Introduction

Fluoride (F−) is an important nourishing constituent for health which is present in groundwater as free fluoride ions. It begins to affect human health as its concentration exceeds World Health Organization (WHO) guideline for F− in drinking water (Saxena and Ahmed 2003). When F− exceeds value of 1.5 mg/L WHO guideline for F− in drinking water, it results in dental and skeletal fluorosis, crippled limbs, rheumatic pain, the calcification of ligaments stiffness, and the rigidity of the joints (Ayoob and Gupta 2006). It can adversely affect the kidneys, liver, gastrointestinal tract as well as the nervous, immune, and reproductive systems disorders (Ibrahim 2011).

The problem of fluorosis develops extensively in many areas of the world, such as in Argentina, Bangladesh, China (Guo et al. 2007), India (Jacks et al. 2005), Japan, Morocco, New Zealand, South African Countries, Thailand and Middle East countries, and USA (Rafique et al. 2009). In Pakistan, the high F− concentrations in groundwater have been reported from Karachi (Siddique et al. 2006), Lahore (Farooqi et al. 2007), Peshawar (Shah and Danishwar 2003), and Thar Desert (Rafique et al. 2008).

Fluoride has affinities for rocks and occurs in F−-bearing minerals that lead to transportation of substantial amount of F− into the groundwater (Lunshan et al. 2007). Hence, groundwater with high F− concentrations is probable in areas where F−-bearing minerals are abundant (Shaji et al. 2007). Common natural sources of F− in groundwater are the dissolution of F−-bearing minerals like fluorspar and fluorapatite and some micas weathered from silicates (Guo et al. 2007). The dissolution of F−-bearing minerals not only produces kaolinite but also discharges cations, such as Ca2+, Mg2+, Na+ and K+, causing an increase in TDS which indirectly affect F− concentration in groundwater (Li et al. 2015). Fluoride-bearing minerals occur in deep soils, when soluble F− concentration is higher in deep soil than in surface soils, and it follows the geological source of F− contamination (Jayawardana et al. 2012).

Higher concentrations of F− usually occur due to water–rock interaction of fluorite, fluorapatite, and cryolite or any F−-bearing mineral which results in Ca2+ depletion of groundwater due to common ion effect (Chae et al. 2007). The levels of dissolved F− are generally linked with the pH value and water type, although the depth and temperature of groundwater and climatic factors like evaporation/precipitation also play significant roles (Rafique et al. 2015). Ion exchange in which OH− replaces F− in groundwater on certain clay minerals or weathered micas (Rafique et al. 2009). Higher F− concentrations are associated with both NaCl- and NaHCO3-type waters, and lower F− concentrations are linked with CaHCO3- and CaMgCl-type waters, where greater Ca2+ levels lead to precipitation of fluorite (Rafique et al. 2015). The clay proportion of soil has positive relation with F− concentrations in soil than silt and negative relation with sand (Farooqi et al. 2009).

The groundwater flow rate is higher in areas nearby river, residence time is lower, and there is a less time of contact between water and F−-bearing rocks. This led to lower concentrations of F− in the groundwater. However, in the central basin areas, groundwater had moved a longer distance and thus had a higher residence time (Li et al. 2015). This indicates that groundwater is more enriched in F− along the flow path away from river. As a result, F− concentrations got increased (Basharat 2012).

Gujrat has been famous for the production of low-temperature pottery and ceramic goods. Gujrat is also known in the world for its clay which is used to produce quality pottery since ages. The clay content of soil of Gujrat varied along with increasing distance from River Chenab as subdistrict Gujrat is located near River Chenab. Sandy loams are more abundant near river that has negative relation with F− concentrations (Farooqi et al. 2009), while moving away from river, silty loams and clayey soil are evident that has capacity to hold higher F− concentrations (Ashraf and Ahmad 2008).

