Introduction

The intrusion of seawater into coastal aquifers is a common problem in coastal zones of the world where increasing water requirements (Masciopinto et al. 1999) and arid climate have induced overexploitation of groundwater (Fidelibus and Tulipano 1992; Kreitler and Richter 1993; Park and Aral 2004; Van Camp et al. 2014). Continued exploitation of limited freshwater resources can also lead to up-flow of relict marine waters underlying the freshwater zone compounding the effects of seawater intrusion (Masciopinto 2006).

Next to admixture of seawater, the main sources of SO4 2− in groundwater are oxidation of sulfide ores (pyrite FeS2), and the dissolution of gypsum and anhydrite (Walraevens 1987; Coetsiers and Walraevens 2006, 2009; Coetsiers et al. 2009). Infiltration of industrial wastes can also add significant amounts of sulfate to freshwaters. Natural concentrations of SO4 2− in freshwaters are usually less than 300 mg/l (Todd 1980).

The ever-growing demand for freshwater for a number of human purposes has become a worldwide cause of concern. Nowadays, groundwater reserves are exposed to intensive exploitation (Walraevens et al. 1994; Van Camp and Walraevens 2009; Van Camp et al. 2010, 2012, 2013), which may create serious problems in coastal areas where some hydraulic connection exists between the freshwater reservoirs and the sea (Masciopinto 2006; Mjemah et al. 2009; Van Camp et al. 2014; Walraevens et al. 2015).

Salinization is the most widespread form of groundwater contamination, especially in coastal aquifers, and is represented by the increase in total dissolved solids (TDS) and some specific chemical constituents such as Cl−, Na+, Mg2+ and SO4 2− (Nadler et al. 1981; Magaritz and Luzier 1985; Dixon and Chiwell 1992; Bencini and Pranzini 1992; Bosch and Custodio 1992; Walraevens et al. 1993a, b; Morell et al. 1996; Sukhija et al. 1996; Chaouni Alia et al. 1997; Giménez and Morell 1997; Walraevens et al. 2007; Mtoni et al. 2012; Da’as and Walraevens 2013; Mtoni et al. 2013). The Jifarah Plain groundwater is affected by different sources of salinization, most serious is the seawater intrusion (Alfarrah 2011; Alfarrah et al. 2011).

Degradation of groundwater quality is usually thought of as the result of direct contamination of groundwater by either point source or diffuse source release of pollutants. In this study, results are presented of an investigation of declining water quality in a heavily dewatered aquifer system; seawater intrusion and water–rock interaction are responsible for the considerable variations in chemical composition of groundwater and for making it non-potable.

The origin of SO4 2− may be traced through identification of hydrogeochemical processes. The aim of this study was to discover what processes have been responsible for variations in the chemical composition of groundwater in the upper aquifer of Jifarah Plain impacted by intense water withdrawal.

Study area

The study area covers the coastal part of the Jifarah Plain in NW of Libya (Fig. 1). The Jifarah Plain is a flat area of triangle shape of about 20,000 km2. It is bordered by the Mediterranean Sea in the north, the Tunisian border in the West and Jebal Naffusah border in the south and east. The study area is a coastal strip of around 105 km length and 18 km wide in the north of central Jifarah, where more than 50 % of the country’s population are concentrated. The climate of the region is semiarid, typically Mediterranean, with irregular annual rainfall; the average annual precipitation is around 250 mm. Jifarah coastal area, like all Mediterranean coastal regions, has experienced considerable periods of drought during the last two decades. The resulting water deficit was compensated through an increased withdrawal from aquifers.

Fig. 1
figure 1

Location and general geological map of the study area

Full size image

The sediments of the Jifarah Plain have been deposited since early Mesozoic times in a near-shore lagoonal environment. Figure 1 represents the location and the general geological map for central Jifarah with the location of the main active pumping wells. The geological substratum, which is playing a role in the hydrogeology of the plain, comprises the Middle Triassic (Al Aziziyah Formation, consisting mainly of bedded limestone) and the Upper Triassic (Abu Shaybah Formation, consisting of continental sandstone). All post-Triassic formations have been eroded prior to the Miocene sedimentation. As a consequence, the Miocene, that covers about two-thirds of the coastal plain, overlays the Abu Shaybah deposits directly. An exception is found in some small down-faulted blocks southwest of Al Aziziyah area where the Lower Cretaceous Kiklah Sandstone and the Upper Cretaceous Ayn Tobi Limestone have been found in some locations.

Geological and hydrogeological setting

The principal aquifer used by the population in the Jifarah Plain is the Upper Miocene–Pliocene–Quaternary aquifer system, called “first aquifer” or “shallow aquifer” or “upper aquifer”; intercalated thin clayey sand and marl series are dividing the aquifer into a number of horizons, all are considered as one unconfined unit (Alfarrah et al. 2013). It is separated from the lower aquifers by Middle Miocene clay.

