Health risk assessment and hydrogeochemical modelling of groundwater due to heavy metals contaminants at Basundhara coal mining region, India

ABSTRACT The enrichment of heavy metals and mixing with groundwater resources, which makes a serious threat to human health and the environment. The present study was conducted to determine the enrichment of heavy metals in groundwater, geochemical modelling, and to estimate the probable health risks (non-carcinogenic and carcinogenic) of a coal mining region, in eastern India. The relative abundance of mean heavy metal concentration was in the order of Fe > Zn > Al > Ba > Mn > Sr > Ni > Ag > Cu > Pb > V > Cr > Co > Cd, and are within the WHO standards limits for drinking water values. Groundwater shows pH values as generally acidic to alkaline in nature, while Eh values indicate an oxidising environment. Principal component analysis indicates, metal concentration in the groundwater to be mainly derived from the geogenic sources and in-situ leaching of mine overburden materials. Hydrogeochemical modelling indicates heavy metals are in supersaturation to near neutral state in the groundwater with oxide and carbonate mineral phases. Based on (HPI), and (HEI) values, the majority of groundwater samples are low to moderately contaminated. The health risk assessment model based on hazard quotient (HQ) and hazard index (HI) values show few groundwater samples as crossing the reference limit and have non-carcinogenic health risks for both children and adults. The study identifies the chemical characterisation and potential health risks due to the presence of heavy metals in the groundwater of the coal mining region.


Introduction
Groundwater resources in coal mining areas are known to be vulnerable to pollution and that may have a serious impact on the environment.Coal mining operations have led to groundwater contamination and that may have risks on human health [1][2][3][4][5][6][7][8][9][10][11].Mining and allied activities like drilling, blasting, and quarrying disturbs the aquifer regime and causes impoundment of water due to overburden and waste dumps.The degradation of water quality, acid water generation, the release of heavy metals, and mixing with shallow water bodies are the common problems in coal mining areas.The groundwater chemistry in the coal mining areas are controlled by various factors such as geological formation, hydrogeological characteristics, and mining conditions of the area, which change drastically from one mine to another [12].
The generation of heavy metals in water is associated with mining operations such as discharge of wastewater, burning of coal, waterlogging in mine pits, and leakage of water from fly ash ponds [13][14][15][16][17][18].In coal mining regions the groundwater and surface water are mainly contaminated by heavy metals from mine drainage, acid mine drainage (AMD) sources, and leachates from mine tailings [19][20][21].The groundwater and surface water were contaminated by acid mine drainage and leachates from the mine tailings, which releases SO 4 2-, Fe, Mn, and many other contaminants to the groundwater [16,19,22,23].
The use of pollution evaluation indices such as heavy metal pollution index (HPI), heavy metal evaluation index (HEI), and degree of contamination (C d ), has become a major component for the evaluation of heavy metal pollutions in water resources [24][25][26].The application of heavy metal pollution index (HPI), and heavy metal evaluation index (HEI), were used to evaluate the metal contaminations in groundwater [21,[27][28][29][30][31][32][33][34].These indices (HPI and HEI) will be helpful, in understanding the level of contamination and the pollution levels of water resources due to the coal mining operations and anthropogenic activities in the study area.The water in these areas has very higher levels of heavy metals load and is highly deleterious for the ecosystem as well as human beings.The health risk assessment model as envisaged by the United States Environmental Protection Agency is an important tool to protect the quality of water supply.The United States Environmental Protection Agency [35] defines human health risk assessment 'as the process of estimating the nature and probability of adverse health effects in humans who may be exposed to chemical contaminants now or in the future'.Heavy metals when present within the permissible limit are not harmful but when accumulated in a particular organ in the human body over a prolonged period can cause bioaccumulation of metals and the concentration of the heavy metal can reach deleterious proportions [35].Ingestion of heavy metals through groundwater and its effect on human health has been studied extensively [36][37][38][39][40][41][42][43][44][45][46][47][48].
The coal deposits of the Ib valley were discovered in the 19 th century and have a reserve of 24.83 billion tonnes of coal which is the third highest in India [49].Basundhara coal mine forms a part of the Ib valley coalfields and lies in the northern part of the basin.Basundhara coal mine is operated by Mahanadi Coalfields Limited (MCL), since the year (1998).It produces around 7 million tonnes of coal per annum (MTPA).The coal mining operation intakes water from the Basundhara River, while the wastewater discharge due to dewatering from coal washeries are released back to the river.Opencast coal mining operations in the Basundhara coal mining area generate huge quantities of solid and liquid wastes due to overburden dumping and from coal washing plants.The shallow aquifer groundwater is the major source of drinking water for the local communities and therefore water quality and its contamination from heavy metals studies are of prime importance in the present endeavour.However, there are no systematic studies on the enrichment of heavy metals, and geochemical characteristics, and probable health risk assessment, in the present study area, due to coal mining operations.So, the objectives of the present studies are (1) study of heavy metal concentrations in water samples using various indexing models, (2) assessment of the potential human health risks due to contamination of heavy metals through ingestion and dermal exposure pathways as evaluated by the model proposed by USEPA, and (3) hydrogeochemical evaluation of heavy metals through multivariate statistical analysis and geochemical modelling to identify the source of origin and to establish the interrelationship amongst the heavy metals in the groundwater of the coal mining region.The results of the present study are expected to deliver a comprehensive baseline report on heavy metals concentrations and its detailed information will be useful for the effective management of water resources in the region due to coal mining operations on a long-term basis.

