Quality Assessment of Groundwater Based on Geochemical Modelling and Water Quality Index (WQI)

Abstract

Potable groundwater contamination through arsenic (As) is a common environmental problem in many developing countries, including Pakistan, with significant human health risk reports. The current research was conducted in District Nankana Sahib, which is a major industrial site in Punjab, Pakistan. According to the Punjab Directorate of Industries in Pakistan, there are more than a thousand industries in this area. These industries produce a lot of waste and effluent, which contaminate the environment with harmful and toxic materials. Continuous irrigation with industrial effluent and sewage sludge may make groundwater sources vulnerable. Therefore, this research was aimed as assessing the hydrochemical profile of groundwater concerning As contamination in the study area using sixty-seven groundwater samples. Multivariate statistical analysis, graphical plots, geochemical modeling, and the water quality index (WQI) were applied to investigate the hydrochemistry of the research area. The outcomes of the WQI revealed that 43% of the samples were of poor quality and not fit for human consumption. About 28% of the groundwater samples showed high arsenic contamination beyond the permitted limit of the World Health Organization (WHO). The piper plot identified three distinct types of water in the research area: calcium, chloride, and calcium-chloride types. The Gibbs diagram illustrated that rock–water interaction influenced the hydrochemistry. Terrigenous sedimentary rocks, such as stream deposits, flood plain deposits, and detrital sedimentary rocks, among other types of sedimentary rocks covered most part the study area. Principal component analysis (PCA) and hierarchical cluster analysis (HCA) indicated that the arsenic in groundwater exhibited a significant positive correlation for pH, Fe and As. Health risk assessments indicated a hazard quotient (HQ) greater than 1, indicating a 28% contribution showing that groundwater ingestion is highly toxic to the local habitats. The results of this study further help in managing future sustainable groundwater management approaches in the Nankana District, Punjab, Pakistan.

1. Introduction

Groundwater is an essential source for domestic and industrial purposes. The extraction of this natural resource is highly increased globally due to the rising population and the easy access to new and cheap extraction technologies [1]. A growing trend toward the subsurface disposal of wastes has accelerated the movement of contaminants in groundwater systems in recent years [2]. The groundwater throughout the regions where industry and agriculture have been delegated could be contaminated by a wide range of pollutants. These contaminants include nitrate, sulphate, chloride, toxic organic compounds released by mineral fertilizers and pesticides, pathogens, and sulphate [3,4]. Groundwater contamination is a critical issue, particularly in rapidly developing cities and industrial areas, where untreated industrial wastewater drainage, excessive fertilizers, improper landfills, and agricultural pesticides are comparatively common [5].

Groundwater exposure to arsenic is a severe environmental hazard to human health. Many individuals are exposed to arsenic through the oral ingestion of water [6]. Arsenic concentration in groundwater has been recorded as more than 10 μg/L in different countries throughout the world, such as Spain, Portugal, and Hungary, and particularly in Asian countries, such as Pakistan, China, Iran, India, and Bangladesh [7,8,9]. As per recent research, human health can even be affected by exposure to arsenic under 10 μg/L; therefore, developed countries are modifying their water treatment plants to provide potable water with lower than 1 μg/L of arsenic [10].

Both geogenic and human activities contribute to arsenic in groundwater. Geogenic sources of arsenic are rock types and sediments, where arsenic occurs naturally as a trace element. Arsenic is typically found in sedimentary or igneous rock, either by itself or in combination with other elements, such as oxygen. Igneous and sedimentary rocks have an average arsenic concentration of 2 mg kg⁻¹. The arsenic concentration in most rocks ranges from 0.5 to 2.5 mg kg⁻¹ [11], whereas higher concentrations have been observed in phosphorites and clayey or argillaceous sediments [12,13]. The transportation of arsenic from natural sources into groundwater depends on the type of arsenic, the environment of the aquifer, and the biogeochemical processes. Arsenic is found in more than 200 minerals on the earth’s surface. Arsenolite, arsenopyrite (FeAsS), anargite (Cu₃AsS₄), and orpiment are the most common sources of groundwater arsenic pollution. Furthermore, anthropogenic activities, mainly mining arsenic-bearing sulfide, pesticides, and improper landfills, push arsenic into groundwater [14].

In Pakistan, recent studies on groundwater contamination revealed that about 70% of the ground and surface water reservoirs are highly polluted with organic, inorganic, and biological pollutants [15,16]. Approximately 47 million inhabitants in Pakistan belong to those areas, where they rely on consuming tube-well water for drinking and domestic purposes, which may contain As concentrations exceeding 10 μg/L in 50% of the total wells [16,17], and as recorded by [18,19], 17% of people drink water containing above 50 μg/L of As. Only 26% have access to clean water [20]. Most of the regions of the Sindh and Punjab provinces of Pakistan are comprised of alluvial and deltaic deposits of the Quaternary (Pleistocene) age, with thicknesses extending from a few meters to hundreds of meters [21,22]. Furthermore, the groundwater in such regions is polluted with significant levels of arsenic [23]. Previous research has been performed to find fresh water aquifers in the District using the electrical resistivity method [24]. To identify the possible sources of arsenic in the District’s groundwater, the aim of this research was (i) to investigate the physicochemical features of the groundwater sources; (ii) to investigate the study area’s particular distribution pattern and the amount of arsenic risk to the local population; (iii) to investigate the relationship between arsenic and other groundwater variables, as well as to identify pollutant sources; (iv) to evaluate the suitability of the groundwater for ingestion using the WQI approach.

