Environmentally Sensitive Elements in Groundwater of an Industrial Town in India: Spatial Distribution and Human Health Risk

This paper presents the results of a study to assess the groundwater quality in an industrial town located in Punjab, India. A total of 99 samples of groundwater were analyzed during the premonsoon and postmonsoon periods of 2018, which revealed the presence of numerous environmentally sensitive elements (ESEs), namely, arsenic (As), aluminum (Al), chromium (Cr), iron (Fe), mercury (Hg), nickel (Ni), selenium (Se), and lead (Pb). Geographic information system (GIS)-based spatial interpolation showed higher contamination levels around the industrial areas and the drainage channel where industrial effluent is generally discharged. Further, groundwater quality was assessed using the heavy metal pollution index (HPI) and the metal index (MI), which indicated poor drinkability of the groundwater. Human exposure to groundwater contaminated with ESEs can pose serious health risks; therefore, noncarcinogenic and carcinogenic health risks due to presence of these elements were also evaluated. Reported health risks to humans from exposure to contaminated groundwater indicate the importance of regular monitoring of groundwater for ESEs vis-a-vis industrial effluent disposal practices.


Introduction
Groundwater is a key natural resource which serves the drinking, agricultural, and industrial needs of one-third of the human population [1,2]. In recent years, the diminishing quantity and degraded quality of surface water bodies have substantially increased the burden on earth's groundwater [3]. Activities such as mining, dumping of solid biomass in landfills, leaching of agrochemicals, and improper disposal of industrial effluents have contaminated the groundwater with various environmentally sensitive elements (ESEs), such as heavy metals, arsenic, pesticides, fertilizers, and so forth. [4]. Groundwater quality is a major concern in both developed and developing nations [5]. Several studies from various parts of the world have highlighted the issue of deteriorating groundwater quality in Bangladesh [6], China [7], South Africa [8], Iran [9], Italy [10], Korea [11], Pakistan [12], and Thailand [13]. Some of these studies have also indicated public health issues arising out of human exposure to contaminated

Study Area Description
The Ludhiana district of Punjab, India, which is one of the industrial hubs of North India, was the study area. Figure 1 shows the geographical location, landuse, landcover, and the sampling locations of the study area. Sutlej River flows along its northern boundary. The entire geographical area is 3767 km 2 and has 3.5 million human inhabitants [29]. The mean annual rainfall is about 577 mm [30]. The mean maximum and minimum temperatures are 45 and 6 • C in the months of June and December, respectively [31]. The major water requirements of the area are for agricultural, domestic, and industrial demands. Groundwater, rainwater, and canal water are used to meet the agrarian needs; however, the domestic and industrial demands are met with groundwater sources only. Major industries located in the study area include micro, small, medium, and large businesses of different types, namely, chemicals, fabricated metals, bicycle and bicycle parts, machine tools, rubber goods, hosiery goods, beverages, textiles, dyeing, paper products, and electroplating. There are 37,047 (micro, small, and medium) businesses and 151 large-scale businesses which are registered in the study area [32]. Studies have reported increasing groundwater quality deterioration in the study area due to activities such as industrial wastewater disposal, disposal of garbage, use of fertilizers, and pesticides on agricultural land [28].
It consists of five prominent sand horizons down to 400 m depth separated by thick clay horizons. The first aquifer generally occurs between 10 and 30 m. The second through fifth occur between 50 and 120 m, 150 and 175 m, 200 and 250 m, and 300 and 400 m, respectively. The depth to water level in the area ranges between 9 and 26 m below ground level (bgl). During the premonsoon period, the depth to water level varies between 4 and 31 m bgl, and postmonsoon, it ranges between 3 and 27 m bgl [33].

