Hydrogeochemistry, water quality and health risk assessment of water resources contaminated by agricultural activities in Korkuteli (Antalya, Turkey) district center

Water is not only the essence of life, but also one of the most important factors determining the quality of human life. The quality of water is extremely important because it is essential for life. The fact that currently 75% of the earth is covered with water gives the idea that there is no water shortage in the world, but the potable water rate is only 0.74% of all waters worldwide (Akın & Akın 2017). In addition, emerging climate changes, developing technology and increasing population are the biggest problems in reaching healthy water today. These problems cause surface waters to gradually decrease and become unusable. A result of this, groundwater is widely used. However, groundwater is not endless and not always suitable for the intended purpose. For this reason, it is very important to determine the amount of groundwater, the quality of the waters and their usage areas. In addition, knowledge of the geochemical evolution of groundwater characteristics is important for the sustainable development and effective management of water resources.

Pollution of groundwater is one of the most significant environmental problems in the world in recent years (Kumar et al. 2014). Groundwater quality depends on the quality of the recharged water, rain and geochemical processes. The quality of groundwater reflects the combined effect of many processes along the groundwater flow path, at any point underground. Geochemical processes are responsible for the seasonal and spatial variation of groundwater chemistry. It is possible to divide the factors affecting these geochemical processes into geogenic and anthropogenic. Geogenic factors are related to geological, hydrological and hydrogeological conditions. Anthropogenic factors that change the chemistry and quality of groundwater are the result of domestic, irrigation and industrial uses. However, at the same time, degradation of the groundwater quality affects its use for drinking, agriculture and industrial activities. Today, approximately 80% of diseases all over the world and one third of deaths in developing countries are caused by polluted drinking water (WHO 2004). Therefore, the determination of the groundwater quality is important for the suitability of the water.

In this study, Korkuteli district in Antalya province, located in the southwest of Turkey, was chosen as a study area. In Korkuteli district, groundwater is the major source of water for human consumption and irrigation. Korkuteli district is an important region of Antalya, especially in terms of greenhousing and other agricultural activities related to livestock. These activities pollute the groundwater anthropogenically. Therefore, groundwater quality in this area is very important in terms of determining the suitability of water for drinking, domestic and agricultural purposes in the study area. In relation to this, the fact that the effect of agricultural activity on the water resources of the region has not been studied in detail before is a major deficiency for the study area. This study aims to resolve this deficiency and the results are important to determine the quality and usability with regard to sustainable management of water resources in the region.

Korkuteli district center is located between latitude 37° 3′ 51.9336″ N and longitude 30° 11′ 47.7132″ E coordinates (Figure 1). Korkuteli district center (area 121 km²) was selected as the study area due to the dense agricultural activity and settlement area. The altitude of the district center is 1,020 m. In general, the Mediterranean climate and the terrestrial climate prevail in the district. The mean rainfall for the study area was calculated as 316.39 mm (between 1969 and 2016) by the isohyet method (a line drawn on a map connecting points that receive equal amounts of rainfall) using the data from the meteorological stations in the study area and surrounds. In addition, the mean evapotranspiration was 324.84 mm. The drinking water of the district center is supplied from the springs and well waters in the study area. Also the irrigation waters are supplied from the well and surface flows. Korkuteli and Salamur streams are the most important surface flows in the study area (APESR 2012).

A total of 30 water samples from wells, springs and tap waters were analyzed in November 2016 (dry period) and May 2017 (wet period) for the determination of their major and minor chemical characteristics. The water samples were collected in two plastic bottles, pre-washed with 0.5% nitric acid (HNO₃) and deionized water from each sampling point. During sampling, two bottles were filled with water from each spot, filtered, and a few drops of HNO₃ were added to one water sample that was used to detect major cations, while the second sample was not acidified and was used to detect major anions. All samples were transported and kept in the dark at 4 °C for analysis. The physical parameters, such as hydrogen ion concentration (pH), electrical conductivity (EC), total dissolved solids (TDS), oxidation–reduction potential (Eh) and discharge temperature (°C), were determined in situ using a YSI multi-parameter water quality sonde (YSI 6050).