2. Area and Physiography

The study area (Nankana Sahib District) is in Punjab province, Pakistan, within a latitude of 31.452097 N and a longitude of 73.708305 E (Figure 1). The district has three subdivisions: Shangla Hill, Shah Kot, and Nankana Sahib Tehsil. Geographically, it is situated about 91 km from the west of Lahore and about 75 km from the east of Faisalabad city. The total population of Nankana Sahib is approximately 1,356,774 inhabitants (2017 census report by Pakistan bureau of statistics). Climatically, it lies within the semi-arid climatic zone. In the summer, the temperature reaches 40–45 °C. The temperature in May and June is quite high, while a minimum temperature 22–25 °C is recorded for December and January. The monsoon season begins in June and lasts for around two months. The average rainfall of Nankana Sahib is 635 mm [25]. The soil of Nankana Sahib is fertile. There are three major crops cultivated in the area: wheat, rice and sugarcane [26]. In the District, most inhabitants use groundwater for domestic use; the tube-well density in the entire District is (19,161 in total). In most areas, the depth of the water table is about 7.5 to 9.14 m, which has no difficulty pumping out [27].

Geology and Hydrogeology

The soil of Punjab is composed of alluvial materials transported from the Himalayan Mountains by tributaries of the Indus River. Across many aspects, the soils are quite similar throughout the region. They are relatively coarse and medium-textured reddish–brown to greyish–brown soils with high clayey sand and silty sand, which are Precambrian metamorphic and igneous rocks or semi-consolidated tertiary rock covered by quaternary alluvium deposits in the Punjab plains [28].

The district is located between the Ravi River and Chenab’s watersheds, also known as Rachna Dhoab. The area is covered with thick alluvial and river deposits. The current study area contains terrigenous sedimentary rocks, such as stream deposits, flood plain deposits, and detrital sedimentary rocks, among other types of sedimentary rocks. Sand and gravels are water-bearing strata. Most aquifers in this area are confined, although certain regions have unconfined aquifers with gravel and pebbles on the surface [24].

Furthermore, the significant sources of aquifer recharge include canals, water channels, and percolation through irrigation and rainfall. The area’s water table fluctuates depending on seasonal recharge. Exploratory drilling in the region of the subsurface bedrock in Chaj, Rechna, and Bari Doabs has discovered that the Precambrian underlying rocks are covered by Quaternary alluvium [29]. Sediments range from sandy clay to clayey sand, predominantly grey to brownish grey, in various proportions. Water-bearing strata are sand, gravel, and coarse aggregates. Ravi and Chenab are the two major rivers that drain the area [30,31]. During the summer, the average monthly discharge is around 15–20 times that in the winter. Upstream groundwater seepage is a common cause of reduced flow from December to the middle of March. During the spring melt of the Himalayan snow, river levels often rise sharply, reaching their highest point in July or August, when the monsoon rains arrive [32].

3. Methodology

3.1. Sampling and Analysis

A total of 67 groundwater samples were taken from different sites in July 2018 in the entire District. Samples were taken for physicochemical analysis from existing tube wells, hand pumps, motor pumps, and the water supply scheme in the District. Pre-cleaned polystyrene bottles were used to store the samples. Samples were taken after the bore holes were purged through continues pumping to remove the stagnant water’s impact. On-site measurements were performed for the variables, such as pH, electrical conductivity (EC), temperature, and total dissolved solids (TDS) measurements. For further analysis, the samples were sent to the Pakistan Council of Research in Water Resources (PCRWR) laboratory. The evaluation of water samples for aesthetic, physicochemical, and bacteriological characteristics was carried out according to the American Public Health Association (APHA). The pH and TDS were measured by a pH meter (Hanna Instrument, Model 8519, Italy) and the 2540 C standard method (2012), respectively. For alkalinity (carbonate and bicarbonate), the 2320 standard method (2012) was employed. To measure arsenic levels, the AAS Vario 6, Analytik Jena AG Germany (3111B APHA) 2012 technique was utilized. Nitrate (NO₃⁻ as N) was determined by using Cd-reduction (Hach-8171) using the spectrophotometer technique [33]. The amounts of calcium, chloride, magnesium, and sulfate were determined using the standard procedures 3500-Ca-D (2012), titration (Silver Nitrate) (2012), 2340-C (2012), and SulfaVer4 (Hach-8051) (2012) by a spectrophotometer. Flame photometer PFP7, UK method was utilized to quantify potassium and sodium. Furthermore, fluoride (F), iron (Fe), and turbidity were measured using 4500-FC Ion-Selective Electrode Method Standard (2012), Ferro Ver Method (HACH Cat. 21057-69), and turbidity meter, LaMotte (Model 2008, USA) method, respectively [34]. The charge balance error (CBE) was estimated to ensure that water samples were accurately analyzed, within ±10% acceptable limits [35]. The following equations were used to calculate charge balance error, where cations, as well as anions, were represented in meq/L [36].

$$\mathrm{CBE}\%=\frac{{\displaystyle \sum }\mathrm{cations}-{\displaystyle \sum }\mathrm{anions}}{{\displaystyle \sum }\mathrm{cations}+{\displaystyle \sum }\mathrm{anions}}\times 100$$

(1)

3.2. Multivariate Statistical Analysis

3.2.1. Correlation Analysis

Correlation analysis is generally utilized to identify the correlation between two variables. We employed Pearson’s correlation to better comprehend the correlation between Hydrochemical parameters, as well as the effect of associated parameters.

3.2.2. Hierarchical Cluster Analysis (HCA)

Hierarchical Cluster analysis is a widely used statistical tool for organizing data based on their characteristics. It divides objects into clusters, so components with the same group show the most similar characteristics [37]. Ward’s method of joining rules and the Euclidean distance were applied to estimate the correlation among water quality measures. The cluster analysis results were then employed to deduce the hydrochemical procedures which occurred in the research area [38].

3.2.3. Principle Component Analysis (PCA)

PCA is widely used to reduce huge datasets with minimum loss of information [39]. PCA was calculated with SPSS software (version 25) to clarify the cluster analysis results and the factors that described the variances in most groundwater quality data [40].