Groundwater in semiarid areas surrounding Gujrat is known to have F− concentrations that go above the guideline for drinking water 1.5 mg/L set by WHO (Rafique et al. 2008). The percentile F− contribution in lowering the water quality of areas around Gujrat, i.e., Sargodha (7 %), Faisalabad (16 %), Lahore (18 %) and Kasur (18 %) is comparatively high than Gujrat (2 %) which indicate certain geological factors that are controlling F− enrichment in groundwater of Punjab (Pakistan Council of Research in Water Resources 2010).

The present study evaluates groundwater quality in a region where screening by Pakistan Council of Research in Water Resources (PCRWR) reported fluoride in groundwater, and the area is known for its high class of clay minerals and was undertaken to define the sources of dissolved fluoride and the geochemical processes that influence its occurrence. The aim is to identify geochemical relationships that influence in understanding the origin of high fluoride established on measures of the overall water quality. This study was the first attempt to analyze F− and to predict its geochemistry controlling the F−concentrations.

Materials and methods

Study area

Location and hydrometeorology

District Gujrat (Fig. 1) including an area of approximately 1254 km2 was selected as the study area for this study. It is located in the semiarid zone of Punjab, Pakistan between 32.5738N and 74.0789E. For political and administrative purposes, the whole district is divided into three subdistricts (Kharian, Sarai Alamgir, and Gujrat); subdistrict Gujrat was selected for study. Climate of the study area is tropical with an average annual rainfall of 670 mm, 70 % of which occurs between June and September. The wet season starts from end June and lasts up to the start of November, whereas the rest of the months are generally dry. The temperature rises to maximum of 48.2 °C in summer and drops to a minimum of 2 °C during winter (Gujrat city profile). The main rivers of the study area are River Jhelum and River Chenab. The study area falls within the area of upper Chaj Doab and sandwiched between these two rivers but more influenced by River Chenab, having a distance of 12 km. Gujrat is populated with 270,000 people and majorly drained by River Chenab (Ashraf and Ahmad 2008).

Fig. 1
figure 1

Map of study area with location of groundwater and soil sampling points, circles represent groundwater sampling points, and triangles represents soil sampling points

Full size image

Geological and hydrogeological setting

The soil loams of the Gujrat are composed of silty and sandy alluvial substantial (Ashraf and Ahmad 2008). Punjab Province is drained by five major rivers that flow down from high range mountains of Himalaya. The sediments carried along with these rivers water from the massive alluvial basin of the aquifer that contains material washed down from the Himalaya Mountains were set down. Study area contains more than 500-m-thick Quaternary deposition (Ashraf and Ahmad 2008). The alluvium material consists of sand and silt and minute amounts of gravel and clay, set down by the Indus River by its present and ancestral offshoots. In harmony with their mode of deposition by large continually flowing rivers, the alluvial depositions are heterogeneous by nature, and individual layers have restricted horizontal and vertical continuity (Greenman et al. 1967).

In general, unconfined aquifers and some confined aquifers are present in the study area (Basharat 2012). Groundwater table is declining with increasing population and over abstraction (81 % annual abstraction) of groundwater than recharge (Mahmood et al. 2011). Aesthetically acceptable water starts below the depth of 50 feet (Ashraf and Ahmad 2008). The direction of groundwater flow is from recharge zone, foothills of Himalaya toward discharge plain area of Gujrat. Flow direction of groundwater in the study area also got affected by River Chenab’s fluctuating flow during rainy and dry season (Mahmood et al. 2011).

Sampling

Groundwater samples were collected from five villages of subdistrict Gujrat during October 2014. To cover entire subdistrict Gujrat, five random villages were selected (Fig. 1). From these villages, in total 70 groundwater samples were collected. In Dhool Khurd (DK), Kalra Khasa (KK), Ghazi Chak (GC), and Kula Chor (KC), 15 groundwater samples were collected from each village at the depth of 14–105, 7–105, 8–122, and 11–75 m, respectively. From village Chhani Nikoani (CN), 10 groundwater samples were collected at the depth of 20–90 m because the area is sparsely populated compared to other villages. From each village Dhool Khurd (DK), Kalra Khasa (KK), Ghazi Chak (GC), and Kula Chor (KC), total 10 soil samples (each composite of 3) were collected in polyethylene zipper bags (five from surface and five from subsurface), while from village Chhani Nikoani (CN), four soil samples were collected (two from surface and two from subsurface). Soil sampler (hand Augur) was used for surface (0–5 cm) and subsurface (35–40 cm) soil samples collection.

