Georesources -Integrated ground water resources study of the Sfax region

Proceedings of the 6th Tunisian-Japanese Seminar on Culture, Science and Technology 7-12 November 2005. Melia El Mouradi Palm Marina. El Kantaoui, Sousse, Tunisia www.ecopark.rnrt.tn/tjasstc.htm 227 Georesources - Integrated ground water resources study of the Sfax region, Tunisia M. Zairi , R. Trabelsi , I. Hentati , H. Smida , I. Karray and Hamed Ben Dhia Georesources Integrated ground water resources study of the Sfax region, Tunisia M. Zairi1, R. Trabelsi2, I. Hentati3, H. Smida4, I. Karray5 and Hamed Ben Dhia6 Abstract— In Tunisia and particularly in southern Tunisia, ground waters are the main water resources. They are principally used in agriculture development (80%), industries (10%) and drinking water supply (5 to 10%). Recently a lot of problems are facing the water management authorities. These problems are associated to water quality degradation and the increase of water demand due to life standard improvement and demographic and economic growth. Thus, the water resource management and particularly the groundwater one have to be optimised on quantitative and qualitative basis. The main objectives of 'The Integrated ground water resources study of the Sfax region in Tunisia' project are to identify the groundwater resources of Sfax region, to define their quantitative and qualitative characteristics, and to build up a rational scheme based on computerized tools for the use and management of these resources. The methodology developed to reach these objectives is based on a multidisciplinary approach using a variety of syntheses and explorations techniques. A lot of researches, studies, field investigations and laboratory analyses were implemented to complete the components of the project. The project aims to define the groundwater safe yield for the superficial and deep aquifers of Sfax region through the synthesis of the existing works and field and laboratory investigations. The study also intends to define the recharge areas to ensure their protection and improve the infiltration by water and soil conservation works, the setting up of a monitoring network for groundwater qualitative and quantitative evolution survey, the identification of the high vulnerable area to proceed to their protection against pollution risks, and the setting up of a procedure for groundwater data management and update. In the mathematical modelling, the provisional uses of groundwater simulations are constructed considering the related socioeconomic aspects. The confrontation of hydrogeological and geochemical data is used for the description of the salinity origin in coastal aquifers. The composition of the aquifer material and the intensive agriculture practices lead to various processes of salinization. Dissolution/ precipitation of minerals, ion exchange and mixing reactions are the main phenomena responsible for 1 Ecole Nationale d'Ingenieurs de Sfax (ENIS). B.P. W 3038 Sfax, Tunisia moncef.zairi@enis.rnu.tn 2 ENIS. Laboratoire Eau, Energie et Environnement (LEEE). B.P. W 3038 Sfax, Tunisia. trabrouaida@yahoo.fr 3 ENIS. LEEE. imenhentati@yahoo.fr 4 ENIS. LEEE. habib_smida@yahoo.fr. 5 ENIS. LEEE. 6 President of Sfax University. Route de l’aéroport Sfax Tunisia: hamed.bendhia@uss.rnu.tn groundwater quality decrease. The cartography of environmental vulnerability to the pollution of the aquifers was considered as an efficient tool to limit and to control the quality degradation risks of these waters. The Geographical Information Systems developed was combined to the DRASTIC method and used for the groundwater vulnerability survey. Indeed they facilitate the superposition of maps from different sources and the punctual data interpolation. Index Terms— Groundwater, resource, modeling, geographic information system, groundwater salinization, groundwater vulnerability. INTRODUCTION he hydrogeologic system of the Sfax region is composed of the deep confined aquifer of Sfax and 15 phreatic watertables defined relatively to their respective catchments limits. T The studies interested to the deep aquifer of Sfax present data on geology, structure of the reservoir, water head and groundwater chemistry [1, 2]. However, the hydrogeologic studies of the superficial aquifers are of preliminary type where it is in general about limiting the geographical extension of the aquifers and their main characteristics. Some indications on the climatology and the local geology are given. The state of the water head and salinity is often appreciated from results of a monitoring network limited to the exploitation