3.3. Groundwater Type and Classification

Common ion concentrations in groundwater samples are typically represented using the piper plot. The piper plot was constructed to analyze the various factors that influenced the hydrochemistry of the research area. In the Gibbs diagram, cations Na⁺/(Na⁺/Ca²⁺) and anions Cl⁻/(Cl⁻ + HCO₃⁻) were placed against TDS. Microsoft Excel (version 2020) was used to construct the Gibbs plot to identify the controlling mechanisms of groundwater of the research area.

3.4. Geochemical Modeling

Chloro-alkaline indices (CAI-I and CAI-II) and saturation indices were computed to identify the ion exchange processes and the status of saturation of different minerals in the groundwater samples, respectively. PHREEQC interactive software (version 3.1) was employed to calculate the mineral saturation phase [41,42], while the following formulas were employed to compute the chloro-alkaline indices.

CAI-1 = Cl⁻ (Na⁺ + K⁺)/Cl⁻

(2)

CAI-2 = Cl⁻(Na⁺ + K⁺)/HCO₃⁻ + SO₄²⁻ + CO₃⁻ + NO₃

(3)

3.5. Water Quality Index (WQI)

The water quality index (WQI) was estimated by adopting the weighted arithmetic index technique [43]. The following steps were followed to calculate the WQI.

Step 1: Each sample’s Wn factor was determined using the following formula:

$$\mathrm{Wn}=\frac{\mathrm{K}}{\mathrm{Sn}}$$

where Sn = standard desirable value of n^th parameters.

Wn = 1 (unity).

K = constant of proportionality.

The value of K = 1; it is a constant value and used to calculate the standard desirable value of each chemical variable.

Step 2: The sub-index (Qn) value was calculated by using the expression

$$\mathrm{Qn}=\frac{\left[\left(\mathrm{Vn}-\mathrm{Vo}\right)\right]}{\left[\left(\mathrm{Sn}-\mathrm{Vo}\right)\right]}\times 100$$

(4)

where

Vn = average concentration of the n^th parameter.

Sn = standard desirable value of the n^th parameter.

Vo = actual values of the parameters in pure water (generally V₀ = 0 for most parameters, except for pH).

$$\mathrm{QpH}=\frac{\left[\left(\mathrm{VpH}-7\right)\right]}{\left[\left(8.5-7\right)\right]}\times 100$$

(5)

Step 3: Step 1 and Step 2 were combined to determine WQI

$$\mathrm{Overall}\mathrm{WQI}=\frac{{{\displaystyle \sum }}^{}\mathrm{WQn}}{{{\displaystyle \sum }}^{}\mathrm{Wn}}$$

(6)

3.6. Heath Risk Assessment

Arsenic can enter the human body in different ways, such as inhalation, dermal exposure, and oral digestion. Among these pathways, direct oral intake is the potential source of human susceptibility to arsenic and heavy metals (more than 90%) [44]. As a result, the acute and cancer risks caused by drinking polluted water containing arsenic were assessed. Exposure risk assessment was utilized to obtain the chronic risk. ADD refers to the average daily arsenic dosage, and it is calculated by applying the formula below:

ADD = C × IR/BW

(7)

C refers to arsenic concentration (μg/L) in groundwater, IR stands for ingestion ratio (2 L/day), and BW represents body weight (70 kg).

A Health Quotient (HQ) calculation can determine non-carcinogenic health risks triggered by consuming high arsenic water levels [45,46]. The HQ of the study area was calculated utilizing the subsequent equation:

HQ = ADD/RFD

(8)

The oral recommended dosage of 0.3 (μg kg/day) is RFD. It is regarded safe when the risk factor of the sample is less than 1. Samples having hazard quotient values below 1 are considered safe, while those with HQ values greater than 1 may reflect health danger. Furthermore, cancer risk analysis was also measured by employing the below formula

CR = ADD × CSF

(9)

where CSF stands for the cancer slope factor (1.5 mg/kg/day) [46].

4. Result and Discussion

4.1. Groundwater Hydrochemical Properties

The physicochemical parameters were statistically analyzed and compared with the standard limits of WHO [47].

Table 1. Descriptive statistics of groundwater samples variables.

Chemical Parameters	W.H.O Standards	Minimum	Maximum	Mean	St. Deviation	Samples E.P.L	% Of Samples E.P. L
pH	6.5–8.5	7.1	8.3	7.8	0.263	0	0%
EC (μS/cm)	<1000	160	3710	1572.03	974.329	40	59%
TDS (mg/L)	<1000	1.67	2226	954.70	600.374	30	44%
Total hardness (mg/L)	100–500	70	710	294.78	134.199	8	11%
HCO3− (mg/L)	150–300	50	945	359.55	179.818	43	64%
Ca2+ (mg/L)	75	20	S120	55.19403	20.594	11	16%
F− (mg/L)	0.6–1.5	0.04	2.5	0.54	0.444	4	5%
Mg2+ (mg/L)	30–150	5	124	37.61	24.074	0	0%
Na+ (mg/L)	50–200	6	700	242.49	191.389	36	53%
K+ (mg/L)	30–200	2.3	95	10.96	16.303	0	0
Cl− (mg/L)	250	2	568	195.10	177.986	24	35%
NO3(N)	45	0.01	18	1.59	2.888	0	0%
PO4−3 (mg/L)	0.1	0.01	0.35	0.13	0.095	34	50%
SO42− (mg/L)	250	11	472	188.45	129.385	21	31%
Fe (mg/L)	1.5	0.03	0.95	0.22	0.198	0	0%
As (ppb)	10	0.13	79.65	16.5	22.276	19	28%
Turbidity (NTU)	<5	0.01	15.61	1.15	2.681	2	2%

Note: E.P.L: Exceeding Permissible Limit.

