Comprehensive assessment of groundwater quality using heavy metal pollution indices and geospatial technique: a case study from Wanaparthy watershed of upper Krishna River basin, Telangana, India

The present study is to characterize groundwater quality using heavy metal pollution indices and geospatial variations. A total of 58 samples from hand pump/submersible bore wells were collected from the Wanaparthy watershed of the upper Krishna River basin according to the grid size (5×6 km2). The trace elements’ concentration in groundwater samples are found in the order of Zn (38.67%) > B (32.67%) > Ba (13.59%) > As (8.49%) > Hg (3.71%) > Cr (1.28%) > Ni (0.52%) > Cd (0.47%). Among these trace elements, arsenic (22.4%) and mercury (5.1%) were found above the permissible limits of WHO drinking water guideline values. A positive correlation between TH versus EC/TDS indicates the presence of trace elements due to chemical reaction (rock–water interaction). Arsenic correlation with EC/TDS/TH indicates artificial intervention. Drainage network analysis enumerates high concentration of parameters at near or joining to upper order of drainage system, which might be due to input of runoff water (interaction of variable rocks composition) and later stage infiltration to subsurface and reached to an aquifer. Heavy metal pollution index (HPI) showed 86.2% of samples are in the category of low class, whereas 12.07% of samples fall within medium class. According to metal index (MI) classification, 12.07% samples are in very pure, 24.14% samples are pure, while the remaining 63.79% samples are in the slightly to strongly affect category. This study suggested the main source of trace elements in groundwater might be from the dominant granitoid rocks because the area is mostly devoid of industrialization.


Introduction
Renewable groundwater is found between layers of impermeable rock formations known as an aquifer. Aquifers are connected through joint, crack, fracture, or structurally deformed zone. It is essential to know the portability of groundwater and the amount of recharge and discharge ratio because the future generation depends on how we utilized and managed the aquifer system. In most parts of India, groundwater uses are at its peak, which gives rise to risk due to limited resources of surface water and a highly dense population. Trace elements in groundwater are generally at low concentration in the undisturbed environment due to natural chemical weathering and rock-water interaction (Tiwari et al. 2016;Karbassi et al. 2007). Groundwater pollution is not only influence water quality but also render chaos to a health problem, economic growth, and social life (Singh and Kamal 2017;Marcovecchio et al. 2007). Trace element concentrations in groundwater are very much concerned about the environmental pollution and wellbeing of humans. This is because of trace elements are non-degradable and once it reaches into body system, it accumulates and causes diseases at very low content, and they are highly dangerous (Zakhem et al. 2015;Vinodhini and Narayanan 2008;Lohani et al. Marcovecchio et al. 2007;Momodu and Aayako 2010;Lee et al. 2007). Besides, heavy metals can be defined as a group of elements with an atomic density above 4000 kg/ m 3 (Hutton and Symon 1986). Trace elements in groundwater can be either from natural or anthropogenic processes; whereas natural processes include weathering, chemical reactions of elements, and soil leaching, while anthropogenic are from domestic waste, fertilizers, urbanization, mining, industrial waste, etc. (Agarwal et al. 2013;Zarazua et al. 2006;Reiners et al. 1975). Among the trace elements, few of them such as Ni, Cr, and Zn are essential as micronutrients for biochemical activities in aquatic life though they become highly toxic in higher concentrations (Kumar et al. 2012;Nurnberg 1982). Arsenic is considered highly hazardous due to its toxic and carcinogenic to human health even at low concentrations , and the mineral source is arsenopyrite (FeAsS). Other trace elements B, Ba and Hg are known to less health importance, but still, their presence in groundwater is considered as toxic at low concentration. Review of the literature showed that many researchers had reported the groundwater quality assessment using heavy metal pollution index (HPI) and metal index (MI) to find out the additive result of/trace metals in water for quick analyses of overall groundwater standard for drinking and effect on human health (Abbas et al. 2021;Wu et al. 2021;Panseriya et al. 2020;Singh and Kamal 2017;Tiwari et al. 2015;Zakhem and Hafez 2015;Goher et al. 2014;Protano et al. 2014;Yankey et al. 2013;Prasanna et al. 2012;Giri et al. 2010;Kikuchi et al. 2009;Zhang et al. 2009;Pandey et al. 2009;Prasad and Sangita 2008). This is the continuous study of major ions hydrogeochemical, seasonal variation and health risk assessment (Vaiphei and Kurakalva 2021;Vaiphei et al. 2020). In addition, since, no report has found on groundwater studies on trace elements' distribution in Wanaparthy watershed, Telangana, India, hence, this study will focus on understanding the groundwater quality with respect to trace elements' concentration and distribution. The methods to be followed are identifying individual trace elements' concentration, HPI, MI, and their spatial distribution in the study area.