area of the aquifers. The superficial aquifers are overexploited particularly in the coastal areas. The groundwater quality is found to be degraded due to the increase of aquifer exploitation by construction of new pumping wells, marine water intrusion in the coastal zones, and the irrigation water return [3]. The manual data archiving, synthesis and interpretation, the lack of a rigorous description of these aquifers induced a weak estimation of water resources in the region and hence the difficulties in establishing any management plan. The regional water resources management authority is facing a main problem of data gathering synthesis, interpretation and update. The safe yield of the superficial or deep aquifers is not well defined. The monitoring wells and piezometers have Proceedings of the 6th Tunisian-Japanese Seminar on Culture, Science and Technology 7-12 November 2005. Sousse, Tunisia 228 Georesources - Integrated ground water resources study of the Sfax region, Tunisia M. Zairi , R. Trabelsi , I. Hentati , H. Smida , I. Karray and Hamed Ben Dhia bad geographical distribution on the aquifers area. They are essentially located at the coastal area and consequently the aquifers hydrodynamic characteristics are not well defined over the aquifers extension. The knowledge of the deep and superficial aquifer systems behaviour is capital to ensure a sustainable management of the groundwater resources in the project zone. Due to the large geographic extension of the project area (15000 Km2) and the great quantity of data needed for aquifer characterization, particularly those concerned with exploitation wells, the use of computerised tools remains the best way to overcome these difficulties. Hence the construction of a hydrogeological database combined with a Geographic Information System was the first stage realised in the project execution. The utilisation of these innovative techniques permitted the best processing and interpretation of the various data gathered during the project. The non homogenous distribution of the hydrogeologic characteristics of aquifers and the lack of investigations in some area having a significant role in the definition of these aquifers limits required the completion of exploration works. These consisted of deep and shallow borings, aquifer tests, and geo-electrical profiles. INVESTIGATION WORKS A. Collection of existing documents The objective of this part is to define the existing hydrogeologic situation of the groundwater in the zone of intervention covering the Sfax region. We focused on all groundwater research results and studies at the public and private organisms in Tunisia. The collected information concerned the existing cartography, aerial view, geology, climatology, groundwater chemistry, hydrology, geophysics, drillings and piezometric fluctuations. B. Systematic inventory of groundwater wells in the region The inventory of water wells activities started in October 2001 and finished in May 2002. A total of 13326 points of water have been inventoried and localized on digitized topographic maps. All collected information has been reported on field cards then elaborated and stocked in the database. Field cards are classified relatively to their respective topographic map in files containing 200 cards each. The basic data collected during the inventory concerned the identification, characteristics and use of the water wells and the assessment of water quantity. The static level, the temperature of water, the electric conductivity and pH were measured in the wells. A ground water sample was also taken from wells with pumping equipment to laboratory for complete chemical analyses. C. Mapping of the geologic outcrops During the execution of this project component, 19 geological maps at the scale of 1:50000 indicating the geological formations limits and the main faults locations in the region were constructed on the basis of aerial photographs calibrated and verified by field outcrops. The constructed geologic maps were also digitised and stored in the geographic information system database. D. Chemical and isotopic investigations The quality of groundwater of the zone has been determined by conductivity, pH and temperature measurement during the wells inventory. The complete geochemical analysis of the major elements has been achieved on 3,000 samples of waters taken from pumped wells and distributed on all the zone of the project. The analyzed elements are the major elements: Ca++, Mg++, Na+, K+, HCO3-, SO4=, and Cl- and trace elements: Mn++, Fe++, Br-, and NO3-. The objective of isotopic investigation was to define