Characterization of groundwater potability and irrigation potential in Uttar Pradesh, India using water quality index and multivariate statistics

Groundwater, as compared to surface water, provides a more hygienic, safe, and portable supply of water. However, this perception may not always hold due to the susceptibility of groundwater to contamination from human activities and natural occurrences (Masindi & Foteinis 2021). Over the past few decades, the excessive use of groundwater has severely threatened water security (Zhengxian & Wang 2021). Population expansion, growing standards of living, and the spread of irrigated agriculture have resulted in increased water demand and the withdrawal of groundwater (Gupta et al. 2023; Levintal et al. 2023). Out of the total water present on earth, less than 1% is present as fresh water for direct use (Dudgeon 2019). It is presumed that groundwater will be declining to an alarming level in the coming decades making it unsuitable for drinking and irrigation purposes (Snousy et al. 2022). According to lithology weathering of rocks, ion-exchange processes, etc., groundwater characteristics change with time and space (Narany et al. 2015). Groundwater quality is also deteriorated due to anthropogenic reasons such as agricultural runoff, untreated industrial and municipal wastewater discharge, and leaching (Gupta et al. 2019; Barbieri et al. 2021). Evaluating the quality of groundwater is an essential part of managing groundwater (Tang et al. 2023). The Sustainable Development Goal 6 (SDG 6) of the United Nations focuses on the provision of clean water and sanitation, aiming to sustainably manage water resources, wastewater, and ecosystems (Zhang et al. 2023). Analysing the hydrochemical composition of the water is a fundamental requirement for assessing its quality and determining the water quality index (WQI).

Uttar Pradesh (UP) is one of the biggest states in India where substantial water resource development projects have been launched since independence in the form of large-scale canal irrigation projects. The most significant shift in the way of irrigation facilities was observed in the early 1980s, when the Green Revolution and low-cost pump set technology triggered a change in irrigation patterns; consequently, a high exploitation of groundwater resources started. To fulfil a range of drinking and irrigation requirements, it is essential to conduct a hydro-chemical characterization of groundwater, which helps to uncover the underlying processes that govern its quality. Overextraction of groundwater may lead to a change in the physicochemical composition of water. For the geochemical identification of groundwater, several methods, including Chadha plot, chloro-alkaline indices, and saturation indices (Kumar et al. 2022), have been employed. Additionally, the evaluation of overall water quality has been done using a variety of indices, including the Arithmetic WQI (Tripathi & Singal 2019), Comprehensive WQI (Kumar et al. 2020), Oregon WQI (Abbasi 2002), and Canadian Council of Ministers of the Environment WQI (Khatri et al. 2020). There are certain limitations associated with calculating the WQI according to Bui et al. (2020), such as the requirement of a significant amount of time, being a lengthy and intricate process, and being inconsistent due to the involvement of multiple equations. During the evolution of WQI, assigning weights to each parameter is a crucial step (Tripathi & Singal 2019). Every specialist may have a different opinion in assigning a weight to the parameter, leading to indiscriminate uncertainties in the WQI value. Various weight-assigning methods, such as relative importance weight, entropy weight, criteria-importance through inter-criteria correlation weight, and integrated weight, have been widely employed to assess water quality through the calculation of WQI (Das & Das 2023). Additionally, groundwater quality needs to be determined for assessing water quality for drinking and irrigation through multiple indices, including the sodium percentage, Kelly index, and sodium adsorption ratio (SAR) (Gupta et al. 2023). In various studies, multivariate statistical methods, such principal component analysis (PCA) and cluster analysis (CA), were used to detect probable contaminants (Fathi et al. 2018). PCA is a multivariate statistical technique that can be used to decrease data when large datasets are available (Noori et al. 2009). Additionally, CA is one of the frequently used multivariate statistical methods to assess the degree of homogeneity in the identified variables (Bagla et al. 2021).