Study area
The Ib river basin famed for its thick coal seams belonging to the Permo-Carboniferous period is located to the northwest of Odisha state.The Basundhara coal mining area occurs in the western part of the Ib Valley coalfields.The study area is located between latitudes 21º 58ʹ 12 ″ -22º 06ʹ N and longitudes 83º 37ʹ 12 ″ N to 83 ° 51 ′ 36 ″ E, spanning over an area of 300 km 2 (Figure 1).The day temperature rises to around 48°C during the months (April -May), and during winter, the temperature falls to around (6-10°C) in the month from December to January.The average annual rainfall of the area is about 1413 mm, reported in the year 2018 (IMD Jharsuguda Station).The southwest monsoon, which is active from the middle of June through late September, brings about 75% of the annual rainfall.
The coal resources of the Ib-Valley coalfields and the Geology of the coaliferous sediments of the Gondwana Basin was reported in detail [50] and shown in (Figure 2).The Ib valley coal seams are laid in a synclinal rift valley which forms a part of the Mahanadi master basin.The Gondwana sediments are deposited unconformably over a basement of Gangpur rocks of the Precambrian age.The Gondwana sediments are made of sandstone, shale, fireclay, conglomerate, siltstone, and coal seams.The Barakar Formation is made up of sandstone, fireclay, siltstone, shale, and thick seams of coal.The Barakar Formation hosts the thickest coal seams and is best exposed in the Basundhara area.The major river, Basundhara River flows in a southeasterly direction before joining the Ib River (Figure 1).Besides, these tributaries, several ephemeral streams become active during the rainy season and usually go dry during the summer season.The Gondwana Supergroup of rocks, mainly, the thick bands of gritty sandstone act as potential aquifers.The sporadically occurring laterite cappings hold good aquifer potential and are suitable for dug wells.The water bearing formations are hydraulically connected by a number of faults and joints which add to the good aquifer potential of the region.

Water sampling and analysis
A total of 56 nos. of water samples (47 groundwater, 6 surface water, and 3 from coal mining sites) were collected with GPS positions from the Basundhara coal mining area and its adjoining areas during the (pre-monsoon period, 2019).For the analysis of the heavy metals, water samples were filtered through 0.45 µm (Whatman filter paper) to eliminate suspended materials and stored in 100 ml highdensity polythene bottles.The depth to the water level of groundwater samples (dug wells, tube wells, and bore wells) were presented in the supplementary table (Table S1).Based on the information the depth of water level varies from 0.97 m to a maximum depth of 12.55 m.The depth to water level data indicates shallow aquifer conditions prevailing in the coal mining area.
The water samples were analysed using the APHA [51] method.pH and Eh were measured by HANNA instruments (Model No: HI991003) and TDS were analysed by using a benchtop metre (Hanna Instruments, Model HI5321-02), and shown in the supplementary table (Table S1).For the heavy metals analysis, filtered groundwater samples were acidified by ultra-pure nitric acid (HNO 3 ), to maintain the pH level < 2. The heavy metals such as Ag, Al, Ba, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, Sr, V, and Zn were analysed using an inductively coupled plasma-mass spectroscope (ICP-MS) (Thermo Fisher Scientific, X Series-II, Sl.No. 0195c) at the Centre for Materials for Electronics Technology (C-MET), Hyderabad.The accuracy and precision of water analyses were cross-checked by using multi-elemental reference standards (NIST-SRM 1643e, USA) and the duplicates were analysed, after every 10 samples to maintain the accuracy and precision of the results.The results of observed values for all the heavy metals were found within the 10% SDV of the reported values.The sensitivity analyses/limits of detection of each individual heavy metal analysed by ICP-MS were presented in (Table 1).

