Jordan is considered among the poorest countries in the world in terms of water resources. The scarcity of water resources in Jordan imposes strategic difficulties for economic development especially for agriculture [1]. The demand on water resources is increasing with time for domestic and agricultural purposes. Jordan is characterized by semiarid climate, which suffers from water shortage and limited water supply. In recent years, water demand increased rapidly through the high rate of population growth and population influxes together with the higher needs for the industry in the Jordan [2]. Several studies were reported to date addressing the quality of water in Jordan. Salameh and Bannayan [3] studied the water quality in southern part of Jordan as a part of a comprehensive report about the water resources in Jordan. Salameh [4] studied the water quality degradation in several sites in Jordan. El Naqa and Al Kuisi [2] studied the hydrogeochemical modeling of the water seepages through Tannur dam in southern Jordan. Al-Khashman et al. [5] studied the environmental assessment of spring water in Tafila district, southern Jordan. Al-Tabbal and Al-Zboon [6] studied the suitability assessment of groundwater for irrigation and drinking purpose in the northern region of Jordan. Al-Khashman and Jaradat [7] studied the assessment of ground water quality and its suitability for drinking and agricultural uses in arid environment. Al-Khashman et al. [8] investigated the monitoring and assessment of spring water quality in southwestern basin of Jordan.

In order to meet such water needs, the Ministry of Water and Irrigation have selected number of sites in the southern area, that are suitable for construction of storage dams to be used primarily for domestic, recharge of groundwater and irrigation purposes. The Tannur dam (Figure 1) is one of these sites and the first roller compacted concrete (RCC) dam. The dam provides irrigation water to the southern Ghors irrigation scheme and industrial water for local area. Construction of the Tannur dam commenced in January 1998 and completion is expected on time in May 2001 (Figure 1). The dam is 60 m high, 250 m long and impounds 18 million m³ of runoff from Wadi Hasa, one of the major wadis discharging to the southern Dead Sea area. The elevation at the dam is approximately 400 m (Table 1). It forms of water supply and irrigation project serving 1200 ha of farmland and industrial demand in southern Jordan [2].

The Tannur dam location is in a steep valley of Wadi El-Hasa at Tannur mountain, approximately 5 km downstream from the bridge of King’s highway, about 40 km south of Karak (Figure 2). The dam is located in a narrow V-shaped valley with abutment sides rising at about 30˚ for up to 100 m above wadi. The rock foundation for half of dam height comprises interbedded limestone and marl, highly fractured in parts, and with the presence of gypsum seams and clay (weathered mudstone) layers.

The general characteristics of the dam are given in Table 1. Tannur dam is located in the southern Jordanian desert where the climate is arid, with rainfall

and runoff only occurring in the winter seasons (October to April). Annual rainfall is low, generally varied between 63 - 113 mm/year. The average temperature is 22.3˚C, with minimum values of 3.4˚C in January and maximum of 35˚C in July [9]. The maximum sunshine duration occurs in June with absolute values of 12.2 hours/day, but in winter (December and January) the average minimum sunshine is only about 4.2 hours/day. The average relative humidity varies from 40.9% to 26.2% in the winter months and from 22.7% to 36.4% in the summer season. The prevailing wind direction is from westerly to southwesterly. This paper examines the hydrochemical properties and quality of water in Tannur dam related to its suitability for agriculture and domestic use. Furthermore, a geochemical model was carried out using the computer software PHREEQC to explain the thermodynamic condition of water during water-rock interaction.

The geology of the study area is shown in Figure 3. The outcropping rocks at the dam location and reservoir area belong to the Ajlun group of Late Cretaceous [2] [10]. Fuheis-Hummar-Shueib (FHS) formations represent the oldest rocks found in the dam site area, representing the dam foundation rocks. These formations

consist of thin to moderately thick bedded (5 - 55 cm thick) limestone, marlstone, marl and clayey marl with gypsum bands [11]. The limestone is moderately hard and moderately weak, and these strata contain small bands of marl and fossiliferous limestone up to 70 cm thick [10]. The Wadi As-Sir Limestone (WSL) formation (A7) is exposed over the whole of the study area. This formation is characterized by predominant hard buff dolomitic limestone, thickly and thinly bedded limestone with nodules of chert interbedded with marly limestone and dolomite wackstone [12]. The alluvium and wadi sediments comprise recent sub rounded to rounded, poorly sorted gravels ranging in size from pebbles to boulders. The clastic materials are composed of regional bedrock, mostly limestone, chert and sand grains [10] [12] [13] [14].