In the state of UP, the major water source for drinking, irrigation, and other domestic needs is groundwater (Tiwari et al. 2017). According to the UP state's groundwater report (2021), the total annual extractable groundwater in India and the UP is 393.70 billion cubic metres (bcm) and 65.32 bcm, respectively. Out of which 40.89 bcm of groundwater is utilized for irrigation in UP every year, thus leaving 19.48 bcm for future use (Yadav et al. 2021). During the past few decades drinking water in this region is associated with many contaminants, such as fluoride and many more (State of Groundwater in Uttar Pradesh 2021). Several studies have reported fluoride contamination in the groundwater of UP. Some of the sampling points from the districts of Agra, Varanasi, Raebareli, and Kanpur with Unnao and Pratapgarh were found most affected (Tiwari et al. 2017; Maurya et al. 2020). Groundwater in the Mathura is hard to extremely hard type and is alkaline (Ahmad et al. 2019).

According to the World Health Organization (WHO), fluoride is a crucial component for healthy bone and tooth growth and development (WHO 2017). The consumption of water with a fluoride concentration greater than 1.5 mg/L has been associated with serious health risks. The consumption of fluoride-enriched water puts nearly 260 million people at risk in 28 countries to health-related problems such as skeletal and dental fluorosis, neurotoxicological complications in children, spontaneous abortion, and severe soft tissue damage (Maurya et al. 2020). Exceeding the permissible limits, fluoride contamination has resulted in significant health issues for over 40% of India's population, with 28 states reporting higher fluoride concentrations (Chakraborti et al. 2016). Understanding the sources of pollutants and the hydrogeochemical processes, and regular water quality monitoring, are necessary for the sustainable development and effective management of groundwater resources in each region. However, information on groundwater quality and the impacts of industrial and urban growth on groundwater resources is accessible only for some of the districts, such as Moradabad, Unnao, Mathura, Fatehpur, Raebareli, Gaziabad, Varanasi, and Sonbhadra (Pathak et al. 2008; Raju et al. 2012; Rawat et al. 2012; Chabukdhara et al. 2017; Chaurasia et al. 2018; Ansari & Umar 2019; Sahu et al. 2019). Very limited research has been done to cover water quality for the entire state. Such information is still lacking in many districts of UP.

We conducted this study to assess the water quality of the entire UP state and develop water quality indices for groundwater to ensure its suitability for various uses such as drinking and irrigation. Considering the limitations of previous studies in terms of addressing the water quality challenges, this study was planned and extended. In the present study, an effort has been made to identify the key ion chemistry that regulates the composition of the groundwater in the 69 districts of UP. Following this, the WQI of groundwater for irrigation and drinking purposes has been calculated. A range of irrigation indicators, such as sodium absorption ratio (SAR), Kelly's ratio (KR), etc., have also been calculated. Several multivariate statistics have been applied to assess the significance of groundwater quality data.

With 243,290 km² of total land area, UP is the fourth-largest state in terms of the geographical area whereas with a population of 199,581,477 it is the most populous state in the country. It lies between the longitudes of 77°3′ and 84°39′E and the latitudes of 23°52′N and 31°28′N. It is situated in the northwest of India and has a border with Nepal on the outside. The state of UP can be divided into three main topographical regions based on its geomorphology: first, the Gangetic plains make up the majority of the central part of the state. Second, the Siwalik foothills of the Himalayas and the Terai region border the state on the north; and third, the Vindhyan Range and plateau lie in a relatively smaller area of the southern region of the state.

