Evaluates the physicochemical characteristics and change in the concentration of arsenic species in groundwater, release from 2nd aquifer of coal mining area, its drain outlet and different sites of reservoir: Application of multivariate techniques


 For energy production activity the ground water of 2nd aquifer (AQ2) of block 2 of Thar coal field is drained out in a reservoir/Gorano dam. In present study the physico-chemical and toxic elemental levels in ground water samples of 2nd aquifer (AQ2) of Thar coalfield at the depth of 100–120 m was carried out. The further sampling of water was carried out from drain outlet (DW) and four sampling sites of reservoir (S1, S2, S3 and S4). The sediment samples of water reservoir were collected from four sites and analyzed for toxic elements. The large data set of water quality of AQ2, DW and four samples sites at different spots of reservoir, carried out during 2019 for 17 parameters in triplicate of each composite samples were further analyzed by multivariate statistical techniques involving cluster (CA) and principal component analysis (PCA). The all water samples were analyzed for physicochemical parameters including cations and anions. The resulted data of cations (sodium, potassium, calcium, magnesium, cadmium, lead, total arsenic and iron), anions (chloride, fluoride, nitrate, sulphate and bicarbonate) and physicochemical parameters were descript in terms of basic statistical parameters, PCA and CA. The chemical correlations among different parameters were observed by PCA, which were employed to categorize the water samples by CA. The positive correlation of fluoride (F−) and total arsenic with sodium and bicarbonate indicated that the GW (AQ2), DW and water samples of reservoir with high conductivity and salinity stabilized the toxic elements in the water samples of different origin. Results showed that all collected water samples have high values of sodium absorption ratio. The water quality assessment indicated that the high levels of physicochemical parameters and toxic elements in ground and reservoir water have concern in and make it unsuitable for drinking and domestic purposes.


Introduction
The maximum content of lignite coal is originated in district Tharparkar, Sindh, Pakistan. At the present time different national and international agencies are working on power plans (Ali et al. 2015;). In ground water of mining areas, numbers of pollutants such as organic and inorganic minerals are dissolved in water and there by adversely distressing its quality and utility (Singh et al. 2012; Kurwadkar, 2019). In adding, water can liquefy lot of the chemicals from soil and rock. The refuse masses containing sulphides (pyrite) be able to rise the acid mine drainage problems in the environment. Departure of huge quantity of water for diverse requirement through mining processes frequently causes considerable changes in its physico-chemical characteristics as well as regularly creates depression of groundwater in the region which may result in its shortage (Ali et al., 2015).
The feature of groundwater based on different factors and their contents, which are generally resulting from the geological information of the mining area. Arsenic (As) in the water is a severe natural misfortune for living and non-livings, which instigate from geological and anthropogenic resources Generally the organic As compounds are less toxic than inorganic (As compounds). For the determining of inorganic arsenic in water the very sensitive methods was developed (Marahel et al., 2011). Cadmium is also dangerous to living beings, it can cause of injures of organs living things such as kidney, Lungs and Liver (Bernard, 2008;Satarug et al., 2011). Intake of cadmium for long time creates more problems for metabolism of calcium in the life system which initiates the damage of cell and its death (Chakraborty et al., 2013). Cadmium also discourages the function of essential elements such as zinc assisted enzymes by exchange. It also causes cancer and dermatogens include radiation, maternal infections, chemicals, and drugs. (Barrera et al., 1997).
Lead is more in quantity in atmosphere, its cause of contamination in water and air and also other things.
It is hazards metal which gradually increases and causes such diseases damage of brain, anemia, kidney disorder and hematological disorder (Shah et al., 2010). Livings beings are affected by lead (Pb) from anthropogenic as well as natural example contaminated beverages, food, drinking water, industrial emissions, soil. To analyze the environmental and biological samples numbers of techniques was comprehensively used including: atomic absorption spectrometry (Detcheva and Grobecker, 2008), inductively coupled plasma optical emission spectrometry (Zhu and Alexandratos, 2007) and others.
For coal mining purposes the elimination of water of aquifer (ground water), its removal devoid of drain system can be expensive. The open cut withdrawal system needs the pro cient elimination of overload and inter-burden stratum in order to access coal seams. The dewatering of the surface mine is carried out by use of an pumping-out via a borehole (entirely piercing the whole depth of the aquifer), then the water of aquifer of coal mining is forced out at a regular ejection rate to reduce the piezometric level of the ground water beneath the coal mining horizon at the pit boundary (Liang et al., 2017).
The current study has been is in connection with Thar coal eld, where the coal mining and power generation plants by Sindh Engro Coal Mining Company is carried out. Initially the ground water of 2nd aquifer of block 2 of Thar coal eld, had started releasing into the reservoir, locally called Gorano dam, to diminish the amount of groundwater that lies underneath of the 2nd aquifer. This study was served during 2019 for determination of trace and toxic elements (As, Cd and Pb) and other parameters such as (pH, EC, TDS, salinity, Na + , K + , Ca 2+ , Mg 2+ , Cl − , SO 4 2− , HCO 3 − , Fe 3+ , NO 3 − , and F − ) in ground water (2nd aquifer of block-2) of coal mining area of Thar coal, drain out let of it and from four sites of reservoir at equal distance (500m). The sediment of four sites of reservoir/gorano dam were analysed for toxic elements. The accuracy of the methodologies was determined by simultaneously analyzed matrices matched certi ed sample of water (Table 2). Table 2 The physicochemical parameters, cation and anions concentration in ground water of aquifer AQ2, drain out let (DW) and sampling sites (S1, S2, S3 and S4) of Gorano reservoir The huge data set obtained was calculated successfully with multivariate techniques, principal component (PCA) and cluster analysis (CA) to estimate the information concerning the similarities /dissimilarities between the diverse sampling sites, to recognize the water quality variables. These methods have been used to assess the effectiveness of water quality monitoring networks of aquifer 2 and drain outlet in reservoir, plans to increase the number of examining sites and select the basic parameters of water quality. Along these lines, it is conceivable to identify and evacuate sites of reservoir as well as repetitive parameters to reduce the economic cost of monitoring plan without sacri cing uctuation in water quality data. The analytical precision for the estimations of cations and anions indicates by anions/cations balance error /ionic balance error, calculated based on the concentration of all ions in meq/L. To verify the appropriateness of drained ground water of 2nd aquifer of coal mining area in a reservoir for irrigation purpose., the sodium absorption ratio of each site were also calculated 2. Material And Methods