Groundwater samples were collected in 300-mL polyethylene bottles that had been pre-washed with dilute (1 %) nitric acid according to standard methods (Greenberg et al. 1998). Prior to sample collection the pump was flushed for 5 min. The bottles were first rinsed with groundwater and then completely filled and sealed with caps to avoid the oxidation of redox-sensitive constituents. At each sampling point, two samples were collected. One was preserved by acidifying to the pH value of <2 with ultra-pure HNO3 for the later analysis of cations, while the other was stored without acid for the later analysis of anions in the environmental geochemistry laboratory. The exact sampling coordinates were identified using a GPS unit.

Analytical techniques

Temperature and pH were measured by digital pH meter, oxidation reduction potential (Eh) by Eh meter, conductivity, and total dissolved solids (TDS) were measured in the field using digital meter. In laboratory, water samples were analyzed for alkalinity as HCO3 − by acid–base titration, sulfates by titration with barium chloride method, chlorides by silver nitrate titrimetric method, nitrates by phenol disulphonic method, and phosphates by ammonium molybdate method (APHA 1995). Similarly, acidified samples were analyzed for major cations (Na+, K+, Ca2+, Mg2+) using Sequential Flame Atomic Absorption Spectrophotometer (Varian Spectra AA-240) under standard operating conditions having r > 0.99.

Fluoride was analyzed by using fluoride ion-selective electrode Orion 9609 with ion analyzer EA 940 by diluting water sample with total ionic strength adjustment buffer (TISAB) in ratio 1:1 to avoid dissociation of fluoride complexes with other metals present in samples that can hinder the F analysis (Greenberg et al. 1998). High levels of precision and accuracy were maintained during all phases of laboratory analysis, which included blank and spiked samples to ensure the reliability of each method. The ionic balance error (IBE) values were within an acceptable limit of ±5 % (Domenico 1990; Mandel 1981) for all samples.

Soil samples were grounded followed by sieving through 200-mm mesh. The pH value of soil samples was determined with pH meter in ratio of 1:5 soil to water, while the soluble fluoride concentrations were determined in same ratio with ion-selective meter same as that of water. Minerals were identified by X-ray diffraction spectroscopy (XRD).

All scientific and statistical computations were made by using IBM SPSS statistics version 20, XLSTAT Pro v7 5.2, PHREEQC, and Arc GIS 10 software. Multivariate statistical methods were used for interpretation and classification of large environmental datasets. The normality of data under variables was checked by Kolmogorov–Smirnov test. Principal component analysis (PCA) was employed on the dataset to reduce the analytical data collected from each village. For mineral speciation, saturation indices were used to analyze the degree of equilibrium between mineral species and water.

Reagents, procedural blanks, and samples were measured three times, and the average of three values was used. For QC/QA, Standard Reference Materials (JSL-1 & JSD-1) was used and the recovery was found 98 % using the same procedure and reagents. Arsenic was measured by using the atomic absorption graphite technique at Geoscience Advanced Research Laboratories.

Results and discussion

Overall groundwater quality

Table 1 shows the chemical parameters measured in the groundwater samples collected at the CN, DK, KK, GC and KC. The pH of the samples varied from 7.8 to 8.7, 7.1 to 8.9, 7.2 to 8.5, 7.0 to 8.0, and 7.1 to 8.6 at CN, DK, KK, GC, and KC, respectively. Since pH usually has no direct impact on consumers, no health-based guideline value has been proposed for pH by WHO (WHO 2011). The TDS of the samples varied from 342 to 1528, 107 to 1152, 127 to 1152, 87 to 1246, and 134 to 2238 at CN, DK, KK, GC, and KC, respectively. The palatability of water with a total TDS level of less than about 600 mg/L is generally considered to be good; drinking water becomes significantly and increasingly unpalatable at TDS levels greater than about 1000 mg/L (WHO 2011).

