Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

Water knowledge for the new millennium

Ramadan A. Awwad T.N. Olsthoorn Y. ZhouStefan UhlenbrookEbel Smidt

Optimum Pumping-Injection System for Saline Groundwater

Desalination in Sharm El Sheikh

WaterMillWorking Paper Series2008, no. 11

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

Authors: Ramadan A. Awwad1,2, T.N. Olsthoorn3, Y. Zhou1, Stefan Uhlenbrook1, Ebel Smidt3

1 UNESCO-IHE, Institute for Water Education, PO Box 3015 DA, Delft, the Netherlands, [email protected] 2 Water Resource Research Institute, National water centre, 13621 El Qanater El Khairiya, Egypt 3 Department of Water Resources, Delft University of Technology, PO Box 5048, 2600 GA Delft, the Netherlands

The WaterMill Working Paper Series

The WaterMill Working Paper Series is part of the WaterMill project (Water Sector Capacity Building in Support of the Millennium Development Goals), which has been implemented by the UNESCO-IHE Institute for Water Education in Delft, the Netherlands, since 2004. The WaterMill Project is a capacity building project for the WaterSector responding to the targets as laid down in the Millenium Development Goals and by the Commission on Sustainable Development. The project offers several advanced training programmes at the post-graduate level to 72 professionals originating from the partner countries of the Netherlands.

As part of their training each of the 72 professionals had undertaken a 6-month research project which focuses on the achievement of the MDGs in their home country. The WaterMill working paper series presents the research outputs of these projects.

Contact information

For more information regarding the WaterMill project and the WaterMill Working Paper Series please contact Prof. Dr. Pieter van der Zaag (email: [email protected]).

Abstract

Tourism is the main economic base in Sharm el Sheikh and is booming. However, no fresh water resources are available. Currently groundwater, which is salt in the entire coastal area of this part of the Sinai, is desalinated to produce drinking water. Ridgewood is the major actor in desalination in this area and is currently running four plants in Sharm el Sheikh. Due to increasing demand the capacity is extended regularly.

Desalination is done by RO (reversed osmosis). This process splits the extracted water into about 30% drinking water and 70% so-called brine, which contains all the salts. This very salt water is discharged back into the aquifer by means of a number of injection wells. Discharge into the sea is not permitted because of the danger feared for the coral reef ecosystem, which is the basis of the tourist economy.

However, three dangers are involved with the method of using the aquifer in this way. First is the very real risk of short-circuiting, in which the injected brine flowing through the aquifer reaches the extraction wells in a limited time. Second problem is that increasing capacity is limited by increasing drawdowns, due to which wells and pumps have to be lowered, and more wells have to be installed. Last, there is the risk of pollution of the groundwater by other sources, for instance, loss of wastewater form sewerage pipes, which may cause the extracted water quality to deteriorate over time, for instance causing increase iron concentrations. The plant studied in this thesis suffers from most of these problems. This thesis analysis the problems for one of the Ridgewood sites, El Hadab, and suggests some solutions. There are also important for the sustainability of water supply in entire region of Sharm El Sheikh and others parts of the Red Sea Coast.

To analyze the situation, data were gathered on site and obtained form Ridgewood. These data were combined with other data and analyzed for the geology and lithology of the aquifer. The pumping tests that have been carried out in the past have also been analyzed and used to obtain the necessary hydraulic data of the aquifer.

In order to analyze the situation systematically, an extended groundwater model was built, that includes flow modeling, solute transport and density of the groundwater. The analyzed data served to construct the model. The model was based on the well-known codes of MODFLOW, MT3DMS and Seawat2000, all public domain products of the USGS. The modeling was done using a graphical user interface (GUI) of Groundwater Vistas, which is a commercial product. In order to be accurate enough the model cell size ranged from 25 to 50 m, while 14 layers were chosen to allow sufficiently accurate modeling of the influence of density on the groundwater flow.

Several pumpage-injection scenarios have been analyzed for El Hadab to examine the groundwater flow and solute transport system and to address the effects of anticipated future demands on the system.

The modeling provided the following results:

• The drawdown reaches a steady state in about one year. This implies that increasing drawdowns can only be due to increasing extractions and possibly due to clogging of wells.

• The conductivity of the aquifer (about 3 m/d) is low. Therefore, the extraction at any location is limited due to large drawdown.