Table 1 gives a description of the Upper Miocene–Pliocene–Quaternary formations which are of most interest in the coastal area.

Table 1 Description of the upper Miocene–Pliocene–Quaternary deposits in the coastal area of Jifarah Plain
Full size table

The upper aquifer is made up of a series of formations of uncertain age. Al Assah Formation is silt, sand and gravels with local occurrences of crystallized gypsum. The Pleistocene formations include terraces, which consist of cemented gravel and conglomerate. Qasr Al Haj Formation is mainly alluvial fans and cones consisting of clastic materials derived from the scarp. Jifarah Formation consists mainly of fine materials, mostly silt and sand, occasionally with gravel caliche bands and gypsum; it covers extensive parts of the Jifarah Plain. Gergaresh Formation, which is known as Gergaresh Sandstone of Tyrrhenian age, occasionally contains silt lenses and sandy limestone.

The Holocene deposits include recent wadi deposits; these deposits consist of loose gravels and loam. Beach sands are represented by a narrow strip at the coast and are made up of shell fragments with a small ratio of silica sands. Eolian deposits are represented by sand dunes and sheets covering large of the coastal strip (coastal dunes). These coastal dunes consist of shell fragments with small amounts of silica sands. It is worth mentioning that the eolian material composing coastal dunes contains a large amount of grains of gypsum. In some places, it is composed of nearly pure gypsum (98 %) especially in the immediate vicinity of the sebkhas, with a silty gypsum filling (LIRC 1995). Fluvial–eolian deposits are found on the plateau surface in central Jifarah. They are made up of silt, clay and fine sands with occasional caliche bands. Sebkha sediments are mainly gypsum deposits and observed along the coastal area of the plain. They occupy the relatively low topographic areas and are separated from the sea by sea cliffs. Some of the sebkhas have occasional incursions of the sea, and others may have subsurface connection with the seawater.

The depth to the bottom of the upper aquifer varies between 30 and 200 m and depths of the wells that are utilizing this aquifer are between 30 and 180 m. Most of the wells tapping this aquifer give productivity of about 20–80 m3/h.

Methodology

Sampling and analytical methods

A regional hydrogeochemical survey and water level measurements were performed during dry periods from March to July in 2007 and from September to November of 2008. In the coastal area, a total of 134 shallow and deep wells (mostly 30–180 m deep), located at different distances from the Mediterranean Sea, were selected for groundwater sampling and water level measurement (see Fig. 1), along profiles perpendicular to the coastline. The samples were collected during pumping, and the water level measurements were performed in static condition.

The sampling points were chosen along vertical lines perpendicular to the coast, with lengths comprised between 8 and 28 km, in order to explore the aquifer from inland to the coast line. Water depth was measured from the ground surface using water level meter and was converted into water level by subtracting from ground elevation. The collected water samples were preserved in polyethylene bottles after filtering with 0.45-µm cellulose membrane filters. Two samples were taken from each well, one for determining anions and the other for determining cations. Samples for cation analysis were acidified to lower the pH to around pH = 2 by adding a few drops of nitric acid. Parameters measured are physical properties such as pH, temperature, water level and electrical conductivity. Cations (Na+, K+, Mn2+, FeTotal, Ca2+, Mg2+, Zn, Si) were analyzed using flame atomic absorption spectrometry (Varian). Anions (Cl−, NO3 −, NO2 −, SO4 2−, PO4 3−) and NH4 + were analyzed using the molecular absorption spectrophotometer (Shimadzu). F− was measured with ion selective electrode. Determination of bicarbonate (HCO3 −) and carbonate (CO3 2−) used the titration method with dilute HCl acid to pH 4.3 and 8.2, respectively. The above-mentioned analytical methods were used at the Laboratory of Applied Geology and Hydrogeology, Ghent University, and were provided in the Laboratory Manual and in Standard Methods for Examination of Water and Wastewater (APHA 1985). Careful quality controls were undertaken for all samples to obtain a reliable analytical dataset with an ionic balance error less than 5 %.

Hydrochemical evaluation methods

The interpretation process is mainly based on the calculation of the ion deviations (Δmi) from conservative freshwater/seawater mixing, the calculation of the saturation indices (SI), graphical illustration methods including Piper diagram, calculation of ionic ratios, and elaboration of maps and cross sections showing the spatial and vertical distribution of water quality parameters in the study area.

Saturation indices

The PHREEQC 2.16 program was used to calculate saturation indices for calcite, dolomite, halite and gypsum based on the chemical analytical results and measured field temperatures for all samples.