includes each variable’s highest and lowest value, mean, and standard deviation. The pH in the research region varied from 7.1 to 8.3, with a median of 7.8. This suggests that the groundwater in the research area is well buffered. The electrical conductivity varied from 160 to 3710 μS/cm, showing a mean of 1572.03 μS/cm, and 59% of the water samples crossed the WHO standards of 1000 μS/cm. Similarly, values of total dissolved solids (TDS) ranged from 1.67 mg/L to 2226 mg/L, having a mean of 954.70 mg/L, where 44% of the water exceeds the standard limit of 1000 mg/L [40,47]. High concentrations of TDS make water saltier, so consuming such water may cause serious health issues, such as gastrointestinal exasperation [48]. The concentrations of Na⁺, Mg²⁺, and K⁺ were in the range of 6–700 mg/L, 5–124 mg/L, and 2.3–95 mg/L, with a median of 242.49, 37.61, and 10.96 mg/L, respectively. All samples of Mg²⁺ and K⁺ were within the permissible limit, while 53% of groundwater samples showed high Na⁺ concentrations. Dissolution of sodium-containing minerals and anthropogenic activities can contribute to sodium enrichment in groundwater [49]. The concentrations of chloride, sulfate, and phosphate in groundwater were in the range of 2–558 mg/L, 11–472 mg/L, and 0.01–0.35 mg/L, respectively, where 35% of the water samples of chloride, 31% of sulfate, and 50% of phosphate crossed the WHO limit. Phosphate in groundwater could be caused by regional sources, such as sewage interacting with shallow groundwater, agricultural runoff discharge and landfill leachates [50]. Sulfate in groundwater might be caused by sulphide mineral oxidation, weathering of gypsum-bearing rocks (CaSO₄²H₂O), and human inputs, such as industrial waste water and fertilizers [51]. Chloride in groundwater could be derived from a number of sources, including home and municipal pollutants, weathering, and leaching from sedimentary rocks [52]. The content of HCO₃⁻ ranged from 50 to 945 mg/L, with a mean of 359.55 mg/L, where 43 groundwater samples enriched with HCO₃^- were above the standard limit of WHO. The oxidation of organic substances present in the aquifer yields carbon dioxide, which is most likely the source of bicarbonate in subsurface water [53]. Nitrate and iron concentrations extended from 0.01 to 18 mg/L and 0.03 to 0.95 mg/L, respectively. All NO₃⁻ and Fe⁻ samples were under the WHO prescribed range. The primary sources of nitrate contaminants in the groundwater are agricultural runoff, wastewater discharge from sewage water, and the degradation of nitrogenous waste from humans and animals [47]. Table 1 shows descriptive statistics of groundwater variables.

4.2. Arsenic Occurrence and Distribution

The concentration of arsenic in water samples varied widely from 0.13 to 79.65 ppb, with a median value of 16.5. About 28% of the water samples showed arsenic levels higher than the specified limitation of WHO 10 μg/L, demonstrating that arsenic in the research area is an alarming situation for human health and natural habitats. Kouzul noted that continuous water consumption with arsenic, even 3 μg/L, can cause serious health issues, such as cancer in the human body [54].

The spatial distribution map was constructed through the interpolation method in Arc GIS to present the spatial distribution of arsenic contamination in the research area (Figure 2). For representing the spatial distribution of elements in groundwater, the inverse distance weighting (IDW) technique is regarded to be a better interpolation technique [55]. The IDW technique is simple and requires minimal inputs for interpolation. IDW interpolation was performed on the dataset utilizing its powers 1–4, as well as the ideal power. It employs a weighted reciprocal of distance from the position where the value is to be acquired to previously known points, by arranging values in a linear way [56]. It is predicated on the premise that a sampled point that is closer to an unsampled point has a greater similarity to that unsampled point than the value of a sampled point [57]. In the current study, high arsenic was recorded in the certain areas of Nankana District, such as Syed Wala, Mandi Faizabad, More Khunda, Rehan Wala, and Khiary Kalan. These areas are located near River Ravi. In Pakistan, the most important aquifer exists in the Indus Plain. Quaternary sedimentary deposits dominate the Indus Plain, mostly of glacial and alluvial origin. The sedimentary aquifer of the Indus Plain of this part has higher concentrations of arsenic [58].

Various scatter plots were plotted to explain the correlation between arsenic and other chemical elements in groundwater samples. The scatter plot (Figure 3a–c) depicted high arsenic concentrations in the pH values between 7.4 and to 7.94. Arsenite (V) and arsenate (III) in groundwater are primarily is the form of oxyanions, so their adhesion to such surfaces decreases when increased pH results in arsenic mobilization [59]. Arsenic can be mobilized in groundwater by the influence of HCO₃⁻ ions through carbonization of As-bearing sulfide minerals [60,61]; however, in our study, HCO₃⁻ showed a weak correlation with arsenic, suggesting no influence of bicarbonate in the mobilization of arsenic in groundwater.

Similarly, NO₃⁻ and SO₄⁻ depicted a weak correlation with arsenic. The weak correlation of arsenic with NO₃⁻ and SO₄⁻ often indicates a strongly reducing environment, which probably increases the arsenic in the groundwater [62]. Moreover, As and Fe did not show a positive correlation in Figure 4b. According to previous research [63,64,65,66,67], the minimum correlation between As and Fe might be due to the reduction of Fe-oxyhydroxides combined with the precipitation of Fe²⁺ as FeCO₃ when an anaerobic environment prevails during bacterial oxidation of sedimentary organic matter. Furthermore, scatter plots of phosphate and sulfate vs. arsenic depicted a weak correlation of phosphate and sulfate with arsenic, indicating no significant role of these ions in the movement of arsenic.