Structurally, the area is intensively faulted and folded as a result of various tectonic activities. The Wadi El Hasa fault is the major E-W faults in the study area. In the west it trends ENE and curves eastwards to the N-W and ESE direction. At the western side it branches into two faults. The Wadi El Has fault strikes generally E-W from the Dead Sea fault, which is about 25 km west of Tannur dam location [10] [12]. The minor faults at the eastern end of the Wadi El Hasa fault are considered to be dextral strike-slip faults as indicated by the peripheral small scale folding. Along the Wadi El Hasa fault, a volcanic plug has erupted.

The aquifer systems in the study area can be divided into three main aquifers; they are Kurnub Sandstone system (lower aquifer), middle aquifer systems (Wadi Es Sir Limestone aquifer) and shallow aquifer systems (upper aquifer).

Reservoir water samples were collected from thirty six locations in the dam, as shown in Figure 4.

The water samples were collected in two periods, the first during May 2015 and the second during September 2015. The collected water samples were stored in polyethylene bottles. The bottles were washed previously with deionized water and diluted hydrochloric acid. Before filling the bottle with the samples, the bottles were rinsed with the samples water and transported to a lab in cold storage.

After collection the water samples were transferred from polyethylene containers (1 L) into polyethylene bottles (250 mL), and then filtered through a 45-µm cellulose nitrate membrane filter using a vacuum pump in order to remove insoluble particles. Each sample was divided into two polyethylene bottles one for major anion analyses and the other was acidified the samples to pH < 2 for cation analysis and all samples were kept in refrigerator at 4˚C until the time of chemical analysis, which was usually performed within one week after bottling. Electrical conductivity, pH, temperature and dissolved oxygen (DO) of the water samples were measured on site by using portable pH meter, EC meter, dissolved oxygen meter and temperature meter. All glassware and polyethylene bottles were soaked in 20% HNO₃ for 1 day and rinsed several times with deionized water before use. Conductivity measurements were carried out with 470 JENWAY conductivity meter with temperature compensation, while the pH values were measured in the field using 370 JENWAY pH-meters equipped with a combination glass electrode. Calibration was always carried out before measurement using standards buffer solutions of pH 4.00 and 7.00. Dissolved oxygen values were measured in the field using field DO-meter (WTW equipment). TDS was measured by the conductivity measurements. The calculation is carried out by the following equation: TDS (mg/l) = EC (µs/cm) × 0.64, where EC is the electrical conductivity (Howari et al., 2005). The bicarbonate concentration was determined by titration of the water samples with 0.02 H₂SO₄ using methyl orange as indicator. Major anions (Cl⁻, NO 3 − and SO 4 2 − ) were analyzed by 100 Dionex Ion Chromatography instruments equipped with AG4A-SC guard column, AS4ASC separating column, SSR1 anion self-regeneration suppresser and conductivity detector. The samples were injected through 25 MI sample loop and eluted at 2.0 ml min-I using 1.7 mµ NaHCO₃ and 1.8 mµ Na₂CO₃. The system was calibrated with a certified standard from Dionex. Major cations (Ca²⁺, Mg²⁺, Na⁺, and K⁺) were measured by 800 Varian flames Atomic Absorption Spectrophotometer. The concentrations of cations were determined using a CS12 analytical column, CG 12 guard column, using 20 mµ CH 4 SO 3 − . The concentration of bicarbonate was determined by titration with 0.01 hydrochloric acid using methyl orange as indicator. The standard solutions of the anions and cations as well as blank samples were prepared with different concentrations. All standard solution were made daily by diluting the stock solutions with 0.01 M HNO₃ [22], which was prepared from analytical grade HNO₃ solution obtained from Merck. A quality control procedure, including, recalibration of the instruments, analysis of triplicate samples and recovery test of standard reference material was used to control data quality [23]. All chemicals and reagents used in this study were of analytical grade unless otherwise stated. Deionized water (Milli-Q 18.2 μs/cm) was used for all dilutions. Standard solution was prepared by diluting the stock solutions.

To prevent the sample contamination with any source of pollution, all the glassware, Pyrex and plastic containers were washed several times with soap, deionized water and treated with 0.01 M HNO₃ and finally rinsed with ultra-pure water. After analysis the accuracy of these standards were within ±7%. The numerical simulation model PHREEQ was used to set up the hydrological components of the groundwater, especially the saturation indices of minerals (calcite, dolomite, gypsum, anhydrate and halite) to test the saturation of minerals.