Study area
Wanaparthy watershed is located in upper Krishna River basin between 16°19ʹ1ʺ to 16°49′53ʺ N latitude and 77°49′21ʺ to 78°12′55ʺ E longitude of Telangana state in southern India (Fig. 1a) covering an area of 1600 km 2 . The geomorphology of the area, North-west and north-eastern region have higher topography, which gradually slopes down towards south-west side (Fig. 1b); the elevation ranges from 309 to 692 m (AMSL) (ESM_Fig. 1). The dominant river system in the area is of dendritic pattern where the higher order streams final reach to major Krishna River. It originates from Western Ghats (Maharashtra) and passes thought the southern part of Telangana, which reaches into Bay of Bengal on the eastern coast (Andhra Pradesh). Krishna River serves as a significant water source for drinking, domestic purposes, and agriculture in the study area. Red soils, lateritic soils, and black cotton soils are the main dominant soil types. The major cultivation in the study is rice, maize, sugarcane, bajra, cotton, mango, grape, lemon, papaya, pomegranate, etc. whereas there is less predominant of mining and industries. The area comes under a tropical semi-arid climate with temperature ranging from between 16.9 °C to 42 °C. The dominant rock formation of the study area belongs to the peninsular gneissic complex and Dharwar group of Precambrian age consisting of grey biotite granite, migmatite, leucogranite, alkali feldspar granite, pink biotite granite, banded migmatite quartzite, amphibolite etc. There is migmatite intrusion which cut across the massive peninsular gneissic complex in the north-west to north-east direction (

Materials and methods
Fifty-eight water samples were collected from hand pump/ submersible bore wells as per the grid (5×6 km 2 ) drawn for the study area during March 2019 from Wanaparthy watershed. Samples were preferably collected either from the center of the village or center of the prepared grid map. Except a few samples from agriculture field where groundwater source point is not accessible within the village area. Before collecting the samples, water is drawn out from the hand pump or submersible bore wells for about 10 min to confiscate out the stagnant water. Using pre-cleaned 500 mL and 60 mL polypropylene bottles with milli-q water, two sets of samples were collected separately. The 500 mL water samples were utilized for titration purpose to measure a total harness using EDTA (Ethylenediamine tetra acetic acid), ammonia buffer, Eriochrome Black-T (EBT) and Total Alkalinity (using sulphuric acid, phenolphthalein indicator and methyl orange), following Apha 2005 standard procedure. The 60 mL containers were acidified with 2-3 mL of conc. HNO 3 to keep dissolved and stabilize the trace elements (Kumar et al. 2012). Quality parameters such as pH, TDS, and EC were measured in situ using portable meters (Hanna, HI98130). Later, samples were transported to the Environmental Geochemistry Lab of CSIR-NGRI, Hyderabad. Trace elements' (As, B, Ba, Cd, Cr, Hg, Ni and Zn) concentration in groundwater was analyzed using ICP-OES (Optima 4300 DV, Perkin Elmer). A calibration curve is constructed by running the freshly prepared working standards and continued testing until the correlation coefficient (r 2 ) close to 1.