Study area
The huge reservoir of natural coal Thar in district Tharparkar, Sindh, Pakistan, detail is reported in our previous work (Ali et al., 2017). Different power generation projects are carried out by different national and international agencies (Ali J et al. 2015;). The discharge of the water of 2nd aquifer of block 2 of Thar coal mining eld in Gorano reservoir, which are at the depth of 120 m. Gorano reservoir is newly developed reservoir in 2017 spread in 1,500 acres. This site completely reserved for dewatering discharge by a pipe which is 27 Km away from coal mine site. The Gorano dam/reservoir is geographically located at the southern part of Pakistan. The water sampling was carried out from drain outlet and from four sites of reservoir as mention in Fig. 1.

Chemical reagents and glassware
Ultrapure water which was used throughout procedure system was taken from ELGA lab water system (Bucks, UK). The acids and oxidizing reagents obtained from Merck (Darmstadt, Germany) analytical reagents were rated for purpose determination of trace and toxic elements in studied samples. The standard solution of As, Cd and Pb were prepared from certi ed standard solution (1000 ppm) obtained from Fluka Kamika (Buchs, Switzerland) and made upto volume with. 0.1 M HNO 3 . Before analysis all the glassware was kept in 5M HNO 3 for 24 hours and then rinsed with deionized water.

Instrumentation
The pH meter (720-pH meter, Metrohm) was applied for the determination of pH, of water samples of different origin. Inductively coupled plasma optical emission spectroscopy (ICP-OES) The Thermo Scienti c™ iCAP™ 7000 Series ICP-OES, A atomic absorption spectrometer Model 700, Perkin Elmer (Norwalk, CT, USA), equipped with hollow cathode lamps with addition of an air-acetylene burner has been used. For the physico-chemical parameters such as, Electrical conductivity (EC), and Total dissolved solids (TDS) conductometer (Ino Lab conduc. 720, Germany) was used.

