Nabeul Unconfined Aquifer, North-Eastern Tunisia

Environ Earth Sci DOI 10.1007/s12665-010-0616-1

ORIGINAL ARTICLE

Investigation of groundwater mineralization in the Hammamet– Nabeul unconﬁned aquifer, north-eastern Tunisia: geochemical and isotopic approach

Amor Ben Moussa • Sarra Bel Haj Salem • Kamel Zouari • Vincent Marc • Fayc¸al Jlassi

Received: 17 July 2009 / Accepted: 12 June 2010 Ó Springer-Verlag 2010

Abstract Detailed hydrogeochemical and isotopic data of Keywords Unconfined aquifer Dissolution groundwaters from the Hammamet–Nabeul unconfined Dedolomitization Cation exchange Return flow aquifer are used to provide a better understanding of the Recent recharge natural and anthropogenic processes that control the groundwater mineralization as well as the sources of different groundwater bodies. It has been demonstrated that Introduction groundwaters, which show Na–Cl and Ca–SO4–Cl water facies, are mainly influenced by the dissolution of evapo- In most arid and semi-arid regions in the world, the avail- rates, the dedolomitization and the cation-exchange pro- ability of sufficient freshwater has become a limiting factor cess; and supplementary by anthropogenic process in for development. In North African regions, where water relation with return flow of irrigation waters. The isotopic scarcity was always a dominant problem, the interference signatures permit to classify the studied groundwaters into with the natural hydrologic cycle as a result of groundwater two different groups. Non-evaporated groundwaters that resources overexploitation and changes in land usage have are characterized by depleted d18O and d2H contents provoked not only the reduction of the available water but highlighting the importance of modern recharge at higher also the deterioration of the water quality. In Hammamet– altitude. Evaporated groundwaters with enriched contents Nabeul basin, north-eastern Tunisia, the unconfined aquifer reflecting the significance infiltration of return flow irri- groundwaters represent the only way to overcome the water gation waters. Tritium data in the studied groundwaters scarcity during dry years. Fast population growth, and lend support to the existence of pre-1950 and post-1960 expanding agricultural and industrial areas has resulted in a recharge. Carbon-14 activities in shallow wells that provide rapid increase in water demand from the shallow aquifer. The evidence to the large contamination by organic 14C cor- long-term withdrawals has engendered several deleterious roborate the recent origin of the groundwaters in the study problems such as water-level decline, salinisation and pol- area. lution of groundwater resources. Within this framework the present investigation, which uses a set of geochemical and isotopic tracers, aims to provide relevant information con- A. Ben Moussa (&) S. Bel Haj Salem K. Zouari cerning the sources of different groundwater bodies as well Ecole Nationale d’Ingeńieurs de Sfax, B. P. ‘‘W’’, as the natural and anthropogenic processes that control their 3038 Sfax, Tunisia mineralization, which will contribute to the sustainable e-mail: [email protected] management of groundwater resources in the study area. V. Marc Laboratoires d’Hydrogeólogie d’Avignon, Avignon, France Physiographic setting

F. Jlassi Commissariat Re´gional au De´veloppement Agricole, The Hammamet–Nabeul basin, which occupies an area Nabeul, Tunisia extent of approximately 450 km2, belongs to the Cap Bon 123 Environ Earth Sci peninsula, north-eastern Tunisia (Fig. 1). It is bordered in sinensisn L., Vitis vinifera L., Olea europea L., Ficus the east by the oriental coastal plain, the south by the Gulf carica L., Fragaria vesca L., Lactuca sativa capitata L., of Hammamet, the west by the Bouficha basin and the Solanum tuberosum L., Solanum lycopersicum L., and north by the Grombalia basin (Fig. 1). Ground elevation in Allium cepa L. in the cultivated areas, a few scattered the Hammamet–Nabeul district ranges from about 10 m halophytes such as Zygophyllum album L., Halocenem amsl near the Gulf of Hammamet to 500 m amsl in the Ed strobilaceum L. near the Gulf of Hammamet, and finally Darbouka Mountain in the north-western limit of the basin. annual terrestrial mesophytes such as Romarinus officina- The climate is semi-arid with a mean annual rainfall lus L., Thymus vulgaris L., Palargonium graveolens L., on ranging from about 500 mm at the hill zones to less than the mountain slopes and seldom at the highest elevations 200 mm in the valley zones. The potential evapotranspi- (Chaieb and Boukhriss 1998). ration exceeds 1,350 mm/year. The mean annual temperature is 20°C with maximum value of up to 31°Cin summer. The surface drainage network, which is very Geology and hydrogeology dense, is constituted by several non-perennial Wadis. It collects surface runoff from the surrounding hills toward The Hammamet–Nabeul basin is located in the western the Gulf of Hammamet. Pediments, in the study district, are coast of the Sicilian Strait (Elmejdoub and Jedoui 2009). mainly composed of poorly evolved alluvial soils with a From geologic point of view, this area shows sedimentary thickness of more than 2 m. These alluvial soils undergo units ranging from the Early Eocene to the Quaternary salinisation processes downstream and along the costal (Fig. 2). The Eocene series that begin with a thick Ypresian plains. However, topomorphic and lithomorphic vertisols limestone unit are overlaid by the Lutetian–Priabonian are present in inland plains and are characterized by the marl of the Souar Formation, which locally outcrops in the presence of swelling clays of dark colour. These organic Ed Darbouka Mountain (Ben Salem 1995). On the Eocene matter-rich soils show cracking during the dry season and sequences reposes the Oligocene coarse- to medium- are generally suitable for horticultural and orchard culti- grained sandstone belonging to the Fortuna Formation vation. Some other parts of the study area are marked by (Burollet 1956; Blondel 1991). This latter largely outcrops the abundance of Mediterranean red soils, which have fine along the El Manchar Mountain in the western part of the texture and a well-developed polyhedral structure and basin. In the eastern part of the basin and in some restricted calcareous accumulations (Elmejdoub and Jedoui 2009). areas along the foot of the surrounding Mountains, the The landscapes and topography, as well as the habitats and Fortuna Formation is covered by the Miocene sandy clay biota, all contribute towards the richness and diversity of and clay series of the El Haria Formation. The Plio-Qua- floral cover in the study area. It changes from a mixture of ternary Formations of the clays of Potters, sands of Nabeul, vegetables, legumes and fruit-bearing trees such as Citrus clays of Sidi Barka, sandstones of Hammamet and sands of the Rejich, which overlaid the El Haria Formation, outcrop largely in the central and south-western parts of the study area (Schoeller 1939; Colleuil 1976; Ben Salem 1995). From tectonic point of view, the major structures rec- ognized in the study area are mainly resulted from the relative movement of the African and the Eurasian plates (Mzali and Zouari 2006). Throughout the Late Pliocene– Early Quaternary period, the region was exposed to complicated tectonic events. Indeed, the Pliocene–Early Quaternary tectonic activity has resulted in NE–SW folds structures, which are associated with East–West to N120 dextral reverse faults and NE–SW faults with a reverse component (Mzali and Zouari 2006; Mzali et al. 2007). Hydrogeologically, the Hammamet–Nabeul unconfined aquifer is hosted in the Plio-Quaternary detrital deposits, which vary from 20 to 300 m in thickness (Fig. 3). It consists mainly of varying proportions of sands, sandstones, clay, evaporates with dominant pedogenic calcite nodules. This unconfined aquifer represents the largest water producer likely due to its high hydraulic conductivity. Fig. 1 Location map It is tapped by high number private and state owned wells, 123 Environ Earth Sci