Heavy metal pollution index (HPI)
HPI is a rating method of water quality utilizing determining individual trace elements' concentration and their composite effects (Zakhem andHafez 2015 andSheykhi andMoore 2012). The rating or weights (W i ) is assigned between 0 and 1 for each trace element, which is based on the importance of individual quality considerations or by taking the inverse proportional value of approved standard to all trace elements considered (Mohan et al. 1996;Horton 1965). According to Tiwari et al. (2015) and Prasad and Bose (2001), different researchers/scholars set their limit values for HPI, and the critical limit value of HPI for drinking water is 100. Edet and Offiong (2002) categorized HPI into three distinct classes: low (HPI < 15), medium (15 ≤ HPI ≤ 30) , and high (HPI > 30). However, this study followed the classification of HPI based on the description of Kumar et al. (2012). These were explained as low (HPI < 19), medium (19 ≤ HPI ≤ 38) , and high (HPI > 38). The concentration limits (i.e., the maximum permissible value S i and maximum desirable value I i for every parameter) are based on the international standard (WHO 2011). Maximum permissible value S i can be defined as beyond this limit it is not advisable to drinking water while maximum desirable value I i is one way of relaxation limit set by the individual organization where it gives the best water to drink purpose (like every country will have different maximum desirable value I i it depends on the economy of the state). The HPI is calculated using the following equations (Mohan et al. 1996): where Q i is sub-index of ith parameter. W i is the unit weight of ith parameter and n is the number of parameters used.
The unit weight (W i ) is obtained using the following formula: where K is the proportional constant (1) and S i is the standard permissible value of ith parameter.
The sub-index (Q i ) of the parameter is given by where M i is the monitor value of the metal in ith parameter. I i is the ideal value of the metal in ith parameter. S i is the standard value of the metal in ith parameter and (−) sign is to have the absolute value of the two numbers.

Metal index
MI is also an important factor in determining groundwater quality for public concern. Metal index is classified as follows: very pure (MI < 0.3), pure (0.3 ≤ MI < 1) , slightly affected (1 ≤ MI < 2) , moderately affected (2 ≤ MI < 4) , strongly affected (4 ≤ MI < 6) and seriously affected (MI ≥ 6). MI value is considered a sign of warning when it is > 1 (Goher et al. 2014;Bakan et al. 2010). Metal index (Tamasi and Cini 2004) is calculated as follows: where C i is the measure concentration of each metal and MAC is the maximum permissible limit.

Geographical information system (GIS) analysis
Geographical information system (GIS) helps in collecting and manipulation of a diverse range of spatial data and generates the spatial distribution phenomena of all individual values (Gupta and Srivastava 2010). GIS manipulates all water quality parameters, gives an overall idea, and provides easy understanding to the people and policymakers (Singh et al. 2013). The inverse distance-weighted interpolation method of ArcGIS-10.7 software (GIS Lab, CSIR-NGRI, Hyderabad, India) was used to generate a spatial distribution map of all parameters.