• The salinity of the extracted water in El Hadab showed a tendency of increase. So their salinity has not increased much yet. However, the model shows that the brine is currently flowing towards the wells and will reach the extraction wells within a couple of years with high probability. Once this happens, salinity will increase to almost that of the injected brine within about two decades. This will increase desalination cost and probably increase the salinity and amount of brine in the future, thus aggravating the problem.

• Based on the flow computations, a scenario was run in which the injection wells have been moved to the very south of the modeled peninsula. This delays short-circuiting for possibly about a century (depending on the increase of the desalination capacity). However, the brine will also reach the wells in this “optimal” case on the long run. By that time, most of the groundwater will have been replaced by brine. Therefore, research is recommended into long-term sustainable solutions, which will probably involve larger desalination plants with injection of the brine into the sea in an environmentally sustainable way. A suggestion of this is presented in the report.

• One problem encountered in El Hadab was high iron concentration in the extracted water, causing of fouling the RO-membranes. The data show that only well 19 has a high Fe concentration, which is probably due to some source of pollution with organic matter (wastewater) upstream of this well. The model path lines show where to look for a possible source. It is recommended that well 19 is not shut off, because then another well will get the iron. Ridgewood may investigate treatment of this well separately.

• When analyzing the data, it was revealed that the manganese concentration is almost doubled between 2005 and 2007. This is a new problem, which should be addressed by Ridgewood in the near future.

Contents

1. Introduction ............................................................................................................................... 1

1.1 Problem Definition................................................................................................................ 2

2. Methodology............................................................................................................................... 3

3. Geological And Hydrological Setting....................................................................................... 4

4. Numerical Simulation Of Coupled Groundwater And Solute Transport ............................ 5

4.1 Model Construction............................................................................................................... 6

4.2 Model Layers ........................................................................................................................ 6

4.3 Spatial Discretization ............................................................................................................ 7

4.4 Boundary Conditions ............................................................................................................ 8

4.5 Source And Sinks.................................................................................................................. 9

4.6 Initial Conditions................................................................................................................... 9

4.7 Numerical Solvers And Accuracy......................................................................................... 9

5. Simulation Of Scenarios.......................................................................................................... 14

5.1 Scenario A: Business As Usual........................................................................................... 14

5.2 Scenario B: Elimination Of Injection In El Hadab.............................................................. 15

5.3 Scenario C: New Extraction Wells...................................................................................... 16

5.4 Scenario D: New Injection Wells........................................................................................ 16

5.5 Scenario E: Pumping-Injection From Seawater .................................................................. 17

6. Manganese (Mn) ...................................................................................................................... 19

7. Conclusions............................................................................................................................... 20

8. References................................................................................................................................. 21

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 1

1. Introduction

Egypt is located in northeastern Africa and includes the Sinai Peninsula (also seen as Sinai), which is often considered part of Asia. Egypt is gift of the Nile .Egypt's natural boundaries consist of more than 2,900 kilometers of coastline along the Mediterranean Sea, the Gulf of Suez, the Gulf of Aqaba, and the Red Sea. Sinai, the triangular-shaped peninsula of Egypt, is situated between Asia and Africa. The separation of the two continents caused the form and geographical shape of Sinai the way it looks today. Sinai is approximately 380 km long (north - south) and 210 km wide (west - east). The surface area has an extension of 61.000 km²; the coasts are stretching about 600 km on the west and on the east (Figure 1-1).

Figure 1-1, map for Egypt

The total length of the Egyptian Red Sea coast is about 1,705 km. The City of Sharm El Sheikh is a main resort city in the Red sea coast. Sharm El Sheikh is an Egyptian city and a tourist center on the Red Sea. Sharm el-Sheikh is overlooking the Straits of Tirana at the mouth of the Gulf of Aqaba. It stretches for about 40 km along the seashore, and it does not reach far into the surrounding desert. Figure 1-2 show Sharm El Sheikh City. Egyptian and foreign investors to become the leading seashore resort on the Red Sea have continually enlarged the city.

Figure 1-2, a map of the great Sharm el-Sheikh is located on the Egyptian Red Sea coast.

2 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

1.1 Problem definition The Red Sea coast and Sinai use the groundwater as main water supply. From start to finish point, only one third of the water is truned into fresh drinking water, while the other two thirds become "brine" and are discharged back into the ground from where it was drawn.

Many desalination plants are along the coastline in Egypt. Ridgewood operates several sites in the south Sinai Peninsula (Sharm El Sheikh), where desalinated water is produced from saline groundwater. The concentrate (often called brine) is rejected into the subsurface through rejecting wells located at some distance from the extraction wells. A quite common problem is that groundwater salinity increases over time, raising the energy and overall cost of the desalination plants in question.