this groundwater reservoir from the underling conﬁned aquifer of the Oligocene. The piezometric map of the shallow water table shows that groundwaters converge northwest–southeast from the north-eastern high lands toward the Gulf of Hammamet, which constitutes the natural discharge area (Fig. 4). This highlights the recharge of the unconﬁned aquifer in the foot of the mountains, limiting the basin in its northern part.

Sampling and analytical procedure

The sampling campaign was undertaken in the study area during March 2007. A total of 56 shallow wells were sampled for geochemical and isotopic analyses (Fig. 5). Water temperature, pH and electrical conductivity (EC) were measured in the ﬁeld. Chemical analysis of the water samples were carried out in the ‘‘Laboratoire de Radio- Analyses et Environnement’’ of the School of Engineers of Sfax, Tunisia. Major cations (Ca, Mg, Na, and K) and

anions (Cl, SO4 and NO3) concentrations were analysed in ﬁltered samples using a Dionex DX 100 ion chromatograph equipped with a CS12 and an AS14A-SC Ion Pac columns

and an AS-40 auto-sampler. The total alkalinity (as HCO3) was determined by titration with 0.01 or 0.1 HCl against methyl orange and bromcresol green indicators. Strontium ion was analysed using an inductively coupled plasma- atomic emission spectrometer (ICP-AES), Liberty 200AX- Fig. 2 Simplified lithostratigraphic column of the Hammamet– Nabeul basin Varian on samples that had been filtered through 0.45-Am filters and acidified to pH 2 using 16 N pure HNO3. Hydrogen and oxygen isotope analyses were performed in located in the costal and the central part of the basin. The the laboratory of the International Agency of Atomic bedrock of the Hammamet–Nabeul unconfined aquifer is Energy (IAEA) in Vienna, by employing, respectively the constituted by the Miocene clayey deposits that separates standard CO2 equilibration (Epstein and Meyada 1953) and Fig. 3 Hydrogeological cross section in the Hammamet– Nabeul unconfined aquifer

123 Environ Earth Sci

Fig. 4 Piezometric map of the Hammamet–Nabeul unconﬁned aquifer

Fig. 5 Sampling map

123 Environ Earth Sci the zinc reduction techniques (Coleman et al. 1982), fol- reﬂects the dominance of sodium and chloride; and the lowed by analysis on a mass spectrometer. Oxygen and inﬂuence of land use activities on groundwater minerali- hydrogen isotopes analyses were reported to d notation zation. Water type II (Ca–SO4–Cl) is dominated by cal- relative to Vienna-Standard Mean Oceanic Water cium and sulphate ions with Cl- concentrations becoming