Sampling and pretreatment
Samples of water were collected in summer of 2019 in the month of May and June. The groundwater was sampled directly from second aquifer with the legal permission and assistance of Sindh coal authority. The high power compressor was operated for withdrawing ground water of second aquifer during construction of power plant. Then further sampling was carried out from the drain outlet of groundwater of second aquifer of block-2 in a gorano dam/reservoir. Then water samples were also collected from reservoir of four sites at equal distance of 500 m (length of dam is 2.0 km), with the help of water grab sampler (1.5 L capacity), equipped with a simple pull-ring that allowed for sampling of water at surface and depths of 20-30 cm, from 10 to 20 spots of each sites (S1, S2, S3 and S4) at random and held in well stopper polyethylene plastic bottles, formerly washed with 10% nitric acid and rinsed with ultrapure water obtained from ELGA Lab water system. Then collected water (AQ2 and DW) and from four sampling sites was made ve composite samples (surface and lower levels of same site). Then ltered through 0.45 microspore size membrane for the removal of suspended particular materials to laboratory carefully and labeled. Finally samples of water were stored at 4 0 C in refrigerator before analysis. The sampling of sediment was carried out simultaneously with water from 5 to 7 spots of each site, SD1, SD2, SD3 and SD4 (except AQ2 and DW sites) by means of an Ekman dredge which was designed as SD1, SD2, SD3 and SD4. The samples were kept tightly closed in polyethylene bags. At the same day the water and sediment samples were transported in the laboratory.
In laboratory sediment samples were dried in an electric oven at 70 C for 24 h. Then crushed and passing through a sieve (200 µm). For the preparation of sediment sample, triplicate of each samples (0.5g) were taken in crucibles and added 3.0 to 5.0 mL of mixture of nitric acid and hydrogen peroxide at 2:1 ratio.
After 10 to 15 minutes the crucibles placed on hot plate at 70 0 C temperature for digestion purpose then add 10 mL of 0.2 M solution of nitric acid (HNO 3 ). Finally lter solution in 100mL sample bottles by using whatman lter paper and samples were prepared for run on instrument.

Analytical procedures
The analysis of water samples of different origin for different physico chemical parameters in water samples of AQ2, DW, S1, S2, S3 and S4, were carried out in triplicate of each composite samples by . Initially 50 mL of each water samples were taken in 100 ml of beakers and with the help of pH-meter and conductometer, different physico-chemical parameters such as pH, TDS, EC and salinity were measured respectively (Table 1S). The TDS were not measured in water samples directly due to its high values, and for that purposes the each composite samples (n = 10) were 10 times diluted with deionized water then conductometer was applied to measure TDS. An ion chromatograph was used for the estimation of anions levels in water samples of different origin using peak area of each with an error < 2% ( . Basic statistics and correlation calculations were conducted in order to give preliminary information concerning the water quality data of six sites (AQ2, DW, S1, S2, S3 and S4).

Application of multivariate technique
Basic statistics was carried out in order to give initial information about the change of water quality of different origin (ground water/reservoir). The multivariate techniques, CA and PCA were applied to evaluate the quality of water samples at different origin (AQ2, DW, S1, S2, S3 and S4) (

Analytical gure of merit
The calibration of standards were prepared for Na, K, Ca, Mg, Fe, As, Cd and Pb at required concentration range, using FAAS and ICP-OES. The limits of detection (LOD) was estimated according to (LOD = x × s/m) where "s" is the standard deviation of 10 determinations of the blank and "m" is the slope of the calibration graph attained, the LODs; 5.52, 14.0, 164.3, 2.46, 69.2, mg/L for Na, K, Ca, Mg and Fe, respectively. The calibration graphs of As, Cd and Pb were prepared their solution starting from the limit of quanti cation up to 20 µg/L using ICP-OES. The LOD were found as 0.02, 0.13 and 0.05 µg/L, respectively. The precision of methods was carried out by analysis of matrices matched certi ed sample of water (SRM 1643e) and sediment (SRM 1944) as shown in Table 1. The prominence of the analytical resulted data was maintained throughout with cautious standardization, measurement of procedural blanks and all the time average resulted data obtained by triplicate analysis of each composite sample. For justi cation of the resulted data of cations and anions in water samples of each (AQ2, DW, S1, S2, S3 and S4) were within the ± 5%., the ionic balances was calculated by equation as revealed in our earlier work Arain et al., 2014) was observed as ± 5%