Collection of Samples and Analysis
Ninety-nine groundwater sampling locations (93 deep tube wells with depths ranging from 90 to 150 m and 6 shallow tube wells from 10 to 20 m deep), as shown in Figure 1, were sampled to cover the entire study area. Samples were collected during premonsoon (April-May) and postmonsoon (November-December) periods of 2018. The premonsoon and postmonsoon periods were defined as 1-2 months before and after the monsoon, respectively. The schedule of the sample collection is given in Supplementary Table S1. The coordinates of all the sampling locations were determined with a portable global positioning system (GARMIN Etrex 10). Standard protocols as prescribed in standard methods were followed during sampling and preserving the samples [34]. Glass bottles of 1000 mL capacity were utilized for sample collection. Sample bottles were washed by 1:1 HNO3 and rinsed three times using triple-distilled water. Further, bottles were dried in a hot air oven at a temperature of 80 °C for 4 h. Cellulose filter paper (Whatman ® filter paper, grade 42, GE Healthcare Companies, UK) was used to filter groundwater samples. For the preservation of groundwater samples, each sample was acidified by adding 2 mL of HNO3 to pH < 2, and acidified samples were put into an ice box. Each sample was labeled properly to prevent misidentification between samples and it was ensured that no air bubbles were present in the samples. Within 12 h of

Hydrogeology
The study area consists of Indo-Gangetic alluvium of quaternary age. Subsurface strata consist of clay, fine sand, medium sand, and hard clay. Moreover, kankar and gravel with sand are also present in some places. Two aquifers, confined and unconfined, mainly exist in the area and the subsurface water flows from N-SW and E-SW. The lithological data of the study area indicate the presence of many sand beds forming the principal aquifers separated by clay beds at various depths. It consists of five prominent sand horizons down to 400 m depth separated by thick clay horizons. The first aquifer generally occurs between 10 and 30 m. The second through fifth occur between 50 and 120 m, 150 and 175 m, 200 and 250 m, and 300 and 400 m, respectively. The depth to water level in the area ranges between 9 and 26 m below ground level (bgl). During the premonsoon period, the depth to water level varies between 4 and 31 m bgl, and postmonsoon, it ranges between 3 and 27 m bgl [33].

Collection of Samples and Analysis
Ninety-nine groundwater sampling locations (93 deep tube wells with depths ranging from 90 to 150 m and 6 shallow tube wells from 10 to 20 m deep), as shown in Figure 1, were sampled to cover the entire study area. Samples were collected during premonsoon (April-May) and postmonsoon (November-December) periods of 2018. The premonsoon and postmonsoon periods were defined as 1-2 months before and after the monsoon, respectively. The schedule of the sample collection is given in Supplementary Table S1. The coordinates of all the sampling locations were determined with a portable global positioning system (GARMIN Etrex 10). Standard protocols as prescribed in standard methods were followed during sampling and preserving the samples [34]. Glass bottles of 1000 mL capacity were utilized for sample collection. Sample bottles were washed by 1:1 HNO 3 and rinsed three times using triple-distilled water. Further, bottles were dried in a hot air oven at a temperature of 80 • C for 4 h. Cellulose filter paper (Whatman ® filter paper, grade 42, GE Healthcare Companies, UK) was used to filter groundwater samples. For the preservation of groundwater samples, each sample was acidified by adding 2 mL of HNO 3 to pH < 2, and acidified samples were put into an ice box. Each sample was labeled properly to prevent misidentification between samples and it was ensured that no air bubbles were present in the samples. Within 12 h of sampling, all water samples were transferred to the laboratory and stored at 4 • C for further analysis.
Analysis of ESEs, namely, As, Al, Cr, Hg, Fe, Ni, Se, and Pb, was accomplished with an atomic absorption spectrophotometer (model AAS 4141, Electronics Corporation of India Ltd., Hyderabad, India). Hollow cathode lamps were used for determining the As, Al, Cr, Hg, Fe, Ni, Se, and Pb at different wavelengths of 193.7, 309.3, 357.9, 253.7, 248.3, 232.0, 196.0, and 217.0 nm, respectively. Cr, Fe, Ni, and Pb were determined by the aspirating direct air-acetylene flame method with lamp currents of 7, 5, 3.5, and 10 mA, respectively. Al and Hg were estimated by the aspirating direct nitrous-oxide-acetylene flame method with a lamp current of 10 mA and the cold vapor atomic absorption spectrophotometer method with a lamp current of 5 mA, respectively. As and Se were determined by the continuous hydride generation method using nitrogen-acetylene gases with a lamp current of 10 mA.