To assess water quality and hydrochemistry of water samples, the parameters such as; Total Hardness (TH_mg/L), Sodium Adsorption Ratio (SAR), Percent Sodium (Na%), Residual Sodium Carbonate (RSC) and Permeability Index (PI) were calculated based on the chemical characteristics of water samples beside the hydrochemical parameters which include hydrochemical evaluation and Saturation Index (SI). These parameters and assessment indices will be defined later.

Thirty six water samples were collected from reservoir water in Tannur dam during the studying period, May 2015 and September 2015. Statistical summary of volume-weighted mean concentrations of physical and chemical parameters are presented in Table 2.

The ratio of total anions to that of cations ((∑anions)/(∑cations)) was an indicator for the completeness of measured parameters [24]. The average equivalent sum of cations to that of anions ((∑anions)/(∑cations)) was 0.89 ± 0.32. Also, for the set of samples considered in this study, linear regression of cation sum on anion sum gave value R² = 0.93 indicating that the quality of the data was good. The in-situ measured temperature for the water samples in the wet season (May 2015) ranged between 9.0˚C and 12.8˚C, with an average value of 10.6˚C. On the other hand, the temperature of water samples in the dry season (September 2015) varied between 11.0˚C and 18.1˚C, with an average value of 13.8˚C. Dissolved oxygen (DO) in water samples in wet season ranged from 3.4 and 6.6 mg/l, with an average value of 5.8 mg/l, but in dry season ranged from 4.1 mg/l to 6.9 mg/l, with an average value 5.41 mg/l. The water samples of the study area have pH values varied from 6.8 to 8.2 (alkaline type) with an average values in wet and dry seasons 7.6 and 7.9, respectively. The high value of pH content has resulted from the increase of calcite from carbonate rocks in the study area. In the water samples, the EC in wet season ranged between 1210 and 1910 µs/cm, with a mean value of 1501 µs/cm, while, in the dry season EC varied from 1800 and 1968 µs/cm, with a mean value of 1903 µs/cm. However, the high conductivity of the water samples in Shoubak basin corresponding to the highest of dominant ions, which is result by ion exchange and solubility of rocks in aquifer [22] [25]. TDS values in water varied from 772 mg/l to 1241 mg/l in wet season, with a mean value of 942 mg/l but in dry season it ranged from 1155 mg/l to 1264 mg/l, with a mean value of 1218 (Table 2). The increase of salinity in water

samples can be explained by the chemical reactions that take place during water-rock interaction and through long period of contact with rocks. Dissolved Oxygen (DO) varied in the range from 3.4 mg/l to 6.6 mg/l in the wet season, with a mean value of 5.8 mg/l, but in dry season it ranged from 4.0 mg/l to 6.9 mg/l, with a mean value of 5.4. The chemical composition of the water in Tannur dam was variable. The chemical parameters including statistical measures were shows in Table 2. The abundance of the major ions in water samples is in the following order: HCO 3 − > Ca²⁺ > Cl⁻ > NO 3 − > SO 4 2 − > Na⁺ > Mg²⁺ > K⁺. The concentration of calcium ion in wet season ranged from 102.1 mg/l to 150.3 mg/l, with a mean value of 126.3 mg/l but in dry season varied from 150.6 mg/l and 182.3 mg/l, with a mean value of 166.3 mg/l. Higher value of calcium in the wet season was due to release of calcium from the sedimentary carbonate rocks and soils to the aquifer (Al-Khashman, 2006). Calcium often comes from carbonate minerals such as calcite and dolomite, which commonly occurs in sedimentary and volcanic rocks, surrounded the study area. In wet season the concentration of Mg²⁺ in water ranged from 43.6 mg/l and 71.4 mg/l with a mean value of 54.8 mg/l but in dry season the concentration of Mg²⁺ varied from 66.9 to 79.6 mg/l, with a mean value of 73.9 mg/l. The maximum content of Mg²⁺ exists on the eastern side of the study area is related to the dissolution of volcanic rock which is present in this area and from chemical fertilizers [26]. The source of Ca²⁺ and Mg²⁺ in the water dam is believed to be limestone, dolomite, gypsum and anhydrite in the Wadi Es Sir and FHS formations [2]. The presence of HCO 3 − in the water samples is derived from carbon dioxide in the atmosphere, soils and by dissolution of carbonate rocks. The increasing of bicarbonate in the dry season is attributed to the dissolution of carbonate rock in the study area (Figure 5). Sodium is the most abundant cation in the hydrosphere and present in natural water. The common source of sodium in the water samples is gypsum in sedimentary rocks, weathering of plagioclase and feldspar, which are typical constituents of volcanic rock surround the study area and evaporate deposits. Nitrate in water generally originates from several natural and human sources on the earth surface [27]. Also, nitrogen value in groundwater is derived from the biosphere [28]. The nitrate content in the water of Tannur dam in wet season ranged from 130.2 mg/l to 198.4 mg/l with a mean value of 161.5 mg/l, whereas, the nitrate in dry season varied from 179.8 mg/l to 243.6 mg/l with a mean value of 223.3 mg/l. The main source of nitrate in the water samples is related to the intensive use of nitrogen fertilizers in the nearby agricultural lands in the eastern and northern side of the Tannur dam, where most of the farms are located and manure and plant decomposition [1] [29].