Ion deviation from conservative freshwater/seawater mixing

Calculation of the ionic deltas Δmi consists of a comparison of the actual concentration of each constituent with its theoretical concentration for a freshwater/seawater mix calculated from the Cl− concentration of the sample (Fidelibus et al. 1993), because Cl− is dominant ion in seawater and can be assumed to be conservative in many natural waters (Appelo and Postma 1993). The ionic deltas quantify the extent of chemical reactions, affecting groundwater composition, next to mixing. The chemical reactions during freshwater/seawater displacement can be deduced by calculating a composition based on the conservative mixing of seawater and freshwater, and comparing the conservative concentrations with those in the samples. The mass fraction of seawater (f sea) in the groundwater can be obtained from chloride concentrations of seawater and freshwater as follows (Appelo and Postma 1993):

$$f{\text{sea}} = \frac{{m_{{{\text{Cl}}^{ - } , {\text{sample}}}} - m_{{{\text{Cl}}^{ - } ,{\text{fresh}}}} }}{{m_{{{\text{Cl}}^{ - } ,{\text{sea}}}} - m_{{{\text{Cl}}^{ - } ,{\text{fresh}}}} }}$$
(1)

where m −Cl ,sample = the concentration of Cl− in the sample expressed in mmol/l, m −Cl ,fresh = the concentration of Cl− in the freshwater expressed in mmol/l, m −Cl ,sea = Cl− concentration in the seawater end member in mmol/l (Cl− = 566 mmol/l for mean ocean water; for Mediterranean seawater (possible end member), m −Cl ,sea = 645 mmol/l).

Based on the conservative mixing of seawater and freshwater, the concentration of an ion i (m i ) in the mixed waters was calculated using the mass fraction of seawater f sea as follows (Appelo and Postma 1993):

$$m_{{i , {\text{mix}}}} = f_{\text{sea}} \cdot m_{{i,{\text{sea}}}} + (1 - f_{\text{sea}} ) \cdot m_{{i,{\text{fresh}}}}$$
(2)

where m i is concentration of an ion i in mmol/l and subscripts mix, sea and fresh indicate the conservative mixture, and end members seawater and freshwater, respectively. Any change in concentration m i,reaction (∆m i ) as a result of reactions (not mixing) then becomes:

$$\Delta m_{i} = m_{{i,{\text{reaction}}}} = m_{{i , {\text{sample}}}} - m_{{i , {\text{mix}}}}$$
(3)

where m i,sample = the actually observed concentration in the sample in mmol/l.

The deviation from the conservative freshwater/seawater mixing is due to chemical reactions. A positive delta means that the ion has been added to the water, e.g., due to desorption from the exchange complex. While adsorption will lead to negative delta.

Ions in infiltrating rainfall near the coast are often derived from sea spray, and only Ca2+ and HCO3 − are added due to calcite dissolution (Appelo and Postma, 1993). All other ions are thus ascribed to seawater admixture. In this case, m i,fresh = 0 for all components except Ca2+ and HCO3 −.

The main end members used in the calculations for this study are the Mediterranean seawater and freshwater from the upper aquifer. For Mediterranean seawater where Cl− = 645 mmol/l, the seawater fraction has been calculated as:

$$f_{\text{sea}} = \frac{{m_{{{\text{Cl}}^{ - } ,{\text{sample}}}} }}{645}$$
(4)

Table 2 shows the ion concentrations in standard seawater, Mediterranean seawater end member and the assumed freshwater end member in the Jifarah Plain. Recharge water in the plain is the water flowing to the aquifer from the high topographic recharge area in the south. As no data were collected from the south border of the plain, the groundwater in the recharge area is expected to have the same composition as the freshwater samples collected from a nearby high topographic region in Janzur, where the freshest water sample (i.e., sample TJ17) is considered as a reference sample to the composition of freshwater coming from the south. The recharge water in sample TJ17 has a high concentration of Ca2+ and HCO3 − as a result of calcite dissolution. The analyzed recharge water in the plain is also showing considerable concentrations of Na+, Mg2+ and SO4 2− as a result of carbonate and evaporite rocks dissolution in the unsaturated zone, and a greater impact of concentration by evaporation, that is characteristic for the study area.

Table 2 Chemical composition of possible end members
Full size table

Stuyfzand classification

The Stuyfzand classification (Stuyfzand 1986, 1992, 1993, 1999) subdivides the most important chemical water characteristics at 4 levels: the main type, type, subtype and class of a water sample (Table 3). Each of the four levels of subdivision contributes to the total code (and name) of the water type.