Reagents, Standards, and Quality Assurance
High-purity, analytical-grade reagents and chemicals were used during the entire process for analyzing the heavy metals in groundwater samples. All the reagents and standards were prepared with triple-distilled water throughout the experimental work. For quality analysis, ready-made standard stock solutions of selected heavy metals of the concentration 1000 mg/L in HNO 3 were used (CDH Pvt. Ltd., New Delhi, India). To prepare calibration curves of each metal, four standard solutions for each metal at different concentrations were prepared. AAS was aspirated with blanks (triple-distilled water) and by zeroing the instrument at regular intervals to ensure its accuracy. Also, after analysis of 10 groundwater samples, 1 sample was examined in triplicate, so that consistent outcomes were maintained. The standard error between standards and measured samples were <5% [18,23].

Statistics and GIS Analysis
For statistical analysis, IBM Statistical Package for Social Sciences (SPSS ® , version 25, IBM, New York, NY, USA) was used. The basic descriptive statistics feature of the package was used to analyze the range, mean, standard deviation, and so forth. For the paired t-test, the normality of the period differences was checked using the Kolmogorov-Smirnov test, and the compare mean module of SPSS was used to calculate the p-value. Pearson's correlation was performed with the assistance of the correlate module/bivariate correlations modules in SPSS. The Ludhiana district base map was digitized using ArcGIS ® (version 10.4), ESRI, India. All the spatial and attribute data were input in the database of ArcGIS. All the thematic maps were prepared using the spatial analysis tools and the inverse distance weighted (IDW) interpolation technique. In this study, the search radius of 12 points was selected. The IDW interpolation technique uses the measured values of the surrounding locations to interpolate the values the prediction location.

Evaluation of Groundwater Quality by Indexing Approach
In this study, two pollution indexing approaches were applied to determine the groundwater quality and they are explained in the following subsections.

HPI
The HPI method was used to calculate the overall heavy metal contamination in groundwater. This method primarily depends on two factors: unit weightage (W i ) and the standard prescribed limit (S i ) of each heavy metal. HPI calculation is based on the weighted arithmetic mean and, in this study, the HPI model, as developed and reported by Mohan et al. [35], described in Equation (1), was applied: where W i and Q i are the unit weight and subindex of the ith parameter, respectively, and n is the number of parameters/ESEs considered. Unit weightage W i is inversely proportional to the S i , the recommended standard permissible limit of ith parameter in milligrams per liter, of all the selected heavy metals and was calculated by using Equation (2): where K is the constant of proportionality, the value of which ranges from 0 to 1.The subindex (Q i ) for each heavy metal in Equation (1) was calculated by using Equation (3): where M i and I i are the analytical concentration and ideal value of the ith parameter/heavy metal in milligrams per liter, respectively. S i is the recommended standard permissible limit of the ith parameter in milligrams per liter. In Equation (3), the (-) sign indicates the numerical difference between two values, eliminating the algebraic sign. The critical value of the HPI is 100, greater than which ingestion of groundwater will pose serious health effects to the human body [36]. The HPI method has previously been used to evaluate the status of metal pollution in surface water bodies, such as the River Bogayi in Turkey [37] and the Harike Wetland in Punjab, India [38]. The HPI in groundwater studies has also been used in Iran [39], Bangalore [40], Jharkhand [41], and West Bengal [42].

MI
The MI demonstrates the overall groundwater quality with respect to the heavy metal pollution. It was calculated using Equation (4), as defined by Tamasi and Cini [43]: where C i is the monitored concentration of the ith heavy metal (mg/L) and (MAC) i is the maximum allowable concentration of the ith metal. The threshold value for the MI quality index is 1. MI > 1 indicates water quality may not be suitable for long-term use due tothe potential for harmful chronic exposure.

Human Health Risk Assessment
The metal-contaminated groundwater could pose critical health risks via two common routes: ingestion (through drinking) and dermal exposure (through skin absorption). The United States Environmental Protection Agency (US EPA) has proposed a model for human health risk. The intake dose of ingestion and dermal activities was calculated using Equations (5) and (6), as suggested by the US EPA [44]: A description of the parameters used in the above equations for adults and children is given in Supplementary Table S2.