Human activities near the Tannur dam site have had direct and indirect effects on the rate of water contamination. The direct effects on water include dissolution and transport of excess quantities of fertilizers with associated materials and hydrological alteration related to agriculture activities. While, the indirect effects include changes in water-rock reactions in soils and aquifers caused by increased concentration of major ions [30] [31]. Groundwater chemistry exchange matter with the various minerals and gases within the aquifer which it resulting from dissolution and precipitation of minerals [32]. Saturation indices of minerals in the water samples can be expressed by the saturation index (SI). The SI is defined as the logarithm of the ratio of the ion activity product of the mineral equilibrium constant at a given temperature [33]. When SI < 1 the water is undersaturated and the minerals will dissolve. If the ratio of saturated indices greater than 1; the water is supersaturated and the minerals tend to be precipitated [32] [33]. On the other hand, if SI is equal to 1, the water is in equilibrium with the

mineral. The statistical of saturation indices of water samples were presented in (Table 3). All water samples have positive calcite and dolomite indices; indicate that the water was oversaturated with respect to calcite and dolomite. On the other hand, all the water samples indicate negative value for gypsum, halite and anhydrate indices; which means that all of the water samples below the saturation state with respect to the gypsum and anhydrate and they are able to dissolve additional amount of these minerals. The saturation indices at different water sampling locations in the study area were plotted in Figure 6. It is noticed that the water samples are oversaturated with respect to calcite and dolomite, slightly under saturated with respect to anhydrate and gypsum, but highly under saturated with respect to halite.

The geochemical evolution of water samples can be understood by plotting the concentrations of major ions on the trilinear diagram of Piper [34] (Figure 7) to determine the water type according to the Langguth classification [35]. Figure 7 shows that most of water samples analyzed during May 2015 and September 2015 in the field of mixed Ca²⁺-Mg²⁺-Cl⁻ type of water, whereas some samples are representing Ca²⁺-Cl⁻ and Na⁺-Cl⁻ types. From the Figure, alkaline earths ( HCO 3 − , Ca²⁺ and Mg²⁺) significantly exceed the alkalis (Na⁺ and K⁺) and strong acids chloride and sulfate. Water chemistry originates from dissolution of carbonate rocks. However, the water was generally classified as Ca-HCO₃ water with low salinity. The hardness of water samples was classified according to Sawyer and McCarty [36]. The hardness of water resulted from the presence of calcium, magnesium, bicarbonate and sulfate concentrations that are the most abundant ions. (Table 4) shows the classification of the water samples based on their hardness. The calculated of hardness of the water samples as CaCO₃ in mg/L according to Todd [27] :

TH = 2.5 Ca 2+ + 4.1 Mg 2+ (1)

It was clearly shown from Table 4 that the water samples in wet season can be classified as hard water, 29 water samples show relatively very hard water. However, in the dry season two water samples are classified as hard water while the most of water samples were considered as very hard water. The general increase of water hardness from wet to dry season can be attributed to the dissolution of carbonate rocks that also includes traces of evaporate deposits and percolation of rainwater to the saturation zone [20].