Results And Discussion
The mean with standard deviation (n = 30) of physicochemical parameters of groundwater samples (AQ2), drain out let and four sites of reservoir (Gorano dam) are given in Table 2. To estimate the correlations using Pearson correlation coe cients (r) among the data of variables, were estimated. The obtained resulted values of different variables/parameters of all collected water samples of different origin, groundwater samples (AQ2), drain outlet (DW) and from different four sites of reservoir (S1, S2, S3 and S4) were extensively deviate from the recommended standard values of WHO, for drinking water.
However the values of pH and SO 4 2− in all water samples except ground water of AQ2 are in agreement with recommended standards of WHO for drinking water. The pH of AQ2, DW, S1, S2, S3 and S4 of was found in the range of 7.75 − 8.5, 7.6-7.92, 7.5-7.75, 7.2-7.64, 7.0 -7.4 and 6.7-7.2 respectively, are consisted with WHO regulated levels. This con rms that it is slightly alkaline in nature, which is attributable to the concentration of HCO 3 ion, which is relatively high in all sites of reservoir, ranged as 249 to 397 mg L − 1 . However, the pH values of water samples of reservoir were diminished with the distance. This reducing in pH is taking placed owing to the decline in dissolved minerals (Tripathy 2002).
The mean values with standard deviation of TDS and EC of water samples of different origin, AQ2, DW, S1, S2, S3 and S4 were given in Table 2. The TDS values in Aq2, DW, S1, S2, S3 and S4 were found in the range of 8940-9710, 8680-9440, 8680-9240, 6980-7840, 5280-5870, 4650-5210 mg/L, respectively. The EC were found in the range of 9.8 − 11.5, 9.46-11.0, 8.86-10.5, 7.68-9.54, 6.75-8.12, 5.89-7.45 mS/cm, respectively. The total dissolved and electrical conductivity enhanced up to ve times than WHO suggested allowable limit for drinking water (Table 3), which may be owing to the elevated salinity and mineral substances in the selected groundwater of 2nd aquifer (AQ2) of coal mining area (Block 2), drain water at initial and different distance of reservoir (S1, S2, S3 and S4, corresponding to average 500 m among each site), these data are consisted with literature reported study ((Brahman et al., 2014; Nag and Saha 2014). It was observed that the EC of initial drained water (DW) have higher values than four sites, decreasing from (6.34 to 37.7% in S1 to S4). The total dissolved salts at S4 were about 47% reduced than initial values of DW. The EC of groundwater (AQ2) samples was found to be 24 to 40 time elevated than WHO guidelines for drinking water (Table 2). It was observed that the change in EC is almost the same as TDS. The resulted values of toxic elements (tAs, Cd and Pb) in groundwater of second aquifer of Thar coal mining area, its drain outlet and four sampling sites of reservoir have high concentration. The obtained results indicated that ground water (AQ2), DW, S1, S2, S3 and S4 were contaminated with larger concentration of As, Cd and Pb, higher than the suggested WHO limit for drinking water (10 µg L − 1 ). The total contents of As in AQ2 was found in ranged as 39.2 − 59.5 µg/L, whereas the drain water (discharge out let) have As concentration slightly lower than AQ2, the difference was not signi cant (p < 0.05). The .5-20.4 µg L − 1 for S1, S2, S3 and S4, respectively. The resulted data indicate two to four folds higher levels of Cd in all water samples than limit for drinking purposes. The levels of Pb in AQ2, DW, S1, S2, S3 and S4 were found in the range of 51. 4-74.6, 43.6-69.4, 54.5-56.4, 36.4-55.5, 28.6-49.5 and 22.4-35.2 µg L − 1 , respectively, these all values were three to six fold higher than those values of water recommended by WHO. The all toxic elements were found in decreasing order based on sequence of sampling sites as AQ2 ≤ DW < S1 < S2 < S3 and < S4 as shown in Table 2.