Computation of Heavy Metal Pollution Index (HPI)
Heavy metal pollution index (HPI), a mathematical tool is used for determining the impact of heavy metals on water quality [33].For HPI calculation, a unit weight (Wi) value is assigned, which is inversely proportional to the standard permissible value (Si) for each heavy metals (Ag, Al, Ba, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, Sr, V, Zn).In the present study, Si refers to the highest permissible value or maximum allowable concentrations; and Ii indicates the maximum desirable limit in drinking water standards suggested by WHO [52].HPI can be calculated using the equation [28].
Where Wi is the unit weight, Qi-sub-index of the i th parameter, and n is the number of parameters considered for calculation.The sub-index can be computed by, Mi is the measured value, Ii is the desirable limit, and Si is the permissible limit for heavy metals of the i th parameter.The maximum critical limit of HPI for water is marked at 100, and above this number, it is treated as unsuitable for drinking.

Computation of Heavy Metal Evaluation Index (HEI)
Heavy metal evaluation index (HEI), is a metal index that helps in determining the level of contaminations in drinking water.It is an indicator of the overall quality of potable water.HEI was introduced by Edet and Offiong, 2002 [53] to surface water (river) and gradually applied to the groundwater studies to study the contamination levels of heavy metals in groundwater [21,24,34].The HEI can be calculated using the formulae, Where, Mi-measured values, Si-permissible limits of heavy metal of the i th parameter

Health risk assessment model
Risk assessment modelling is accomplished in a number of steps such as hazard identification, exposure assessment, and toxicity (dose-response) assessment, and finally, risk characterisation.For determining the amount of exposure, it is imperative to calculate the average daily dose (ADD) of contaminant through different pathways viz., soil, groundwater, and food.Ingesting directly, inhaling through the mouth and nose and absorption through the epidermis are the crucial pathways that can aggravate the levels of exposure of metals in the human body.Amongst these three pathways, the most dominant and active pathway appears to be the ingestion pathway [35,40,43,48,54].The toxicity factors which underscore the degree of risk may be classed as reference dose (RfD) for noncarcinogenic risk and slope factor (SF) for carcinogen risk calculation [55].The risk is based on the average daily dose (ADD) and RfD or SF values of selected heavy metals.In order to find out the exposure dose through ingestion pathways, USEPA has recommended the following formula [35,[56][57][58] Where ADD stands for the average daily dose (mg/kg/day), C represents the concentration of the heavy metals (mg/l), IR is the average daily water intake or consumption rate (L/day: Children-0.78L/day, Adults-3 L/day [59]), EF is the exposure frequency (days/year, 365 days), ED is the exposure duration (Years: Children-12 years, Adults-66.4years) [60,61], BW is the average body weight (kg: Children:18.7 kg, Adults-57.5 kg [48,62,63]), AT is the average exposure time (days: Children-4380 days, Adults-24236 days).The health risk assessment is measured by using the hazard quotient (HQ), which can be calculated from the average ingestion of ions in water over a considerable period of time with reference to the corresponding reference dose (RfD) using the following formula Where HQ is the hazard quotient and RfD is the reference dose of individual heavy metals (mg/kg/day).The RfD values of individual heavy metals [64] are represented in supplementary (Table S2).A hazard index (HI), involving summation of all the calculated HQ values is employed to assess the risk associated with multiple metals in the drinking water.
Where HI is the hazard index and HQi is the hazard quotient of individual metals.The deleterious effect of drinking water on human health is accentuated when HI > 1 [35,64].