El Hadab station is the first and largest Ridgewood production site in the Southern Sinai for water desalination. The capacity of El Hadaba desalination plant is about 6,600 m3/day. This is the amount of drinking water produced. The station has 19 production wells, and 7 injection wells. The plant was built and started in 1996. The station has a pure water recovery 32%. The plant and the extraction wells are located about 1,500 m from the sea. The injection wells, however, are located close to the sea shore (Figure 1-3).

Figure 1-3, locations of production and rejection wells in El Hadab station

The problems encountered in El Hadab site are:

1. The Static water level depth in July 1996 was bout 42m and by March 2007 had fallen to about 75m, a drop of 33 m in 11 years.

2. The conductivity of the pumped groundwater in 12/10/2007 was 53,000 μS/cm, but the conductivity of the well was about 52,000 μS/cm in 20/8/2002, an increase of salinity.

3. The new well 19 with a depth of 140 m yields water with a high iron (Fe) Concentration of about 7.2 mg/L. This iron concentration causes problems in the desalination process.

4. Due to the decline of water level, many wells out of service.

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 3

2. Methodology

The density-flow and transport model used in this paper based on the finite difference approximation. The analysis was done with the help of a groundwater model of the site and its surroundings allowing simulation of the development of the groundwater salinity at the site and the extraction wells. For this purpose, the density dependant groundwater model SEAWAT 2000 and a graphic user interface Groundwater-Vistas were used.

In summary, the model constructed based on the following steps:

1. Establish the minimum area to be represented by the model.

2. Determine the hydrological features that can serve as boundaries to the model.

3. Compile the hydrogeogical information.

4. Compile the geological information.

5. Determine the number of physical dimensions needed for the model.

6. Define the size of the model.

7. Define the model discretization.

8. Input the model parameters.

9. Input the model stresses.

10. Run the model.

11. Output the calculated hydraulic heads.

12. Calibrate as far as possible using the groundwater depth and salinity data from the site

13. Using the model and the data, analyze the past hydraulic and salinity development of the site

14. Develop a series of scenarios, of which the prediction of the development of the site when continuing business as usual is the so-called 0-scenario with which all other scenarios will be compared

4 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

3. Geological and hydrological setting

Sharm El Sheikh (Figure 3-1) geology stems from the following eras: Precambrian, lower Miocene, Cretaceous, Pliocene and Quaternary (based on geologic map of Sinai 1:250,000 published by the Egyptian Geological Survey and Mining Authority (EGSMA, 1994 and WRRI)). However, the most important part is the coastal zone geology. This will be discussed briefly in the following section.

The rocks exposed north and west of Sharm El-Sheikh forming the northern part of the rocky land at the coastal plain. These outcrops are composed mainly of alternate beds of marl and sandstones with fossillerferous carbonate beds in the lower part. Pliocene rocks are exposed along the coastal plain from Ras Mohammed (in the south of the Sharm El Sheikh) to Ras Nusrani forming the rocky lands of the coast. They comprise dark colored conglomerates that alternate with sandstone beds. Quaternary deposits represented by continental to littoral sediments. These deposits are made up of coralline limestone, Wadi deposits, and alluvial deposits (Omara et al, 1959).

Figure 3-1, Rock units of the model area (Sources, EGSMA, 1994, WRRI, 2008)

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 5

4. Numerical simulation of coupled groundwater and solute transport

The model area, in Sharm El Sheikh South Sinai, located between longitudes 34º16´00´´to 34º19´30´´ and latitudes 27º51´00´´to 27º54´40´´, as shown in Figure 4-1. The surface area of the aquifer system is about 40.32 km2 with length 8.5 km parallel of shore line. The east boundary of the study area is Red Sea, the west boundary is Jebel Ruweisat EL Nimr, the North border Jebel Safra Al Elat, and the south border is Jebel El Safra. The study area lies within the geographical UTM zone number 36 N.

Figure 4-1, model domain

Hydraulic parameters were estimated for aquifer with step pumping test for ten wells (W3, W4, W5,W11.W12,W13,W16,W18,W19,Winjection 1) and these field experiments performed by many drilling companies in shown Figure (5-1) illustrate the distribution locations of the tests. Formulas of Theis and Birsoy and Summers (1980) were used for the analysis and evaluation of the pumping test data.