(VSMOW), where d = [(RS/RSMOW) - 1] 9 1,000; RS increasingly important according to the groundwater flow represents either the 18O/16O or the 2H/1H ratio of the path. This water type may represent a mixture of calcium 18 16 2 1 sample, and RSMOW is O/ O or the H/ H ratio of the sulphate water and sodium chloride water. SMOW. Typical precisions are ±0.1 and ±1.0% for oxygen-18 and deuterium, respectively. Tritium (3H) Origin of mineralization analyses were performed in the laboratory of the IAEA by electrolytic enrichment and liquid scintillation counting To recognize the elements contributing toP the groundwater method (Thatcher et al. 1977). Tritium contents were mineralization, plots of the sum anionsP ( anions) versus reported in Tritium Unit (TU), in which one TU equals one each anion and the sum of cations ( cations)P versus each 18 tritium atom per 10 hydrogen atoms. Carbon-14 analyses cation were established. The plots of theP cations versus were carried out in the Sfax School of Engineers by using Na, Ca and Mg on one hand and the anions versus Cl liquid scintillation counting. Radiocarbon data were and SO4 on the other hand display well-defined correla- expressed as percent modern carbon (pmC) with an ana- tions, which may indicate that the mentioned ions con- lytical uncertainty of 0.3 pmC. tribute significantly to the groundwater salinization (Fig. 7). In order to precisely determine the origins of the referred ions and the processes that control their concen- Results and discussion trations in Hammamet–Nabeul groundwaters, several bivariate diagrams were completed. In the Na versus Cl In situ measurements interpretation diagram, the majority of samples displays a well-defined relationship (R2 = 0.9), suggesting the same origin of In situ parameters such as pH, temperature, electric con- sodium and chloride is likely related to the halite dissolu- ductivity (EC) and total dissolved solids (TDS) together tion (Fig. 8a). This dissolution is, however, verified, with analytical data of the major ions in groundwater through both the undersaturation state of the groundwaters samples are represented in Table 1. The groundwater pH with respect to the halite; and the parabolic proportional values range from 7.3 to 9.1 and the temperature varies evolution of the negative saturation indexes when corre- within a wide range of 14.6–22.5°C, indicating the com- lated to the sum of ions deriving from the NaCl eventual bination effects of numerous factors, i.e. the depth to dissolution (Fig. 8b). groundwater, the residence time in the flow system and/or On the other hand, the majority of groundwater samples the groundwater flow time from the recharge area. The EC have saturation indexes (SI) versus calcite and dolomite and the total dissolved solids (TDS) range from 0.73 to ranging from 0.2 to 1.5 (Table 1), indicating a saturation 10.12 mS/cm and from 0.43 to 7.32 g/l, respectively. state with respect to those mineral phases (Parkhurst et al. Higher values of these parameters characterize wells 1992). Therefore, the referred minerals are not likely to located in the southern parts of the basin, especially in the dissolve. However, all groundwater samples are undersat- vicinity of the Gulf of Hammamet, suggesting both the urated with respect to gypsum and anhydrite, indicating the insufficiency of recharge in these parts and the relatively eventual dissolution of these sulphate minerals. This dis- long-term water–rock interaction. solution is highlighted trough the 1:1 ratio of most samples

in the plot of Ca versus SO4 (Fig. 9a) and the proportional, parabolic, trend observed in the correlation of the satura- Major elements geochemistry tion indexes, with respect to the referred minerals, versus the sum of ions resulting from the eventual dissolution Water types (Kamel et al. 2005) (Fig. 9a, b). The increase in Ca2? concentration due to gypsum dissolution supersaturates the The Piper diagram (Piper 1944) was used in order to define groundwater with respect to calcite and causes its precip- the different water types in the Hammamet–Nabeul itation. Furthermore, as calcite precipitates the concentra- unconfined aquifer. Nitrate concentration was taking into tion of bicarbonate decreases in groundwaters, which account when plotting this diagram because of its relative causes the undersaturation with respect to dolomite and abundance in the groundwater (Fig. 6). The data plotted in enhances the incongruent dissolution of this mineral known Piper classification diagram display two chemically dif- as dedolomitization (Appelo and Postma 1993). This pro- ferent groundwater facies. Water type I (Na–Cl–NO3) cess that contributes to the increase of magnesium 123 Environ Earth Sci

Table 1 In situ measurements, geochemical and isotopic data of Hammamet–Nabeul shallow groundwater

Well number T (°C) pH EC TDS Na Mg Ca Cl NO3 SO4 HCO3 (lS/cm) (mg/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L)