Heavy metal concentrations in groundwater
The results for the analysis of the heavy metals with their descriptive statistics are presented in (Table 1).The data were analysed for the water samples for drinking water specifications, based on the standard guidelines [52,65,66].The pH in the groundwater varied from 5.55-8.3,with a mean value of 7 ± 0.64, indicating an acidic to an alkaline environment and spatial distribution map shown in (Figure 3).Total dissolved solids (TDS) vary between 151-1088 mg/l, with a mean value of 503 ± 209 mg/l.The mean value and relative abundance of heavy metal concentration in groundwater decrease in the order of Fe > Zn > Al > Ba > Mn > Sr > Ni > Ag > Cu > Pb > V > Cr > Co > Cd.Silver (Ag) concentration was much below the desirable limit of 100 µg/l, and ranged from 3.4 to 29 μg/l, a mean value of 8.4 ± 5.4 µg/l.Ag in groundwater is due to the presence of sulphide bearing minerals such as galena, arsenopyrite, pyrite and argentite.Aluminium (Al) concentration in groundwater breached the permissible limit of 200 μg/l, and ranged from 216 to 807 μg/l, a mean value of 391 ± 108 μg/l.Al in groundwater could have been derived from the feldspathic rocks like microcline, orthoclase present in sandstone, shale, laterite cappings, sedimentary clays, and coal bearing sediments [67].Barium (Ba) varies from 18.4 to 2189 μg/l, a mean value of 146 ± 304 μg/l, and reports below the permissible limit except for Sl no.27.Cadmium (Cd) concentration varies between 0 to 4.52 μg/l, having a mean of 0.6 ± 1.1 μg/l.Cobalt (Co) concentration varies from 0 to 16.1 μg/l, with a mean value of 0.9 ± 2.5 μg/l.Chromium (Cr) concentration varies from 0 to 4.04 μg/l, with a mean value of 1.1 ± 1.04 μg/l.Copper (Cu) concentration varies from 1.12 to 15.5 μg/l, and a mean of 4.1 ± 3.1 μg/l.The heavy metals, Cd, Co, Cr, and Cu concentration are below the permissible limit.Iron (Fe) concentration in groundwater varies from 67 to 8356 μg/l, a mean value of 1618 ± 2111 μg/l, and 43% of samples are above the permissible limit of 1000 μg/l.Fe concentration in groundwater is due to mixing with mine tailings, weathering of the lateritic profile, and waste effluents from coal mines.Fe concentration in groundwater is largely due to oxidation and dissolution of iron bearing minerals such as pyrite present in coal seams, iron shales, ferruginous sediments, and laterite capping in the coal mining areas [68].The leachate from overburden dumps and waste effluents of coal mine and coal mine drainage contribute considerable amount of heavy metals to the groundwater [68,69].Manganese (Mn) concentration varies between 11 to 1040 μg/l, a mean value of 90 ± 164 μg/l, and 4% of samples are found to be above the permissible limit of 300 μg/l.Mn in groundwater is due to leaching from overburden dumps, coal mine effluents, and leaching from the laterite, shale, and greywacke sandstones.Minerals like pyrolusite (MnO 2 .H 2 O) and rhodochrosite (MnCO 3 ) contribute Mn values to the groundwater under near neutral pH conditions [70].Nickel (Ni) concentration varies from 0 to 26 μg/l, a mean value of 8.6 ± 5.4 µg/l, and all are below the permissible limit of 70 µg/l.Pyrite oxidation combined with carbonate dissolution gives rise to a low pH condition which helps in mobilising nickel in groundwater [71].Lead (Pb) concentration varies from 0 to 10.4 μg/l, a mean value of 2.4 ± 2.6 µg/l, which is below the desirable limit of 10 µg/l [66].Lead (Pb) can dissolve in sulphide phase minerals at lower alkalinity and pH conditions, which can retain larger concentrations in groundwater [72].
In coal mining areas, leachate is generated from overburden dumping materials, dissolution of sulphate bearing minerals like galena (PbS), pyrite (FeS 2 ), hydrocerussite (Pb 3 (CO 3 ) 2 (OH) 2 ) contribute Pb to the groundwater.Strontium (Sr) concentration varies from 17 to 370 μg/l, a mean of 91 ± 73 µg/l, and is below the desirable limit of 600 μg/l.Strontium (Sr) occurs as sulphate as in celestite (SrSO 4 ) and to a lesser extent carbonate strontianite (SrCO 3 ), which can dissolve in water [73].The water-soluble compounds of strontium can pollute groundwater and have a deleterious effect on human health.Vanadium (V) concentration varies from 1.09 to 8.25 μg/l, mean value of 2.03 ± 1.02 μg/ l, and is below the permissible limit of 50 μg/l.Vanadium in groundwater can be derived from leaching from laterite cover, bauxite, coal, and petroleum deposits [74].Zinc (Zn) concentration varies from 36 to 3341 μg/l, mean value of 425 ± 646 μg/l, and below the desirable limit of 5000 μg/l.Zinc minerals like Sphalerite (ZnS), Willemite (Zn 2 SO 4 ), and Smithsonite (ZnCO 3 ) can contribute Zn to the groundwater in near neutral conditions.