Table 3 Water type classification (Stuyfzand 1986)
Full size table

The major type is determined based on the chloride content, according to Table 3. The type is determined on the basis of an index for hardness (see Table 3), which can be expressed in French hardness degrees:

$$TH = 5 \times \left( {Ca^{2 + } + Mg^{2 + } } \right)\;{\text{in}}\;{\text{meq}}/l.$$

The classification into subtypes is determined based on the dominant cations and anions. First the dominating hydrochemical family (and groups within families between brackets) is determined both for cations (Ca + Mg (Na + K) + NH4 or (Al + H) + (Fe + Mn)) and anions (Cl, HCO3 + CO3 or SO4 + (NO3 + NO2)). The most important cation and anion (group: within a group: the dominant ion in that group) determine the name of the subtype. Finally, the class is determined on the basis of the sum of Na+, K+ and Mg2+ in meq/l, corrected for a sea salt contribution (Eq. 5). This indicates whether cation exchange has taken place and also the nature of the exchange, by assuming that all Cl− originates from seawater, that fractionation of major constituents of the seawater upon spraying can be neglected and that Cl− behaves conservatively.

$$\left\{ {{\text{Na}}^{ + } + {\text{K}}^{ + } + {\text{Mg}}^{2 + } } \right\}_{\text{corrected}} = \left[ {{\text{Na}}^{ + } + {\text{K}}^{ + } + {\text{Mg}}^{2 + } } \right]{\text{ measured}}{-}1.061{\text{Cl}}^{ - }$$
(5)

where − = often pointing at a saltwater intrusion; + = often pointing at a freshwater encroachment; and 0 = often pointing at an equilibrium.

A positive value greater than the error margin √0.50 Cl− delivers “+” and refers to marine cation surplus, indicating freshening. A negative value (<−√0.50 Cl−) delivers “−”: marine cation deficit pointing to salinization. A value in between both negative and positive error margin delivers “0”: equilibrium.

Each of the subdivisions contributes to the total code (and name) of the water type; for example B4-NaCl- reads as: “brackish extremely hard sodium chloride water, with a {Na+ + K+ + Mg2+} deficit.” This deficit is often due to cation exchange during saltwater intrusion (salinization). It is well known that the hydrogeochemical composition of coastal groundwater affected by seawater intrusion is mainly controlled by cation exchange reactions next to the simple mixing process (Appelo and Postma 1993). These processes can explain deviations of the concentrations of cations from conservative mixing of both waters.

Results and discussion

Water level

The overexploitation of the Jifarah upper aquifer is characterized by the falling of piezometric head over a wide region, reducing the outflow rate to the sea, and the continuous degradation of the chemical quality of water. Depression cones in various places have dropped from 25 to 35 m below sea level (Fig. 2a), which testifies the inversion of the hydraulic gradient and the intrusion of seawater. This was mainly observed in Sabratah region and southern Tripoli.

Fig. 2
figure 2

a Piezometric map for the coastal area of central Jifarah. b The spatial distribution of Cl− concentrations for the analyzed samples. c Spatial distribution of SO4 2− in the upper aquifer

Full size image

Major hydrochemical parameters

Major anions and cations, pH, conductivity, total dissolved solids, as well as temperature, were assessed on all samples using WTW instrument. The results show that temperature ranges between 18 and 26 °C, pH range is 6.71–9.94, conductivity ranges between 510 and 15,650 µS/cm (25 °C), TDS range is 360–11,141 mg/l and chloride concentration ranges from 2 to 5285 mg/l. The high Cl− concentration is due to mixing with seawater. High concentrations of chloride occur in all wells at a few kilometres from the coast, where the concentration of Cl− increases downstream along the flow path. However, in many farther inland wells, it is still the dominant anion. Figure 2b shows a contour map with the spatial distribution of concentrations of Cl− for the analyzed samples. Table 4 shows descriptive statistics for physico-chemical parameters of groundwater for selective representative samples in the Jifarah coastal area.

Table 4 Analytical results for physico-chemical parameters of groundwater for representative samples in the Jifarah coastal area
Full size table

Figure 2c shows the spatial distribution of SO4 2− for the analyzed samples. SO4 2− concentration in the study area ranges from 27 to 2238 mg/l. Along the shoreline, the main source for increasing SO4 2− is mixing with seawater, where seawater intrusion can add significant amounts of sulfate to freshwaters. This can be linked to the high Cl− concentration in the upper aquifer along the coastal area, where the highest concentrations were recorded to the west in Sabratah and in Tajura eastwards.

More than 1000 mg/l SO4 2− is observed in the south of Sabratah region, at 11 km from the coast to about 18 km inland. The probable source of SO4 2− is the dissolution of gypsum from the lower part of the upper aquifer, where gypsiferous sandstone and gypsiferous limestone are found in the formations that belong to the Upper Miocene, where low Cl− is recorded, excluding seawater intrusion as the source. Increased seawater admixture increases SO4 2− along the coast, where high Cl− concentrations have been recorded.