The State of UP, which is primarily covered in Gangetic alluvium, is distinguished by a variety of hydrogeology formations that range in geological age from Archean to Recent and were created because of various topographical, climatic, and geological conditions. These formations have an impact on groundwater repositories across the state's river basins, together with the space-time variable yearly water cycle and almost the entire state is surrounded by the Ganga basin. The Ganges River system, comprising the Ganga and Yamuna rivers along with their tributaries, originates from the Himalayas and contributes to the deposition of alluvial soil in the Gangetic plains. This study region belongs to the Central Gangetic Plain and is distinguished by Quaternary alluvial deposits that comprise both older and younger alluvium. This area's geology is mostly made up of alluvium, which is made up of sand, silt, clay, and kankar. This location has alluvial soil, clayey loam, and sandy loam among its soil types (Ekbal & Khan 2022). This fertile soil is conducive to the cultivation of various crops, such as rice, wheat, barley, and gram. Because of varied hydrogeological and geomorphological contexts, ranging from copious in the alluvial plain to limited in Bundelkhand, the regional and temporal distribution of groundwater availability is non-uniform. According to the Ministry of Water Resources, River Development and Ganga Rejuvenation, five hydrogeological units: -Bhabar, Marginal alluvial plains, Tarai, Central Ganga plains, and Southern Peninsular Zone, can be distinguished within the states. In UP, the mean annual rainfall is 798.0 mm in the east UP and 817.1 mm in the west UP region according to the IMD Report in 2018 (IMD 2018).

The chemical composition of groundwater within aquifers is heavily influenced by a combination of natural and human-driven factors (Hussainzadeh et al. 2023). To understand the hydrochemical processes, a hydrochemical scatter plot created by Chadha (1999) was used. The groundwater's cations and anions were denoted by [(Ca²⁺ + Mg²⁺) − (Na⁺ + K⁺)] and [(CO₃ + ⁠) − (⁠ + Cl⁻)], respectively. The graph was divided into four quadrants, with the first one representing recharging water (Ca–MgHCO₃ type), the second base ion-exchange water (NaHCO₃ type), the third sea water (NaCl type), and the fourth reverse ion-exchange water (Ca–Mg–Cl type).

Groundwater suitability for drinking and irrigation was evaluated using the WQI, a composite rating indicator. Using WQI, a layman may readily comprehend the groundwater quality for various purposes. WQI's main objective is to condense complex and enormous amounts of physicochemical data (Sutradhar & Mondal 2021). The entire calculation was founded on average values for each variable that have been suggested by numerous organizations, including the Bureau of Indian Standards (BIS 2012), the Indian Council of Medical Research (ICMR 1975), the WHO (2017), and Food Agriculture Organization (FAO 2013).

The following steps were involved in the WQI's calculation and formulation:

• Step 1. Each of the selected physicochemical variables is assigned a weight (w_i) on a scale of 1–5 based on the sample's percentage that falls within the allowable limit as per the prescribed standards, according to its relative importance in the overall water quality for drinking and irrigation needs. For drinking water, weight to each variable was assigned based on their human health impact, its influence on groundwater quality, and with the help of previous studies (Kumar et al. 2015; Vaiphei et al. 2020; Chaurasia et al. 2021) but in the case of irrigation water, samples that were having values higher than the standards, the weight of 5, have been allocated to them whereas the weight of 1 has been given to the samples having values less than the standards. Each variable had been given a relative weight which was calculated by summing the assigned weights (reflecting their relative importance), and dividing them by the total weight (Equation (1)). The sub-index values were aggregated using arithmetic calculation. The overall index value, which determines whether water quality is classified as ‘Good or Bad,’ is also influenced by the specific numerical values assigned to water quality variables during the aggregation process (Sargaonkar et al. 2008).

• Step 2. To calculate a relative weight (Rw_i), by using the following formula:

  

  (1)

  where n is the total number of chosen variables and Rw_i denotes the relative weight of individual variables. w_i denotes the weight of individual variables.
• Step 3. Using Equation (2), a scale of quality has been developed

  

  (2)

  where q_i refers to the quality rating, C_i represents the concentration of each physiochemical variable in each water sample, and S_i refers to standard limits proposed by various organizations for drinking water or irrigation needs.
• Step 4. Equation (3) has been used to calculate the WQI

  

  (3)

WQI stands for the water quality index, Rw_i (Equation (1)) denotes the relative weight of each physicochemical variable, and q_i (Equation (2)) stands for the quality rating. The WQI was developed for drinking water and irrigation needs. WQI values for drinking water uses have been categorized into five classes as follows; class I (excellent, <50), class II (good, 51–100), class III (poor, 101–200), class IV (very poor, 201–300), and class V (unsuitable, >300). But for irrigation water WQI values had been grouped into three classes as follows: class I (none restriction, ≤150), class II (slight restriction, 152–300), class III (moderate restriction, 301–450), class IV (severe restriction, ≥450).