1 20.2 7.7 2,570 1,940 11.7 4.5 8.2 12.5 3.0 3.4 4.0 2 19.8 7.6 1,827 1,370 5.0 5.0 5.6 4.9 2.0 4.2 4.6 3 20.1 7.6 1,795 1,350 6.0 3.2 6.1 6.5 1.1 3.1 4.2 4 18.5 7.8 8,190 6,660 41.3 26.7 24.3 47.6 0.5 44.6 10.2 5 19.2 7.3 2,780 2,160 12.5 7.6 9.5 12.4 2.1 7.1 6.2 6 20.7 8.12 3,040 2,520 15.1 6.8 11.0 16.5 3.3 10.4 3.0 7 19.8 7.32 3,620 2,480 13.1 8.3 9.0 17.1 5.6 4.3 4.4 8 18.8 8.03 3,800 2,840 18.0 10.9 14.9 18.2 2.2 16.4 4.6 9 19.4 7.9 2,220 1,760 8.0 4.7 9.0 9.7 1.3 6.8 3.0 10 21.2 7.7 3,810 2,680 15.0 9.0 8.6 13.3 3.1 10.5 6.4 11 20 7.8 1,800 1,210 7.7 4.0 6.0 6.0 0.2 4.4 7.0 12 19.6 7.9 1,339 1,020 6.6 3.6 4.5 7.5 0.1 2.0 4.0 13 19.7 7.8 5,130 3,820 24.4 12.2 19.0 23.7 0.9 26.7 5.3 14 16.2 8 4,830 3,790 22.9 16.9 17.9 19.1 2.7 24.8 6.8 15 20 9.16 733 430 2.2 2.1 2.0 1.9 0.0 1.2 3.2 16 20 7.4 4,860 3,580 27.1 12.3 17.3 24.5 4.3 19.2 7.8 17 20.8 7.8 4,410 3,150 22.1 10.5 15.3 22.1 2.9 17.3 5.0 18 19.4 7.6 6,970 5,240 41.5 19.4 30.3 37.6 6.9 40.8 5.0 19 19.7 7.5 7,930 6,720 47.5 22.6 29.4 48.2 2.9 42.6 6.2 20 21.3 7.9 1,336 970 6.3 2.5 4.0 4.3 1.3 2.9 3.4 21 20.6 7.7 886 660 2.8 2.5 4.2 2.2 0.1 2.7 4.3 22 22.5 7.8 1,069 700 3.2 2.7 2.8 2.2 0.0 2.5 4.7 23 22.6 7.2 4,730 3,170 25.8 11.3 13.0 22.5 0.5 17.9 5.0 24 21.6 7.7 1,261 780 7.9 3.1 2.5 5.1 0.0 3.0 5.4 25 19 7.5 4,380 2,800 6.4 2.9 2.5 4.2 0.0 2.9 4.8 26 24.7 7.7 1,534 880 22.8 9.0 7.7 19.8 0.0 10.9 6.6 27 22.4 7.5 2,190 1,430 5.5 6.0 6.3 10.3 0.3 1.6 5.4 28 22 7.5 1,548 1,090 6.3 4.6 7.2 10.3 0.0 3.1 4.0 29 23.2 7 1,424 1,090 5.0 4.4 7.4 3.5 0.0 9.2 1.0 30 19 7.7 5,770 4,040 44.4 10.2 8.1 27.1 0.0 27.3 6.6 31 22.5 7.5 4,950 3,660 33.4 8.0 7.7 26.8 3.1 20.3 4.6 32 19.8 8 1,606 1,190 8.5 3.2 3.3 7.2 1.5 2.5 3.9 33 17.4 8.01 1,928 1,360 8.6 4.1 3.4 6.4 0.2 4.7 5.2 34 19.7 7.8 1,337 980 6.8 2.6 3.9 3.1 0.6 4.9 5.4 35 18.9 8 1,463 970 7.7 3.3 3.1 4.0 1.3 2.9 5.0 36 21.1 7.4 3,210 1,790 17.4 6.8 4.7 15.7 1.1 6.4 5.4 37 19 7.5 9,640 6,520 56.1 12.5 30.0 64.8 5.3 25.4 4.7 38 19.8 8.24 6,000 4,810 39.4 15.4 18.6 31.9 2.2 19.5 15.8 39 20.5 7.6 3,920 2,570 25.8 10.8 9.0 17.4 4.5 13.4 6.0 40 19.8 7.6 10,120 6,700 65.6 17.8 24.6 71.5 0.5 27.6 8.4 41 19.7 7.5 3,650 2,420 19.9 8.8 8.7 21.4 1.5 9.0 4.4 42 14.6 7.64 5,370 3,800 31.5 15.6 10.0 24.3 0.0 24.1 8.8 43 20.7 7.6 2,870 1,970 15.1 5.8 7.6 12.6 0.3 9.9 4.8 44 19.8 7.6 3,230 2,320 16.1 6.9 8.8 13.4 2.1 11.5 5.0 45 20.8 7.41 5,130 4,010 27.6 11.8 15.9 28.7 2.5 20.5 5.2 46 19.7 8.1 5,730 3,640 31.6 10.3 17.4 38.4 1.9 15.0 3.0 47 17.8 8.3 3,730 2,260 21.8 4.9 5.2 15.9 0.2 2.4 12.0 48 20.7 7.4 8,910 7,320 64.6 23.4 31.5 54.3 0.0 53.7 5.4

123 Environ Earth Sci

Table 1 continued

Well number T (°C) pH EC TDS Na Mg Ca Cl NO3 SO4 HCO3 (lS/cm) (mg/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L)

49 21 7.9 4,440 4,140 34.2 11.3 15.0 30.3 3.4 20.9 6.4 50 18.5 7.71 3,420 2,610 18.6 5.3 9.7 15.1 1.7 7.3 5.2 51 20.5 7.5 3,800 2,870 22.6 6.8 11.6 16.5 3.0 15.7 5.4 52 18 8.18 9,320 6,180 66.6 16.4 14.8 64.2 0.0 24.6 7.4 53 18.4 8.4 4,990 3,350 31.9 12.5 9.1 25.7 3.9 18.4 5.4 54 21.7 7.3 2,160 1,830 24.7 5.2 4.7 20.1 0.0 8.1 5.2 55 22 7.8 5,560 3,270 9.5 1.9 1.3 5.5 0.0 2.7 4.9 56 22.7 7.9 1,671 1,020 9.5 3.0 3.0 10.2 0.0 9.6 5.3 Well number Saturation indexes 18O 2H l4C (pcm) 3H (UT) (% vs. SMOW) (% vs. SMOW) Anhydrite Calcite Dolomite Gypsum Halite