Table 1 Basic statistical parameters in groundwater of subdistrict Gujrat
Full size table

In almost all of the groundwater samples, the EC and TDS measurements were higher than the WHO guidelines. The EC measurements ranged from 0.13 to 3.34 mS/cm, and the TDS measurements ranged from 87 to 2238 mg/L. Soluble electrolytes and high salinity are generally the main cause of elevated EC and TDS in groundwater (Brahman et al. 2013). It is clear that much of the groundwater in this region is unacceptable for drinking in terms of high EC and TDS.

Sodium was the major cation, with concentration ranging from 73.9 to 529 mg/L. The average concentration of Na+ in groundwater samples from Chhani Nikoani (398 mg/L) and Kula Chor (291 mg/L) was higher than WHO guideline (200 mg/L), while the average concentration of Na+ in samples of other villages was within the WHO guideline. The Na+ naturally occurs in water; sometimes, mineral deposits enhance its concentration in groundwater to unacceptable levels (Rafique et al. 2015). Potassium concentration varied from 0.04 to 52.9 mg/L with 95 % of the samples were below the WHO guideline for K+ in drinking water (12 mg/L) and 5 % of samples contained K+ higher than the WHO guideline. The Ca2+ concentration ranged from 1.12 to 256 mg/L in groundwater of present study area. Most of the samples (90 %) had Ca2+ concentrations under the WHO guideline (100–150 mg/L), while few (10 %) had Ca2+ values above the WHO guideline. Magnesium concentration ranged from 0.38 to 98.4 mg/L. Around 8 % of the samples had Mg2+ concentration above WHO guideline, while 90 % had Mg2+ concentration below WHO guideline. Ca2+ and Mg2+ are relatively lower in the groundwater of the study area. Chloride (Cl−) and HCO −3 were the dominant anions in groundwater samples. Cl− ranged from 200 to 950 mg/L and HCO −3 ranged from 17.8–958 mg/L. Fifty percent of the groundwater samples contained Cl− above the WHO guideline. The HCO −3 concentrations were more than WHO guideline (500 mg/L) in 33 % of samples. Higher values of Cl− might be due to the presence of geologic deposits containing halite in the groundwater of study area (Srinivasamoorthy et al. 2008). And high concentrations of HCO −3 in few samples are linked to pH and weathering of carbonaceous minerals (Kumar 2014). Sulfate (SO 2−4 ) concentration ranged from 33.3 to 133 mg/L.

Data were plotted on diagrams developed by Gibbs (1970) for the purpose of understanding the dominant control(s) on water quality (Fig. 2a, b; Table 1). These plots suggest 60 % contribution of rock weathering and 40 % evaporation in controlling groundwater chemistry for an area which is semiarid. It signifies the water–rock interaction phenomena which got affected by residence time. The dominance of NaCl-type groundwater is the resultant of evaporation, which assists to increase the ionic strength of the groundwater. Primarily, it also increases the minerals solubility in groundwater, including those that release fluoride into groundwater. However, as the dissolved species concentrations continue to increase, the groundwater becomes supersaturated with respect to minerals that are less soluble, which ultimately precipitate out from groundwater (Rafique et al. 2015).

Fig. 2
figure 2

Gibbs plot which illustrate the dominant processes that control the chemistry of groundwater samples, a for cations, b for anions. Symbols are as described in Fig. 3