the recharge rate and their distribution zones. They concerned the analyses of 19 samples for Tritium, 50 samples for Oxygen isotopes, 50 samples for Deuterium; and 16 dating with Carbone14. E. Re-interpretation of geophysical investigations The re-interpretation of previous geophysical investigations is related to the seismic profiles for the deep aquifer characterisation and the geoelectrical ones for the superficial aquifers reservoir definition. Thus, the seismic sections have been interpreted permitting the definition of the extension of the different reservoir layers and their structures. A total of 18 seismic lines and 9 composite oil well logs were used for this re-interpretation. They permitted the cartography of reservoir through construction of isochrones, depth and thickness contour maps. The second part concerned the reinterpretation of the existing geo-electrical investigations based on the new borings results. An investigation of 100 vertical electric polls was also realised during the present study. The results allowed the definition of the superficial aquifer geometry and the definition of the ground water zone affected by marine water intrusion. Proceedings of the 6th Tunisian-Japanese Seminar on Culture, Science and Technology 7-12 November 2005. Sousse, Tunisia 229 Georesources - Integrated ground water resources study of the Sfax region, Tunisia M. Zairi , R. Trabelsi , I. Hentati , H. Smida , I. Karray and Hamed Ben Dhia F. Execution of piezometers and well tests A total of seven deep piezometers were completed to a depth varying from 400 to 800m. Also 48 superficial piezometers with a depth from 50 to 120 m were bored. These piezometers served to define hydrogeologic and hydrodynamic characteristics of the deep and superficial aquifers . They also constituted the water level and quality monitoring network for groundwater evolution survey. G. Database and digital maps construction The database was constructed using the Microsoft Access software and permitted the storage of the collected information on wells from superficial and deep aquifers. The groundwater qualitative and quantitative historic data are also recorded in the database. A cartographic database was constructed by digitising topographic maps covering the project zone. We used a 1:50000 scale and we digitised topography, hydrographical network, road network, superficial aquifers limits, geological formations, faults and soil and water conservation works locations. H. Mathematical modelling The main objective of this activity is to represent the behaviour of the aquifer system of the Sfax region in order to plan and optimise its management according to the region water demands. The modelling of the deep and superficial aquifers of Sfax was done using the FEFLOW software [4]. This software allows the modelling of water flow and mass and heat transport in two or three dimensions and uses the finite element method for the resolution of the governing equations. After the models calibration for steady and non steady states, they were used to simulate the aquifers response to the exploitation increases. The provisional scenarios were constructed on the basis of real agricultural, industrial and potable water demand until the year 2030. The simulations served also to define the safe yield corresponding to the optimal aquifers exploitations. RESULTS AND DISCUSIONS A. The aquifer system of the Sfax region The Analysis and synthesis of the collected data and their comparison to the various investigation results allowed to define the aquifer system of the Sfax region. It is composed of the deep confined aquifer and the phreatic system of Sfax. The deep aquifer covers an approximate area of 14,000 km2 (including an off-shore part of about 3,500 km2). It is contained in the sand bodies of the middle Miocene with a varying thickness from the North to the south and from west to the East with a maximum recorded in the central part of the basin. It has a flow direction from NW to the SE. The deep waters salinity varies from 3 to 10 g.l-1 and they are of Na-Cl type. On the basis of the groundwater salinity distribution, two domains are identified. The first, where the salinity is as low as 4.5 g.l-1, covers the major part of the basin of the deep aquifer, and the second, with a salinity of 9 g.l-1, concerns the zone of Skhira mainly in the south of the sector (Figure 1). The Tritium analysis showed the absence of an actual refill of the aquifer. The C14 measured concentrations are weak and correspond to ages between 38,000 and 14,000 years BC. These ages confirm an old origin of the deep