Noncarcinogenic Risk Assessment
The noncarcinogenic health risk of each metal was assessed by calculating the hazard quotient (HQ). The HQ was calculated by dividing the intake value by the reference dose (RfD) using Equations (7) and (8). RfD ingestion represents the reference dose for ingestion of a selected metal, which was obtained from US EPA IRIS [45] and Li et al. [46], and RfD dermal is the dermal reference dose calculated according to Equation (9) using the gastrointestinal absorption factor (GIABS) of selected metals, as suggested by the US EPA [47].
Further, the hazard index (HI) was calculated to determine the total noncarcinogenic effect of metals in groundwater by adding the HQ ingestion and HQ dermal intake pathways of all selected metals using Equation (10) and then calculating HI total using Equation (11), as detailed by the US EPA [44]: RfD ingestion (7) HQ dermal = Intake dermal RfD dermal (8) HI ingestion = HQ ingestion ; HI dermal = HQ dermal (10) HI total = HI ingestion +HI dermal .
All the RfD ingestion and RfD dermal values are presented in Supplementary Tables S3 and S4, respectively. Groundwater samples having HI total > 1 suggest detrimental noncarcinogenic health effects to humans, whereas HI total < 1 suggests that the groundwater would not pose any impact on human health [48].

Carcinogenic Risk Assessment
Exposure to polluted groundwater might create cancer-causing effects to humans. In this study, As, Cr, and Pb were considered as carcinogenic substances and the total carcinogenic risk (R total ) was calculated using Equations (12)-(15) [49]. As there was no specified cancer slope factor (CSF) for metals such as Al, Fe, Hg, Ni, and Se, carcinogenic risk was not calculated for these metals.
If the R total value is less than 1 × 10 −6 , the chances of cancer risk are considered to be negligible, while an R total value more than 1 × 10 −4 indicates a substantial cancer risk [45]. CSF values of ingestion and dermal exposure were obtained from the Li et al. [46] and US EPA [47] (Supplementary Table S3). Table 1 depicts the level of contamination of the groundwater samples with various ESEs, during premonsoon and postmonsoon periods, in comparison to the Bureau of Indian Standards (BIS) [50]. Further, the geospatial variation of ESEsin the groundwater of the study area is shown in Figure 2. The IDW interpolation technique of ArcGIS software (version 10.4), was used to prepare the spatial distribution maps. It can be seen from Table 1 that in the case of Pb contamination, 88 and 91 samples exceeded the permissible limits during premonsoon and postmonsoon periods, respectively. The concentration of Pb in the NE-SE parts of the study area was found to be higher during both periods. In addition, the concentration of Ni exceeded the permissible limit in 72 samples during the premonsoon period and in 79 samples during the postmonsoon period. The concentration of Al also exceeded the permissible limit in 57 and 62 samples during the pre-and postmonsoon periods, respectively. Shrivastava [26] also reported high concentrations of Al, Ni, and Pb in groundwater samples collected from other districts of Punjab, which are located in the vicinity of the study area. The mean concentration of the ESEs was observed to increase during the postmonsoon period. The paired t-test for pre-and postmonsoon periods showed significant variation during the postmonsoon period, as there was a p-value < 0.05 for seven parameters, namely, Fe, Pb, Ni, Al, Se, Cr, and As. However, Hg did not show a significant variation. The probable reason could be the lesser/statistically insignificant number of samples (two premonsoon and three postmonsoon samples) indicating Hg contamination. Paired t-tests for pre-and postmonsoon periods for all heavy metals are shown in the Supplementary Table S5. The reasons for the increased concentration postmonsoon could be due to dissolution of the metals from parent bedrock into groundwater at higher temperatures during the monsoon season [18]. In general, concentrations of most of the ESEs were found to be high in samples collected in and around industrial areas. Therefore, it could also be attributed to the seepage of ESEs from the surface to the aquifer during monsoon season in the industrial areas. Similar studies from other parts of India have also reported that groundwater beneath industrial establishments is generally contaminated with heavy metals [51][52][53]. Although Fe is an essential element for human beings, a higher Fe concentration in drinking water can cause certain health impacts [54]. In 38 and 45 samples, the Fe concentration was above the permissible limit during premonsoon and postmonsoon periods, respectively. For Cr total , only 6 samples were above the permissible limit during the premonsoon period, while in the postmonsoon period, the concentration of Cr total in groundwater samples exceeded the limit in 25 samples. Brindha and Elango also reported the higher concentration of Cr near the industrial sites of Chennai, India [55]. The concentration of As exceeded the acceptable limits in 10 and 15 groundwater samples collected during pre-and postmonsoon periods, respectively. Hg was found to be above the permissible limits in two and three samples from shallow wells located in and around the industrial area during preand postmonsoon periods, respectively. Higher concentrations of Hg and As in groundwater samples around the industrialized region of Maharashtra, India were reported by Bhagure and Mirgane [56]. Further, mean concentrations of the various ESEs in groundwater samples were observed to be in the following order: Premonsoon-Fe > Al > Ni > Pb > Cr > Se > As > Hg; Postmonsoon-Al > Fe > Ni > Pb > Cr > Se > As > Hg.