Figure 4-2, illustrate distribution locations of step pumping tests.

Groundwater flow and transport modeling used Geographical information. The construction, calibration, evaluation of the transient numerical flow and solute transport model of Hadaba are described in this paper. A sensitivity and pumping tests analysis was used to calibrate the model. No rain recharge is assumed.The assumptions made as following:

1. The Horizontal hydraulic conductivity is assumed isotropic within a model cell. Heterogeneity is simulated by varying horizontal hydraulic conductivity of model cells or layers. The vertical hydraulic conductivity is based on specified values of horizontal hydraulic conductivity.

2. Pre-pumping conditions are assumed steady-state. During 1996 to 2007, groundwater pumping is assumed to be the only transient stress on the system to cause the observed decline in groundwater levels in wells.

3. The Pumping-injection system the desalination plant (El Hadab station) was modelled.

6 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

4. The main recharge of the aquifer occurs by coastal salt-water intrusion from east and south (Red Sea coast) and injected brine water.

4.1 Model construction The model is a finite difference approximation of the three dimensional (groundwater system). The hydrological data sets for the Umm Said Plateau (El Hadab) described below in this paper were discretized to develop the input arrays required for this model. The data sets were prepared at grid-cell resolution ranging from 25 to 50 meters (m).

All map layers were converted to shape-file format. In order to achieve a correct

overlay analysis all the input layers must have the same coordinate system. Therefore, with the use of ARCGIS9.2 projection tools, all maps and images were assigned the same coordinate system with the following parameters:

Projection: UTM

Units: Meters

Zone: 36N

Datum: D_WGS_1984

Spheroid: WGS_1984

4.2 Model layers For sufficient accuracy of the transport density flow model, 14 layers with distinctive properties were used to model the aquifer. The hydraulic properties of the model come from the analyzed pumping test and calibration for SEAWAT2000 model density driven flow and solute transport. The specific yield also comes from the pumping test. The hydraulic conductivity obtained from step drawdown tests but with small modifications to match the measured drawdown and salinity as shown in Table 4-1 below. Aquifer thickness is assumed 150 m. The horizontal hydraulic conductivity was set to the pumping test value (3 m/day) in the first run but due to the difference between observation and calculated value, it had to be changed to the values in Table 4-1. Each of these material types assigned have a degree of uncertainty. Figure 4-3 shows an west-east cross section. Layer thicknesses are shown in this figure as well as different hydraulic conductivity.

Table 4-1, Layer thickness, hydraulic conductivity, and specific storage used in the model.

layer Thick-ness (m)

hydraulic conductivity

(m/day) porosity

Specific storage × 10-4 (1/m)

1 16 10,1.1,3 0.14, 0.32,0.21

1.33, 24.1,

3

2 12 1.5, 1.1 0.1 0.0107

layer Thick-ness (m)

hydraulic conductivity

(m/day) porosity

Specific storage × 10-4 (1/m)

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 7

3 12 0.5, 1.1 0.1 0.0107

4 12 1.48, 1.1 0.1 0.0107

5 12 1.43 0.14 24.1

6 9 1.58 0.14 24.1

7 9 1.5 0.14 24.1

8 9 1.6 0.14 24.1

9 9 4.5, 1.8 0.25,0.21 1.33, 24.1

10 9 4.5, 1.7 0.25,0.21 1.33, 24.1

11 12 4.5, 1.8 0.25,0.21 1.33, 24.1

12 9 4.5, 1.9 0.25,0.21 1.33, 24.1

13 9 4.5, 2 0.25,0.21 1.33, 24.1

14 10 4.5, 2 0.25,0.21 1.33, 24.1

Figure 4-3, Finite-difference vertical resolution

4.3 Spatial discretization The model uses 14 layers to simulate the flow and solute transport in the model area (Table 4-1, and Figure 4-4). The model thickness extends as deep as 100 m below sea level. With the exception of model layers 1, and 2, which have some thicker parts locally, model layer thickness generally decreases with depth. This allows greater resolution near the bottom of the transport model, where movement of brine and salt-water intrusion are dominant.

8 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

Figure 4-4, model grid

4.3.1 Discretizing time Nevertheless, the model makes several trial runs with different times steps. It is impractical to use minor steps e.g. 0.1 days. Changes were made to allow the processes to interact and to implement a logical time step mechanism between flow and transport. Time step lengths are restricted by stability criteria that are necessary for accurate transport solutions. Finally, the time step was chose equally to 3 days with the multiplier 1.2. It makes the model more realistic.