Sodium absorption ratio (SAR)
On the basis of these results sodium adsorption ratio was calculated for the selected water samples by the equation given below and also indicated in previously published work (Brahman et

Multivariate statistical analysis
Cluster analysis (CA) was also applied for qualitatively analyze the difference and similarity among the water samples of ground water of aquifer 2, its drain outlets and at different sites of reservoir. The statistical technique, cluster analysis of water samples obtained from different origin was performed on all selected physico-chemical variables (physico-chemical parameters and toxic elements), and to evaluated the dissimilarity/similarity on the resulted data of all variable in water samples of six sampling sites ).
The dendrogram indicates the grouping of all six sampling sites, based on resulted data of all variables, into two statistically important clusters (Fig. 2). The results of all physico chemical parameters of water samples of all selected origin after their scaling by z-transformation, to measures the % of similarity and to construction of a dendrogram (Chen et al. 2007). A percent similarity level at 100%, two clusters was formed that is further divided into two groups: from the left, the rst cluster (cluster A) is composed of two groups, one comprises of site-3 and 4.Whereas in second group, and site-2 was found. The second cluster (B) on the right side is further divided in to two groups. One comprised of AQ2 and DW (where no signi cant difference was observed in levels of different toxic elements, cations, anions and physico chemical parameters. Whereas the second group is involved in sight (S1), that is located at 500 meter distance from drain outlet, where the levels of all physico-chemical data including toxic elements were considerably decreased from AQ2 and DW, though the dissimilarity was not signi cant (p > 0.05).
The Principal component analysis (PCA) is intended to change the original data of variables into new, unassociated variables (axes), designated as principal components (PCs), which are linear arrangements of the original data of the variables. The novel a liation lies beside the paths of highest variance. The PCA provide an objective way of nding manifestations of difference in the resulted data be accomplished for description as probable summarizing (Sarbu and Pop 2005). The spatial correlation matrix of the water quality parameters is obtained by PCA. The PCA was carried out on the resulted data set (17 parameters/variables) independently for water samples of different origin (DW, Aq2, S1, S2, S3 and S4), to regulate/ identify the compact set of variables that might be capture the variance of data set.
The primary/ rst principal component (PC1) indicated more than 56.88% of the whole discrepancy in the set of resulted data of water samples obtained from different origin is shown in Table 3.

Correlation among different variables
The Pearson correlation among physicochemical factors of water samples obtained from AQ2 collected from 2nd aquifer of coal eld, DW and four sites have equal distance of reservoir are indicates in The high exposure of F − (> 1.5 mg/L) creates adverse impact on the humans especially dental and skeletal uorosis and in addition high intake of F − for long time might be causes impairment in muscles, kidneys and nerves system (Ayoob and Gupta, 2006). The dilemma of uorosis involving dental and skeletal system is frequent occur in the areas of crystalline basement rocks, predominantly those have

Sediment Analysis
The average concentration of the iron, sodium, potassium, arsenic, cadmium and lead in the sediment samples collected from four sites of reservoir (SD1, SD2, SD3 and SD4) for subsequent two month, are shown in Table 4.The resulted data based on dry weight basis. It was observed that the concentration all elements were inversely proportion to those values obtained from corresponding water samples, might be due to settling of them with distance of reservoir (2.0 km). The concentration of As was found in the range of 6.7-11, 10.

Conclusion
It was observed that the discharge of groundwater of aquifer 2 of coal mining area due to construction of power plant, in a newly prepared reservoir/gorano dam. The variation in physico-chemical characteristics, anion, cation and toxic elements was determined. The resulted data indicated that the all parameters in water samples are sequentially decrease from AQ2 to S4. The multivariate technique, principle component analysis helped in recognized the causes responsible for the variations water quality from AQ2 to S4 are only geologic in nature. The cluster analysis grouped six sampling sites into two clusters on the basis of similar characteristics of water samples of different origin. It was observed that all selected physico-chemical parameters, cations, anions, and toxic elements were higher than the suggested limits by WHO. The high values of sodium absorption ratio of water samples of reservoir indicates that they are not suitable for agricultural purposes. Further studies should be focused on the bio-accumulation of As and F − in aquatic biota and hazards associated with their consumption.

Con ict of interests
The all authors declared not any con ict of interest Availability of data and materials The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.