Elsewhere, high concentrations over 1000 mg/l occur in the coastal area for several wells at the east of the study area in the area nearby Surman. In addition to mixing with seawater and/or upconing of deep saline water, the probable source of SO4 2− for these wells is the dissolution of gypsum from the sebkha deposits in those areas, where these wells are located in the immediate vicinity of sebkha. The increase in SO4 2− in the south of Sabratah is accompanied by an increase in NO3 − concentrations. Part of the SO4 2−-rich groundwater is derived from agricultural drainage water that flows through the sebkha deposits in Sabratah coast or through the lower part of the upper aquifer’s sediments in the southern part, inducing dissolution of gypsum. Besides, the high SO4 2− can be due to evaporation of highly concentrated irrigation water. High NO3 − concentrations in these waters are a result of intensive agricultural activities in an area, which has been extensively irrigated during the last two decades (Alfarrah et al. 2011). Additional SO4 2− might also add from the mountain aquifer as a significant amount of gypsum is present in its host rocks, which might contribute to the increase in SO4 2− toward the south.

Out of 134 samples analyzed, 54 % have SO4 2− higher than the highest desirable level of 200 mg/l (WHO 1993) and 16 % have SO4 2− higher than the maximum permissible value of 500 mg/l according to WHO (1993), with a maximum of 2238 mg/l for sample S22 south of Sabratah.

The dissolution of gypsum in groundwater found in several wells raises Ca2+ to significant levels up to 782 mg/l in groundwater. Further, next to the seawater intrusion and/or upconing of deep saline water, additionally, the dissolution of sebkha deposits (mainly consisting of gypsum) contributes to the increase in Ca2+ in several wells in the immediate vicinity of these deposits (e.g., A23 with Ca2+ of 796 mg/l). Other wells, where sebkha deposits are playing a very important role in increasing the Ca2+ concentration, show concentrations between 200 and 652 mg/l. Further inland at more than 20 km south of Tripoli city, no data are available.

Water types and Piper diagram

Classification of hydrochemical facies for groundwaters according to the Piper diagram is represented in Fig. 3a, b.

Fig. 3
figure 3

a Water types according to Piper diagram. b Spatial distribution of water types according to piper diagram

Full size image

In the Piper diagram, almost all water samples are plotted above the seawater–freshwater mixing line (comprising freshwater sample TJ17 and Mediterranean Sea water), due to Ca2+ surplus. Although various hydrochemical facies were observed (NaCl, CaCl, MgCl, CaHCO3, NaHCO3 and CaSO4), CaCl and NaCl types are dominant. Large proportions of the groundwaters show NaCl type, which generally indicates a strong seawater influence (Pulido-Leboeuf 2004) or upconing of deep saltwater, while CaCl water type is indicating salinization and cation exchange reaction (Walraevens and Van Camp, 2005). The region of the CaCl type water may be a leading edge of the seawater plume (Vengosh et al. 1991; Appelo and Postma 1993; Jeen et al. 2001). Furthermore, sources of CaSO4 water type are the dissolution of gypsum from the deeper parts of the aquifer and, to some extent, the dissolution of the scattered sebkha deposits.

Furthermore, in the south toward the recharge area and in the Janzur coast, groundwater of NaHCO3 and CaHCO3 type is associated with the lowest mineralization. This reflects the dissolution of carbonate minerals in carbonate aquifers which make up most of the watershed boundaries.

In the south of Janzur, MgSO4 and NaSO4 water types are observed for some samples, which might be due to upward flow from the lower aquifer, which has high sulfate content in this region (Alfarrah et al. 2011).

Hydrochemical profile

Salinization is induced as the new saline end member is introduced into the freshwater aquifer. The main chemical reaction is cation exchange, resulting in deficit of Na+ and surplus of Ca2+:

$${\text{Na}}^{ + } + 0.5{\text{Ca}} - {\text{X}}_{2} \to {\text{Na}} - {\text{X}} + 0.5{\text{Ca}}^{2 + }$$

where X represents the natural exchanger in the reaction. During cation exchange, the dominant Na+ ions are adsorbed and Ca2+ ions released, so that the resulting water moves from NaCl to CaCl water type, which is typical for salinization (Jones et al. 1999). The salinization process can be schematized as follows (Walraevens and Van Camp 2005):

$${\text{CaHCO}}_{3} 0 \Rightarrow {\text{CaCl}}^{ - } \Rightarrow {\text{NaCl}} - \Rightarrow {\text{NaCl}}0$$

F (fresh) ⇒ Fb (fresh-brackish) ⇒ B (brackish) ⇒ Bs (brackish-saline) ⇒ S (saline).