4.4 Boundary conditions The east boundary of the study area is Red Sea; the west boundary is Jebel (Mountain or hill) Ruweisat EL Nimr, the North border Jebel Safra Al Elat, and the south border is Jebel El Safra.

1. East (Red sea shoreline): constant zero head boundary (sea shore), concentration 45800 mg/L.

2. West (Jebel Ruweisat El Nimr): no-flow boundary (physical boundary represents impermeable boundary rock. No flow boundary when flux across the boundary is zero.

3. North (Jebel Safra Al Elat): no-flow boundary

4. South (Red sea coastline and Jebel El Safra): constant head boundary (seashore), concentration 45800 mg/l, specified head equally zero.

5. The part southwest was specified as no-flow boundary.

6. No-flow boundary condition was specified in the first two layers where there are east and south cover cliffs.

An overview over the boundary conditions is given in Figure 4-5, shows Boundary conditions for flow model.

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 9

Figure 4-5, shows Boundary conditions for flow model (from layer 3 to 14)

4.5 Source and sinks Both injection and pumping wells are point source and sink respectively. The injection and pumping wells are represented as a point. Due to low rainfall and high evaporation rates, recharge to groundwater within the Umm Said Plateau is almost non-existent so rain recharge is around zero.

An overview over source and sinks is given in Figure 4-6.

Figure 4-6, shows source and sinks for model domain.

4.6 Initial conditions Initial conditions refer to the head distribution and concentration distribution everywhere in the model domain at the beginning of the simulation. It was standard choose the initial condition that generated after calibrated model in steady state. In this case, we assume the initial head is zero depended on records in 1996.SEAWAT-2000 model done with uniform initial head equal zero. In order to match with observed concentrations in 2005 and 2007, the initial concentration set to 30000 mg/L.

4.7 Numerical solvers and accuracy For this study area, the solution of the groundwater flow was done with preconditioned conjugate gradient solver (PCG2). The solver calculates iteration parameters internally. The Cholesky

10 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

method is an option in PCG2. The Cholesky method was used as precondition method. The head change criterion was to 0.001 m.

4.7.1 Sensitivity Analysis The model sensitivity analysis consisted of the determination of the effects of varying the following parameters with respect to the simulated water table elevations. Most of the parameters estimated during model calibration were related to hydraulic conductivity. The model started using all varying hydraulic conductivity, vertical hydraulic conductivity, and specific storage in different zones. The results show that the horizontal hydraulic conductivity was significantly affected. Therefore, next run sensitivity analysis started by selecting Kx as the parameter to vary. The number of simulations was four. The hydraulic conductivity was multiplied by values 0.1, 0.5, 1, 2, 2.25.

Figure 4-7, results of transient-state sensitivity analysis

From the sensitivity analysis (Figure 4-7) the optimal Kx and Ky value multiplier to be 2.25, which reduced the sum squared values from 13200 to 79. Table 4-1shows the parameters used best results were obtained a 14-layer model.

4.7.2 Flow model results The computed iso-piezometric map of end 2007 shows inland with an average gradient of 1 to 2 % (Figure 4-9). Potentiometric gradient is almost a flat of the model area except extraction area (Figure 4-8).

Figure 4-8, simulated potentiometer surfaces in end 1996 (layer 10).

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 11

Figure 4-9, Simulated potentiometer surfaces in end 2007(layer 10)

4.7.2.1 Temporal variations of heads in extraction well and injection wells

Figure 4-10, shows the computed head in well 6 over the period 1996-2007. The head becomes steady in the first five years. The head decrease further only due to the increased pumping rate. Form this figure we can conclude for that drawdown is sensitive for the extraction rate.

Figure 4-10, Calculated equivalent head for well 6 versus time

Figure 4-11shows computed water level for one of the brine injection wells. It shows a large water level rise and so a high rich overflowing. Figure 4-12 shows that computed heads in 2007 in an east west cross section through the wells. The west boundary is a NO-flow (Mountain) considered as No-flow boundary. It is reflecting the drawdown that occurs approximately 10 m below sea level. The east boundary is Sharm El Sheikh Coastline that was considered as constant head, with no drawdown. The maximum drawdown in the extraction areas.

