Assessment of groundwater quality and human health risk from nitrate contamination using a multivariate statistical analysis

Groundwater plays an essential role in providing a major drinking water sources in arid and semi-arid areas worldwide. About 2.5 billion people worldwide fulfill their drinking needs using groundwater as the only source (UNESCO 2012; Aasif et al. 2023). Many studies have indicated that the quality of groundwater has been deteriorating over time due to various factors such as rapid industrialization, population growth, anthropogenic and geogenic activities, excessive exploitation, the intensive use of fertilizers and pesticides in agriculture, and the discharge of untreated municipal and industrial effluents into water sources (Gao et al. 2020; Adimalla & Qian 2022). Consumption of contaminated drinking water poses a threat to human health, resulting in various health issues in many regions around the world (Adimalla et al. 2020; Rajmohan et al. 2021). Over the past few decades, extensive research is reported about the geochemistry of groundwater, groundwater quality, sources of groundwater pollution, and the interrelated health risks. For example, Adimalla et al. (2020) evaluated the groundwater quality in the Nalgonda district, Telangana state, India. They found that health risk was higher in infants and children as compared to adults. However, while evaluating the comprehensive chemical quality of groundwater, Krishan et al. (2023) employed an integrated approach for studying groundwater quality for drinking through hydrochemistry and water quality index in the Mewat district, northern India. (Subba Rao 2012; Rao et al. 2018; Adimalla et al. 2020; Raheja et al. 2024) employed the pollution index of groundwater (PIG) to estimate the relative influence of each physicochemical parameter on the overall chemical quality of groundwater. Their results showed that hydrogen ion concentration (pH), electrical conductivity (EC), total dissolved solids (TDS), total hardness (TH), calcium (Ca²⁺), magnesium (Mg²⁺), bicarbonate (HCO₃⁻), chloride (Cl⁻), sodium (Na⁺), potassium (K⁺), sulfate (SO₄²⁻), nitrate (NO₃⁻), and fluoride (F⁻) were used as pollution indicators to estimate groundwater quality for drinking. Raheja et al. (2023a) investigated the groundwater quality in Kurukshetra, India, using multivariate approaches (Irrigation Water Quality Index (IWQI), Wilcox diagram, and principal component analysis (PCA)). PCA is an effective tool that divides various hydrochemical parameters into distinct principal components (PCs) based on the relations among different variables. These components represent the basic factors influencing groundwater chemistry (Lanjwani et al. 2020; Zhang et al. 2022).

Several studies have highlighted the significance of employing various methods not just for measuring groundwater quality but also for evaluating pollution sources and identifying susceptible distribution patterns (Najafzadeh et al. 2021; Sunitha & Reddy 2022; Zhang et al. 2022; Raheja et al. 2023c). Kumar et al. (2019) applied various techniques to identify the origin of nitrate contamination in the groundwater of the middle Gangetic plain, India, and suggest that nitrate pollution in groundwater is primarily caused by human activities, including the application of fertilizers, livestock practices, the presence of wastewater ponds, and seepage from septic tanks. For example, the study by Bazeli et al. (2022) in rural areas of Khaf County, Iran, assessed non-carcinogenic human health risks and revealed that the hazard quotient (HQ) values of nitrate for infants, children, teenagers, and adults in 32, 32, 14, 10, and 7% of the groundwater samples were higher than the desired limit. Another study carried out by Singh et al. (2020) in the alluvial plains of Punjab, northern India, found that children are more exposed through direct ingestion compared to other age groups. Prolonged ingestion of high nitrate levels can pose instantaneous health hazards, including the risk of methemoglobinemia (commonly known as blue baby syndrome) in newborn babies and a sensitive potential for stomach cancer in men and women (WHO 2017; Karnena et al. 2022; Sun et al. 2023). The World Health Organization (WHO 2017) has set a maximum allowable nitrate concentration in drinking water of 50 mg/L. In India, the Bureau of Indian Standards (BIS 2015) recommends a permissible nitrate level in drinking water of 45 mg/L. The issue of high salinity in groundwater affects the northwestern regions of India, including Haryana, Delhi, Punjab, and Rajasthan, where these problems originated from terrestrial sources (Krishan et al. 2023). In recent years, several studies on groundwater have focused on issues such as monitoring pollution sources, examining geochemistry, and assessing the quality of water for both drinking and irrigation uses (Howladar et al. 2018; Toure et al. 2019; Adimalla 2020; Aouiti et al. 2021; Patel et al. 2022; Dey & Raju 2023; Raheja et al. 2023b). Factually, this research primarily highlighted a particular problem-solving approach within specific regions. Consequently, there exists a research gap that can be addressed by presenting a novel combined approach. This study reveals the effectiveness of such an innovative integrated method.