waters and the recent refill absence. The water pumped from the deep aquifer of Sfax is used for the irrigation, the industry or as a drinking water. The volumes pumped from the aquifer were about 10 millions m3 for the period of 1978 to 1986. Since 1987 this quantity increased to reach values of the order of 26 millions m3 in 2003. In the deep aquifer modelling, the initial condition for head distribution corresponds to the initial situation to the year 1970. The piezometric heads for 1988 have been considered for the calibration of the steady state. Simulations in transient state are done for the period from 1988 to 2002 taking account the historic set of the piezometric and exploitation monitoring. The provisional simulations permitted to test the aquifer capacity to satisfy water needs for industrial and agricultural sectors and the drinking waters. The deep aquifer model exploitation is based on seven scripts of exploitation and well localization. Assessments are done by the analysis of the head abatement during the time until the 2030 year. Before the present study the phreatic groundwater system of Sfax was composed of 15 watertables whose limits correlated to those of catchments. This number is reduced to a two watertables with a limits defined on the basis of their hydrogeologic characteristics. These watertables were called the coastal aquifer and the continental aquifer with regard to their respective main flow directions. The coastal aquifer is constituted by the 10 watertables of: 1. Bled Rgueb, 2. Bir Ali Ouadrane, 3. Sebkhat Mechguigue, 4. Skhira, 5. Chaffar, 6. Mahrés, 7. El Amra, 8. Sidi Salah, 9. Djebeniana, and 10 Sfax-Agareb with a total surface of 6,477 km2. The aquifer reservoir is composed of sand, silt and clayey sand multilayered with a frequent lateral and vertical variations. Its thickness varies from 8 to 60 m with an average of 30 m. The water flow is Proceedings of the 6th Tunisian-Japanese Seminar on Culture, Science and Technology 7-12 November 2005. Sousse, Tunisia 230 Georesources - Integrated ground water resources study of the Sfax region, Tunisia M. Zairi , R. Trabelsi , I. Hentati , H. Smida , I. Karray and Hamed Ben Dhia oriented from NW to SE in the sea direction and the water salinity varies from 0.3 to a maximum of 23 g/l in the parts affected by marine water intrusion at the sea shore. The balance between inflow and outflow showed that the watertable is overexploited with a deficit of 4 millions m3/year. The piezometric survey during a period of 10 years indicated a general decrease of heads of 0.3 m/year. For the model of the coastal watertable, the heads recorded in 1975 were considered for the steady state calibration. The wedging in transient regime is done for the period from 1975 to 2002. The simulations bring to the construction of a piezometric reference map for 2002 compatible with the imposed conditions for 1975. In the provisional simulations we tested the effect of the exploitation increase until 2030. The continental watertable corresponds to the 5 following aquifers: El Hencha, Meddelia, Bou Jeml, Hadj Gacem, and Bouthadi, with a total surface of 1838 Km2 (Figure 2). It is multilayer watertable lodged in sands and silts of the Quaternary and the sandy lentils inserted in clays and marls of the MioPliocene. The thickness of the aquifer formation is of 28 m with a minimum of 20 m and a maximum of 42 m. It is currently underexploited with 6.5 millions m3/year positive balance. The groundwater salinity varies from 4 /l to more than 10 g.l-1. We note that during the deep borings execution a medium aquifer was encountered in three wells at a depth of 150 m with salinity between 2.5 and 5 g.l-1 and pumping flow between 2 and 10 l.s-1. B. The groundwater salinization Representative groundwater samples were selected from 68 pumping wells. The electrical conductivity, temperature and pH were measured in situ and the chemical analyses concerned Na, Ca, Mg, HCO3, Cl, SO4, K, Mn and NO3. The measured physical and chemical parameters showed large spatial variations. The electrical conductivity varied between 1748µS/cm and 18190µS/cm and the salinity between 0.7 and 13.6 g.l-1. The various Stiff diagram shapes indicate heterogeneous water chemistry [5]. Based on this representation, four groundwater quality groups are identified: group I, accounting for 41% of the samples, group II for 30%, group III for 21% and group IV for 7%. Group I is found mainly in the northern part of the zone, group II in the southern part and group IV in the coastal area. The expanded Durov diagram was used to identify processes and