The ratio of Ca²⁺/Mg²⁺, Na⁺/Cl⁻, Mg²⁺/Ca²⁺ + Mg²⁺, SO 4 2 − /Cl⁻ and Ca²⁺/Na⁺ were calculated for water samples (Table 5). In the water samples, Ca²⁺/Mg²⁺ equivalent ratios were >1.0, the Ca²⁺/Mg²⁺ was higher in water samples due to higher carbonate concentration in the water. High correlation was found between

calcium and magnesium of water (R² = 0.81), suggesting that the common source of these ions from carbonate dissolution in the water. While the median Na⁺/Cl⁻ equivalent ratios were <1, the lower ratio of Na⁺/Cl⁻ in water can due to dissolution with clay minerals exchanging Na⁺ for Ca²⁺. On the other hand, strong correlation was found between sodium and chloride of water (R² = 0.91) (Table 6), suggesting that the common source of these ions from salt dissolution. The possible sources of these ions were anthropogenic sources and natural sources. The median Mg²⁺/Ca²⁺ + Mg²⁺ ratios were < 1 indicating weathering of limestone and dolomite, while the median Ca²⁺/Na⁺ equivalent ratios in two periods varied between 1.48 to 1.82, this ratio is high in water as a result of solubility of carbonate in the study area. The median SO 4 2 − /Cl⁻ ratios were <1 indicating the dissolution of gypsum and limestone in the water. Chloride ion is highly correlated with sodium (R² = 0.91), while they are better correlated with HCO 3 − , SO 4 2 − and Mg²⁺ (R² = 0.74, 0.72 and 0.68) respectively.

The suitability of water for irrigation is dependent on the effects of its mineral constituents on both plants and the soil [27] [37]. The quality of water for irrigation water is based on the total salt concentration of the water, the concentration of specific ions that may be toxic to plants. Sodium adsorption ratio (SAR) is an important parameter for determining the suitability of water for irrigation because it is a measure of alkali and sodium hazard to plant [38]. The sodium adsorption ratio (SAR) is defined as follows where the concentrations of the constituents are expressed in milliequivalent per liter (meq/l).

SAR = Na / { sqrt ( Ca + Mg ) / 2 } (2)

The SAR values ranged from 1.48 to 2.29 with a mean value of 1.75 during the wet season (May 2015) and 2.08 to 2.44 with a mean value of 2.20 during dry season (September 2015). The dam water of Tannur was classified with respect to SAR (Richards, 1954). All analytical data plotted on the US salinity diagram [39] shows that most of water samples classified in the field of C3S1, indicating high salinity with low sodium water. This type of water is suitable for irrigation on almost all types of soil (Figure 8). Another parameter can be used for classification of irrigation water is the sodium percentage while, sodium reacts with

soil to reduce its permeability. The sodium percentage (Na%) is calculated using the formula given below:

Na % = { Na + K / ( Ca + Mg + Na + K ) } × 1 00 (3)

where all of the concentrations are expressed in meq/l.

The sodium percentage indicates that the water is good to permissible for irrigation (Table 7). All samples in the wet season are classified as good water for irrigation, while in dry season 69.4% of water samples are classified as good water, and others are classified as permissible water for irrigation (Table 7). When the concentration of sodium is high value in irrigation water, sodium ions tends to be adsorbed by clay minerals, displacing Mg²⁺ and Ca²⁺ ions. This exchange process of sodium in water calcium and magnesium in soil reduced the permeability and eventually results in soil with weak internal drainage [18] [28].

In addition to the sodium adsorption ratio (SAR) and sodium percentage (%Na), the residual sodium carbonate (RSC) represents the excess sum of carbonate and bicarbonate over the sum of calcium and magnesium also influences the unsuitability of water resources for irrigation. The residual sodium carbonate (RSC) is calculated by the following equation:

RSC = ( HCO 3 − + CO 3 2 − ) − ( Ca 2+ + Mg 2+ ) (4)

where the concentration of ions is expressed in meq/l.

The classification of irrigation water in the Tannur dam according to the residue sodium carbonate (RSC) values is shows in Table 8. The first category has (RSC) < 1.25 epm (good quality). The second category has ranged from 1.25 to 2.5 epm (doubtful), while the third category has more than 2.5 epm (unsuitable). Table 8 shows the samples in two time periods and their relevant class. Results shown in this table indicate that the water of Tannur dam is mostly of good quality.

The permeability index (PI) values also indicate that the water of Tannur dam is suitable for irrigation. The permeability index is shown in the following equation:

PI = ( Na + + √ HCO 3 − ) × 100 / ( Ca 2 + + Mg 2 + + Na + ) (5)

where all values of ions are expressed in meq/l.

The permeability index (PI) in water from Tannur dam ranged from 34% to 43% during May 2015. On the other hand, in dry season the PI varied from 45% to 60% during September 2015, which found under class-1 of Doneen’s chart [40].