Pearson's Inter-ESE Correlation
The interrelationship between the ESEs was evaluated using Pearson's correlation. The correlations of heavy metals (As, Al, Cr, Hg, Fe, Ni, Se, and Pb) during pre-and postmonsoon periods are presented in Tables 2 and 3, respectively. The critical values for Pearson's coefficient (r) for 99 groundwater samples were 0.197 and 0.257 at p ≤ 0.05 and p ≤ 0.01, respectively. Correlation analysis at the p = 0.01 and 0.05 levels indicated significant relationships between the metals Al-Se,

Pearson's Inter-ESE Correlation
The interrelationship between the ESEs was evaluated using Pearson's correlation. The correlations of heavy metals (As, Al, Cr, Hg, Fe, Ni, Se, and Pb) during pre-and postmonsoon periods are presented in Tables 2 and 3, respectively. The critical values for Pearson's coefficient (r) for 99 groundwater samples were 0.197 and 0.257 at p ≤ 0.05 and p ≤ 0.01, respectively. Correlation analysis at the p = 0.01 and 0.05 levels indicated significant relationships between the metals Al-Se, As-Ni, Fe-Ni, Ni-Pb, Ni-Se, and Pb-Se and Al-Fe, As-Fe, As-Se, Cr-Ni, Cr-Pb, Fe-Pb, Fe-Se, Hg-Pb, respectively, during the premonsoon period. Also, during the postmonsoon period, the correlation results indicated a positive relation between metals such as Al-Se, As-Se, Fe-Ni, Fe-Pb, Ni-Pb, Ni-Se, and Pb-Se at the 0.01 level and Cr-Hg, Cr-Se, and Fe-Se at the 0.05 level. Strong positive relations indicate that Pb, Ni, and Se have the same source of pollution during both periods, which is evident from the outcomes correlating higher concentrations of these metals in and around the industrial areas. It was implied that, due to the strong correlation between Fe and Ni (r = 0.362) and Ni and Pb (r = 0.417) during both periods, their source of contamination could be from dumping of electroplating industrial waste, which is very common in the study area [57].  Note: ** Correlation significant at the 0.01 level (two-tailed); * Correlation significant at the 0.05 level (two-tailed). Table 4 presents the classification of groundwater in terms of the indices HPI and MI. The classification was adapted from Sankar (2019) [40]. The geospatial representation of water quality indices based on HPI and MI during the premonsoon and postmonsoon periods are shown in Figures 3 and 4, respectively. The HPI-based groundwater assessment revealed that 35.4% of the groundwater samples during the premonsoon period and 52.5% of the groundwater samples during the postmonsoon period were above the critical pollution index of 100. This indicates that at these sampling points, the groundwater may pose health risks to humans if used for drinking. It can also be seen from Figure 3 that the sampling wells near the industrial area and along the channel (BudhaNullah) which receives the industrial effluent show higher HPI values. On the other hand, the analysis using the MI indicated that 79.8% and 86.9% of the samples showed values greater than 1, which is the critical MI. Figure 4 depicts the spatial variation of the MI in the study area. The MI also exhibited a similar geographical pattern as that of the HPI in terms of level of heavy metal pollution. However, the variations in the geographical spread of each class of water quality seen in Table 4 and Figures 3 and 4 could be due to the difference in the definition of the indices. The HPI provides an aggregated influence of individual heavy metals on the total water quality, whereas the MI gives only an exceedance value from the standards. The quality of groundwater evaluated using both the HPI and the MI strongly suggests that exposure to this groundwater may cause detrimental effects to human health. provides an aggregated influence of individual heavy metals on the total water quality, whereas the MI gives only an exceedance value from the standards. The quality of groundwater evaluated using both the HPI and the MI strongly suggests that exposure to this groundwater may cause detrimental effects to human health.