Figure 4-11, Calculated equivalent head for injection well 1 versus time

12 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

water level vs. distance (West-East)

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

0 1000 2000 3000 4000 5000

Distance(m)

wat

er le

vel (

m)

layer2layer3layer4layer5layer6layer7layer8layer9layer10layer11layer12layer13layer14

Figure 4-12, water level versus distance in west-east direction (2007)

4.7.2.2 Flow budget

Flow budget calculated from the 1996 to end 2007. Figure 4-13 shows the changing injection rate with time, flux ramming in the aquifer with time, and flux discharge to the sea. Form figure above, we can see clear the volume of brine goes to the sea approximately equal the volume of water still in the aquifer. The brine water discharge to the sea start small and after 1998 go up until reached to constant rate but in 2003 started increased again. However, the discharge started going down in 2007. Obviously, the flux remained in the aquifer going up with time.

0

2000

4000

6000

8000

10000

12000

14000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

time(year)

flux(

m3/

day)

Distrage to sea(m3/day) Flux remain in aquifer(m3/day) Injection rate(m3/day) Figure 4-13, Time series for water balance

4.7.3 Transport model results The close relationship between advective transport and groundwater flow means that the factors considered for flow, the location and quantity of the inflow and outflow to the flow system, the hydraulic conductivity distribution, and the presence of pumping-injection wells determine to where brine migrate. The process of advection is so dominant that the mean velocity predicted by flow models can be used estimate patterns of brine transport with good accuracy. The transport model was run with longitudinal, transverse, and vertical dispersivity of 5, 0.5, and 0.5 m respectively to match the observation of total dissolved solids in the wells.

Results of run with SEAWAT2000 are model show the development of TDS plumes in the aquifer. The simulation results show that brine water does not yet reach the extraction wells yet but will likely do so in a couple of years (Figure 4-14).

Breakthrough curve

34.5

35

35.5

36

36.5

37

37.5

38

38.5

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

Time (days)

conc

entr

atio

n (g

/L)

well 15 w9 w19 w2 w15 w13 w14 w16w12 w10 w17 w6 w1

Figure 4-14, Breakthrough curves for TDS (g/L) most of extraction wells.

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 13

4.7.4 Particles tracking for pumping and injection wells Particle tracking is a form of transport modeling in which only the bulk movement of groundwater is investigated. Particle tracking neglects the effects of chemical reactions, dispersion, and diffusion. Particle tracking was done with the modelling code Modpath.

Figure 4-15, tracking the source of brine water

Figure 4-15 shows particles rolls from the injection to the extraction. It is indicating that the brine it will be in extraction location 18 years from starting injection.

Particle tracking around the coastline for El Hadab shows that seawater will reach the extraction intrudes after 54 years. Results also showed seawater from the south does not reach the wells within 100 years. We recommend this south location as new injection location Ras Umm Said (Figure 4-16).

Figure 4-16, tracking the source of sea water (southern-east, east)

4.7.4.1 Tracking the source of water in well 19

Modpath was also used to search for a possible source of the high iron concentration in well 19 assumed to be done to the presence of organic matter (Figure 4-17).

Figure 4-17, Illustration for reverse particle tracking for well 19

It is recommended to search for a source of organic pollution in the area indicated by the particle tracking from this well. It is further advised to not shutdown this well and treat the iron of this well separately.

14 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

5. Simulation of scenarios

After the model calibration, five scenarios were simulated to predict the future impact on groundwater system. Scenario A simulates constant pumping-injection rate for 100 years. In scenario B, the brine water injection was stopped (assuming injection into the sea) for 50 years to see effects on water level and TDS in the groundwater. Scenario C, deals with the pumping from new extraction wells that Ridgewoods plans in the future. Scenario D, deals with the pumping from the current location while the injection wells are relocated southward near the seashore. The model was run for 100 years to see the effects on flow and transport. Scenario E, represents the situation in which the pumping and injection take place from and to the sea to serve all Sharm El Sheikh and locate the project is outside of protection areas.

5.1 Scenario A: Business as usual Scenario A assuming no change in the pumping rates between 2007- 2096 in the model area. The response on the groundwater flow and transport is given in figures 1-5 to 5-3. Figure 5-2 shows constant head after 13 years (steady state).

Figure 5-1, simulated potentiometer surfaces in end 2096 (layer 10)

Figure 5-2, Simulated head for well 10 versus time (layer 10) from 1996 to 2096

The Figure 5-3 shows TDS breakthrough curve in all model layers.