reaction paths such as mixing, ion exchange and dissolution affecting groundwater composition [6, 7]. The major part of the samples from group I are affected by aquifer material dissolution. A part of group I, II and III are close to conservative mixing waters. Most samples in group II and III are indicating that both mixing and ion exchange are responsible for their quality. A maximum increase in the salinity should produce water of group IV and five samples from group II and III are affected by mixing and reverse ion exchange affect their composition. The Na-Cl diagram indicates, for a part of group II samples, enrichment in Na relative to Cl concentrations. The high Na concentration is related to direct ion exchange reactions between groundwater and the clay particles of the aquifer material. Groups I and IV show Na deficiency with respect to the conservative mixing line due to reverse ions exchange and dissolution reactions [4]. A strong correlation between NO3 and Cl is observed: it is derived from the evaporation processes and fertilizers dissolution, the salinity is affected by irrigation return flow [9, 10]. The Mg/Ca ratio increases with the proportion of seawater introduced in the mixture [11]. The waters of group IV have the highest ratio proving a seawater origin. Group I waters present the weakest values of the ratio following dissolution of gypsum and calcite present in the reservoir. The SO4/Cl ratio decreases as seawater proportion in the mixture increases [12, 13]. The group IV presents the weakest ratio verifying a marine origin of the mixed waters. However, group I have a large chloride concentration and a high SO4/Cl ratio indicating a possible gypsum dissolution. C. Environmental vulnerability of groundwater In order to ensure the groundwater protection against their quality degradation, a qualitative concept was used to evaluate the pollution risk of these waters. The hydrogeology, physical, geographical and climatic characteristics of an aquifer are analysed to elaborate a vulnerability map representing the geographic distribution of the intensity of the risk of the aquifer contamination. Through the variety of existing techniques for the assessment of environmental vulnerability of aquifers, we used the DRASTIC method, in this study. DRASTIC stands for the different parameters considered in the development of the vulnerability map, i.e. Depth (depth of the water plan), Refill (net recharge), Aquifer (nature of the reservoir formation), Soil (type of soil), Topography, Impact of the unsaturated zone (type of layers of the unsaturated zone), Conductivity (permeability of the aquifer layer). The DRASTIC index is calculated from the sum of weighted scores corresponding to the individual parameters. For every parameter, we assign a score between 1 and 10 according to the degree of its impact on the pollution of the aquifer. These scores are weighted by factors between 1 and 5. Proceedings of the 6th Tunisian-Japanese Seminar on Culture, Science and Technology 7-12 November 2005. Sousse, Tunisia 231 Georesources - Integrated ground water resources study of the Sfax region, Tunisia M. Zairi , R. Trabelsi , I. Hentati , H. Smida , I. Karray and Hamed Ben Dhia Figure 1 Groundwater salinity of the Sfax deep aquifer (Symbols on the map indicate the well location and the relative measured salinity). (a) (b) Figure 2. (a) Presumed limits of the water tables of Sfax, determined by the limits of watersheds. (b) Limits of the two new water tables, the continental (green) and the coastal (purple), defined by hydrogeological characteristics. The DRASTIC method was pied to three catchments in the region. The weak vulnerability of groundwater in the region is related to the fact that precipitations are very weak and therefore the reduction of the net recharge of the aquifer. The depth of the water plan is important (except along the coasts) induced an efficient natural protection of e groundwater. The zone of survey is also dominated by gypsum, halomorphic, hydromorphic soils limiting the infiltration. Expected from all the others parameters (soil, precipitation, depth of the water plan, permeability…), the topography, which is not important in the DRASTIC method, increases the vulnerability. In fact, the study zone presents a very weak slope favourable to the infiltration. Proceedings of the 6th Tunisian-Japanese Seminar on Culture, Science and Technology 7-12 November 2005. Sousse, Tunisia 232 Georesources - Integrated ground water resources study of the Sfax region, Tunisia M. Zairi , R. Trabelsi , I. Hentati , H. Smida , I. Karray and Hamed Ben