Full size image

Chadha (1999) proposed a hydrogeochemical classification scheme for water by plotting the difference between (CO3 + HCO3) and (Cl + SO4) concentrations in meq/L expressed as percent versus (Ca + Mg) and (Na + K) concentrations in meq/L also expressed as percent (Chadha 1999). This classification is a modified version of the Piper diagram (Piper 1944). The resulting diagram has four fields, representing four types of hydrogeochemical processes (Fig. 3). Nearly, 43 % (N = 30) of the samples collected from the study area belong to NaCl-type water affected by evaporation. Rest 21 % (N = 15) belong to CaMgCl-type waters, 14 % (N = 20) with are NaHCO3-type water, and 16 % (N = 11) classified as CaHCO3-type waters (Fig. 3). Water quality produced by either base ion exchange (NaHCO3) or reverse ion exchange (CaMgCl) type waters would only occur where the aquifer includes the proper clay minerals. Carbonate minerals are prevalent within the study area (Naseem et al. 2010). Moreover, high Na/Ca ratio for Field 4 samples means that Na increased and calcium decreased through base ion exchange or calcite precipitation. The relatively low Ca/(Ca + SO4) ratios and high sodium for Field 3 and Field 4 samples support the conclusion that ion exchange is influencing groundwater chemistry at the locations where these samples were collected (Edmunds and Smedley 2005; Chae et al. 2006; Guo et al. 2007). The NaHCO3-type water that is undersaturated with respect to calcite which is due to the exchange of calcium and magnesium ions in solution for sodium on clays or other secondary minerals.

Fig. 3
figure 3

Chadha diagram showing groundwater samples classification taken from the study area (after Chadha 1999). Field 1: CaHCO3-type waters, reflecting recharge and weathering. Field 2: CaMgCl-type waters, reflecting reverse ion exchange. Field 3: NaCl-type waters, reflecting evaporation or mixing with seawater. Field 4: NaHCO3-type waters, reflecting base ion exchange (Symbols same as used in Fig. 2)

Full size image

Spatial distribution of fluoride and role of geochemical control

The F− concentrations from study area Gujrat varied from 0.3 to 6.4 mg/L. The mean F− concentration in groundwater samples from village Chhani Nikoani (1.8 mg/L) and Dhool Khurd (2.2 mg/L) was exceeding WHO drinking water standard of 1.5 mg/L. According to classification used by (Nagendra Rao 2003), about 48 % of groundwater samples from village Chhani Nikoani and Dhool Khurd had F− concentrations below WHO standard, while 32 % of samples were within range of 1.6–3.0 mg/L which can cause dental fluorosis, and remaining 20 % of samples had F− concentrations of 3.1–7.0 mg/L which can cause skeletal fluorosis in humans (Nagendra Rao 2003). The mean F− concentration in groundwater samples from village Kalra Khasa (1.3 mg/L), Ghazi Chak (0.9 mg/L), and Kula Chor (1.45 mg/L) was mostly below WHO standard. Overall, about 73 % of samples had F− concentrations below WHO standards, while remaining 22 % of samples from these villages were within range of 1.6–3.0 mg/L which can cause dental fluorosis, and remaining 5 % of samples were within the range of 3.1–7.0 mg/L, which lead to skeletal fluorosis (Nagendra Rao 2003).

A slight positive correlation (R 2 = 0.30) between F− concentration and TDS was observed (Fig. 4a). In the study area, high TDS values were observed in few samples from village Chhani Nikoani, Dhool Khurd, and Kula Chor with comparatively high F− concentrations, while village Kalra Khasa and Ghazi Chak had low TDS values with comparatively lower concentrations of F− to other villages. Similar results of correlation between F− and TDS were reported from Nagar Parkar (Rafique et al. 2009).

Fig. 4
figure 4

Relationship of F− with a TDS, b with pH+ in groundwater of study area Gujrat. Symbol for different villages is as described in Fig. 2

Full size image

Figure 3 shows that higher F− concentrations are associated with both NaCl (Field 3) and NaHCO3 (Field 4), and the samples plotted in Fields 3 and 4 of Fig. 3 are relatively depleted in calcium and show evidence of base ion exchange (as described in “Overall groundwater quality” section), and lower F− concentrations are linked with CaHCO3- and CaMgCl-type waters, where greater Ca2+ levels lead to precipitation of fluorite (Rafique et al. 2015).

The NaHCO3-type water is result of the removal of Ca2+ and Mg2+ from groundwater by Na+ ion exchange with F− on clay minerals. This shows cation exchange reaction is important in controlling F− chemistry in groundwater (Chae et al. 2007; Rafique et al. 2015) where Ca2+ has been replaced with Na+ in waters containing high F−. The dominance of NaCl-type groundwater (Fig. 3) is due to evaporation, which may increase the ionic strength of the water. The concentration of dissolved mineral got increased because of evaporation, and groundwater becomes saturated with respect to the least soluble minerals, which ultimately precipitate out from groundwater. Hence NaCl water type can affect F− concentration in groundwater due to minerals precipitation (Rafique et al. 2015). In the current study, NaCl type of water contains the highest concentration of F− as indicated before in Fig. 3.