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 15

Figure 5-3, TDS breakthrough curve (g/L) in well 17

Figure 5-3 shows that with continued pumping the extraction water will become almost brine water (TDS 53.9 g/L) in about 20 years from 1996, which is due to short circuit between injection and extraction wells. This is not sustainable, as the brine will become more saline due to the shortcut that makes desalination more and more expensive

5.2 Scenario B: Elimination of injection in El Hadab During 1996, Sinai Environmental Service Company started construction Raja Station. They sold the Raja desalination plant to Ridgewood Egypt Company. Scenario B was designed to simulate the effect of this reduction (injection well shutdown).

Results of scenario B from the SEAWAT2000 model are shown Figure 5-4.

Figure 5-4, Simulated TDS (g/L) concentration at layer 10 for scenario B

Figure 5-5, TDS concentration (g/L) versus time at well 17 for scenario B

The breakthrough at well 17 shows that mix of brine water with seawater reached the extraction from 7200 to 9000 days and it goes down until brine is replaced by seawater (Figure 5-5).

16 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

5.3 Scenario C: New extraction wells Scenario C concerns new extraction wells at the location investigated by Ridgewood for the purpose. Two wells were placed in the model at the target location, called the Towers area near the coastline. These wells extract 2400 m3/day together. The total injection in this location is 12600m3/d with salinity of 53.9 TDS g/L. The extraction injection wells were kept constant during the simulation run from 2007 to 2046. The results are given in Figure 5-6 shows the TDS concentration for 2046.

Figure 5-6, Simulated TDS (g/L) concentration in groundwater in 2046 (layer 10)

The Figure 5-7 shows TDS concentration versus time in well 17 (Scenario C).

Figure 5-7, Concentration versus time in well 17(breakthrough curve)

Brine water mixed with seawater arrives after 14.3 years. With continued pumping the extraction water will become almost brine water in about 25 years from start (Figure 5-7).

5.4 Scenario D: New injection wells In this scenario, new injection wells were drilled in the south part of the model area, where the particle tracking technique indicated that little intrusion occurs. The injection depth was set at 140 m. The pumping and injection were kept constant during this scenario D.

Figure 5-8 shows the TDS concentration after 100 years. The arrows show the injection location in the south.

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 17

Figure 5-8, Simulated TDS (g/L) concentration in groundwater after 100 year with new injection location

(layer14)

Figure 5-9 shows the TDS concentration (g/L) versus time (day) in well 17.

Figure 5-9, breakthrough curve for well 17

Figure 5-9 shows the TDS variation with time (day) for well 17 (breakthrough curve). The results from scenario D show initial increase due to brine already in the aquifer later on TDS becomes essentially equal to seawater. The figures show the TDS intrusion from the south increase with increasing depth. The brine water did not reach to the extraction in 100 years after the start (Figure 5-8, Figure 5-9).

Results of scenario D solute-transport show the development of a TDS plume in the aquifer from the south. This indicates that the proposed new injection wells field is a good location for injection in the long term. If demand goes up, brine may reach extraction wells, making even this design unsustainable. Ridgewood should look for sustainable solution

5.5 Scenario E: Pumping-injection from seawater Desalination along the Red Sea coast is a growing industry. Seawater desalination is the most sustainable source for potable water in the area. Seawater desalination faces problems with the disposal of its brine. If the brine is discharged into the sea, the density difference between brine and seawater induces the formation of a stratified system. Limited scientific papers deal with the issue. However, in scenario E, the intake are will be near the coast and injection will be 500 m away the coastline, where corals reefs are not affected disposal from desalination process(depending on sea currents).

Neodren is a technology developed by drilling experts based on experiences with dredge-less pipeline. A first step to fulfil this requirement is the use of directed drilled horizontal drains below the seabed. With the use horizontal drains the raw and fine screening, as well as conventional filtration of seawater will not necessarily due to the fact that the sand of the seabed will act as a natural pre-filter, separating all kind of solids and particles down to micron range from the

18 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

seawater to be fed to the desalination plant. This technology can be operated in sandy and karstic seabed as ecological and economical alternative for conventional open seawater intake systems and has even advantages in comparison with beach well installations (Peters, 2006). Neodren technology minimizes the environmental impact and is safe against the destroying forces of waves in the coastline. The technology is for desalination plants between 100 m3/day to 100000 m3/day. Neodren technology reduces of the costs of the infrastructure, logistics and operation.