The crop productivity and its impact on the soil properties are affected by the different quality parameters of water used for irrigation. Toxic and inadequate nutrient phases in the water and excess concentration of different ionic species have a negative impact on crop yield. As a result, several variables have been considered to determine if the ground water is suitable for irrigation, as mentioned below (measurements in square brackets which represent the appropriate element's respective ionic concentration levels in meq/L). The suitability of groundwater for irrigation was assessed using different indices i.e. sodium percentage (Na%), Kelly ratio (KR), sodium absorption ratio (SAR), magnesium absorption ratio (MAR), residual sodium bicarbonate (RSBC), potential salinity (PS), and Wilcox diagram (Wilcox 1955). A description of all the irrigation indices is given in Table 1.

Table 1

Irrigation indicators with description

S.N. .	Indicators .	Description .	Formula .	References .
1	SAR	The high sodium component of irrigation water creates an alkali risk and lowers soil permeability.		Roy et al. (2018) and Ravikumar et al. (2013)
2	KR	It has been developed to assess the suitability of irrigation water based on sodium-ion compared with calcium and magnesium ions.		Doneen (1964)
3	RSBC	Due to the scarcity of carbonate ions in large concentrations and the fact that bicarbonate ions do not precipitate magnesium ions, RSBC was proposed.		Gupta (1983)
4	MAR	The increased alkalinity of soil caused by the higher magnesium concentration in irrigation water has an impact on crop production.		Raghunath (1987)
5	Na%	The excess sodium combined with the carbonate ion will help in converting the soil to an alkaline state. Instead, mixing sodium with chloride ions would hasten the development of saline soil, which eventually degrades the soil's ability to absorb water and inhibits plant growth.		Rao & Latha (2019)
6	PS	It is determined as the concentration of chloride ions plus half of the sulphate ions.		Doneen (1964)

S.N. .	Indicators .	Description .	Formula .	References .
1	SAR	The high sodium component of irrigation water creates an alkali risk and lowers soil permeability.		Roy et al. (2018) and Ravikumar et al. (2013)
2	KR	It has been developed to assess the suitability of irrigation water based on sodium-ion compared with calcium and magnesium ions.		Doneen (1964)
3	RSBC	Due to the scarcity of carbonate ions in large concentrations and the fact that bicarbonate ions do not precipitate magnesium ions, RSBC was proposed.		Gupta (1983)
4	MAR	The increased alkalinity of soil caused by the higher magnesium concentration in irrigation water has an impact on crop production.		Raghunath (1987)
5	Na%	The excess sodium combined with the carbonate ion will help in converting the soil to an alkaline state. Instead, mixing sodium with chloride ions would hasten the development of saline soil, which eventually degrades the soil's ability to absorb water and inhibits plant growth.		Rao & Latha (2019)
6	PS	It is determined as the concentration of chloride ions plus half of the sulphate ions.		Doneen (1964)

View Large

Water quality data were subjected to different statistical tests, such as Wilcox, Chadha diagram, etc., using MS Excel 2021 R studio was used for creating a heatmap with dendrogram and corrplot, and PCA was done by Origin 2023. Based on the WQI, the suitability of groundwater has been assessed for drinking and irrigation needs. Using the digital elevation model in ArcGIS 10.5 software, spatial mapping for irrigation and drinking has been conducted.