4.2.1. Groundwater Type

The piper diagram was used to understand the hydrochemical evolution and the impact of significant ion processes in the groundwater of the study area [68]. In the piper diagram, water samples are generally divided into six major categories: (i) Ca-HCO₃⁻ type, (ii) Na-Cl type, (iii) Ca–Mg–Cl type, (iv) Ca-Na-HCO₃⁻ type, (v) Ca–Cl type, and (vi) Na-HCO₃⁻ type [69]. The piper diagram (Figure 5) reflects three prominent water types in the study: calcium type, chloride type, and calcium chloride type, whereas water of sulfate type, mixed CaMgCl type and no dominant type was also found in the area. The diagram indicates that calcium and chloride are the principal cation and anion, respectively. The most probable sources of calcium in groundwater are the dissolution of carbonate rocks by carbonic acid. Human activities might also impart the enhancement of calcium in the groundwater by using lime as an agricultural fertilizer [70,71].

4.2.2. Groundwater Chemistry Controlling Mechanisms

Precipitation, evaporation, and rock–water interaction are prominent hydrogeochemical processes in semi-arid areas, and these processes mainly govern the water chemistry of these regions [72]. The Gibbs plot explains the mechanisms of groundwater of the study area. The two semi-logarithm diagrams showed the precipitation dominance. Chloride contamination in groundwater could be caused by either natural or anthropogenic sources, such as weathering of chloride-bearing rocks, evaporation, intrusion of seawater, leakage of industrial or domestic waste rock dominance, and evaporation dominance [73]. In the Gibbs plot, the ratios of significant cations [(Na + K)]/(Na + KCa)] and anions [(Cl/Cl + HCO₃)] were plotted individually against TDS (Figure 6). The plot demonstrates that many water samples fall in the rock-dominance zone. These findings imply that the process of rock and mineral weathering is the most significant factor influencing groundwater quality in the research area. Some samples in the cation section fall beyond the plot preview, indicating anthropogenic contamination [74]. Furthermore, several bivariate plots are also constructed to better comprehend the weathering mechanisms of the rock-water interaction processes [75,76].

(1)

Silicate Weathering

Silicate weathering considers the primary geochemical process influencing the hydrochemistry of significant ions, particularly in aquifers of hard rocks [77]. The concentrations because of silicate weathering could be evaluated by the bivariate plot of Ca⁺ + Mg⁺/TZ⁺ (7d). The scatter plot (Figure 7d) shows that most samples fall above and towards the Ca²⁺ + Mg²⁺ = 0.5 line, indicating the predominant role of silicate weathering in geochemical processes. In contrast, some samples lie below the Ca²⁺ + Mg²⁺ = 0.5 line, which indicates that the abundance of Ca²⁺ + Mg²⁺ is probably due to the contribution of reverse ions exchange and weathering of carbonate minerals [78,79].

(2)

Carbonate Weathering and Dissolution

The scatter plots of HCO₃⁻ + SO₄²⁻/Ca²⁺ + Mg²⁺ and Ca²⁺/HCO₃⁻ were used to analyze the carbonate weathering in the research area. Figure 7c depicts the prevalence of bicarbonate over calcium. According to Das et al., if the ratio between bicarbonate and calcium is close to 1:2, the line suggests silicate weathering, while a ratio close to 1:4 indicates carbonate weathering [78]. As plot 7c shows, most samples fall beside the 1:4 line, demonstrating carbonate weathering in our study area. In the HCO₃⁻ + SO₄²⁻ vs. Ca²⁺ + Mg²⁺ plot (Figure 7c), many samples lie above the 1:1 line, implying the predominance of carbonate weathering, and samples that fall below the line suggest silicate weathering [80,81]. Furthermore, the Ca²⁺ vs. SO₄²⁻ scatter plot (Figure 7a) shows the dominance of sulfate over calcium. Samples falling above the 1:1 line imply the dissolution of gypsum or anhydrite, whereas the fall on the 1:2 lines suggests carbonate weathering. [80].

(3)

Evaporation

The evaporation process could be predicted by illustrating the Na/Cl vs. EC plot. No mineral species are expected to precipitate if the dominant controlling process is evaporation; hence, the Na/Cl ratio remains constant. If the ratio of sodium and chloride increases with electrical conductivity (EC), it reveals a predominance of silicate weathering. [50]. The Na/Cl vs. EC plot of our study area illustrates that the Na/Cl ratio is inversely proportional to EC, while some samples are positively correlated with EC. If Na/Cl increases with EC, the predominant mechanism is silicate weathering, whereas decreasing this ratio would promote salt dissolution [50,78].

5. Multivariate Statistical Analysis

5.1. Pearson’s Correlation Analysis

Table 2. Chemical composition of groundwater in the research area as determined using Pearson’s correlation matrix.

Parameters	EC	pH	Turbidity	TDS	Alkalinity	HCO3−	Cl−	SO42−	Ca2+	Mg2+	Hardness	Na+	K+	NO3−	PO4−	F−	Fe	As
EC	1
pH	−0.18	1
Turbidity	0.21	0.12	1
TDS	0.95	−0.18	0.20	1
Alkalinity	0.76	−0.14	0.17	0.79	1
HCO3−	0.76	−0.14	0.17	0.79	1.00	1
Cl−	0.86	−0.21	0.12	0.92	0.52	0.52	1
SO42−	0.88	−0.08	0.28	0.94	0.69	0.69	0.85	1
Ca2+	0.34	−0.56	−0.01	0.36	0.19	0.19	0.40	0.29	1
Mg2+	0.53	−0.58	0.01	0.61	0.47	0.47	0.61	0.48	0.48	1
Hardness	0.54	−0.65	0.00	0.60	0.42	0.42	0.62	0.49	0.75	0.94	1
Na+	0.93	−0.01	0.21	0.97	0.78	0.78	0.88	0.94	0.19	0.43	0.40	1
K+	0.45	−0.27	0.10	0.53	0.48	0.48	0.50	0.38	0.22	0.55	0.49	0.44	1
NO3−	0.34	−0.41	−0.09	0.36	0.33	0.33	0.34	0.19	0.25	0.54	0.49	0.23	0.56	1
PO4−	0.86	−0.09	0.25	0.81	0.64	0.64	0.73	0.82	0.17	0.38	0.36	0.83	0.41	0.12	1
F−	0.46	0.26	−0.02	0.48	0.61	0.61	0.29	0.49	−0.13	0.06	−0.01	0.56	0.14	0.13	0.52	1
Fe	0.28	−0.06	0.10	0.25	0.15	0.15	0.29	0.20	0.22	0.00	0.09	0.28	−0.11	−0.01	0.24	0.08	1
As	−0.38	−0.17	0.01	−0.41	−0.26	−0.27	−0.43	−0.35	0.10	−0.19	−0.10	−0.44	−0.25	−0.19	−0.31	−0.27	−0.08	1