Hydrogeochemical studies and correlation matrix
The correlation matrix for heavy metals are presented in supplementary table (Table S3).
The parameters like (pH, TDS, SO 4 ) and heavy metals were analysed for establishing the relationship between different elements, their mode of occurrence, and identify their possible sources of origin [75].The correlation matrix for Al vs Fe (−0.21),Mn (0.29), Co (0.50), Ni (0.33), V (0.34), respectively, suggest weak to moderate correlation and indicates a varied source of origin.This shows that the source of Al might be related to the dissolution and leaching from the host rocks, weathering of aluminium silicate minerals, and lateritic soils in coal mining areas.The lateritic rocks may contain a significant concentration of Al, Fe, Co, Ni, V, and other metals [76].The correlation matrix between Mn vs. Co (0.70), Cd (−0.13),Ni (0.41), indicates different sources of origin.A strong correlation between Mn and Co indicates a common source of origin and similar geological provenance for these heavy metals.Higher Mn concentration might be related to the Mn ore bearing provinces of Gangpur Group sediments and Mn rich sediments from the Bonai -Keonjhar belt of Odisha, which hosts rich iron and manganese ore deposits [77].The correlation matrix between Fe vs Co (−0.11),Mn (0.10), Ni (−0.07), and Cd (0.44), Cr (0.48), indicates weak to moderate correlation with different heavy metals.A strong correlation amongst the elements indicates that they have a common origin and similar geochemical behaviour.A negative and weak correlation between Fe and SO 4 2-(−0.18),indicates that pyrite (FeS 2 ) dissolution is not the major factor contributing Fe to groundwater and both are independent parameters having different sources of origin.Al is having a weak to negative correlation with pH, and as it is near neutral/saturation conditions since it precipitates as amorphous Al(OH) 3 compound, at pH higher than 4.0.In a strongly alkaline environment, Al becomes mobile and higher concentrations was observed in the coal mining region.The Fe and Mn concentrations are inversely correlated to Eh, which determines their oxidation state and, their mobility increases [78].
The weathering of pyrite (FeS 2 ), arsenopyrite (FeAsS) which commonly occurs in coal bearing strata, as well as by dissolution of other sulphate bearing minerals, gypsum (CaSO 4 .2H 2 O) and anhydrite (CaSO 4 ) releases sulphate (SO 4 2− ) in the mine water and its surrounding streams.pH (5.55-8.3)does not encourage high oxidation and dissolution of metal sulphides such as arsenopyrite (FeAsS) and pyrite (FeS 2 ) which are likely to be formed in the pH range (3.6 to 5.7) [79].At higher levels, carbonate phase minerals under a highly alkaline pH environment and low SO 4 2− concentration could have slowed the dissolution of metals as clay minerals provide a good adsorbing surface for many trace elements [10].This may be the possible reason for the low metal concentration in the Basundhara coal mining area.In the present study, metals like Al, Fe, and Mn show a little higher concentration than the other heavy metals in the groundwater, which could be attributed to the weathering of coal bearing sediments, the release of metals from black shales, and oxidation of pyrite and iron hydroxide present in the aquifer [68].Both Fe and Mn concentrations show a negative correlation with SO 4 2-such as (R 2 = −0.18for Fe) and (R 2 = −0.02with Mn), respectively.The precipitation of Fe-oxyhydroxides and Mn-oxyhydroxides, explains the weak to negative correlation with the neutral to alkaline pH conditions.The correlation between Fe vs. SO 4 2-(−0.18)and Mn vs. SO 4 2-(−0.02), is very negative.
However, a strong correlation exists in the case of Mn and SO 4 2-in samples collected from the leachate of the coal mine or in proximity to the mine.Here the pH is subdued due to the generation of AMD close to the coal mine (Sl No. 20,22,23).In (Sl.No. 56) sample, shows the high concentration of Fe and SO 4 2-with low pH indicates the area has been affected by AMD [80].Precipitation of Fe oxyhydroxide in form of goethite explains the poor correlation between Fe and SO 4 2-.The low concentration of Fe may be attributed to its precipitation in the aquifer even when Mn remains in the dissolved state in the aquifer [81].The presence of carbonate bearing local host rocks also has a significant impact on the neutralisation of acid mine drainage [82].The surface runoff in the region takes its natural course and flows to the surrounding streams which finally fall into the Basundhara River.During high base flow aided by high rainfall (1400 mm/ year), and the impact of dilution causes significant neutralisation of the mine drainage in the coal mining areas.Due to near neutral pH values, low levels of SO 4 2-and heavy metals concentrations in the Basundhara coal mining area, the possibility of generation of acid mine drainage is negligible.