The major chemical constituents (cations) were analyzed by ICP-MS (inductively coupled plasma-mass spectrometer) at the Bureau Veritas Commodities Canada Ltd (ACME Laboratory Vancouver, Canada, an ISO 9002 accredited company). Bicarbonate (HCO₃), carbonate (CO₃) concentrations were determined by the titrimetric method; chlorine (Cl), sulphate (SO₄), nitrate (NO₃), nitrite (NO₂) and ammonium (NH₄) were determined using ion chromatography in the Hacettepe University Water Chemistry Laboratory (Ankara, Turkey) (Table 1). The charge–balance error in the water samples was less than 5%, which is within the acceptable limits. All mathematical calculations were calculated using Excel 2007 (Microsoft Office). Statistical analysis, such as the correlation matrix, was performed using the SPSS software version 15.

The lithological units and interaction time of water with these units control the chemical composition of the groundwater. Firstly the lithological units in the study area have been investigated. All of the lithological units in the study area are autochthonous. These units are Beydağları autochthonous and quaternary alluvium (Qal), slope debris (Qym), old stream terrace fillings (Qt) and old alluvium-slope debris (Q1ym) (Şenel et al. 1989; Şenel 1997; Figure 1, Table 2). Beydağları autochthonous units are composed of Beydağları formation (Kb), Tekkeköy members (Kbt), Çamlıdere olistostrome (Tpç), Küçükköy formation (Tek), Karabayır formation (Tmk), Karakuştepe formation (Tmkt) and Kasaba formation (Tmka) (Şenel et al. 1989; Şenel 1997).

Secondly, lithological units in the study area were evaluated according to their hydrogeological properties. The general characteristics and hydrogeological properties of the lithological units are given in detail in Table 2. Alluvium and quaternary units represent the most important granular aquifer in the study area (Aqf-1) and have an area of about 45 km². Limestones can contain significant amounts of groundwater in cracks and melting spaces which allow water to move. For this reason the Beydağları formation, consisting of neritic limestones, and the Karabayır formation, consisting of algae limestones, are defined as ‘karstic aquifer’ (Aqf-2). The Karstic aquifer is the other important aquifer in the study area. The majority of water resources in the study area are discharged from the alluvium and limestones (Figure 1). The Kasaba formation, which consists of conglomerates and sandstones, is classified as ‘aquitard’ (Aqt). Çamlıdere olistostrome, Karakuştepe formation and Küçükköy formation in the study area are classified as ‘aquifuge’ (Aqj) due to their lithological characteristics (Table 2).

Groundwater chemistry is controlled by hydrogeochemical processes. There are many chemical processes during the movement of the groundwater from the recharge area to the discharging area. Some of these processes include precipitation, ion exchange, redox condition, leaching and dissolution. The chemical properties of the water determine the usage status for domestic, industrial or agricultural activities. In many developed and developing countries water pollution is one of the most significant factors for the spread of diseases and infant deaths (Kumar et al. 2016). The causes of water pollution can be classified as natural and anthropogenic sources. Accordingly, it is important to know the chemical properties and quality of water for the management and evaluation of groundwater (Kumar et al. 2016). For this reason, in this study we have attempted to determine the physicochemical properties, chemical properties and quality of groundwater in the study area.

The pH of spring waters (according to the measurement method used for pH, the lower limit of quantitation is ±0.01) varied between 8.71–9.36 and 8.88–9.60 for the dry and wet season, respectively. The pH values of the well waters were measured at 8.54–9.33 in the dry season and 8.89–9.30 in the wet season. Also, the pH of the tap waters varied for the dry and wet season between 8.76–8.93 and 8.85–8.92, respectively. The pH values of all water samples increased in the wet season (Table 3).

The EC values, EC, of spring waters (according to the measurement method used for EC, the lower limit of quantitation is 0.0001–0.01 μS/cm.) were measured to be between 288.80–695.00 μS/cm in the dry season and 279.80–708.00 μS/cm in the wet season. The EC values of well waters also varied in the range of 317.10–794.00 μS/cm in the dry season and 309.90–809.00 μS/cm in the wet season. Also, the EC values of tap waters were measured in the range of 474.60–638.00 μS/cm in the dry season and 424.00–744.00 μS/cm in the wet season (Table 3). The high EC values in the wet season indicated the spatial variability of leaching and dilution with recharging rainfall. Also, the higher EC values in the wet season might be attributed to enhanced chemical weathering and lengthier residence time of groundwater in the aquifers (Oinam et al. 2011; Alam et al. 2016).