The chloride ion concentration is taken as a reference parameter (Jones et al. 1999). Therefore, as saltwater intrudes coastal freshwater aquifers, the Na/Cl ratio decreases and the Ca/Cl ratio increases.

Upon the inflow of freshwater a reverse process takes place:

$$0.5{\text{Ca}}^{2 + } + {\text{Na}} - X \to 0.5{\text{Ca}} - X_{2} + {\text{Na}}^{ + }$$

Flushing of the saline aquifer by freshwater will thus result in uptake of Ca2+ by the exchanger with concomitant release of Na+. This is reflected in the increase in the Na/Cl ratio, and formation of the NaHCO3 water type, which is typical for freshening. The anion HCO3 − is not affected because natural sediments behave as cation exchanger at the usual near-natural pH of groundwater (Appelo 1994). The freshening process can be schematized as follows (Walraevens and Van Camp 2005):

The hydrogeochemical profile in Sabratah (Fig. 4) is selected as an example showing the evolution of water salinity along the flow path. In this profile, the water type based on Stuyfzand (1986) is shown, including the specification of the Mix water type. The position of the names of the water types that are indicated above the water table has no relation with the sampling depth. In Sabratah, the waters are brackish and enriched with Cl− and Ca2+. The water types change from NaCl and CaCl in the north, to CaMix in the central zone and further to CaSO4 southwards. Close to the shoreline, the water is NaCl type, due to strong effect of seawater. CaCl results from cation exchange, due to mixing with seawater. In the central zone (Suq al Alalqah), CaMix(ClHCO3) evolving further inland to CaMix(SO4Cl), indicates the location of the transition zone, where the groundwater changes from CaMix enriched with Cl− ion to CaMix with SO4 2− as the dominant anion. The CaSO4 water type observed in the upstream zone to the south, up to 18 km inland, shows the existence of the evaporitic rocks intercalated within the sandstone layers of the aquifer.

Fig. 4
figure 4

Hydrochemical profile in Sabratah area (profile S–S”)

Full size image

The Mg2+ content found in all wells in the upstream zone is mainly resulting from the freshwater end member coming from the recharge area, where Mg2+-containing carbonate is dissolved (Alfarrah et al. 2011). The positive cation exchange code in the classification name is in this case thus not indicating freshening as Mg2+ is not supplied by the marine end member. From the central zone and downstream, cation exchange equilibrium (cation exchange code “0”) exists for most of wells, which in this case indicates the onset of the salinization process ((Na+ + K+ + Mg2+)corrected is decreasing as its cations are adsorbed). The water level is showing low heads, and the EC remains above 1000 µS/cm (25 °C) in all wells, indicating a high level of salinity for the aquifer in this area.

The salinization in this area is mainly due to seawater intrusion and upconing of deep saline water (e.g, S3 & S5 in Fig. 4) in the downstream and central zones. Besides, the salinization may be linked to interaction between groundwater and gypsum in the upstream zone.

Saturation indices

The saturation indices (SI) for calcite, dolomite, halite, aragonite, gypsum and anhydrite were calculated to verify precipitation and dissolution of these minerals. The selected minerals were based on the major ions in groundwater from the study area. Figure 5 is a diagram showing SI values for calcite, dolomite, gypsum, anhydrite, halite and aragonite organized from west to east in the x axis. The sample numbers are sorted according to their location from west to east. The figure also represents an approximate location of samples for each area.

Fig. 5
figure 5

Calculated saturation indices of groundwater samples with respect to selected minerals

Full size image

Out of 134 groundwater samples, 86 % of groundwaters seem to be supersaturated (SI > 0) in respect to calcite (CaCO3), whereas 9 % are undersaturated in respect to calcite (SI < 0) and 5 % are at equilibrium (SI = 0). Dolomite (MgCa(CO3)2) seems to be oversaturated in 81 % of groundwater samples analyzed, 14 % are undersaturated and 5 % are at equilibrium. 98 % of groundwater samples in the study area are undersaturated in respect to gypsum (CaSO4·2H2O) and anhydrite (CaSO4).

In general most of the analyzed samples have saturation indices close to saturation in respect to calcite (SI mostly 0–1) and dolomite (SI mostly 0–2). This supersaturation with respect to calcite and dolomite rather points to water in equilibrium with those minerals. During sampling, most often dissolved CO2 gas escapes, slightly raising pH and thus shifting carbonate equilibrium (more CO3 2−), such that SI > 0 is obtained, whereas in water in the aquifer SI with respect to calcite is close to zero. So, the water is not really oversaturated, it just seems to be oversaturated or has some tendency to oversaturation.