Results and discussion
Physicochemical analysis Table 1 describes the overall statistical arrangement of all water quality parameters investigated in the study area. All considered parameters permissible limit is with reference to WHO (2011) standards. The spatial distribution maps of physicochemical parameters for Temperature, pH, EC, TDS, TH, and TA were shown in ESM_Fig. 2 (a-f). Temperature and pH: The water temperature in the study ranged between  29 °C and 30 °C. Temperature is very important due to tendency to cause chemical reaction and also control oxygen level in water. pH values obtained in this work is between 7.28 and 8.43 which is between desirable limits of WHO 2011 drinking water guideline values. The value of pH in the studied is not corrosive. pH has positive correlation with EC and Alkalinity. 7 for instance, is the pH neutral value of water. If pH value is low than high corrosive properties and vice versa. Electrical conductivity (EC) values range from 640 µS/cm to 5890 µS/cm, whereas 18.97% of samples are above the permissible limit of 1500 µS/cm. Total dissolved solids (TDS) are found high in the area as most of the samples (81.03%) are above permissible limit 1000 mg/L. Total hardness can be evaluated from the following equation (Kumar et al. 2012 andSawyer et al. 2003): Total hardness (TH) ranges between 58.5 and 1480.5 mg/L where 46.55% of samples are beyond the permissible limit (500 mg/L) as prescribed by WHO. The average concentration of TH found to be 526.57 mg/L. Total Alkalinity (TA) is due to the contribution of carbonate (CO 3 2− ) and bicarbonate (HCO 3 − ) ions in water and the main controlling factor. Alkalinity in the study area found to be in the range of 60-580 mg/L, which is within the permissible limit.
Trace element concentrations in groundwater samples found descending order are as Zn (38.67%) > B (32.67%) > Ba (13.59%) > As (8.49%) > Hg (3.71%) > Cr (1.28%) > Ni (0.52%) > Cd (0.47%). Zn has the maximum concentration of all analyzed trace elements in groundwater from the study area. According to the WHO limit, none of the samples found is above 3000 µg/L permissible. The concentration of Zn is within limit, i.e., between 11.69 and 130.2 µg/L, as shown in Fig. 3a. Though Zn is harmful in higher concentration, it is also one of our essential trace elements. They are usually consumed as organic compounds and salts in food, medicine and drinking water. Zn gives good health through proper diet. Rusting of pipes which results to concentrate with water through household tap is one major cause of high Zn in drinking water and effecting human health. Generally, zinc is associated with lead and cadmium in groundwater. Boron concentration varies between 5.722 and 64.61 µg/L in Fig. 3b. Boron (B) distribution is found under permissible limit, i.e., below 2400 µg/L. Boron gets into groundwater due to water percolate into fractures/cracks in rock and soil. Naturally, Boron occurs in oceans, sedimentary, shale, coal, and soils. Barium (Ba) is in the third spot regarding the amount of percentage present in groundwater. It constitutes about 8.49% of the total content of metals. Though it is in second place, the concentration found is within permissible limit, i.e., 0.033-26.71 µg/L (Fig. 3c) where maximum allowable is up to 700 µg/L. TH as CaCO 3 mg = (Ca 2+ + Mg 2+ )meq∕L*50 Generally, Barium is high in regions where pH is low and associate with granite rocks, alkaline igneous rock, volcanic rocks, and sedimentary (Mn-rich) rocks. Arsenic (As) is one of the dangerous trace elements because they are toxic even at very low concentration. It ranks 4th highest in order of composition in groundwater samples, contributing 8.49%. Many samples (22.41%) found to be is above the permissible limit of 10 µg/L. The minimum and maximum range of As is between 0.308 and 38.38 µg/L, as shown in Fig. 3d. The natural source of arsenic is arsenopyrite (FeAsS); its association minerals are Fe and sulfide. Anthropogenic sources are from chemical pesticides, industrial waste, coal burning, lead, and gold mining. In contrast, mining and industrial activity in the study area are not prominent, so it imparts the idea of contamination through geogenic processes. Therefore, the chances of anthropogenic dominancy in the study area found to be of using pesticides for agriculture activities. Mercury (Hg) constitute 3.71% and is found to contain between 0.412 and 8.988 µg/L (Fig. 3e). The permissible limit of Hg is 6 µg/L, where 3(5.17%) samples are found to exceed the limit. Natural sources are volcanoes, hot-springs, serpentinite changes to silica-carbonate in the presence of carbonate-rich water (White 1957). While natural processes are hydroelectric plants, mine, paper industry wastes, etc. Chromium (Cr) concentration is found between 0.123 and 2.496 µg/L, as shown in Fig. 3f and constitute 1.28%. No samples exceed permissible limit 50 µg/L. Chromium in water is due to waste products from electroplating, leather tanning, and textile factories (Fetter 1993), while leaching of soils and rocks are the major sources. Nickel (Ni) concentration range from 0.012 to 0.865 µg/L (Fig. 3g), where no sample exceeds the permissible limit i.e., 70 µg/L. Generally, nickel is contaminating water bodies due to iron leaching. Sources of nickel are steel factories, non-ferrous alloys (aluminum, copper, zinc, tin, lead, brass, etc.) and superalloy (having high melting, less corrosion, strength etc.). Cadmium (Cd) is the least concentration among all trace elements ranging from 0.277 to 0.544 µg/L (Fig. 3h), constituting 0.47%. No samples exceed permissible limit 3 µg/L. Cadmium is a trace metal present widely in nature (Xie et al. 2021) in crust and water. They are used as rusting resistance as well as to stabilized plastic. Cadmium is known for their uses in making batteries. Major environment contaminant medium is through water and air (release fossil fuels combustion). Cadmium is released to water from galvanized plumbing (coated with zinc), water pipes, domestic wastes, and fertilizer factories. Cadmium is generally associated with ores of zinc, copper, lead. When the groundwater in highly acidic, a chemical reaction takes place, which results in high Cd concentration in groundwater. Krishna et al.'s (2009) work on similar lithology and region of south India peninsula, i.e., granitoid terrain of Medak district, Andhra Pradesh (presently under Telangana State, India)  explains that trace elements such as Ba, Ni, and Cr were controlled and associated by mixed origin from both natural and anthropogenic inputs, while, As, Zn, Mn, Fe, Pb and Co were due to anthropogenetic sources.