In addition, the temperature, T (°C), values of spring waters (according to the measurement method used for T (°C), the lower limit of quantitation is 0.001 °C) were measured in the range 12.10–15.50 °C in the dry season and 11.80–15.30 °C in the wet season. The T (°C) values of well waters also varied in the range 11.80–17.60 °C in the dry season and 14.70–17.80 °C in the wet season. Also, the T (°C) values of tap waters were measured in the range 13.10–17.10 °C in the dry season and 15.00–16.60 °C in the wet season (Table 3).

Oxidation–reduction potential, Eh, is a measurement that indicates the degree to which a substance is capable of oxidizing or reducing another substance (APHA 1998). In the present study, Eh values were measured to range from 319.50–400.80 mV in the dry season to 298.90–388.70 mV in the wet season (according to the measurement method used for Eh, the detection limit is 0.1 mV) (Table 3).

The most common cations in the groundwater were calcium (Ca) and magnesium (Mg). Subsequently, sodium (Na) and potassium (K) ions were present in the composition of the water in lesser amounts (Tables 3 and 4). The sources of Ca and Mg in water were generally carbonate-rich rocks such as limestone and dolomitic limestone. An increase in Mg ion in the dry and wet seasons was observed at several locations (K1, K4, K5, K7, K11, K12, and K13). The major source of Mg in the groundwater was from interaction of dolomitic limestone with water.

Bicarbonate (HCO₃) was the main anionic constituent of the groundwater samples ranging from 225.70 to 506.30 mg/L in the dry season and from 236.65 to 517.80 mg/L in the wet season (Tables 3 and 4). HCO₃, representing the major source of alkalinity, generally prevails due to the dissolution of CO₂ and carbonates, reaction of silicates with carbonic acid (Ranjan et al. 2013) and oxidation of organic matter (Jeong 2001). Sulphate (SO₄) ion, which originates from the oxidation of sulphite, was the most abundant after bicarbonate (Ranjan et al. 2013). Another reason for the excessive amount of sulfate in the groundwater is the presence of anthropogenic inputs. SO₄ concentrations in the study area ranged between 21.07 and 132.73 mg/L in the dry season and between 30.09 and 155.80 mg/L in the wet season (Tables 3 and 4). This situation also indicates the entrance of anthropogenic sulphate related to fertilizers in the study area.

In the study, NO₂, NO₃, NH₄ and heavy metal analyses (aluminum (Al), As, cadmium (Cd), copper (Cu), Cr, iron (Fe), manganese (Mn), nickel (Ni), lead (Pb), zinc (Zn)) were made to determine pollutant types and sources (Table 3). NO₂, NO₃, and NH₄ are pollutants which come mainly from agricultural and industrial effluents, contain a high concentration of nitrogen and are some of the largest contributors to groundwater pollution in the world (Li et al. 2016). According to the results of analysis, NO₂ concentrations were determined as 0.00 mg/L in both periods. The detection limit for nitrite is ±0.001 mg/L. For this reason, the mean and standard deviation values for nitrite were not calculated (Table 3). NO₃ concentrations ranged from 2.19 to 21.34 mg/L in the dry season and ranged from 4.90 to 51.00 mg/L in the wet season. In addition, NH₄ concentrations were 0 mg/L in the dry season and 0–0.13 mg/L in the wet season (Table 3).

According to the heavy metal analysis results, Al concentrations were determined to be 0 mg/L in the dry season, and 0–0.05 mg/L in the wet season. As concentrations ranged from 0.08 to 0.17 mg/L in the dry season and ranged from 0.09 to 0.17 mg/L in the wet season. Cd concentrations were determined to be 0.05 mg/L in the dry season and 0 mg/L in the wet season. Cu concentration ranged from 0 to 0.04 mg/L in the dry season and from 0 to 0.87 mg/L in the wet season. Besides this, Cr concentrations were determined to be between 0.01 and 0.07 mg/L in the dry season and between 0.01 and 0.08 mg/L in the wet season. Fe concentration ranged from 0.01 to 0.11 mg/L in the dry season and ranged from 0.01 to 0.12 mg/L in the wet season. Mn concentration was determined as 0 mg/L in both seasons. Ni concentrations were determined to be 0 mg/L in the dry season, and between 0–0.09 mg/L in the wet season. Pb concentrations were determined to be 0 mg/L in the dry season, and 0–0.01 mg/L in the wet season. Finally, Zn concentration ranged from 0 to 0.17 mg/L in the dry season and 0 mg/L in the wet season (Table 3). Accordingly, seasonal variation in NO₂, NO₃, NH₄ and heavy metals concentrations is thought to be related to infiltration of rainwater and agricultural activities.