Fluoride-rich groundwater in many regions of the world is often of NaHCO3 type with low concentrations of Ca2+ and alkaline pH values (Li et al. 2011). The present study shows positive correlation with pH A positive correlation (R 2 = 0.85) between F− concentration and pH was observed (Fig. 4b). Similar positive correlation of pH with F− concentration was observed in different regions of the world (Valenzuela-Vasquez et al. 2006; Farooqi et al. 2007; Rafique et al. 2009; Wang et al. 2009; Mamatha and Rao 2010; Li et al. 2011). Groundwater samples from present study area have pH range from 7.1 to 8.6. Water with pH value <7 rarely have detectable F− concentration (Genxu and Guodong 2001). The F− and OH− have similar sizes of their ionic radii and could play substitute role for each other in minerals. Clay particles have ability to adsorb F− ions, with increasing pH the F− ions got displaced by OH− ions, ultimately releasing F− ions into groundwater (Genxu and Guodong 2001). This relationship might explain the high F− concentrations in samples from village Chhani Nikoani and Dhool Khurd with pH value from 7.8 to 8.9.

High pH would also favor calcite precipitation (thereby increasing fluorite solubility). The solubility of calcite (CaCO3) is of particular importance, because it shares a common ion with fluorite (CaF2), which normally controls the upper limit of dissolved fluoride. Figure 5a–d plots the calcite and fluorite saturation indices. This plot shows that groundwater in the study area has evolved to the point where all samples with high fluoride are supersaturated with respect to calcite, which should favor the removal of calcium from solution and allow fluoride concentrations to increase (Kundu et al. 2001). Assuming that NaHCO3-type waters were produced by base ion exchange (the removal of Ca and Mg from solution in exchange for Na on clay minerals), the net result is the same as described above for NaCl-type waters. As Ca2+ concentrations are decreased, the groundwater is able to contain higher levels of fluoride before saturation with respect to fluorite is reached. This interpretation is supported by the fact that fluoride concentrations are generally lower in the CaHCO3- and CaMgCl-type waters, where higher Ca2+ levels lead to fluorite precipitation.

Fig. 5
figure 5

Relationship between various chemical components of groundwater samples. a Fluorite and calcite saturation index, b fluorite and dolomite saturation index, c fluorite saturation index and F−, d fluorite saturation index and Ca2+ −

Full size image

Statistical verification of correlation among physicochemical parameters

The correlation matrix for physicochemical parameters is given in Table 2. For statistical verification of these correlations, PCA was applied on groundwater dataset. The PCA plot is powerful recognition technique that explains characteristics of samples and distribution of intercorrelated parameters (Brahman et al. 2013). PCA plot was drawn for dataset of present study to verify the intercorrelation with physicochemical parameters. The PCA plot with two component axes explained 71.4 % of variance in data set. The first component axes explained 50.86 % of variance in data, while second component axes explained 20.54 % of variance in data. Parameters distribution in upper right side of plot was more enriched with Na+, K+, F−, pH, EC, and TDS which indicate geological sources of these components (Fig. 6). High F− groundwater is generally characterized by Na+ rich and Ca2+ poor with comparatively high pH values and HCO −3 concentrations (Li et al. 2015), as a result of natural processes of evaporation, dissolution of minerals, and cation exchange. The results of present study show high values for Na+, TDS, EC, pH and HCO −3 which might lead to F−-rich groundwater. The first component axes showed negative loadings for Ca2+, Mg2+, and Cl− which indicate climatic effects. The current study was observed to have lower values for Ca2+ and Mg2+ which lead to F−-rich groundwater (Li et al. 2011). Water type is very important in controlling F− enrichment; NaHCO3 type support high concentrations of F− in groundwater (Li et al. 2011). Figure 6 illustrates the Na+ value was more inclined toward Cl− than toward HCO −3 , indicating dominance of NaCl water type that can limit F− concentrations in groundwater. NaCl water type can affect F− concentrations in groundwater due to minerals precipitation (Rafique et al. 2015). The second component axes of PCA showed positive loadings for all parameters which indicate natural source, i.e., mineral weathering as source of contamination (Mamatha and Rao 2010).