1 -1.5 0.6 1.0 -1.3 -5.5 -5.0 -27.7 84.6 ± 1.1 2.5 ± 0.4 2 -1.5 0.4 0.8 -1.3 -6.3 -4.8 -27.9 96.2 ± 1 1.7 ± 0.3 3 -1.6 0.4 0.7 -1.3 -6.1 -4.9 -29.5 83 ± 1.1 1.5 ± 0.4 4 -0.4 1.2 2.5 -0.2 -4.5 -5.1 -30.6 – – 5 -1.2 0.4 0.7 -1.0 -5.5 -4.5 -26.7 124.8 ± 1.2 3.5 ± 0.4 6 -1.0 0.9 1.7 -0.8 -5.3 -4.8 -27.5 – – 7 -1.5 0.3 0.6 -1.2 -5.4 -4.7 -27.5 – – 8 -0.8 1.1 2.1 -0.5 -5.2 -5.2 -27.7 – – 9 -1.2 0.7 1.1 -0.9 -5.8 -4.8 -24.6 – – 10 -1.1 0.7 1.6 0.9 -5.4 -4.8 -24.6 – – 11 -1.5 0.8 1.5 -1.2 -6.0 – – – 2.9 ± 0.4 12 -1.9 0.6 1.1 -1.6 -6.0 – – 89 ± 1.3 0 ± 0.4 13 -0.6 1.0 1.9 -0.3 -5.0 – – – – 14 -0.7 1.2 2.4 -0.4 -5.1 – – – – 15 -2.4 1.4 2.8 -2.1 -7.0 -2.3 -8.1 – – 16 -0.7 0.7 1.4 -0.5 -4.9 -4.4 -26.25 – – 17 -0.8 0.9 1.7 -0.5 -5.1 -4.4 -23.2 – – 18 -0.4 0.8 1.6 -0.1 -4.6 – – – – 19 -0.4 0.8 1.6 -0.1 -4.5 -4.1 -23.1 – – 20 -1.8 0.5 0.9 -1.5 -6.2 – – – – 21 -1.7 0.4 0.7 -1.5 -6.9 -5.05 -28.61 – – 22 -1.9 0.4 1.0 -1.7 -6.8 -5.2 -28.93 – – 23 -0.8 0.3 0.7 -0.6 -5.0 -5.32 -32.33 – – 24 -1.9 0.5 0.8 -1.7 -6.3 -5.25 -30.28 – – 25 -1.2 0.5 1.2 -0.9 -5.1 -5.21 -28.33 – – 26 -2.2 0.1 0.5 -2.0 -6.0 – – – – 27 -1.6 0.4 0.9 -1.3 -5.7 – – – – 28 -1.7 0.3 0.6 -1.5 -6.1 – – – – 29 -0.1 0.4 0.8 0.1 -4.5 -5.31 -32.16 – – 30 -0.9 0.6 1.3 -0.7 -4.7 -4.7 -23.5 117.6 ± 1.3 4.3 ± 0.4 31 -1.0 0.3 0.7 -0.7 -4.8 -4.2 -24.0 – – 32 -1.9 0.5 1.1 -1.7 -5.9 – – 111.7 ± 1.1 0.7 ± 0.4 33 -1.7 0.6 1.4 -1.4 -5.9 – – 115.4 ± 1.3 2.9 ± 0.4 34 -1.6 0.5 1.0 -1.3 -6.4 -4.6 -24.6 35 -1.9 0.7 1.4 -1.6 -6.2 -4.9 -30.8 73 ± 1.1 1.1 ± 0.3 36 -1.5 0.2 0.6 -1.3 -5.3 – – – – 37 -0.5 0.8 1.2 -0.3 -4.2 -3.6 -25.2 – –

123 Environ Earth Sci

Table 1 continued Well number Saturation indexes 18O 2H l4C (pcm) 3H (UT) (% vs. SMOW) (% vs. SMOW) Anhydrite Calcite Dolomite Gypsum Halite

38 -0.8 1.8 3.6 -0.5 -4.7 – – – – 39 -1.1 0.6 1.3 -0.8 -5.1 -4.8 -26.3 40 -0.6 1.0 2.0 -0.3 -4.1 – – – – 41 -1.2 0.4 0.8 -0.9 -5.1 -4.5 -25.5 – – 42 -0.9 0.7 1.6 -0.6 -4.9 – – 98 ± 1.4 3.5 ± 0.4 43 -1.2 0.5 0.9 -903.0 -5.4 -4.7 -24.5 – – 44 -1.1 0.5 1.0 -0.8 -5.4 -5.1 -27.0 – – 45 -0.7 0.5 1.0 -0.5 -4.9 -4.5 -27.2 – – 46 -0.8 1.1 2.1 -0.6 -4.7 -4.8 -24.5 – – 47 -1.9 1.4 2.8 -1.6 -5.2 -3.3 -20.0 – – 48 -0.3 0.7 1.4 -0.3 -4.3 -4.2 -24.3 – – 49 -0.8 1.1 2.1 -0.5 -4.8 -4.0 -21.9 – – 50 -1.2 0.8 1.3 -0.9 -5.3 -5.2 -30.0 – – 51 -0.9 0.5 1.0 -0.6 -5.2 – – – – 52 -0.8 1.3 2.7 -0.5 -4.2 -4.6 -27.1 106.4 ± 1.2 1.3 ± 0.4 53 -1.0 1.2 2.7 -0.7 -4.8 – – – – 54 -1.9 0.3 0.6 -1.6 -6.0 -5.22 -28.95 – – 55 -1.5 0.5 1.2 -1.3 -4.7 – – 56 -1.4 0.4 1.1 -1.2 -6.6 -5.23 -31.04 – –

concentration is represented by the following reaction Nitrate content (Mokrik and Petkevieìus 2002): þ In the Hammamet–Nabeul unconfined aquifer, about 75% CaMgðÞ CO3 ðÞþs CaSO4 2H2OsðÞþH 2 of samples have nitrate concentration that exceeds the ! CaCO ðÞþs Ca2þ þ Mg2þ þ SO2 þ HCO 3 4 3 drinking water standards of 50 mg/l (WHO 2006). The þ 2H O 2 average value of nitrate in the whole groundwater samples As the reaction proceeds in groundwaters, Ca/Mg is 120 mg/l. These high nitrate concentrations provide ratio decreases and sulphate concentration increases. On evidence for the significance of the return flow waters the plot of [Ca ? Mg] versus [SO4 ? 0.5HCO3], the contribution in the recharge of the unconfined aquifer. dedolomitisation reaction yields a straight line with a Indeed, ammonium nitrate, liquid fertilizer and other slope of 1 (Jacobson and Wasserburg 2005) (Fig. 10). commercial complex nitrogen fertilizers are used in large Some samples display a deficiency of Ca2? versus scale in the agricultural regions, where flood irrigation is 2- ? SO4 which is compensated by an excess of Na with applied. In these regions, NO3 contents are up to 300 mg/l - respect to Cl , suggesting supplementary modification by (Fig. 12). The excessive use of Ca(NO3)2 fertilizers is cation-exchange process. During this process Ca2? and verified through the well-defined relationship between 2? ? - 2? Mg in the waters are exchanged with Na previously NO3 and Ca (Stigter et al. 2006) (Fig. 13a). Similarly, adsorbed on the surface of clay minerals in the aquifer the well-defined relationships in the plots of NO3 versus matrix as shows the following equation (Guo and Wang SO4 and Mg versus SO4 suggest that N and S are used in 2005): the study area in the form of (NH4)2SO4 and MgSO4 fertilizers (Gi-Tak et al. 2004; Bohike et al. 2007) Na -clayðÞþ s Ca2þ þ Mg2þ 2 (Fig. 13b, c). ! Ca2þ þ Mg2þ -clayðÞþ s 2Naþ