Correlation analysis was applied to determine the relationship between physicochemical properties of water samples. It is possible to obtain information about the mineral and chemical processes, and the chemical constituents of water with this relationship were determined by correlation analysis (Varol & Davraz 2015).

In this study, the Kolmogorov–Smirnov (K-S) test was initially used to determine the compatibility of the data to normal distribution (Varol & Şen 2009). According to the K-S test, all the variables were normally distributed with 94% confidence. For this reason, to evaluate the potential relationship between various physiochemical parameters and trace elements, Pearson correlation analysis (PCA) was carried out.

In correlation analysis, if the correlation coefficient is close to 1 or 1, it means a good positive relationship between the two variables. The near zero values are significant if p < 0.05, but there is no relationship between them. Thus, while it is assumed that there is strong correlation between the parameters with r > 0.7, it is said that the r value is moderately correlated between 0.5 and 0.7 (Manish et al. 2006).

All the processes were performed using SPSS software version 15.0 for Windows. In addition, the Pearson correlation matrix was applied separately for the dry and wet seasons in order to study the changes in the relations between the parameters. The correlation matrix of the parameters is given in Table 5 (dry season), and Table 6 (wet season). The analysis results of CO₃, NO₂, NH₄, Cd, Mn, Ni and Pb in the dry season and CO₃, NO₂, Cd, Mn, Ni in the wet season were 0.00 mg/l, and owing to this results were not evaluated in the correlation matrix.

According to the physicochemical PCA results, EC showed positive strong correlation with TDS, Mg, Na, HCO₃, Cl, SO₄, NO₃, and Cr in dry and wet seasons. Also EC showed negative strong correlation with pH in the dry season. T (°C) showed negative moderate correlation with pH in the wet season. pH showed negative strong correlation with TDS, Ca, K, HCO₃, NO₃ and showed negative moderate correlation with Mg, Cl and Cr ions in the dry season. In the wet season, pH showed negative moderate correlation with only Mg ion. Eh showed positive moderate correlation with As in the wet season only. TDS showed positive strong correlation with Mg, Na, HCO₃, Cl, SO₄, and NO₃, Cr ions in the dry and wet seasons (Tables 5 and 6).

Ca showed positive moderate correlation with K, HCO₃ and NO₃ in the dry season and also showed positive moderate correlation with K. Mg showed positive strong correlation with Na, HCO₃, Cl, SO₄, NO₃, Cr in the dry season. In the wet season it showed positive strong correlation with Na, HCO₃, Cl, NO₃ and Cr, and showed positive moderate correlation with SO₄ the Na showed positive strong correlation with HCO₃, Cl, SO₄ and Cr in the dry and wet seasons. It also showed positive moderate correlation with NO₃ and Cu in the dry season and showed positive moderate correlation with NO₃ and Cl. While K showed a positive moderate correlation with Al in the dry season and negative correlation with Al in the wet season (Tables 5 and 6).

In addition, according to the physicochemical PCA results, HCO₃ showed positive strong correlation with Cl, SO₄, NO₃, Cr in the dry and wet seasons. Cl showed positive strong correlation with SO₄, NO₃, and Cr in the dry season, and showed positive strong correlation with NO₃ and Cr, and positive moderate correlation with SO₄ in the wet season. SO₄ showed positive strong correlation with Cr in the dry and wet seasons and also showed positive moderate correlation with NO₃ and Cr in the dry season (Tables 5 and 6).

According to the PCA results of the heavy metal and pollution parameters, NO₃ showed positive moderate correlation with Cr in the dry and wet seasons. While Al showed positive moderate correlation with Fe in the dry season, it showed positive strong correlation with Fe in the wet season. Cu showed positive moderate correlation with Ni and Pb ions. Ni showed positive moderate correlation with Pb ion (Tables 5 and 6). Statistical analysis results indicated that water resources were affected from rock-water interaction, climatic conditions, ion exchange processes and excess application of fertilizer in the study area.