Table 2 Correlation matrix of physicochemical parameters in groundwater of study area Gujrat
Full size table
Fig. 6
figure 6

PCA plot for distribution of F− and physicochemical parameters in groundwater of Gujrat

Full size image

Spatial distribution of fluoride and role of local geology

The spatial distribution of categorized F− concentrations along with River Chenab is presented in Fig. 7. The F− concentrations in groundwater of study area were decreased while moving away from the River Chenab. Northwestern villages Chhani Nikoani and Dhool Khurd had more F− concentrations, located away from River Chenab. While southwestern villages Kalra Khasa and Ghazi Chak were observed to have low F− concentrations, located nearer to River Chenab. A northwestern village also observed to have comparatively more F− concentrations, which was located away from River Chenab. The higher F− concentrations in groundwater were analyzed in villages away from River Chenab, while villages nearer to River Chenab had lower concentrations. This obvious spatial distribution patterns reflect the geochemical controls on its occurrence and are clearly linked to the interpolation of groundwater flow regime, high rate of groundwater flow made harder to lose F− from its mineral rock (Kausar et al. 2003; Chae et al. 2007; Currell et al. 2011; Dey et al. 2012; Li et al. 2015). Sites nearer to River Chenab had high groundwater flow rate, lower residence time, and less time for contact between water- and F− bearing minerals. Hence, decreased concentrations of F− were observed in nearby sites of river. However, when moved away from River, residence time was increased which led to increased F− concentrations in groundwater (Li et al. 2015). Similar trend of F− concentrations with flow regime of groundwater was reported from many regions of the world (Suthar et al. 2008; Rango et al. 2009; Naseem et al. 2010; Yitbarek et al. 2012; Su et al. 2013; Li et al. 2015).

Fig. 7
figure 7

Spatial distribution of F− concentrations in groundwater samples from study area Gujrat

Full size image

XRD and thin sections were used to analyze minerals present in the soils of the study area (Fig. 8a, b). The results inferred that F−-bearing minerals were not identified by XRD pattern. The main mineral was quartz, with muscovite and biotite identified in three villages with relatively higher F− concentrations and not detected in Ghazi Chak and Kula Chor. Muscovite and biotite might be the source of F− in these villages by the following reaction (1 and 2). High alkalinity values and Na+ in groundwater have more OH− ions which can replace with F− content of F−-bearing minerals, increasing its concentrations in groundwater (Guo et al. 2007).

Fig. 8
figure 8

Relationship between a F− and pH, b F− and sand %, c F− and silt %, d F− and clay % of soil samples. Symbols are same as used in Fig. 3

Full size image
$$ {\text{KAl}}_{2} \left[ {{\text{AlSi}}_{ 3} {\text{O}}_{10} } \right]{\text{F}}_{2} + 2{\text{OH}}^{ - } {\text{KAl}}_{2} \left[ {{\text{AlSi}}_{ 3} {\text{O}}_{10} } \right]\left[ {\text{OH}} \right]_{2} + 2{\text{F}}^{ - } $$
(1)
$$ {\text{KMg}}_{3} \left[ {{\text{AlSi}}_{ 3} {\text{O}}_{10} } \right]{\text{F}}_{2} + 2{\text{OH}}^{ - } {\text{KMg}}\left[ {{\text{AlSi}}_{ 3} {\text{O}}_{10} } \right]\left[ {\text{OH}} \right]_{2} + 2{\text{F}}^{ - } $$
(2)

Weathering of muscovite and biotite was major source of F− content in groundwater other than geochemical factors (Kumar et al. 2015). The muscovite and biotite which is considered as the major source of F− in groundwater is detected by the petrographic thin sections (Fig. 8a) and by XRD (Fig. 8b).