The majority of samples in the study area are undersaturated with respect to gypsum and anhydrite. Gypsum comes close to saturation (SI > −0.76) in many wells. The dissolution of gypsum from the deeper part of the aquifer and from superficial sebkha deposits for many wells in the coastal area raises the SI. In several samples at the south of Sabratah, groundwater was found to be saturated (for one sample) or close to saturation toward gypsum and anhydrite, where dissolution of gypsum from the upper aquifer influences the chemical groundwater composition. Other samples are close to the saturation indicating evaporite rocks dissolution from sebkha deposits. Hence, if gypsum is present, its dissolution will play a role in defining the groundwater composition, especially in systems with limited groundwater flow.

Deviation from conservative mixture of end member fraction

Figure 6 shows the ionic deltas calculated for Na+, Ca2+, Mg2+, K+, HCO3 − and SO4 2− for all analyzed samples. The first thing to note is that the processes in the mixing zone of this aquifer are complex and do not show a homogeneous pattern. For example, in Fig. 6 mNa+,reaction (ΔNa+) is plotted in the secondary axis; the ΔNa+ is usually positive for freshwater, but a large number of samples have negative values particularly in the highly saline water, down to −24 mmol/l. The most logical explanation for this deficit of Na+ is that a reverse cation exchange reaction is taking place during the salinization processes, which releases Ca2+ to the solution and captures Na+. A reverse relationship between the two ions (Na+ and Ca2+) is noticed particularly in the highly saline groundwater, where samples with large negative values of ΔNa+ generally show strong positive ΔCa2+. Furthermore, potassium shows negative or low positive deltas characteristic for marine cations as a result of the salinization process.

Fig. 6
figure 6

Diagram of ionic delta for all analyzed samples

Full size image

ΔMg2+ is mostly positive, due to more Mg2+ added by dissolution of Mg2+-rich carbonate than adsorbed at the clay exchange complex during salinization. Only very few samples show a deficit of Mg2+. Figure 6 also shows that the ionic delta of ∆HCO3 − is close to zero or positive for most water samples. This fact suggests some dissolution of carbonate minerals in the aquifer deposits. In general, most samples show positive ∆SO4 2−. Only a few samples show negative ionic delta of SO4 2−, which indicates sulfate reduction. The gypsum dissolution increases ∆SO4 2− to high positive values up to 22 mmol/l.

Dissolved ion ratios

Figure 7 shows scatter plots of Ca2+ vs Mg2+, Na+ versus Mg2+, Cl− vs SO4 2− and Ca2+ versus SO4 2−. The upstream direction of the upper aquifer in Sabratah region is accompanied by a general increase in Ca2+, Mg2+ and SO4 2− (see Fig. 5), where sulfate and calcium become dominant. As mentioned above, in Sabratah area, the brackish extremely hard CaSO4 type is frequently found related to gypsum dissolution. The high salinity and relatively high concentrations of SO4 2− found in some samples taken from inland wells can be related to this process. On the basis of currently available data, we cannot discard the possibility that additionally, a process of marine intrusion is affecting areas of the aquifer close to these points (south of Sabratah), where the concentration of Cl− is relatively high for many wells.

Fig. 7
figure 7

Ca2+ versus Mg2+, Na+ versus Mg2+, Mg2+ versus SO4 2− and Ca2+ versus SO4 2−. The lines represent a linear regression

Full size image

From Fig. 7, it can be concluded that Ca2+ and Mg2+ correlate positively and significantly with SO4 2− (r = 0.62 for Ca2+ and r = 0.79 for Mg2+; Fig. 7d, c). The high correlation coefficient r = 0.62 between Ca2+ and SO4 2− is related to the dissolution of gypsum from the lower part of the aquifer and from the superficial sebkha deposits in the region. A significant Mg2+ contribution is added from the intrusion of seawater next to some contribution from carbonate dissolution. Low positive correlation is found between Ca2+ and Mg2+ (r = 0.37, Fig. 7a), where the concentration of Ca2+ in seawater is much lower than Mg2+ and most of Ca2+ is derived from the dissolution of calcite in the freshwater aquifer. This is further reflected by the high correlation coefficient between Mg2+ and SO4 2− (r = 0.79, Fig. 7c), and Mg2+ and Cl− (r = 0.95). The lower concentration of Ca2+ compared to Na+ close to the seaside and at the depression cones for some samples, is the result of strong seawater influence.

Also the high correlation coefficient for the scatter plot with the linear regression line of Na+ versus Mg2+ (r = 0.94, Fig. 7b) is explained by marine intrusion.

Figure 8a, b shows scatter plots of TDS vs Cl− and Cl− versus SO4 2−.

Fig. 8
figure 8

TDS versus Cl− and Cl− versus SO4 2−. The lines represent the mixing curve of recharge water and Mediterranean seawater

Full size image

From Fig. 8a, less Cl− in the mixture and the high TDS is explained due to the extra source of SO4 2− coming from the dissolution of gypsum from the upper aquifer’s formation in the southern study area. Furthermore, the scattered sebkha deposits produce high SO4 2− waters for several wells.

