desert Res., 56, No.1, 17-46 (2006). El-Shazly, E.M., Abddel-Hady, M.A., El-Ghawaby, M.A., El-Kassas, I.A., El-Shazly, M.M., 1974. Geology of Sinai Peninsula from ERTS-1 Satellite Images, Acad. Sci. Res. Tech., Remote Sence. Project. Eyal, M., Bartov, Y., Shimron, A.E., Bentor, Y. K., 1980. Sinai geologic map, Scale 1:500,000. Ministry of Energy and Infrastructure. Admin. For Res. In Earth Sci., Israel. Fishman, M.J., and Friedman, L.C., 1985. Methods for determination of inorganic substances in water and fluvial sediments. U.S. Geol. Surv. Book 5,Chapter A1.Open File Report 85-495, Denver, Colorado, U.S.A.

168

Frape, S.K., A. Blyth, R. Blomqvist, R.H. McNutt, and M. Gascoyne. 2004. Deep Fluids in the Continents: II. Crystalline Rocks, pp 541–580. In Surface and Ground Water, Weathering, and Soils (ed. J.I. Drever), v. 5 Treatise on Geochemistry (eds. H.D. Holland and K.K. Turekian), Elsevier — Pergamon, Oxford. Gelhar, L. W., C. Welty, and Rehfeldt, K. W., 1992. A critical review of data on field- scale dispersion in aquifers, water resource. Research, 28(7), 1955-1974. Geo W. and Langevin D. C., 2002. User's guide to seawat: A computer program for simulation of three-dimensional variable-density ground-water flow, U.S. Gological Survey technical water-resourec investigations 6-A7. Harbaugh, A.W., Banta, E.R., Hill, M.C., and McDonald, M.G., 2000. MODFLOW- 2000, the U.S. Geological Survey Modular Ground- Water Model—User Guide to Modularization Concepts and the Ground-Water Flow Processes. U.S. Geological Survey Open-File. Hershey R. L., Heilweil V. M., Gardner P., Lyles B., Earman S., Thomas J., and Lundmark K.W., 2007. Ground-water Chemistry Interpretations Supporting the Basin and Range Regional Carbonate-rock Aquifer System (BARCAS) Study, Eastern Nevada and Western Utah. Prepared by Desert Research Institute, Nevada System of Higher Education and U.S. Geological Survey. DHS Publication No. 41230 Hassan, A.A., 1967. A new Carboniferous occurrence in the Abu Durba, Sinai, Egypt. 6th Arab. Petroleum Conference, Baghdad, 2, 8p. Himida, I.H., 1997. Water resources of Wadi Watir. Internal report, Desert Research Center (In Arabic). Ismail Y.L., 1998. Hydrogeological and hydrochemical studies on Wadi Watir area, South Sinai, Egypt. PhD Thesis, Fac Sci, Suez Canal Univ, Egypt. Issawi, B. and Jux, U., 1982. Contribution on the stratigraphy of the Paleozoic rocks in Egypt. Geological Survey of Egypt, 64, 28.

169

Kaufmann.R., 1984. Austin Long, Harold Bentley, Stanely Davis: Natural chloride isotope variations, Nature, Vol. 309 24 May 1984. Khalil H.M., 2010. Hydro-geophysical Configuration for the Quaternary Aquifer of Nuweiba Alluvial Fan, Journal of Environmental and Engineering Geophysics, Volume 15 Issue 2 pp. 77-90 2010. Langevin D., Christian, and Guo W., 2002. User's Guide to SEAWAT: A Computer Program for Simulation of Three-dimensional Variable-Density Ground-Water Flow U.S. Geological Survey Techniques of water resource investigations 6-A7. Langevin, C.D., and Guo, W., 2006. MODFLOW/MT3DMS-based simulation of variable density ground water flow and transport Ground Water vol. 44, no. 3:339-351. Lyons, W.B., S.W., Tyler, H.E., Gaudette, D.T., 1995. The use of strontium isotopes in determining groundwater mixing and brine fingering in a playa spring zone, Lake Tyrrell, Australia. Long d. Journal of Hydrology 167, 225-239. Liebscher, A., Lüders, V., Heinrich, W., Schettler, G., 2006a. Br/Cl signature of hydrothermal fluids: liquid–vapour fractionation of bromine revisited. Geofluids 6, 113–121. Liebscher, A., Barnes, J., Sharp, Z., 2006b. Chlorine isotope vapor–liquid fractionation during experimental fluid-phase separation at 400 °C/23 MPa to 450 °C/42 MPa. Chem. Geol. 234, 340–345. Mabrouk M.A., and Nasr I.M., 1997. The Hydrogeologic Setting In The Delta Of Wadi Watir, Gulf of Aqaba, A Geoelectrical Resistivity Sounding Approach. Desert Inst. Bull., Egypt. 47 No. 1, 149-162. McNutt, R.H., Frape, S.K., Fritz, P., Jones, M.G., and MacDonald, I.M., 1990. The 87Sr/86Sr values of Canadian Shield brines and fracture minerals with applications to groundwater mixing, fracture history, and geochronology. Geochim. Cosmochim. Acta 54, 202–215.