Heavy metal pollution index (HPI)
Eight trace elements (Zn, B, Ba, As, Hg, Cr, Ni, and Cd) were consider to determine the HPI for groundwater. Table 2 for groundwater samples. According to Kumar et al. (2012), the three main classes low (HPI < 19), medium (19 ≤ HPI ≤ 38) , and high (HPI > 38) were followed, respectively, as shown in Table 3. From the HPI values, it is clear that 40 samples fall under the low category, which means less contaminates, and it contributes above 86.21% of the total area. 7 (12.07%) samples fall within medium group while only 1 (1.72%) sample is found to have high concentration of metals in the area. For a better understanding of trace elements in the study, a scatter plot and spatial  Fig. 2, which suggests that migmatite rock might be the source of origin. Since the area is dominant with granitoid composition, it can play a vital role as a source of trace elements in groundwater through natural processes because it is mostly devoid of industrialization. This will help in the future investigation to find the geochemical study of rocks and the amount of spread for better management.

Metal index (MI)
The metal index is another way to check water quality by determining the concentration of trace elements. Metal index help in classifying groundwater quality, and is considered to be hazardous for drinking and aquatic life when MI > 1 (Goher et al. 2014). The metal index is grouped into six types based on their concentration such as very pure (MI < 0.3), pure (0.3 ≤ MI < 1) , slightly affected (1 ≤ MI < 2) , moderately affected (2 ≤ MI < 4) , strongly affected (4 ≤ MI < 6) and seriously affected (MI ≥ 6) as shown in Table 4. Accordingly, it is found that 7 (12.07%) samples are in very pure category, 14 (24.14%) samples are in pure category while the remaining 37 (63.79%) samples contribute slightly affected to strongly affected category. A spatial distribution map of the metal index is shown in Fig. 6. Though, one sample collected from Chegunta village fall within the seriously affected category of MI, it is clear that most of the groundwater in the area followed the trend to be affected. Therefore, it is the right time to take up control measurements so that groundwater can be saved and used in the long run.

Correlation analysis
Correlation index defines the relation of two variables which describe the sufficiently of a variable to relate the other (Davis 1986), such that high correlation coefficient values show good relation and vice versa. Determination of variables, i.e., the dependent (x) is solely control by independent (y) and vice versa (Voudouris et al. 2000). In addition, if the correlation coefficient value is close to zero, then it has no correlation. A high positive value (r) means a good relationship, but if it is negative than the relationship is an inverse correlation. To better understand the source of pollution, either anthropogenic or natural, a statistical calculation was done according to Pearson's correlation matrix. The correlation coefficient of all parameters (physicochemical parameters and trace elements) is shown in Table 5. It is observed that TH with EC/TDS shows the chemical reaction through the natural process; As with EC/TDS/TH indicates the source of chemical fertilizers from agricultural infiltration, domestic waste, and industrial discharge. While parameters that show negative or no relation indicates the source of trace elements in groundwater is not an individual contribution of natural or anthropogenic. Detailed summary of groundwater quality from all samples in the study area with respect to HPI and MI is shown in Table 6.

Conclusion
This study integrated the trace elements, physicochemical parameters, multivariant statistical methods, and ArcGIS to evaluate the groundwater quality and sources of pollution. Concentration of trace elements related to abundance in samples are in order: Zn (38.67%) > B (32.67%) > Ba  . Though Zn has the maximum concentration of all analyzed trace elements in groundwater from the study area, no samples are above the permissible limit of 3000 µg/L (observed values are between 11.69 and 130.2 µg/L). Except for two trace elements, namely As with 22.41% of samples and Hg with 5.17%, samples are found to have above WHO drinking water permissible limits. According to HPI classification, most of the samples (86.21%) fall under low category of HPI (< 19), revealed no significant health issue due to trace elements. Correlation coefficient analyses suggest that the contribution of trace elements in groundwater is due to both natural (chemical reactions) and anthropogenic sources. From spatial distribution maps, HPI and MI have a higher concentration in small patches of the north-east side of Wanaparthy watershed. Considering the overall view of the spatial variation of the parameters investigated, it found that a higher concentration of trace elements is found near or joining to upper order drainage system. This might be due to the input of runoff water (interaction of variable rocks' composition) from a higher elevation and later get infiltration (low lying/depression areas) to subsurface and reach to the aquifer. Heavy metal pollution indices (HPI and MI) suggested that the main source of trace elements in groundwater might be from the dominant granitoid rocks because the area is mostly devoid of industrialization. The present work suggests the MI value in 63.79% of the samples in Wanaparthy watershed is slightly affected to strongly affected category, which means majority of groundwater samples need proper treatment before it uses for drinking purposes.