The referred exchange is conﬁrmed through the plot of Isotopic study ? - 2? 2? 2- - (Na –Cl ) versus [(Ca ? Mg )–(SO4 ? HCO3 )] that exhibits an inverse proportional evolution with a The use of oxygen-18 and deuterium isotopes in hydro- slope of about -1 (Mc Lean et al. 2000; Garcia et al. 2001; geology offers information on the origin and movement of Dassi 2004; Carol et al. 2009) (Fig. 11). groundwater. It can offer an evaluation of physical 123 Environ Earth Sci

general estimate, however, can be derived from data and study results from nearby areas. Data from the nearest Global Network for Isotopes in Precipitation (GNIP) station number 7622500, located at Sfax city, were used to establish the Local Meteoric Water Line (RMWL), that follows the linear regression: d2H = 8d18O ? 13.5 (IAEA/ WMO 1999; Maliki et al. 2000), and the Regional Pre- cipitation Mean Value (RPMV) for d18O and d2H(-4.59 and -23.30%, respectively) (IAEA/WMO 1999). The d18O and d2H contents of the studied groundwater range from -5.66 to -2.27% and from -30.78 to -8.13%, respectively. In the standard diagram of d18O versus d2H (Fig. 14), the whole groundwater samples plot globally on and below the RMWL, suggesting that they are derived from Mediterranean origin rainfall. However, when examined in the detail, these groundwater samples can be divided into two groups. The ﬁrst group, includes 14 samples (about 40%) pre- senting a relatively wide range of d18O and d2H values, varying from -5.22 to -4.6% and from -28.95 to -23.5%, Fig. 6 Piper diagram of the shallow aquifer groundwaters respectively. These values deﬁne a regression line with processes that affect water masses, such as evaporation and the equation: d2H = 7.83 d18O ? 12.25 (R2 = 0.86). The mixing (Geyh 2000). Although, the major constraint in the slope of 7.83, which is very similar to that of the RMWL, use of these isotopes is the availability of long-term stable suggests that groundwaters of this group originate from isotope records of local rainfall that is fundamental for direct condensation from atmospheric moisture (Gat and understanding the relationship between isotopic composi- Issar 1974). Similarly, the intercept of 12.25, which is tions of groundwater and precipitation input function. A very comparable to the d-excess of the RMWL, lends

P P Fig. 7 Plots of anions versus each anion and cations versus each cation 123 Environ Earth Sci

Fig. 8 Plots of Na versus Cl (a) and (Na ? Cl) versus SI of halite (b)

2? 2- 2? 2- Fig. 9 Plots of Ca versus SO4 (a), (Ca ? SO4 ) versus SI of anhydrite (b) and (Ca ? SO4 ) versus SI of gypsum (c)

? ? - 2- - Fig. 10 Plot pf (Ca ? Mg) versus (SO4 ? 0.5HCO3) Fig. 11 Plot of (Na ? K –Cl ) versus [(SO4 ? HCO3 )– (Ca2? ? Mg2?)] support to the modern origin of these groundwaters aquifer has been recharged at altitudes higher than the (Edmunds et al. 2003). On the other hand, the isotopic elevation of the GNIP station (EGNIP = 10 m asl). Con- values of the group 1 water samples are slightly depleted sidering the d18O altitudinal gradient (G) of 0.3%/100 m, than the RPMV, by up to 0.63% for d18O and 5.65% for used by Blavoux (1978), Zuppi and Bortolami (1983), d2H, suggesting that the Hammamet–Nabeul unconﬁned Maliki et al. (2000) and Hamed et al. 2008, the estimation

123 Environ Earth Sci

Fig. 12 Spatial distribution of nitrate

2? - 2- - 2? 2- Fig. 13 Plots of Ca versus NO3 (a), SO4 versus NO3 (b) and Mg versus SO4 (c) of the unconfined aquifer recharge altitude (RA) accord- zones where flood irrigation is applied in large scale. In 18 ing to the equation: RA = EGNIP ? 100 9 (d ORPMV - these zones, the long-term practice of flood irrigation 18 d OGroundwater/G) provides values (between 0 and 220 m causes infiltration of water that largely fractionates in the asl) that coincide with the elevation of the aquifer out- ground surface and in the irrigation channels due to their crops in study area. long exposure to the atmosphere. The second groundwater group comprises 22 samples (about 60%) that plot below the RMWL with a regression Radiogenic isotope of water molecule (3H) line of d2H = 5.7d18O - 0.92 (R2 = 0.87). The low values of the slope and the intercept of this regression line Tritium is a radioactive isotope of hydrogen that has a half- provide insight to the evaporation of this groundwater life of 12.32 years (Lucas and Unterweger 2000). Most of samples form the group 2. This evaporation is likely in tritium that was present in the atmosphere prior to ther- relation with the return flow process relatively abundant is monuclear testing in the 1950s and 1960s was the result of