170

McNutt, R.H., Frape, S.K., Fritz, P., 1984. Strontium isotopic composition of some brines from the Precambrian Shield of Canada. Chem. Geol. Isot. Geosci. 2, 205–215. Plummer, L.N., Prestemon, E.C., and Parkhurst, D.L., 1992, NETPATH: An interactive code for interpreting NET geochemical reactions from chemical and isotopic data along a flow PATH, in Kharaka, Y., and Maest, A.S., eds., International Symposium on Water-rock Interaction, 7th, Park City, Utah, July 9–23, 1992, Proceedings: Rotterdam, Balkema, p. 239–242. Plummer, L.N., Prestemon, E.C., and Parkhurst, D.L., 1994. An interactive code (NETPATH) for modeling net geochemical reactions along a flow path, version 2.0. US Geol Surv Water-Resour Invest Rep 94-4169:130. Rainwater, F.H., and Thatcher, L.I., 1960. Methods for collection and analysis of water samples. U.S.Geol. Survey, Water Supply, 1454, 301p. Said, M.M., 2004. Geochemistry of groundwater in coastal areas, South Sinai, Egypt. Ph.D. Thesis, Fac. Sci., Ain Shams Univ., Egypt.

Schoeller, H., 1962. Les eaux Souterraines. Masson & Cie, Paris, 642p.

Shalaby, A.I., 1997. Geomorphology and hydrogeology of Wadi Watir basin, S. E. Sinai, Egypt‖. M.Sc. Thesis, Fac. Sci., Mansoura Univ., Egypt, 185p. Shmulovich, K.I., Landwehr, D., Simon, K., Heinrich, W., 1999. Stable isotope fractionation between liquid and vapour in water-salt systems up to 600 degrees C. Chem Geol., 157, 343-354. Shouakar-Stash, O., Drimmie, R.J., Frape, S.K., 2005a. Determination of inorganic chlorine stable isotopes by continuous flow isotope ratio mass spectrometry. Rapid Commun. Mass Spectrom. 19, 121–127. Shouakar-Stash, O., Frape, S.K., Drimmie, R.J., 2005b. Determination of bromine stable isotopes using continuous-flow isotope ratio mass spectrometry. Anal. Chem. 77, 4027–4033.

171

Stotler, R. L., Frape, S. K., Shouakar-Stash O., 2010 An isotopic survey of δ81Br and δ37Cl of dissolved halides in the Canadian and Fennoscandian Shields Chemical Geology 274, 38-55. Sulin, 1946 V.A. Sulin Vody Neftyanykh Mestorozhdenii v Sisteme Prirodnikh Vod Gostoptekhizdat, Moscow. United States Geological Survey (USGS), 2004. Reprocessing by the Global Land Cover Facility, 2004, (30) Arc Second SRTM Elevation. College Park, Maryland: the GLCF, Retrieved from World Wide Web. http://www.landcover.org. Zheng, C., and Bennett, G.D., 1995. Applied contaminant transport modeling, theory and practice: Van Nostrad Reinhold, 440 P. Zheng, C., and Wang, P.P., 1999. MT3DMS Documentation and User’s Guide. Contract Report SERDP-99-1, U.S. Army Engineer Research and Development Center, Vicksburg, MS.