An extensive literature review suggests that so far no research study has evaluated the groundwater quality of Rohtak district (Haryana, India) based on PIG and human health risk from nitrate pollution using an integrated approach. Hence, there is a need to understand the groundwater quality and assess the human health risk in the study area. Keeping this in view, the present study is carried out (a) to assess the chemical characterization of groundwater in the Rohtak district, (b) to evaluate groundwater quality for drinking purpose using PIG and their spatial distribution, (c) to recognize the hydrogeochemistry of groundwater by using multivariate techniques (PCA, correlation), and (d) to estimate the concentration of nitrate and its potential non-carcinogenic health risk for children, women, and men. The results of this research will offer valuable scientific insights into the health risks associated with nitrate present in groundwater, contributing to the maintenance of water quality standards in this region.

The Rohtak district in Haryana, India, is located between 28°40.46′ and 29°06.08′ North latitude and 76°12.40′ and 76°52.00′ East longitude. The district covers an area of 1,745 km² and has an elevation range varying from 216 to 275 m above sea level. The average temperature in the area varies from 7 °C in January to 40.5 °C in May and June, and the annual average rainfall is about 592 mm. The study area falls within a subtropical semi-arid region, which is known for its hot summers and cold winters. The depth of groundwater levels fluctuates between the range of less than 1.72–10.75 m below ground level (BGL) during the pre-monsoon period and 1.46–9.07 m BGL during the post-monsoon period (CGWB 2019).

All the cation and anion concentrations are in mg/L, and all groundwater samples fall within the acceptable limit of ±10%.

The PIG, initially proposed by Subba Rao (2012), is an effective tool for evaluating the overall water quality pollution based on hydrochemical data and water quality criteria. PIG calculation involves five different steps, which are explained below:

Step 1: Calculate the relative weight (⁠⁠) of each groundwater quality parameter. The varies from 1 to 5 and is assigned to each chemical parameter, depending on the level of importance of groundwater for drinking purposes, as shown in Table 1.

Table 1

Relative weight of groundwater quality parameters

Sr. No. .	Name of parameters .	Relative weight (⁠⁠) .	.
1	pH	3	0.0652
2	TDS	4	0.0870
3	EC	4	0.0870
4	TH	3	0.0652
5	Ca2+	2	0.0435
6	Mg2+	4	0.0870
7	HCO3−	1	0.0217
8	Na+	3	0.0652
9	K+	3	0.0652
10	Cl−	4	0.0870
11	SO42−	5	0.1087
12	NO3−	5	0.1087
13	F−	5	0.1087
= 46

Sr. No. .	Name of parameters .	Relative weight (⁠⁠) .	.
1	pH	3	0.0652
2	TDS	4	0.0870
3	EC	4	0.0870
4	TH	3	0.0652
5	Ca2+	2	0.0435
6	Mg2+	4	0.0870
7	HCO3−	1	0.0217
8	Na+	3	0.0652
9	K+	3	0.0652
10	Cl−	4	0.0870
11	SO42−	5	0.1087
12	NO3−	5	0.1087
13	F−	5	0.1087
= 46

View Large

PIG value can be classified into five categories (Subba Rao 2012). If PIG value is less than 1.0, it falls in the insignificant pollution class. If PIG values lie between 1.0 and 1.5, then it falls in the low pollution class. If PIG values lie between 1.5 and 2.0, then it falls in the moderate groundwater pollution class. If PIG values lie between 2.0 and 2.5, then there is high pollution, and if the PIG value exceeds 2.5, groundwater has very high pollution, which means that the groundwater in a specific region is at a higher risk of contamination (Subba Rao 2012; Adimalla et al. 2020).