Figures 9 and 10 show the ion ratio diagrams (Na+/Cl− versus Cl− and SO4 2−/Cl− versus TDS) based on the analytical results shown in appendix 1.

Fig. 9
figure 9

Molar ratio of Na+/Cl− versus Cl− concentrations

Full size image
Fig. 10
figure 10

Molar ratio of SO4 2−/Cl− versus TDS concentrations

Full size image

Lower ratios of Na+/Cl− (meq l−1/meq l−1) than the Mediterranean seawater ratio (0.88, Fig. 9) indicate seawater encroachment, whereby Na+ becomes adsorbed to sediment during the cation exchange reaction occurring when seawater intrudes freshwater aquifers, resulting in the deficit of Na+ and surplus of Ca2+. The Na+/Cl− ratios in the study area range from 0.38 to 1.68. Fifty-seven percent are lower than the Mediterranean seawater ratio, 74 % are lower than or slightly higher than the Mediterranean seawater ratio (Na+/Cl− < 1), whereas the other 26 % are clearly higher values. Toward the recharge area, the Na+/Cl− ratio is rising gradually, where it reaches more than 1.20 for some wells. Most of water samples in the downstream direction show low Na+/Cl− ratio, except samples from Janzur, where the Na+/Cl− ratio is more than 1.0. Very low (<0.88) ratios are recorded just few meters from the coast and in the depression cones.

Furthermore, to evaluate the effect of seawater mixing qualitatively, the ionic ratio SO4 2−/Cl− (mg l−1/mg l−1) has been examined versus TDS (Fig. 10).

There is a wide range of the SO4 2−/Cl− ratio versus TDS suggesting that the Jifarah Plain groundwater is controlled by other processes in addition to cation exchange during seawater mixing, where samples from the southwestern part of the coastal area of Sabratah and the ones collected at the immediate vicinity of sebkhas have a higher SO4 2−/Cl− ratio than the fresh recharge water ratio (0.82 TJ17). Most of the analyzed samples are plotted between the ratio of 0.1 and 1.0, indicating mixing of the water with the Mediterranean seawater end member having ratio of 0.103. Few samples with very high salinity are having SO4 2−/Cl− ratio less than the Mediterranean Sea; this can suggest sulfate reduction. The gypsum dissolution-affected waters show the highest SO4 2−/Cl− ratio. This has further been illustrated in Fig. 11, where SO4 2− versus x-coordinate in the coastal area (from west to east using the x-coordinates) depicts the distribution of SO4 2−. The effect of gypsum dissolution is clear in sample S22, where the concentration is more than 45 meq/l (2238 mg/l).

Fig. 11
figure 11

SO4 2− concentration (meq/l) versus x-coordinate

Full size image

Conclusion

The overpumping for groundwater has contributed to the deterioration of the water quality by marine intrusion and upconing of the deep saline water. Cl− and SO4 2− are the major pollutants of the aquifer.

Great part of the observed high concentration of sulfate is coming from the effect of seawater intrusion. Another source of SO4 2− is dissolution of gypsum from the upper aquifer’s formation in the south of the study area. Furthermore, the scattered sebkha deposits, containing large amounts of gypsum, produce high SO4 2− waters, and the infiltration of these concentrated waters has affected many wells and has lead to the development of CaSO4 water type for these wells. These processes were also studied by calculation of the saturation index with respect to gypsum, which increase with increasing Ca2+ and SO4 2− in groundwater.

The hydrochemical interpretation also indicates that the dissolution of calcite, dolomite and/or Mg2+-bearing calcite is an important process in most of the groundwaters. The saturation index shows mostly a tendency of precipitation of calcite and dolomite in the aquifer system due to lowering of CO2 pressure at sampling.

Seawater intrusion is accompanied by other processes, which modify the hydrochemistry of the coastal aquifer. The most remarkable process is that of the inverse cation exchange, characteristic of the changes in the theoretical mixture of seawater–freshwater, which is carried out between clays and the aquifer water. This exchange consists in the release of Ca2+ and the adsorption of Na+.

In the study area, the salinization phenomenon occurred in the downstream zones, where the NaCl and CaCl water types have mostly been developed. Somewhat less brackish water quality was identified in the upstream direction, in zones where pumping is less of a problem, indicating smaller seawater influence, whereby recharge from the south is limiting the degree of salinization.

Deterioration in groundwater quality in Jifarah Plain is mainly resulting from aquifer overexploitation; therefore, reducing the pumping rate is the first step to satisfy and recover the groundwater quality and quantity.