Although the F− concentrations exceeded the WHO guideline in 23 % of samples, the concentrations are lower as compared to other reported areas in Pakistan (Table 3). Water-soluble F− concentration was measured in soil samples. The descriptive statistical summary of surface and subsurface soil samples from each village is given in Table 4. The mean value of F− in surface soil (0–5 cm) samples from all villages was 1.29 mg/Kg, while in subsurface soil (35–40 cm) samples was 1.53 mg/Kg. Water-soluble F− from all villages also showed a positive correlation with pH (R 2 = 0.64, Fig. 9a) as that in groundwater. Soil pH is an important factor in controlling F− concentrations (Loganathan et al. 2003) The F− enrichment of groundwater occurs with increasing alkalinity of soil and releases F− into water as result of ion exchange reaction (Sheikhy et al. 2014). Increasing alkalinity of soil causes desorption of Ca2+, and data points shifted to right after ion exchange reaction with Na+.

Table 3 Reported F− concentrations in different areas of Pakistan
Full size table
Table 4 Statistical summary of pH, EC, F− and soil fractions from surface and subsurface soil samples from Gujrat
Full size table
Fig. 9
figure 9

Combined trend of F− concentration in groundwater and soil samples taken from same points

Full size image
$$ 1/2{\text{ Ca}}^{2 + } - {\text{Clay}} + {\text{Na}}^{ + } \to 1/2{\text{ Ca}}^{2 + } + {\text{Na}} - {\text{Clay}} $$
(3)

Mostly, sandy loams were found in all villages except Chhani Nikoani which is silty loam. Different soil fractions have different effects on F− concentrations. Fluoride showed a strong negative correlation with sand (R 2 = −0.58, Fig. 9b), weak correlation with silt (R 2 = 0.08, Fig. 9c), and clay fraction was positively correlated (R 2 = 0.35) with F− concentrations (Fig. 9d). This positive correlation indicates that clay particles, which occur at low concentrations in sandy loam soils, have great capability to bound F− on their surface. The clay content was evenly distributed in all villages except Kula Chor which have less percentage of clay particles in its soil samples.

A negative correlation was observed between groundwater F− concentrations and soil F− concentrations from same study area (Fig. 10). More F− concentration in soil samples as compared to water samples from village Chhani Nikoani, Dhool Khurd, and Ghazi Chak was observed. This is due to more percentage of clay content in soil of these villages, which have capability to adsorb F− on surface (Guo et al. 2007). Less F− concentration in soil sample as compared to water samples from village Kalra Khasa and Kula Chor was observed. The presence of more percentage of sand particles as compared to silt and clay particles have less holding capability and might result in F− release from soil into groundwater (Li et al. 2011).

Fig. 10
figure 10

Comparison of fluoride concentartions in soil and water of the study area

Full size image

Conclusions

The groundwater occurring in district Gujrat has been found unsuitable (23 %) for drinking as a result of high total dissolved solids and the occurrence of elevated fluoride concentrations; 23 % of the samples collected in this study contain fluoride concentrations that exceed the WHO permissible limit of 1.5 mg/L. This study reveals the fact that fluoride originates from the weathering of minerals in the aquifer matrix and possibly the exchange of OH− for F− on clay minerals and weathered micas under high pH conditions. Rock weathering is the dominant process affecting groundwater quality, of 57 % of the samples, and second process is evaporation. As the ionic strength of groundwater increases through evaporation, minerals are initially more soluble, but finally the solubility limits for non-evaporite minerals are reached, causing precipitation which is best shown by the saturation indices of the samples from the study area which are supersaturated with respect to calcite containing high fluoride, and it is likely that the removal of calcium from solution and permits fluoride concentrations to increase. Some calcium is removed from solution through base ion exchange, thus increasing the fluoride concentrations associated with NaCl- and NaHCO3-type water. Spatial distribution and relatively lower concentrations in the area and presence of clay minerals also suggest the role of local geology and rock–water interaction in controlling the fluoride concentrations.