Inverse distance weighting (IDW) and Kriging are the primary interpolation methods for creating spatial representations of water quality assessments across different locations (Kumar & Sangeetha 2020; Fatima et al. 2022; Azhari et al. 2023). However, IDW can also detect parameters that are potentially not distributed in normal ways, while Kriging is only capable to perform well with normally distributed samples (Kamaraj et al. 2021; Raheja et al. 2022). For this study, IDW is used to prepare the spatial distribution maps. In addition, PCA is also used to identify the correlations among various physicochemical parameters in the study area. Its aims are to moderate the complexity of a multivariate dataset by capturing its information through a limited set of PCs, thereby preserving the essential characteristics of the environment. This reduction in information allows for the extraction of primary factors containing the necessary information. Therefore, to monitor the groundwater quality of the study area and to observe the major influencing factors, PCA is applied on the obtained results of physicochemical parameters.

Table 2 provides a comprehensive statistical summary of the analysis of 13 physicochemical parameters (including pH, EC, TDS, Ca²⁺, Mg²⁺, HCO₃⁻, Na⁺, K⁺, Cl⁻, SO₄²⁻, NO₃⁻, and F⁻) across 47 groundwater samples. Moreover, these results are compared with the drinking water quality standards recommended by the Bureau of Indian Standards (BIS 2015) and the World Health Organization (WHO 2017) to estimate their suitability for drinking purposes in the study region. The standard deviation (SD) values vary from 0.33 to 3,717.13, and this considerable difference in SD may stem from various hydrogeochemical reactions and substantial differences in the distribution of salts within the groundwater. The coefficient of variation (CV) ranges from 4.92 to 107.42%, and the highest value of CV is of Mg²⁺ (107.42%) followed by EC (87.39%), Ca²⁺ (70.66%), TDS (68.41%), SO₄²⁻ (50.04%), Cl⁻ (48.93%), Na⁺ (40.24%), K⁺ (40.19%), NO₃⁻ (37.86%), F⁻ (37.35%), HCO₃⁻ (34.69%), and pH (4.92%). This observation recommends that Mg²⁺, EC, Ca²⁺, and TDS might be the primary key factors that naturally govern the groundwater chemistry in the study area.

The pH value in the study area lies between 7.05 and 8.7, with a mean of 7.66, suggesting that the study area's groundwater is neutral to slightly alkaline. The value of pH is mainly within the permissible limit (6.5–8.5) recommended by BIS (2015) and WHO (2017) for drinking (Table 2). TDS concentration ranges between 229 and 8,544 mg/L with a mean of 2,681.89 mg/L. The maximum permissible limit is 2,000 mg/L, as suggested by BIS (2015). The value of TDS in about 51.06% of groundwater samples exceeded this permissible limit (BIS 2015). EC measures a material's capacity to enable the flow of electric current through the dissolved salts in water. EC can be classified as follows: low salt enrichment if EC value is less than 1,500 μS/cm, medium salt enrichment if EC falls in between 1,500 and 3,000 μS/cm, and high salt enrichment if EC is more than 3,000 μS/cm. In the present study, EC value of the study area ranges from 329 to 18,400 μS/cm with a mean of 3,717.13 μS/cm. However, only 8.5% of groundwater samples fall under the Class I type, and 40.4 and 51.1% are found in the Class II and Class III types, respectively. This observation strongly suggests that interactions between rocks and water, geochemical reactions, and anthropogenic influences are the predominant factors influencing groundwater chemistry in the study area. The TH value was found to vary from 110 to 3,360 mg/L with a mean of 794.38 mg/L. The TH is divided into four different classes: TH <75 mg/L indicates soft water; when the TH value lies between 75 and 150 mg/L, moderately hard water; between 150 and 300 mg/L, hard water; and more than 300 mg/L, very hard water (Sawyer & McCarty 1978). According to this classification, no groundwater samples fall under soft water, and 14.89 and 80.85% fall under the hard and very hard water category. About 48.93% of groundwater samples have exceeded the maximum permissible limit of 600 mg/L, which is unsuitable for drinking purposes (BIS 2015). The predominant of cations and anions based on their mean values are in the following order: Mg²⁺ > Ca²⁺ > Na⁺ > K⁺, and Cl⁻ > SO₄²⁻ > NO₃⁻ > HCO₃⁻ > F⁻. The concentration of Ca²⁺ varied from 60 to 1,060 mg/L with a mean of 294.33 mg/L. Of the groundwater samples, 60% have a higher value than the recommended limit of 200 mg/L (BIS 2015; WHO 2017). The Mg²⁺ value was found to vary from 50 to 2,900 mg/L with a mean of 500.1 mg/L. The HCO₃⁻ concentration ranged from 1.8 to 14.9 with a mean of 7.65 mg/L. A comparison with the permissible limit of HCO₃⁻ for drinking water mentions that all the groundwater samples are within acceptable limits (WHO 2017). The concentration of Na⁺ ranges from 115 to 636 mg/L for all samples collected in the study area. This study found that 68.08% of groundwater samples exceeded the permissible drinking limit recommended by WHO (2017). The value of K⁺ ranges from 29 to 163 mg/L, with a mean value of 90.94 mg/L. Keeping in view of the excess amount of both Na⁺ and K⁺ in the study region, it is recommended that groundwater should be treated before using it for drinking purposes. The Cl⁻ concentration was found to vary from 85 to 921 mg/L with a mean of 462.30 mg/L (Table 2). Results suggest that 76.5% of the total groundwater samples have a higher value of Cl⁻ content than the recommended limit of 250 mg/L for drinking water (BIS 2015). The SO₄²⁻ concentration in the study area was between 75 and 965 mg/L, and about 50% of the groundwater samples exceed permissible limits (BIS 2015). The F⁻ concentration ranges between 0.25 and 1.65 mg/L with a mean of 0.87 mg/L (Table 2). In this study, only 6.3% of the samples exceeded the maximum permissible limit (1.5 mg/L) for drinking purposes (BIS 2015).