Health Risk Assessment
The concentration of ESEs in groundwater was examined to assess the health risk through ingestion and dermal pathways for adults and children, and the outcomes of this analysis are presented in Table 5. Table 5 shows that the HQingestion and HQdermal during the premonsoon period followed the order of Pb > As > Ni > Cr > Se > Fe > Al > Hg and Cr > Ni > As > Hg > Se > Pb > Fe > Al, respectively, for both adults and children. However, during the postmonsoon period, the order of HQingestion and HQdermal was Pb > As > Cr > Ni > Se > Fe > Hg > Al and Cr > Ni > As > Hg > Pb > Se > Fe > Al, respectively, for both adults and children. Pb and Cr were found to be the main contributors to HQ. The HQ values for all heavy metals were found to be < 1 for both ingestion and dermal pathways during both seasons, except for Pb. Moreover, Pb was found to be > 1 for children through the ingestion pathway, and mean observed values were 1.32 and 1.87 during the pre-and postmonsoon periods, respectively (Table 5). It can be inferred that, in the study area, Pb contamination has a higher potential to pose noncarcinogenic health risks to children. Similar health risk assessment

Health Risk Assessment
The concentration of ESEs in groundwater was examined to assess the health risk through ingestion and dermal pathways for adults and children, and the outcomes of this analysis are presented in Table 5. Table 5 shows that the HQ ingestion and HQ dermal during the premonsoon period followed the order of Pb > As > Ni > Cr > Se > Fe > Al > Hg and Cr > Ni > As > Hg > Se > Pb > Fe > Al, respectively, for both adults and children. However, during the postmonsoon period, the order of HQ ingestion and HQ dermal was Pb > As > Cr > Ni > Se > Fe > Hg > Al and Cr > Ni > As > Hg > Pb > Se > Fe > Al, respectively, for both adults and children. Pb and Cr were found to be the main contributors to HQ. The HQ values for all heavy metals were found to be < 1 for both ingestion and dermal pathways during both seasons, except for Pb. Moreover, Pb was found to be > 1 for children through the ingestion pathway, and mean observed values were 1.32 and 1.87 during the pre-and postmonsoon periods, respectively (Table 5). It can be inferred that, in the study area, Pb contamination has a higher potential to pose noncarcinogenic health risks to children. Similar health risk assessment studies reported from China found higher health risks due to heavy metals in children than adults [58].  Further, the total hazard index (HI total ) was calculated for both children and adults during both periods and it was observed that HI total mean values for adults and children were > 1 in the premonsoon (1.76 and 3.57, respectively) and postmonsoon (2.41 and 4.90, respectively) periods, as shown in Table 6. Figure 5 shows that during the premonsoon period, 76.5% and 93.5% of the study area had HI total > 1 for adults and children, respectively, and during the postmonsoon period, 89.3% and 98.9% of the study area had HI total > 1 for adults and children, respectively. It was also observed that HI total values of children were higher than adults for ingestion and dermal pathways. This is because children weigh less than adults. A similar observation from Punjab was reported by Sharma et al. [18]. Further, the total hazard index (HItotal) was calculated for both children and adults during both periods and it was observed that HItotal mean values for adults and children were > 1 in the premonsoon (1.76 and 3.57, respectively) and postmonsoon (2.41 and 4.90, respectively) periods, as shown in Table 6. Figure 5 shows that during the premonsoon period, 76.5% and 93.5% of the study area had HItotal > 1 for adults and children, respectively, and during the postmonsoon period, 89.3% and 98.9% of the study area had HItotal > 1 for adults and children, respectively. It was also observed that HItotal values of children were higher than adults for ingestion and dermal pathways. This is because children weigh less than adults. A similar observation from Punjab was reported by Sharma et al. [18]. The total carcinogenic health risk (Rtotal) was also calculated through ingestion and dermal pathways during pre-and postmonsoon periods, as presented in Table 6, and its geospatial distribution is presented in Figure 6. The calculated mean values for Rtotal, as shown in Table 6, are greater than 1 × 10 −4 , which indicates potential carcinogenic effects. The total carcinogenic health risk (R total ) was also calculated through ingestion and dermal pathways during pre-and postmonsoon periods, as presented in Table 6, and its geospatial distribution is presented in Figure 6. The calculated mean values for R total , as shown in Table 6, are greater than 1 × 10 −4 , which indicates potential carcinogenic effects.