172

CHAPTER V

Conclusion

The Wadi Watir delta is a tourist area in the arid southeastern part of the Sinai

Peninsula, Egypt, where development and growth of the community is constrained by the amount of groundwater that can be withdrawn sustainably. To effectively manage groundwater resources in the Wadi Watir delta, the origin of groundwater recharge, groundwater age as related to the timing of groundwater recharge, groundwater chemistry, upwelling of deep saline groundwater in the well field area and seawater intrusion along the coast needs to be understood. Mineral identification, rock chemistry, water chemistry, and isotopes (δ2H, δ18O, 14C, δ13C, 87/86Sr, δ37Cl and δ81Br) were used to identify recharge sources, mixing of different waters (including a saline groundwater and seawater), and groundwater age. The geochemical model NETPATH and groundwater flow and solute transport models on the SEAWAT platform were used to: (1) determine the evolution of groundwater chemistry in the study area, including mixing of deep saline water in the well field caused by over pumping and seawater intrusion along the coast;

(2) estimate average annual groundwater recharge to the alluvial aquifers of the Wadi

Watir delta; and (3) simulate annual groundwater pumping, evaluate upwelling of saline water from beneath the well field, and determine the extent of seawater intrusion along the coast for different pumping scenarios.

The El Shiekh Attia, Wadi El Ain and main channel areas are the primary areas in the upper Wadi Watir watershed that supply recharge to the alluvial aquifers of the Wadi

Watir delta. Groundwater in the El Shiekh Attia area has isotopic signatures that are

173 similar to winter rain, indicating that recent precipitation is the main source of groundwater recharge to this area. The same storms that recharge the El Sheikh Attia area also recharge the Wadi El Ain area, but the isotopic signature of the groundwater in the

Wadi El Ain area is more depleted in δ18O and δ2H than the El Sheikh Attia area groundwater due to the rainout effect because these storms travel farther to reach the

Wadi El Ain area and this area is higher than the El Sheikh Attia area. The main channel area of Wadi Watir, up gradient from the delta, receives groundwater flow primarily from the El Shiekh Attia area as indicated by isotopic and chemical data. The isotopically depleted groundwater in the Wadi El Ain area is not evident in the main channel area.

Groundwater chemistry in the study area evolves as it flows from the El Shiekh

Attia area down gradient through the main channel area to the Wadi Watir delta.

Dissolution of gypsum, halite, and igneous rocks are the primary contributors of major ions to the groundwater, while precipitation of calcite and formation of clays remove ions from the groundwater. Corrected 14C groundwater ages in the study area range from modern to 4600 years for one sample from the lower Cretaceous aquifer. Also, two groundwater samples in the well field that contain some saline water from upwelling due to over pumping have average groundwater ages of 2400 and 4000 years. These ages are affected by mixing with saline water that likely contains little or no carbon-14. The modern age of alluvial aquifer groundwater in the delta area, that is unaffected by upwelling of saline groundwater, has very important implications for groundwater management in the area because this modern groundwater is recently recharged and is not thousands of years old. Thus, this resource can be managed in a sustainable way by not

174 pumping more than is being recharged on an average annual basis. If this groundwater resource were paleo-water recharged thousands to tens of thousands of years ago, then it could only be managed for a one-time use and not as a sustainable resource.

Groundwater in the Wadi Watir delta alluvial aquifers fall into two groups based on their isotopic and chemical content. Group I groundwaters are characterized by relatively depleted δ18O and δ2H values , as compared to Group II groundwaters, ranging from -3.82 to -2.58‰ and -19.5 to -13.4‰, respectively. Most Group I groundwaters are located away from the coast with many of the sites along the mountain front particularly where the Wadi Watir drainage enters the delta. The isotopic values for Group I groundwaters are similar to the δ18O and δ2H of groundwater in the upper watershed recharge areas of El Sheikh Attia and the main channel. Group II groundwaters are located in shallow wells along the coast and near the wetted Sabkha area and their δ18O and δ2H values are enriched compared to Group I groundwaters, ranging from -0.66 to

+6.86‰ and -5.9 to +22.5‰, respectively. Group II groundwaters are evaporated as indicated by their plotting along an evaporation line originating from winter rain and

Group I groundwaters (δ2H = 4.7 x δ18O – 2.6). Two wells that are located along the coast

(sites 35 and 42) have elevated ion concentrations and isotopic data that indicate they are mixing with seawater. δ18O versus Cl and δ18O versus Br plots show that these two waters plot on a mixing line between Group I groundwaters and seawater.