illustrates the calculated Pearson’s correlation matrix. The estimated matrix revealed that EC and TDS have a positive correlation with HCO₃⁻, Cl⁻, SO₄⁻, Na⁺ and PO₄^-, showing that these dissolve ions strongly influence TDS and EC values [82]. This implies that the sources of these ions in groundwater are probably fluid-rock interaction and/or anthropogenic practices, such as agricultural activities and sewage from the household [83]. Furthermore, a positive correlation between Na⁺ and Cl⁻ (r = 0.88) reflects that the source of both sodium and chloride in groundwater might be the weathering of halite. Similarly, Ca²⁺ and Mg²⁺ are positively correlated with total hardness = 0.75 and r = 0.95, respectively. The positive correlation is because hardness is the estimated measure of calcium and magnesium [84,85,86].

The present study explains the insignificant correlation between various chemical parameters and arsenic in groundwater. It suggests that Pearson’s correlation matrix could not relate As level to different geochemical variables. As a result, the physicochemical parameters and As were processed for PCA to observe any correlations among the possible As sources in the aquifers.

5.2. Principal Component Analysis

This analysis has become an extensively employed multivariate analytical approach that reduces a range of original variables, such as geochemical data, to a minimal figure of indices (i.e., principal components or factors) for interpreting variations between observed data that are not observable from the simple correlation analysis [23].

A PCA was used to identify differences among the physicochemical variables and evaluate potential hydrogeochemical factors that cause arsenic enrichment in groundwater. The results of five principal components, PC1-PC5, are presented in

Table 3. PCA for arsenic along with other chemical variables present in the groundwater of the study area.

Variable	PC1	PC2	PC3	PC4	PC5
EC (μS/cm)	0.33	0.10	0.09	0.03	−0.04
pH	−0.11	0.47	−0.09	−0.12	−0.12
TDS (mg/L)	0.35	0.08	0.04	0.00	−0.06
HCO3 (mg/L)	0.28	0.11	−0.11	0.33	0.25
Cl (mg/L)	0.32	0.02	0.10	−0.23	−0.19
SO4 (mg/L)	0.32	0.15	0.11	0.08	−0.22
Ca (mg/L)	0.15	−0.37	0.35	0.06	−0.04
Mg (mg/L)	0.25	−0.33	−0.13	−0.03	−0.10
Hardness (mg/L)	0.25	−0.39	0.05	0.00	−0.11
Na (mg/L)	0.33	0.21	0.06	0.00	−0.06
K (mg/L)	0.21	−0.14	−0.43	−0.03	0.03
NO3 (N)(mg/L)	0.16	−0.26	−0.41	−0.14	0.53
PO4 (mg/L)	0.29	0.19	0.10	0.13	−0.11
F (mg/L)	0.17	0.34	−0.15	0.36	0.37
Fe (mg/L)	0.09	0.07	0.59	−0.35	0.61
As (PPb)	−0.15	−0.19	0.26	0.72	0.07
Eigenvalue	7.89	2.75	1.36	0.90	0.78
% variance	49.32%	17.24%	8.55%	5.65%	4.90%
Cumulative	49.32%	66.56%	75.11%	80.75%	85.65%

and (Figure 8a–d). The screen plot depicts that three component factors have an eigenvalue greater than 1 (Figure 8). The principal component of PC1, contributing approximately 49.32% of the total variance, showed positive results with EC, HCO₃, Cl, SO₄, Mg, hardness, Na, and PO₄ (Table 3). The result of PCA1 predicted that the main cations and anions were obtained by rock–water interaction and mineral weathering, such as silicate, carbonate, or evaporation [87,88]. Similarly, in PC2, positive loading was attributed to Ca, Mg, hardness, NO₃, and Fe, whereas PCA₃ showed 8.55% total variance with positive loading on Ca, K, NO₃, Fe, and As. PCA₄ explained the 5.65% total variance with positive loading on As, HCO₃, and Fe, while PCA₅ accounted for 4.90% of the total variance, showing a positive correlation with HCO₃, NO₃, Fe, and F. The positive correlation of NO₃ specifies that the source of nitrate concentration is possibly anthropogenic.

The score plot, loading plot, and biplot illustrated that pH and As were clustered in one group, whereas all other parameters were assembled on the right side. This suggests that the variables clustered with arsenic play a prominent role in carrying arsenic in groundwater [23]. According to this phenomenon, pH is probably the potential factor controlling arsenic mobilization in the groundwater of our research area.

5.3. Hierarchical Cluster Analysis (HCA)

In hydrogeology, HCA is a broadly utilized cluster analytical technique. It can determine the initial correlation between water samples. It usually employs a dendrogram to depict the relationship graphically [89,90]. The Euclidean distance was used to identify different groundwater samples utilizing hierarchical cluster analysis. The result of cluster analysis was then plotted in the dendrogram (Figure 9a,b). The dendrogram classified the geochemical parameters into two major clusters, i.e., Cluster 1 (sample 1–62) and Cluster 2 (sample 20–67) (Figure 9a). Cluster 2 contained water samples with high arsenic contamination, whereas Cluster 1 included water samples having less arsenic pollution. Clusters of similar properties and correlations were connected at low linkage distances, whereas dissimilar samples showed high linkage distances [91]. Figure 10b classified chemical parameters into three groups. The first group represented EC, TDS, Na, SO₄, HCO₃, and F. This group suggests fluid–rock interaction or mineral dissolution [39,76] and these variables also showed the same result in PCA1 (Table 3).