123 Environ Earth Sci

Fig. 15 Tritium contents in precipitation in Tunisian GNIP stations

which groundwater residence times extend to ca. 30 ka (Clark and Fritz 1997). The interpretation of radiocarbon data to obtain groundwater absolute ages is, however, largely complicated by the potential mixing with younger and Fig. 14 d18O/d2H diagram older sources of carbon (carbon-14 originating from widespread nuclear testing and from interaction of groundwater with carbonate mineral, respectively) (Clark natural production by the bombardment of nitrogen by and Fritz 1997; Edmunds and Smedley 2000). neutrons in cosmic radiation in the upper atmosphere All analysed samples show carbon-14 activities greater (Solomon 2000). The natural background activity of 3Hin than 70 pmC; and about 45% of these samples have precipitation and surface water prior to 1951 varied activities greater than 100 pmC. These relatively high between 2 and 10 TU (Von Buttlar and Libby 1955; activities lend support to the existence of an important Thatcher 1962; Roether 1967). However, between 1951 fraction of organic 14C, likely in relation with the infiltra- and 1980, tritium from anthropogenic atmospheric testing tion of return flow waters, which largely affects the initial of nuclear weapons overwhelmed the natural cosmogenic 14C contents. However, they corroborate the recent origin origin (Plummer 2005). As a result, tritium input from of the shallow groundwaters in the study area. precipitation is represented by a series of pulses, with the largest pulse occurring during 1963–1964. The spiked nature of the tritium input function results in considerable Conclusion ambiguity in age interpretation; nevertheless, the presence of tritium in water samples is a reliable indicator of sam- Major elements concentration, stable (d18O, d2H) and ples that contain at least a fraction of post-1950s water radiogenic (3H, 14C) isotopes are used to provide reliable (Plummer 2005). Nowadays, due to the radioactive decay, information about the processes that control the ground- groundwater derived from precipitation that fell before the water mineralization within the Hammamet–Nabeul onset of atmospheric testing of nuclear weapons would unconfined aquifer (Cap Bon peninsula, north-eastern have contained less than 0.75 TU. Hence, if we consider Tunisia). The geochemical exploration shows the domi- the data of the tritium contents in precipitation collected in nance of Na–Cl and Ca–SO4–Cl water types resulting from the Tunisian GNIP stations (Tunis-Carthage, no 6071500 the dissolution of halite and gypsum, the dedolomitization and Sfax city, no 7622500), the 3H contents in groundwater and the cation-exchange process. Additionally, the return from the study area, which vary from 0 to 6.1 TU flow process in relation with the long-term flood irrigation (Table 1), suggest two recharge periods (Fig. 15). Waters practice contributes to the mineralization by producing with 3H contents below 2 TU likely represent pre-nuclear high amounts of nitrate. The stable isotope signatures recharge, i.e. recharge prior to thermonuclear testing in the reveal the existence of two groundwater groups. The non- 1950s and 1960s. However, waters with 3H contents evaporated groundwaters with relatively depleted contents, between 2 and 6 TU originate either from post-nuclear reflecting recharge at higher altitudes; and, evaporated recharge, that infiltrated just before nuclear weapon tests, groundwaters with enriched contents highlighting the or from recharge occurring during the last two decades. influence of return flow of irrigation waters. Tritium contents in these two groups provide evidence to the presence Radiogenic isotope of DIC (14C) of pre-1950 and post-1960 recharge periods. Carbon-14 activities in shallow wells, in spite of their contamination The carbon-14 isotope has a half-life of about 5,730 years. by organic 14C, confirm the recent origin of the ground- Thus, it is used for the study of old hydrological systems in waters in the study area. 123 Environ Earth Sci