Hydrogeochemical facies are a useful tool for determining the chemical history and origins of groundwater. They are used to show similarities and differences in the chemistry of groundwater samples based on dominant cations and anions (Piper 1944, 1953). The hydrogeochemical evolution of groundwater in the study area was prepared separately for both dry and rainy periods and was evaluated using a Piper diagram using the concentrations of major cations (Ca, Mg, Na and K) and anions (HCO₃, SO₄, and Cl) in meq/l.

According to the Piper diagrams, in the study area Ca-Mg-HCO₃ and Mg-Ca-HCO₃ as the dominant water types were observed in both the dry and wet seasons (Figure 2). Ca-Mg-HCO₃ and Mg-Ca-HCO₃ facies represented 53.33% and 46.66% of the total water samples, respectively (Table 4; Figure 2). These water types originated from the dissolution of limestone and dolomitic limestone in the aquifer material.

Some diagrams have been proposed by Gibbs (1970) to determine groundwater chemistry, rock–water interaction in aquifers, and sedimentation mechanism in groundwater. There are two types of diagrams. The parameters used in these diagrams are dominant anions (Gibbs Ratio I) and cations (Gibbs Ratio II) and TDS values. These diagrams are extensively used to determine the chemical composition of the water in relation to processes such as precipitation, rock and evaporation dominance.

In this study, Gibbs Ratio I values varied from 0.01 to 0.04 with an average value of 0.02, and Gibbs Ratio II values varied from 0.08 to 0.52 with an average value of 0.24 in the dry season. Gibbs Ratio I values varied from 0.01 to 0.05 with an average value of 0.02, and Gibbs Ratio II values varied from 0.06 to 0.53 with an average value of 0.25 in the wet season. According to the Gibbs diagrams, water samples from dry and wet seasons fall in the rock dominance zone (Figure 3). The diagram indicates that chemical weathering of the rock-forming minerals is the main process through which ions are contributed to the groundwater.

In this section of the study, the water quality assessments were carried out to identify its suitability for drinking, irrigation, and industrial purposes.

Water quality index (WQI) is defined as a method to determine the influence of each water quality parameter on the overall quality of water for human consumption (Varol & Davraz 2015). WQI is an important parameter for determining groundwater quality and its suitability for drinking water. The WQI index is a measure of the composition of water in light of certain defined objectives. In general, the purpose of the index is to compare each of the main parameters of drinking water with the World Health Organization (WHO) and national standards (Yidana et al. 2010). In this study, WQI was computed in three steps. The analysis results were evaluated and compared with WHO (2011) and TSI 266 (2005) (Table 7).

Each of the 21 parameters (pH, TDS, Ca, Mg, Na, K, HCO₃, Cl, SO₄, NO₂, NO₃, NH₄, As, Cr,) has been assigned a weight (wi) according to its relative importance for drinking water quality (Table 7). Greater weight is assigned to parameters which have critical health effects and whose presence above certain critical concentration limits could limit the usability of the resource for domestic purposes (Yidana et al. 2010). Therefore in this study, the maximum weight of 5 has been assigned to parameters including TDS, Cl, SO₄, NO₃, NO₂, NH₄, As and Cr due to their major importance for water quality assessment (Srinivasamoorthy et al. 2008). The minimum weight (1) has been assigned to the parameter HCO₃ due to its having the least importance for water quality assessment. Other parameters, including pH, Ca, Mg, Na, and K, were assigned to weights between 1 and 5 depending on their importance in water quality determination.

Finally, the WQI values obtained from the calculations were evaluated together with the water quality types in Table 8. In addition, the WQI values for each sample and their types are shown in Table 9. According to Table 9, the calculated WQI values ranged from 113.15 to 177.67 and from 103.84 to 204.00 for dry season and wet seasons, respectively. In addition, during the dry season, 100% of groundwater samples represent ‘Poor water’; during the wet season 93.33% of groundwater samples represent ‘Poor water’ and 6.66% indicate ‘Very Poor water’. This indicates that the quality of wet season samples is very low. So, the increase of ion concentrations is related to infiltration of rainwater in the farmland.