The spatial distribution of non-carcinogenic risk (HI) suggests that approximately 80% of the study region's population, including children, women, and men, face a higher health risk. As evident from Figure 6, the prominently green areas in the spatial map of HI across the study area specify no significant non-carcinogenic health risk. However, the yellow and red zones highlight areas where people may potentially face health risks in this area. The yellow and red regions shown in Figure 6 specify that the groundwater samples from these areas were unsuitable for direct consumption due to their intolerable non-carcinogenic values. Consuming an excessive amount of nitrate in drinking water can result in adverse health effects (Singh et al. 2020; Adimalla et al. 2021). Even so, the non-carcinogenic risk of nitrate contamination in drinking water has become a big concern in several areas globally, particularly in areas where a large population depends on groundwater for drinking without prior quality assessment (Adimalla et al. 2020; Singh et al. 2020; Verma et al. 2023). However, the results of this study demonstrate that the health risks of nitrate are most severe for children. Nevertheless, the current study's findings clearly suggest that the state government should promptly implement plans to reduce nitrate contamination in groundwater as well as ensure the provision of safe drinking water in the affected areas.

In this research, the nitrate concentration in groundwater was measured, and a PIG and human health risk assessment for nitrate exposure was conducted. Furthermore, spatial distribution maps were generated, and PCA was conducted to identify contamination zones and potential sources of pollution. Groundwater generally exhibits a neutral to slightly alkaline nature. The cations and anions based on their average values are in the following order: Mg²⁺ > Ca²⁺ > Na⁺ > K⁺, and Cl⁻ > SO₄²⁻ > NO₃⁻ > HCO₃⁻ > F⁻. The spatial maps demonstrate that a higher nitrate contamination is observed in the northern part, while the lowest value is observed in the western part of the study region. However, the PIG values specify that most of the study area is occupied by moderate to high pollution zones. PC1 has positive loadings of TDS, TH, Mg²⁺, Ca²⁺, and EC and suggests that groundwater chemistry primarily replicates the influence of natural sources, rock and mineral weathering, and anthropological activities. PC2 has positive loadings with NO₃⁻ and F⁻, indicating ions are primarily associated with human inputs, domestic sewage infiltration, unexpected irrigation practices, industrial wastewater, septic tanks, animal waste, and landfill leachate. The mean value of non-carcinogenic risk is 1.191 for men, 1.408 for women, and 2.069 for children, respectively. The health risk assessment specified that children in the study area face a higher susceptibility to health risks compared to women and men.

HR is grateful to the Ministry of Education, Government of India (Grant No. 2K19/NITK/PHD/61900011), for funding this study. The authors acknowledge the National Institute of Technology, Kurukshetra, for providing various research facilities.

HR was responsible for sample collection and assessment, data processing, writing the original draft, methodology, conceptualization, analysis, editing, and funding acquisition. AG was responsible for the visualization, conceptualization, and supervision of the whole research work. MP was responsible for visualization, conceptualization, data processing, and reviewing and editing the writing of the entire research work.

The authors declare there is not conflict.

All relevant data are included in the paper or its Supplementary Information.