Geochemical modelling and saturation indices (SI) analysis
The saturation index of the groundwater samples were analysed by using USGS hydrogeochemical modelling software, PHREEQC for Windows [83].The saturation indices of Al, Fe, and Mn for various mineral phases and their summary statistics are presented in (  S1).The saturation state for gibbsite along with iron minerals like goethite, haematite, Fe(OH) 3 except melanterite indicates their leaching from a lateritic crust.Mn minerals like manganite, jarosite, hausmannite are present in an undersaturated condition indicating that Mn minerals can still be undergoing dissolution in the aquifer.The negative correlation between SO 4

2-
vs. Fe, suggests that the dissolution of pyrite bearing mineral phases is very negligible.In the present scenario, mineral stability phases suggest that reducing conditions may continue in the groundwater system in the coal mining regions.

Principal component analysis (PCA)
Statistical tools like Correlation matrix and Principal component analysis (PCA) can help in predicting the potential sources of the physicochemical parameters and the heavy metal load.In this study, PCA was performed for the heavy metals in water samples in the Basundhara coal mining area to understand the heavy metal association and their origin.The varimax rotation method was also applied to maximise the sum of the variance of the factor coefficients which would help in better explaining the possible sources and contributors of heavy metals.The scree plot has been used to decipher the number of principal components required to recognise the heavy metals (Figure 4(a)).The three most significant factors as is evident from the loading map for PCA are presented in (Figure 4(b)).Based on the scree plots, six factors with eigenvalues >1 were extracted from the PCA matrix, exhibiting 72.96% of the total variance.The derived factor loadings, percentage of variance, and cumulative percentage from the data set are presented in (Table 3).The first component (PC1) has 23.17% of the variance with an eigenvalue of 3.24 and is strongly positive loaded on Sr (0.85) and Mn (0.78) while a moderate load is seen in the case of Co (0.72) and Ni (0.64).It illustrates that Sr, Co, and Ni are associated with the earth's crust and the geological formation of the area [40,84].Mn is associated with coal as well as the rocks and ore deposits in the vicinity of the study area [40,85].The second component (PC2) explains 16.14% of the variance with an eigenvalue of 2.26 and strongly positive load on Fe (0.78), Cr (0.77), and moderate load on Cd (0.75).Fe is associated with coal and is also  contributed by weathering of rocks and laterite cappings of the coal mining areas.Fe is derived from weathering and leaching of mine effluents [86,87].Cr is associated with anthropogenic sources like industrial and vehicular pollution.Cd is associated with the burning of coal, domestic sewage, and fertilisers from agricultural runoff [44,87].The third component (PC3) shows a 9.41% variance with an eigenvalue of 1.32 and a strongly positive load on Al (0.81) and moderate load on V (0.69).Al is associated with laterite capping, sedimentary clays, and coal bearing sediments [67].V is associated with laterite cover, bauxite, coal, and petroleum deposits [74].The fourth component (PC4) contributes 8.66% variance with an eigenvalue of 1.21 and moderate positive load on Cu (0.72) and Pb (0.58).Cu and Pb are associated with the leaching of minerals and mine wastes [88].The presence of Pb could have been derived from storage tanks, water pipes, abandoned waterfilled containers [89].The fifth component (PC5) explains 8.41% variance with an eigenvalue of 1.18 and strongly loads on Ba (0.93).The sixth components show a 7.17% variance with an eigenvalue of 1 and a strong load on Ag (0.90).Ag is associated with sulphide bearing minerals such as galena, arsenopyrite, pyrite and argentite.The pairs may be described as Sr vs Mn, and Fe vs Cr (Figure 4(b)).The Sr concentration shows positive loading with Mn in PC1 whereas Fe shows positive loading with Cr in PC2.This shows that similar pathways of contaminants contribute to groundwater contamination.The PCA analysis shows that the source of heavy metals has been derived from six different sources.

HPI and HEI assessment
The HPI values of water samples and their descriptive statistics are presented in Table 1a and spatial distribution map is presented in Figure 5.The HPI levels of water samples are classified under these categories: low < 15, moderate 15-45, moderate to heavy 45-90, and high > 90 (Table 4).In the present study, 16% of water samples belong to moderate HPI, and 84% belong to moderate to heavy HPI values respectively, and none of the samples exceeds the critical pollution index level of (100).The highest HPI values are observed at locations (Sl.No 20) and the lowest at (Sl.No. 10).Relatively higher HPI values are related to the higher contents of Al, Fe, and Mn concentrations in groundwater samples.The majority of locations having high HPI values are located in the proximity of the Basundhara opencast coal mines.
The higher HPI values in groundwater are due to coal mining related activities, metals released from the overburden dump sites, and various coal processing activities.The HEI values with descriptive statistics of the water samples are presented in supplementary table (Table S1) and the spatial distribution map is presented (Figure 6).HEI scores can be classified as, low (<10), medium (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20), and high (>20), category contaminations respectively (Table 4).HEI values of the water samples varied from 2 to 11, mean value of 4.4 and about 95%, and 5% samples are under low to medium level of contamination respectively.Based on the HPI and HEI analysis, the overall groundwater quality with respect to heavy metal concentrations was found to be safe from contamination.The coal mining operation has not substantially contributed heavy metals to the groundwater in the region.