Dhia CONCLUSONS The project permitted to give an assessment of the available groundwater in the deep and superficial aquifers. The comparison of this resource in relation to needs of the agricultural, industrial sectors and drinking water allowed to construct scenarios based on mathematical models for optimal exploitation of this resource. On the other hand, the setting up of the database and the geographical information system and the seizure of information concerning all wells of water in the region permit a better management of resources through the great capacities of treatment and visualization of data and the gain of time in their processing and interpretation. The scheduling of the sustainable utilization of resources is the basis of all economic, environmental and social success. From the economic point of view, durability results in the availability of the water resource for the current industrial and agricultural projects and their growth without compromising the quality and the quantity of groundwater resources. Thus, the provisional modellings allowed to test the different scripts of resource exploitation on the basis of social and economic developments projections until 2030. The environmental durability is ensured by the setting up of a development plan that respects the present works of soils and avoids the deterioration of the water quality by the intrusion of the sea waters in the inshore zone or again by the intensive utilization of chemical elements in the agricultural exploitations. The confrontation of hydrogeological and geochemical data is used for the description of the salinity origin in coastal aquifers. The composition of the aquifer material and the intensive agriculture practices lead to various processes of salinization. Dissolution/ precipitation of minerals, ion exchange and mixing reactions are the main phenomena responsible for groundwater quality decrease. These are enhanced by geochemical interaction of the sediment with water, by seawater intrusion and by seepage of irrigation water excess. Eventhough, the groundwater environmental vulnerability mapping using the DRASTIC techniques showed a weak average value, the established maps remain of great interest for groundwater quality protection and particularly a basic document for the land development planning. REFERENCES [1] Maliki My A. (2000), Etude Hydrogéologique, hydrochimique et isotopique de la nappe profonde de Sfax Thèse Doctorat. Fac .Sc.Tunis. 301p. [2] Amouri M. (1998), Etude hydrogéologique de la nappe profonde de Sfax. Rapport CRDA de Sfax. 42p. [3] R. Trabelsi, M. Zairi H. Smida et H. Ben Dhia Salinisation des nappes côtières : cas de la nappe Nord du Sahel de Sfax, Tunisie. Comptes Rendues Géoscience, 337 (2005), pp: 515-524. [4] Wasy, Feflow Finite Element subsurface Flow and Transport Simulation System. 2001. Berlin, Germany [5] H.A. Stiff, The interpretation of chemical water by means of patterns, J Petrol Technol 3 (1951) 15-17. [6] W.J. Lloyd, J.A. Heathcotte, Natural inorganic hydrochemistry in relation to groundwater, Clarendon Press, Oxford, 1985. [7] C.P. Petelas, I.B. Diamantis, Origin and distribution of saline groundwater in the upper Miocene aquifer system, coastal Rhodope area, Northern Greece, Hydrogeol J 7/3 (1999) 305-316. [8] A. El Achheb, J. Mania, J. Mudry, Processus de salinisation des eaux souterraines dans le bassin Sahel -Doukkala (Maroc occidental), First international conference on saltwater intrusion and coastal aquifers: monitoring, modeling, and management, Essaouira, Maroc, 2001. [9] A. Cardona, J.J. Carrillo-Rivera, R. Huizar-Alvarez, E. Garniel-Castro, Salinization in coastal aquifers of arid zones: an example from Santo Domingo, Beja California Sur, Mexico, Environmental Geology 45 (2004) 350-366. [10] B.C. Richter, C.W. Kreitler, Geochemical techniques for identifying sources of groundwater salinization, CRC Press, Boca Raton, 1993. [11] A. Vengosh, E. Rosenthal, Saline groundwater in Israel: its bearing on the water crisis in the country, J. Hydrol 156 (1994) 389-430. [12] P. Pulido-Leboeuf, A. Pulido-Bosch, M.L. Calvache, A. Vallegos, J.M. Andreu, Strontium, SO42-/Cl- and Mg2+/Ca2+ ratios as tracers for the evolution of seawater into coastal aquifers: the example of Castell de Ferro Aquifer (SE Spain), C.R. Geoscience 335 (2003) 10391048. [13] J.H. Tellam, J.W. Llyod, Problems in the recognition of seawater intrusion by chemical means: an example of apparent equivalence, Q.J. Eng. Geol 19 (1986) 389-398. Proceedings of the 6th Tunisian-Japanese Seminar on Culture, Science and Technology 7-12 November 2005. Sousse, Tunisia