In many parts of the world, groundwater is exposed to pollution from fertilizers and pesticides due to human activities, mostly involving intensive agriculture and urbanization. Fertilizer usage varies according to soil structure, climate and plant species and nowadays, the most widely used fertilizers are nitrogen and its derivatives urea, phosphate and potassium fertilizers. One of the most common ways in which nitrogen and its derivatives pass into groundwater is leaching from agricultural lands. Accordingly, the nitrate level in the groundwater is greatly affected by fertilizer quantity and types (Rahmati et al. 2015). Chemical or artificial fertilizers contain heavy metals and metalloids at lower amounts compared with nitrogen and its derivatives. The continuous application of these fertilizers over the last fifty years may have contributed to increasing heavy metals and metalloids in soil and groundwater (Jayasumana et al. 2015).

In addition, chemicals associated with the use of excessive fertilizers and pesticides may cause negative effects on human health. Notably, excess nitrate in the drinking water can cause health risks such as methemoglobin, which depletes blood oxygen levels in infants, causes gastric problems in adults, decreased functioning of the thyroid gland, and cancer due to formation of nitrosamines, as well as multiple sclerosis (Zhaia et al. 2017). Other agrochemicals, such as arsenic (As), cadmium (Cd), chromium (Cr), copper (Cu), nickel (Ni) and lead (Pb), have toxic properties and pollute groundwater. Phosphorus-containing artificial fertilizers are the main source of agricultural As pollution in groundwater (Sönmez et al. 2008). Triple Super Phosphate (TSP) type phosphate fertilizers are one of the main sources of As. The amount of As in the TSP is 15 times greater than the insecticide Dimethoate, which has contained the highest level of As among insecticides. Arsenic in soil can contaminate drinking water and food crops including vegetables, fruits and field crops (Jayasumana et al. 2015). Groundwaters containing As considerably affect human health and cause fatal consequences due to their carcinogenic effects. Epidemiologic studies have shown that As exposure is related to skin, liver, bladder, and other cancers (Lin et al. 2016).

The study area is an important region of intensive agricultural activity, particularly when it comes to greenhousing (Figure 4; ADPFAL 2011; APESR 2012). Agricultural products and cultivated areas in the region are given in Table 10 (ADPFAL 2016). The use of artificial fertilizers is very high in the study area. The fertilizer types and consumption quantities in the study area are given in Table 11 (ADPFAL 2016). According to Table 11, nitrogen and its derivatives urea, phosphate and potassium are the main groups of chemical fertilizers used in Korkuteli district.

WHO has specified the maximum safe nitrite, nitrate and ammonium concentration in drinking water to be 0.2, 50 and 1.5 mg/L, respectively (WHO 2011). Nitrate concentrations exceeding the 50 mg/l limit value in drinking water can endanger human health, especially in infants (Vidyalakshmi et al. 2013). According to the analysis results, nitrate concentrations in the study area exceed the limit value only in the wet season (K2, K5 and K7). In addition, a nitrate level of 3 mg/L or greater indicates contamination. Water with a nitrate concentration of greater than 10 mg/L should not be used to prepare infant formula or other foods or be given to a child younger than 1 year to drink (Greer & Shannon 2005; PWP 2009). As water in the study area is used by people of all ages living in the region, this underlines the importance of this study. The analysis results showed that all waters in the study area both in dry and wet seasons were contaminated and these waters are not suitable for drinking owing to their nitrate concentrations (Table 3).

In addition, arsenic concentrations in all samples exceeded the threshold value determined by WHO (2011) in both periods. Cr concentrations exceeded the limit values at K1, K4 and K5 samples in the dry season and exceeded the limit values at K1, K2, K4, K5, K7, K11 and K12 samples in the wet season. In addition, Ni (K7 and K12) and Pb acceptable concentrations (K1, K7 and K12) were determined to be exceeded only during the wet season. Accordingly, it is not appropriate to use as drinking water according to the heavy metal analysis results (Table 3).

The sodium adsorption ratio (SAR), permeability index (PI), sodium percentage (Na %), magnesium hazard (MH), and residual sodium carbonate (RSC) indices were used to determine the suitability of groundwater for agricultural irrigation and the results are indicated in Table 12.