Non-carcinogenic risk
Studies have not been initiated so far to assess the health risk due to heavy metals in water on the local populace in the Basundhara area.The direct ingestion or oral uptake of heavy metals in water is more lethal than through dermal absorption.The carcinogenic risks and non- carcinogenic risks to human health are calculated from the study area based on the health risk model suggested by USEPA [35].The examined values of HQ and HI for children and adults are presented in (Table 5).In some samples, the HQ and HI values for children and adults exceed the safe limits (>1).Based on HI values the adults are more prone to health risk problems as compared to children.The non-carcinogenic risks of heavy metals are expressed as HQ values.
The HQ mean values of heavy metals are shown in descending order of Mn > Co > Fe > Zn > Cd > Cu > Pb > Ni > Al > Cr > Sr, for both children and adults.The HQ value of Mn exceeds the safe limit in the sample (No. 1 and 4) which indicates that Mn is the major contributor of noncarcinogenic risk in both the above two locations in the case of adults as well as children.In the case of Co, samples (4 and 27) for adults and sample (No. 4) for children exceed HQ >1, which suggests that the location was affected by non-carcinogenic risk.Except for these two heavy metals i.e.Mn and Co, having high HQ values, the rest of the heavy metals do not have any non-carcinogenic risk.This study confirms that Mn and Co are the two major heavy metals that have non-carcinogenic risks but have affected a few locations in the study area.

Carcinogenic risk
In the study area, amongst the investigated heavy metals, Cd, Pb and Ni are the elements that pose carcinogenic risks.The study is focused on the average daily dose (ADD) for Cd, Pb, and Ni present as heavy metal load in the water of the study area.The ADD results of Cd in water samples vary from 0 to 0.0001 and 0 to 0.0002 for children and adults respectively.The ADD values of Cd for children and adults are below the RfD value in the study area which indicates that there is no carcinogenic risk due to the incidence of Cd.The ADD values of Pb for children and adults range from 0 to 0.0004 and 0 to 0.005 respectively indicating that it poses no carcinogenic risk due to the presence of Pb in water.The ADD values of Ni are between 0 to 0.001 for children and adults, respectively and therefore there is no carcinogenic risk ascribed to the element Ni.This study shows that the above selected three heavy metals present in the water have no carcinogenic adverse effects on the health of children and adults.

Total health risk assessment
Generally, the major ions do not pose a serious threat to human health based on water quality results of similar values as compared to the WHO and BIS limits.So the major carcinogenic and non-carcinogenic health related issues are due to the presence of heavy metals in drinking water.The Health risk assessment following the USEPA model for groundwater was carried out in the study area.The health risks due to heavy metal contaminations for adults through drinking pathways in the study area have been analysed and shown in (Table 5).The reference dose of individual selected heavy metals (Al, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, Sr, Zn) are presented in supplementary (Table S2).However, to determine the portability of the water and the risk associated with the ingestion of chemical toxins through drinking water pathways, HQ was estimated.The Hazard Quotient (HQ) value (mg/kg/day) of heavy metal for Co ranges from 0 to 2.8, with a mean of 0.162 whereas Mn varies from 0.023 to 2.260, with a mean of 0.194.The HQ values of Co and Mn in 5% and 4% samples exceed the health risk limit of 1 respectively.The mean value of HQ for selected heavy metals (Al, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, Sr, Zn) is <1, suggesting that individual heavy metals have little effect on human health (Table 5).Heavy metals bioaccumulate in the human body and therefore have long term effects on health.The hazard index (HI) method is a tool that flags the risk of each component in water [43,90, 91-94,] and helps in determining the cumulative risk from ingestion pathways to habitants in a locality.The HI value varies from 0.104 to 4.577, with a mean of 0.731.The USEPA [64], mandates that the health risk limit for HI should be <1 for water to be potable.Around 20% of the water sample exceed the safe limit for HI and may have serious implications for human health as these have a strong potential non-carcinogenic quotient.