Furthermore, the second group showed close similarities between Ca, Mg, hardness, K, and NO₃, while the third group illustrated the close correlation of As, Fe, and pH. In the As vs. Fe scatter plot (7b), some samples showed a positive correlation between As and Fe. Hence, the results of cluster analysis in the dendrogram (9b) also validate the observation of the predominant reducing environment in the aquifer, which accelerates arsenic extraction from Fe oxides into the groundwater.

6. Geochemical Modeling

6.1. Mineral Saturation Index

Saturation indices (SI) are useful for determining mineral reactivity in groundwater. The SI was computed by utilizing the equation below:

$$\mathrm{SI}=\log \left(\frac{\mathrm{IAP}}{\mathrm{Ks}\left(\mathrm{T}\right)}\right)$$

(10)

In Equation (10), IAP represents the ionic activity product and K signifies the equilibrium constant. Positive SI values specify mineral oversaturation, which leads to precipitation, while negative SI values specify mineral unsaturation, which results in mineral dissolution. Saturation indices were computed and are presented in (Figure 10a,1). The scatter plot (Figure 10b,d) of calcite vs. Ca and dolomite vs. Mg illustrate positive SI values for calcite and dolomite, suggesting that the minerals are oversaturated in the preponderance of the samples. The study area has a semi-arid climatic condition. According to [92], due to low rainfall, the precipitation of calcite and dolomite occurs in the groundwater of semi-arid regions. The saturation indices of halite showed a negative value where all the samples fell below the 0 lines, implying the dissolution of halite mineral in the groundwater samples (Figure 10c). The scatter plot of gypsum vs. SO₄ + Cl shows that only a few samples fell below the 0 lines, suggesting that gypsum dissolution is not a predominant process affecting the groundwater chemistry of the research area.

6.2. Ion Exchange

The chloro-alkaline index (CAI) measures the amount of ion exchange amongst aquifer minerals and groundwater. An ion-exchange mechanism in groundwater can be evaluated using this method. Whenever the direct ion exchange process occurs, chloro-alkaline exhibits a negative value. During the direct ion exchange process, K⁺ and Na⁺ are released from the aquifer and exchanged with trapped Mg and Ca in groundwater. Low salinity and freshwater often show a negative chloro-alkaline value [93]. Positive chloro-alkaline indices explain the reverse ion exchange process, where Ca²⁺ and Mg²⁺ are released from the aquifer and exchanged with K⁺ and Na⁺ trapped in groundwater. Reverse ion exchange could result from the intrusion of seawater and sewage water into the aquifer [94].

In the current study, the CAI-I and CAI-II values extended from −6.83 to 14.89, exhibiting an average of 2.90, and −0.09 to 14.71 showing an average of 4.74, respectively. The chloro-alkaline index results revealed that direct and reverse ion exchange processes were involved in controlling groundwater chemistry in this area.

7. Health Risk Assessment

The health risk assessment ingestion was performed for the District Nankana Sahib.

Table 4. Average daily intake (mg/kg/day), hazard quotient, and cancer risk assessment results calculated for the groundwater of the research area.

Sample	As (ppb)	ADD	HQ	CR
1	4.9	0.14	0.47	2.1 × 10−1
2	3.1	0.09	0.30	1.32 × 10−1
3	0.13	0.00	0.01	5.57 × 10−3
4	6.21	0.18	0.59	2.66 × 10−1
5	4.91	0.14	0.47	2.10 × 10−1
6	4.91	0.14	0.47	2.10 × 10−1
7	4.5	0.13	0.43	1.92 × 10−1
8	4.76	0.14	0.45	2.0 × 10−1
9	4.49	0.13	0.43	1.92 × 10−1
10	4.45	0.13	0.42	1.90 × 10−1
11	12.1	0.35	1.15	5.2 × 10−1
12	7.51	0.21	0.72	3.21 × 10−1
13	7.38	0.21	0.70	3.16 × 10−1
14	9.48	0.27	0.90	4.06 × 10−1
15	3.01	0.09	0.29	1.29 × 10−1
16	4.06	0.12	0.39	1.74 × 10−1
17	7.86	0.22	0.75	3.33 × 10−1
18	6.22	0.18	0.59	2.66 × 10−1
19	2.57	0.07	0.24	1.11 × 10−1
20	2.67	0.08	0.25	1.11 × 10−1
21	2.81	0.08	0.27	1.20 × 10−1
22	2.87	0.08	0.27	1.23 × 10−1
23	3.3	0.09	0.31	1.41 × 10−1
24	1.65	0.05	0.16	7.07 × 10−2
25	2.12	0.06	0.20	9.08 × 10−2
26	2.35	0.07	0.22	1.07 × 10−1
27	1.94	0.06	0.18	8.31 × 10−2
28	1.98	0.06	0.19	8.48 × 10−2
29	1.78	0.05	0.17	7.62 × 10−2
30	2.36	0.07	0.22	1.01 × 10−1
31	3.23	0.09	0.31	1.38 × 10−1
32	9.94	0.28	0.95	4.26 × 10−1
33	1.03	0.03	0.10	4.41 × 10−2
34	9.96	0.28	0.95	4.26 × 10−1
35	47.94	1.37	4.57	2.05 × 100
36	33.68	0.96	3.21	1.44 × 100
37	39.67	1.13	3.78	1.7 × 100
38	1.03	0.03	0.10	4.0 × 10−2
39	0.36	0.01	0.03	1.54 × 10−2
40	3.21	0.09	0.31	1.37 × 10−2
41	74.23	2.12	7.07	3.18 × 100
42	52.59	1.50	5.01	2.25 × 100
43	56.31	1.61	5.36	2.41 × 100
44	57.41	1.64	5.47	2.46 × 100
45	55.87	1.60	5.32	2.39 × 100
46	66.34	1.90	6.32	2.84 × 100
47	79.65	2.28	7.59	3.41 × 100
48	61.73	1.76	5.88	2.56 × 100
49	51.66	1.48	4.92	2.21 × 100
50	45.62	1.30	4.34	1.96 × 100
51	44.59	1.27	4.25	1.96 × 100
52	48.67	1.39	4.64	2.09 × 100
53	2.84	0.08	0.27	1.21 × 10−1
54	3.59	0.10	0.34	1.53 × 10−1
55	14.2	0.41	1.35	6.08 × 10−1
56	44.12	1.26	4.20	1.89 × 100
57	33.28	0.95	3.17	1.43 × 100
58	1.43	0.04	0.14	6.12 × 10−2
59	1.35	0.04	0.13	5.78 × 10−2
60	1.69	0.05	0.16	7.24 × 10−2
61	0.97	0.03	0.09	4.15 × 10−2
62	5.04	0.14	0.48	2.16 × 10−1
63	8.93	0.26	0.85	3.82 × 10−1
64	8.59	0.25	0.82	3.68 × 10−1
65	0.86	0.02	0.08	3.68 × 10−2
66	4.39	0.13	0.42	1.88 × 10−1
67	0.9	0.03	0.09	3.85 × 10−2