Acknowledgments The authors acknowledge the International Guo HM, Wang Y (2005) Geochemical characteristics of shallow Atomic Energy Agency (IAEA) for the financial support provided to groundwater in Datong Basin, northwestern China. J Geochem the Technical Cooperation project (TUN 8/019) within which was Explor 87(3):109–120 carried out this work. Hamed Y, Dassi L, Ahmadi R, Ben Dhia H (2008) Geochemical and isotopic study of the multilayer aquifer system in the Moulares- Redayef basin, southern Tunisia. J Hydrol Sci 53(5):1241–1252 References IAEA/WMO (1999) Global network for isotopes in precipitation, the GNIP database, Release 3. Available at http://www.iaea. Appelo CAJ, Postma D (1993) Geochemistry, groundwater and org/programs/ri/gnip/gnipmain.htm pollution. Balkema, Rotterdam Jacobson AD, Wasserburg GJ (2005) Anhydrite and the Sr isotope Ben Salem H (1995) E´ volution de la peńinsule du cap Bon (Tunisie evolution of groundwater in a carbonate aquifer. Chem Geol orientale) au cours du Neóge`ne. Notes Service Geólogique, 214:331–350 Tunisie, 61:73–84 Kamel S, Dassi L, Zouari K, Abidi B (2005) Geochemical and Blavoux B (1978) E´ tude du cycle de l’eau au moyen de l’oxyge`ne 18 isotopic investigation of the aquifer system in the Djerid– et du deute´rium, the`se d’E´ tat, Universite´ Paris-6, 316 pp Nefzaoua basin, southern Tunisia. Environ Geol 49:159–170 Blondel JAT (1991) Les se´ries a` tendance transgressive marine du Lucas LL, Unterweger MP (2000) Comprehensive review and critical Mioce`ne infe´rieur a` moyen de la Tunisie centrale (Trangressive evaluation of the half-life of tritium. J Res Natl Inst Stand series of the lower and middle Miocene in central Tunisia). PhD Technol 105(4):541–549 thesis, University of Genova, Italy Maliki MA, Krimissa M, Michelot JL, Zouari K (2000) Relation entre Bohike JK, Smith RL, Hannon JE (2007) Isotopic analysis of N and O nappes superficielles et aquife`re profond dans le bassin de Sfax, in nitrite and 383 nitrate by sequential selective bacterial Tunisie (Relation between the superficial aquifers and the deep reduction to N2O. Anal Chem 79(384):5888–5895 aquifer of the Sfax Basin, Tunisia). C R Acad Sci 331:1–6 Burollet PF (1956) Contribution a` l’e´tude stratigraphique de la Mc Lean W, Jankowski J, Lavitt N (2000) Groundwater quality and Tunisie centrale. Annale des mines et de la geólogie, 345 sustainability in alluvial aquifer, Australia. In: Sililoo O et al Carol E, Kruse E, Mas-Pla J (2009) Hydrochemical and isotopical (eds) Groundwater, past achievement and future challenges. evidence of 386 ground water salinization processes on the Balkema, Rotterdam, pp 567–573 coastal plain of Samboromboń Bay, 387 Argentina. J Hydrol Mokrik R, Petkevieìus R (2002) Baltijos baseino Ðventosios— 365:335–345 Arukiulos, franio ir apatinio karbono-famenio kompleksø paleo- Chaieb M, Boukhriss M (1998) Flore suscinte et illustreé des zones hidrogeologine¨ raida. Litosfera 6:48–58 arides et sahariennes de Tunisie. APNES, Tunis, p 29 Mzali H, Zouari H (2006) Caracte´risation geóme´trique et cine´matique Clark I, Fritz P (1997) Environmental isotopes in hydrogeology. des structures lieés aux phases compressives de l’E´ oce`ne au Lewis Publishers, Boca Raton, FL Quaternaire infe´rieur en Tunisie: exemple de la Tunisie nord- Coleman ML, Shepherd TJ, Durham JJ, Rouse JE, Moore GR (1982) orientale. C R Geosci 338:742–749 Reduction of water with zinc for hydrogen isotope analysis. Anal Mzali H, Gabtni H, Zouari H, Hadj Sassi M, Gharsalli J (2007) Chem 54:993–995 Evidence of N120 shear corridors and associated tectonic Colleuil B (1976) Etude stratigraphique et neótectonique des forma- structures in north-eastern Tunisia after geological and geophys- tions neóge`nes et quaternaires de la re´gion de Hammamet– ical data. C R Geosci 339:358–365 Nabeul (Cap Bon) D.E.S, Universite´ de Nice Parkhurst DL, Christenson S, Breit GN (1992) Groundwater-quality Dassi L (2004) Etude hydrogeólogique, geóchimique et isotopique du assessment of the central Oklahoma aquifer, Oklahoma: geo- syste`me aquife`re du bassin de Sbeı¨tla (Tunisie Centrale). The`se, chemical and geohydrologic investigations. US Geol Surv Open University Sfax, Tunisie, 180 pp File Rep, 92-642 Edmunds WM, Smedley PL (2000) Residence time indicators in Piper AM (1944) A graphic procedure in the geochemical interpre- groundwater: the East Midlands Triassic sandstone aquifer. Appl tation of water-analyses. Trans Am Geophys Union 25:914–923 Geochem 15:737–752 Plummer LN (2005) Dating of young groundwater. In: Aggarwal PK, Edmunds W, Guendouz A, Mamou A, Moulla A, Shand P, Zouari K Gat JR, Froehlich KFO (eds) Isotopes in the water cycle: past, (2003) Groundwater evolution in the Continental Intercalaire present and future of a developing science. IAEA, Vienna, aquifer of southern Algeria and Tunisia: trace element isotopic Springer, Netherlands, 381 indicators. Appl Geochem 18:805–822 Roether W (1967) Estimating the tritium input to groundwater from Elmejdoub N, Jedoui Y (2009) Pleistocene raised marine deposits of wine samples: groundwater and direct run-off contribution to the Cap Bon peninsula (N-E Tunisia): records of sea-level Central European surface waters. In: Isotopes in hydrology, highstands, climatic changes and coastal uplift. Geomorphology IAEA, Vienna, pp 73–79 112:179–189 Schoeller H (1939) Le Quaternaire du Golfe ancien de Grombalia, Epstein S, Meyada TK (1953) Variations of 18O content of waters Tunisie, pp 14–30 3 3 from natural sources. Geochim Cosmochim Acta 4:213–224 Solomon DK (2000) H and He, chapter 13. In: Cook PG, Herczeg Garcia MG, Del Hidalgo M, Blesa MA (2001) Geochemistry of AL (eds) Environmental tracers in subsurface hydrology. groundwater in the alluvial plain of Tucuman Province, Argen- Kluwer, Boston, 529, pp 397–424 tina. Hydrogeol J 9:597–610 Stigter TY, Carvalho AM, Ribeiro L, Reis E (2006) A impact of the Gat JR, Issar AS (1974) Desert isotope hydrology: water source of the shift from groundwater to surface water irrigation on aquifer Sinai Desert. Geochim Cosmochim Acta 38:117–1131 dynamics and hydrochemistry in a semi-arid region in the south Geyh MA (2000) An overview of 14C analysis in the study of of Portugal. Agric Water Manag 85:121–132 groundwater. Radiocarbon 42(1):99–114 Thatcher LL (1962) The distribution of tritium fallout in precipitation Gi-Tak C, Kangjoo K, Seong T, Kyoung H, Soon O, Byoung Y, over North America. International Association of Scientific Hyoung S, Chul W (2004) Hydrogeochemistry of alluvial Hydrology VII 2:48–58 groundwaters in an agricultural area: an implication for ground- Thatcher LL, Janzer VJ, Edwards RW (1977) Methods for determi- water contamination susceptibility. Chemosphere 55:369–378 nation radioactive substances in water and fluvial sediments.

123 Environ Earth Sci

In: Techniques of Water Resources Investigations. U.S. Geo- at http://www.who.int/water_sanitation_health/dwq/gdwq3rev/ logical Survey Washington, 79–81 en/index.html Von Buttlar H, Libby WF (1955) Natural distribution of cosmic-ray- Zuppi GM, Bortolami GC (1983) Hydrogeology: a privileged ﬁeld for produced tritium. J Inorg Nucl Chem 1:75–91 environmental stable isotope applications, some Italian exam- WHO (2006) World Health Organization. Guidelines for drinking ples. R Soc Ital Miner Petrog 383:1197–1212 water quality, 3rd edn, incorporating ﬁrst addendum. Available

123