Figure 5-10, seawater intake (Neodren technology)

Figure 5-11 provides a schematic of the outlet that may prevent impact brine disposal near of coastline beyond the coral reef live. The direction flow depends on currents.

Figure 5-11, schematic of the outlet

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 19

6. Manganese (Mn)

Manganese concentration was compared for 13 wells in 12/10/2007 and fourteen wells in 12/2005(Figure 6-1). Manganese is a new problem, not mentioned by Ridgewood but shown by the data. Manganese goes up in all wells. Manganese concentration essential was doubled from 2005 to 2007.

Figure 6-1, shows manganese concentration during10/2007, and 12/2005

20 / Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh

7. Conclusions

The problems (drawdown, salinity, Fe) were analyzed. Using information from geology and borehole data, the step drawdown tests, and brine transport into the area of El Hadab was modelled and simulated using SEAWAT-2000. The main conclusions and recommendations are as following:

• The model matches the available data (drawdown and salinity) quite well

• Steady state drawdown is reached in 1 year. Further drawdown must be due to increased pumping or clogging of wells

• The injected brine has not reached the wells yet but will probably do so in a couple of years. The salinity shows tendency of increase.

• Continued pumping is not sustainable due to short-cut of brine.

• Moving the injection wells to the south of the peninsula may postpone shortcut of brine for a century (not taking into account capacity increases in the future)

• Alternative extraction and especially discharge of brine into the sea beyond the reef may be unavoidable in the future to obtain a sustainable situation

• The salinity increased in abstracted wells due to salt water intrusion from sea which mix with brine water

• Probable impacts of brine injection on the desalination processes and environmental as follows:

- Brine water leaky to the sea that affects the coral reefs life (it is very important in valuable location).

- Increase the salinity in groundwater that makes desalinisation costly.

• A high Fe concentration was found in one well only. It is recommended to search for a source of organic pollution in the area indicated by the particle tracking from this well. It is further advised to not shutdown this well and treat the iron of this well separately.

• Manganese has doubled in almost all wells during the last 2 years. This is a new problem indicated by the data that deserves further study.

• Brine water may be injected through deep wells. However, design and model require more information of detailed from test drilling.

• Ridgewood plans to extract in the future near existing injection wells which will come short circuiting of brine water, making the desalination more expensive.

• It is recommended to carry out chemical or isotope analysis to increase the accuracy of transport model.

• It is recommended measure water level in running wells and the non-running wells to conclude about well clogging and real drawdown.

• The rining Mn concentration in all wells over the last two years in a point of concerns requiring attention from Ridgewood.

The current problem should be addressed to ensure sustainable drinking water service in Sharm El Sheikh.

Optimum Pumping-Injection System for Saline Groundwater Desalination in Sharm El Sheikh / 21

8. References

Ramadan .A. A.A, (2008). “Optimum pumping-injection system for saline groundwater desalination in Sharm El Sheikh”, MSc Thesis

Peters, T., D. Pinto, et al. (2006). “ improved seawater intake and pre-treatment system based on Neodren technology”

Birosoy, Y.K. and W.K. Summers ,(1980).” Determination of aquifer parameters from step tests and intermittent pumping data”, Groundwater.v.18.no.2

Omara, S. (1959). “ The Geology of Sharm El-Sheikh sandstone, Sinai, Egypt”. Egyptian Journal of Geology, Vol. 3, No. 1.

El-Baz, F. (1984). “The Geology of Egypt: An Annotated Bibliography”

Guo, W., and Langevin, C.D., 2002. “User’s guide to SEAWAT, a computer program to simulate variable-density groundwater flow”, US Geological Survey, Techniques of water resources investigation 6-A7.

The WaterMill Working Paper Seriesis part of the WaterMill project (Water Sector Capacity Building in Support of the Millennium Development Goals), which has been implemented by the UNESCO-IHE Institute for Water Education in Delft, the Netherlands, since 2004. The WaterMill Project is a capacity building project for the WaterSector responding to the targets as laid down in the Millenium Development Goals and by the Commission on Sustainable Development. The project offers several advanced training programmes at the post-graduate level to 72 professionals originating from the partner countries of the Netherlands. As part of their training each of the 72 professionals had undertaken a 6-month research project which focuses on the achievement of the MDGs in their home country. The WaterMill working paper series presents the research outputs of these projects.

Contact informationFor more information regarding the WaterMill project and the WaterMill Working Paper Series please contact Prof. Dr. Pieter van der Zaag E-mail: [email protected]

Water knowledge for the new millennium