Geochemical modeling also supports groundwater falling within two distinct groups in the Wadi Watir delta as indicated by isotopic data. Group I groundwaters are low saline groundwaters, as compared to Group II groundwaters, which have evolved due

175 to dissolution of salts and minerals of the aquifer matrix along the flow path. Group II groundwaters are characterized by higher salinity than Group I groundwaters because they have been concentrated by evaporation and likely have undergone additional solution of salts and minerals along flow paths. Geochemical modeling shows that the observed water chemistry for wells in the well field, as represented by sites 17 and 18, is produced by mixing 6 to 12% saline water with groundwater in the well field. For the two wells along the coast that contain some seawater (sites 35 and 42), the geochemical modeling shows that these groundwaters contain 6 to 12.5% seawater.

The majority of shallow groundwaters in the Wadi Watir alluvial aquifers are a

[Mg:Cl] water type reflecting water-rock interactions with carbonate rocks of marine origin. Groundwater in the deep drilled wells in the well field and some wells along the coast are a [Ca:Cl] water type. This water type reflects upwelling of deep saline water beneath the well field and seawater intrusion along the coast. The environmental isotopes

87/86Sr, δ81Br, and δ37Cl support that groundwater in the well field contains water from upwelling of deep saline water. Samples from the well field are depleted in 87/86Sr and

δ81Br and enriched in δ37Cl because of their old marine genesis associated with the deep saline water. These environmental isotopes also support mixing of seawater along the coast with groundwater. Wadi Watir alluvial aquifers also include four samples with relatively low salinity that are a [Na:SO4] water type.

The groundwater flow and solute transport modeling results show that: (1) the extent of seawater invasion along the coast for the period 1982 to 2009 is about 200 m;

(2) the main factors controlling groundwater salinity are the pumping stresses and the

176 availability of recharge; (3) for the period 1982 to 2009 the daily extraction rate from the individual deep drilled wells of the main well field ranged from 200 to 1400 m3/day with an annual average rate of 3100 m3/day for all the wells; and (4) the estimated annual average recharge to the delta using different model scenarios ranged from 4200 to

6000 m3/day for the period 1982 to 2009. The model results also showed that the pumping rates for the hand dug wells range from 0.5 to 5 m3/day with a total average pumping rate for all the hand dug wells of only 60 m3/day, except for the hand dug wells located inside the hotels and resorts near the coast (sites 34 and 35). These two wells had pumping rates of 7 to 30 m3/day throughout the last decade. Pumping rates are linked to development activities along the coast, resulting in some rates doubling over the last ten years.

There are five primary sources of salinity that affect groundwater in the Wadi

Watir delta aquifers: (1) dissolution of minerals and salts from the aquifer matrix;

(2) upwelling of deep saline groundwater due to over pumping; (3) seawater intrusion along the coast; (4) leaching of salts from sabkha deposits near the coast; and

(5) evaporation of shallow groundwater. The problem of upwelling saline groundwater could be reduced by decreasing pumping rates and/or adjusting well screen intervals to be shallower in future drilled wells. Seawater intrusion along the coast could be lessened by drilling new wells at least 300 m from the shore line and the Sabkha area, and reducing pumping rates or keeping them low. In the Wadi Watir delta, groundwater located between the well field area and Sabkha area near the coast is characterized by a relatively low salinity groundwater as compared with the other groundwaters in the delta area.

177

Thus, constructing more shallow wells in this area may provide more potable water.

However, even groundwater in this area cannot be extensively developed on a sustainable basis because of the small rate of average annual recharge to the Wadi Watir delta aquifers.