Conclusions
Based on the findings of the study, it can be concluded that the groundwater of the study area can pose serious health impacts due to presence of ESEs. The spatial distribution of the ESEs in the groundwater and the Pearson's correlation of different metals clearly indicate the contribution of industrial activities to contaminating the groundwater. HPI-based water quality analysis revealed that 1734 and 2219 km 2 of the area were above the critical pollution index during pre-and postmonsoon periods, respectively. Further, MI-based analysis indicated that 3003 and 3328 km 2 of the area were above the threshold limit during the pre-and postmonsoon periods, respectively. The presence of ESEs in the groundwater of the study area can pose noncarcinogenic health risks. The carcinogenic health risks through ingestion and dermal pathways are mainly due to the presence of As, Pb, and Cr contamination. Rtotal for adults revealed that 167 and 635 km 2 of the study area posed carcinogenic health risks during pre-and postmonsoon periods, respectively. Rtotal for children

Conclusions
Based on the findings of the study, it can be concluded that the groundwater of the study area can pose serious health impacts due to presence of ESEs. The spatial distribution of the ESEs in the groundwater and the Pearson's correlation of different metals clearly indicate the contribution of industrial activities to contaminating the groundwater. HPI-based water quality analysis revealed that 1734 and 2219 km 2 of the area were above the critical pollution index during pre-and postmonsoon periods, respectively. Further, MI-based analysis indicated that 3003 and 3328 km 2 of the area were above the threshold limit during the pre-and postmonsoon periods, respectively. The presence of ESEs in the groundwater of the study area can pose noncarcinogenic health risks. The carcinogenic health risks through ingestion and dermal pathways are mainly due to the presence of As, Pb, and Cr contamination. R total for adults revealed that 167 and 635 km 2 of the study area posed carcinogenic health risks during pre-and postmonsoon periods, respectively. R total for children showed that 1316 and 2314 km 2 of the study area could pose carcinogenic health risks during pre-and postmonsoon periods, respectively. The outcomes of this study will be helpful to understand the extent of ESE contamination in the groundwater of the study area and its effect on consumers due to long-term exposure. Therefore, it is recommended that ESEs should also be included in the routine monitoring of the groundwater, and evidence-based policies should be framed to mitigate ESE contamination of the groundwater in the region.

Supplementary Materials:
The following are available online at http://www.mdpi.com/2073-4441/11/11/2350/s1, Tables S1-S5 present the supplementary data used in the analyses. Table S1: Schedule of sampling during premonsoon and post monsoon, Table S2: Description of exposure parameters for calculation of health risk assessment, Table S3: Reference dose ingestion value for heavy metals, Table S4: Reference dose dermal, skin permeability constant and cancer slope factor value for heavy metals, Table S5: Paired sample t -test for heavy metals during pre-and postmonsoon periods.
Author Contributions: D.D.S. planned the research methodology after extensive discussions with M.S. and S.J. D.D.S. also collected the samples and performed the analysis. Further, the findings of the analysis were statistically interpreted and presented in the form of a scientific manuscript after considerable input from P.S.T., M.S., and S.J. S.S. assisted in the generation of geospatial maps using ArcGIS.

Funding:
No external funding and the APC was funded by first author D.D.S. from his personal funds.