Uncertainty analysis
The present work suffers from certain limitations.Uncertainty analysis demonstrates the reliability of the risk evaluation outcomes.A health risk appraisal may require consideration of three major pathways for exposure viz., ingestion, inhalation, and dermal contact, via air, water, soil, and the food chain.In the present study, the ingestion of heavy metals through drinking water was evaluated.The other pathways like dermal and inhalation pathways were ignored.The weight and daily water consumption of children and adults were obtained from the USEPA and ICMR standard which may be a departure from actual field observations.The study area since located in a tropical area where subsistence agriculture is practiced, the staple diet of the local habitants is based on carbohydrates only and barely any protein which also adds to the health woes of the habitants.Further, the study group has not been classified based on gender.The human migration in search of livelihood has also not been accounted for the risk analysis evaluation exercise and will be more robust if a more number of variables is considered.

Conclusion
The heavy metal concentrations in the groundwater were studied and results were discussed based on the heavy metal concentrations, water chemistry, geochemical characteristics, principal component analysis, saturation indices, redox conditions, and assessment of human health risks in the Basundhara coal mining region.The groundwater shows, in majority of the locations, Al, Fe, Mn concentrations exceeded the WHO guidelines values and would require treatment for the drinking water needs.The presence of laterite cappings and nearby mineral rich ore bearing provinces also strongly contributes to the enrichment of heavy metals like Al, Fe, and Mn concentrations, in the coal mining region.The mobilisation of heavy metals are governed by the oxidising and reducing conditions in the aquifer.The interrelationship between heavy metals has been characterised by six principal components.Heavy metals like Al, Fe, and Mn are the major contributor to high HPI values.HPI values show that most of the water samples are under the medium category and none of the samples exceeds the critical pollution index level (100).Based on HEI values, about 95% of samples are of low level, and the rest 5% are moderate level contamination indices.The exposure risk assessment model indicates the level of risk to humans from Co and Mn, which are important pollutants causing noncarcinogenic risks.About 20% of water samples exceed the HI reference value of 1.
The study can be of immense value to the community and the habitants dependent on groundwater for drinking.The sensitivity analysis shows that the Mn and Co concentration in the groundwater are important parameters to the risk assessment.
Overall, the results shows that the groundwater from the Basundhara coal mining area is moderately contaminated by heavy metals Al, Fe, and Mn to different degrees; in addition, exposure to the groundwater does not cause carcinogenic risk.However, a low non-carcinogenic risk remains.Therefore, for the maintenance of public health, more attention needs to be paid to the health risks posed by heavy metals and ensure that the groundwater quality meets drinking water standards as outlined by WHO and BIS guidelines values.The long term monitoring and assessment of heavy metal contaminations of groundwater will help in spreading awareness amongst communities residing in the vicinity of coal mining areas.The outcome of the present study will be helpful to create a baseline database, which can be used for systematic investigation, and to utilise the knowledge and resources for further improvement and protection of water resources in the coal mining regions and elsewhere in the country.

Figure 1 .
Figure 1.Sample location map of the study area.

Figure 3 .
Figure 3. Spatial distribution of pH of groundwater samples.

Figure 5 .
Figure 5. Spatial distribution of heavy metal pollution index (HPI) map.

Table 1 .
Summary statistics of heavy metals concentrations in the coal mining region.
Std. Dev-Standard Deviation, LOD -Limit of Detection, HPI-Heavy metal Pollution Index, HEI-Heavy metal Evaluation Index.

Table 2 )
. Speciation analysis for Fe mineral phases shows, oxide phases (Goethite: FeO(OH); Haematite: Fe 2 O 3 ), are in the supersaturated state, while the sulphate and carbonate bearing phases (Melanterite FeSO 4 • 7H 2 O; Siderite FeCO 3 ) are in undersaturated to near saturation state in groundwater.Mn bearing minerals of oxide phases such as Manganite: MnO(OH); Hausmannite: Mn 3 O 4 and Pyrolusite: MnO 2 .H 2 O are highly undersaturated and with the carbonate phase mineral Rhodochrosite (MnCO 3 ), are in the undersaturation to a near neutral state.Al bearing minerals of oxides phases Al(OH) 3 and Gibbsite (Al(OH) 3 ) is in a saturated state in groundwater.The Eh (v) values fluctuate from −107 to 143, indicating a wide range of geochemical conditions prevailing in the aquifer with a more redox state of conditions and presented in the supplementary table (

Table 2 .
Summary statistics of saturation index of metals.

Table 3 .
Factor loadings of each heavy metal with their variance and Eigen values.

Table 4 .
Water quality classification based on different heavy metal indices values.
Figure 6.Spatial distribution of heavy metal evaluation index (HEI).

Table 5 .
Summary statistics of HQ and HI value of Basundhara coal mining area.