presents the assessment result concerning the average daily dosage, hazard quotient, and cancer risk assessment for individuals consuming arsenic-rich contaminated groundwater. The computed ADD levels in the study region varied from 0 to 2.28 (μg/kg/day), with a median of 0.47g/μg/day, whereas the values of HQ varied from 0.01 to 7.59, with a mean value of 1.57 μg/kg/day (Figure 11); 28% of the groundwater samples showed a value of HQ greater than 1, exceeding the toxic limit. According to HQ’s findings, 28% of the samples are highly hazardous to inhabitants on continued consumption of polluted water. The carcinogenic risk (CR) due to consuming arsenic polluted water varied from 5.57 × 10⁻³ to 3.41 × 10^⁰, with an average value of 7.07 × 10⁻¹, as shown in Table 4. The prescribed limit of CR by USEPA (US Environmental Protection Agency) is 1 × 10⁻⁶ (1 chance in 1,000,000 lifetime exposure) to 1 × 10⁻⁴ (1 chance in 10,000-lifetime exposure), and unacceptable if the CR value exceeds 1 × 10⁻⁴ [95]. These findings demonstrated that drinking arsenic-contaminated water poses a significant health risk to most inhabitants in the research area. As a result, areas exposed to arsenic should take considerable measures to protect inhabitants from arsenic exposure.

8. Water Quality Index

The WQI is commonly applied to assess drinking water quality. Using the Brown weighted arithmetic index, the water quality indices were determined. In contrast, the WQI spatial distribution map was constructed by employing the interpolation method in Arc GIS to examine the distribution pattern of water quality in the present research. The WQI value is used to classify water quality into different categories. Water quality is considered excellent if the WQI value is 0–25, good if the WQI value is 26–50, poor if the WQI value is 51–75, very poor if the WQI value is 76–100, and unfit for drinking if the WQI value is >100 [43,96].

Table 5. Classification of water quality based on the WQI.

Category	WQI	Water Quality	Water Sample
A	0–25	Excellent	19%
B	25–50	Good	32%
C	51–75	Poor	22%
D	76–100	Very poor	10%
E	>100	Not suitable	11%

illustrates the WQI categorization of the research area, including a list of the various types of groundwater based on WQI sources. In contrast, the WQI spatial distribution map is also presented in (Figure 12). The water quality indices indicate that 19% of groundwater samples were considered excellent for human consumption. In comparison, 32% were good, 22% and 10% were poor and very poor, whereas 11% of the water samples were unsafe for consumption.

9. Conclusions

This study involved the evaluation of the groundwater quality of the District Nankana Sahib according to the concentration of arsenic and its spatial distribution. The pH values stated that the groundwater is well buffered. The water quality index demonstrated that 32% of groundwater samples belong to poor quality, whereas 11% of water samples are unsafe for consumption. Furthermore, the predominant water types in the research area were calcium type, calcium–chloride type, and chloride type, with calcium as the dominant cation and sulfate as the dominant anion. The Gibbs plot indicated that fluid–rock interaction is the most important controlling process influencing the groundwater quality.

Moreover, arsenic levels in 28% of the water samples surpassed the suitable levels provided by WHO guidelines. Stations of Syed Wala, Mandi Faizabad, More Khunda, Rehan Wala, and Khiary Kalan areas reveal maximum arsenic concentration. The correlation between arsenic and pH was performed because arsenic solubility reduces as pH increases. PCA analysis showed that the correlation between arsenic and other chemical parameters was weak, except for pH, whereas cluster analysis detected a positive correlation between pH, Fe, and As, implying the reducing environment of the aquifer. Moreover, the positive and negative values of chloro-alkaline indices revealed that both direct and reverse ion exchange processes influenced the groundwater chemistry of the research area. The saturation mineral indices indicated the oversaturation of dolomite and calcite, but showed a undersaturation of halite and the dissolution of some gypsum samples. Health risk assessment outcomes exhibited that 28% of the groundwater samples are highly hazardous (HQ) to the residents on lifetime exposure to polluted water.

The present research concluded that due to arsenic contamination and poor groundwater quality, most of the groundwater samples were not appropriate for human consumption without proper treatment and remediation of arsenic. Therefore, it is recommended that proper water quality control monitoring of arsenic contamination be implemented to protect residents from arsenic contamination.