The average physicochemical composition of groundwater samples from different sampling sites of the study area is presented in Table 2. The mean value of different variables, such as pH, EC (μS/cm), TH (mg/L), Ca²⁺ (mg/L), Mg²⁺ (mg/L), Na⁺ (mg/L), K⁺ (mg/L), (mg/L) Cl⁻ (mg/L), (mg/L)m and F⁻ (mg/L) was 8.10, 999.33, 313.75, 49.87, 51.43, 97.87, 7.77, 322.41, 115.83, 73.39, and 0.63, respectively, whereas the percentage of samples having greater values than the standard limit for drinking water was 5.52, 98.98, 24.14, 10.34, 52.07, 13.10, 5.525, 68.97, 6.55, 6.90, and 4.48%, respectively for these variables. The pH levels varied between 7.35 and 8.93 across the study area, indicating a predominantly neutral to slightly alkaline characteristic of the groundwater. EC and TH were very high in some of the sampling sites in the Mathura and Agra districts. The values of different variables within the standard limit prescribed by regulatory agencies suggest an overall good water quality, whereas a greater mean value of variables indicates poor water quality. Excess potassium (K⁺) concentrations in groundwater samples could be the result of anthropogenic activity such as fertilizer use. One likely source of bicarbonate in the groundwater is the presence of molecules of organic matter that are oxidized to create carbon dioxide, which encourages mineral dissolution (Saha et al. 2019). As shown in Table 2, we observed that some variables, i.e. EC, TH, Ca²⁺, ⁠, Cl⁻ have higher concentrations than their permissible limits and cause several impacts on human health. The standard concentrations of the physicochemical variables in the samples and how these chemicals harm human health when they surpass the limit are given in Table 2. The pH values of different samples were within the acceptable range (6.5–8.5) for approximately 95% of samples (Table 2). Whereas, for EC, 98.97% of the samples in the study area had values higher than the permissible limit. EC is an indirect way to measure the dissolved substances in an aqueous solution and higher values of EC indicated higher concentration dissolved material in a water sample (Chaurasia et al. 2018). Table 2 provides the descriptive statistics of physiochemical variables and their influence on human health. Table 3 summarizes the weight and relative weight of the variables used to calculate the WQI.

Table 2

Descriptive statistics of physiochemical variables in groundwater in UP and their impact on human health

Variable .	Mean ± Std deviation .	% of samples having a higher value than the allowable limits (drinking water) .	% of samples having a higher value than the allowable limits (irrigation water) .	Impact on human health when the limit is exceeded .
pH	8.10 ± 0.26	5.52	5.52	Skin and eye discomfort, mucous membrane irritation, and exacerbation of skin problems
EC	999.33 ± 1,343.43	98.97	21.38	Gastrointestinal conditions
TH	313.75 ± 452.60	24.48	3.79	Problems with the stomach and kidneys, as well as artery calcification
Ca2+	49.87 ± 53.28	10.69	0.34	Kidney stones, hypercalcaemia, improper brain, and heart function
Mg2+	51.43 ± 110.15	52.07	17.24	The effects of laxatives and hypermagnesemia
Na+	97.87 ± 146.28	13.45	0.34	Blood pressure elevation, coronary heart disease, and hypertension
K+	7.77 ± 22.54	5.52	82.76	Hyperkaleamia can lead to a heart attack, as well as problems with the digestive system and brain systems
	322.41 ± 176.12	68.97	2.41	Respiratory and metabolic acidosis
Cl−	115.83 ± 325.21	6.90	1.03	Cause heart disease, asthma, eczema, dry, itchy skin, and even cancer
	73.39 ± 197.36	7.24	0	The risk of dehydration from diarrhoea, as well as the laxative effect
F−	0.63 ± 0.65	4.48	4.48	Fluorosis in the skeleton and teeth

Variable .	Mean ± Std deviation .	% of samples having a higher value than the allowable limits (drinking water) .	% of samples having a higher value than the allowable limits (irrigation water) .	Impact on human health when the limit is exceeded .
pH	8.10 ± 0.26	5.52	5.52	Skin and eye discomfort, mucous membrane irritation, and exacerbation of skin problems
EC	999.33 ± 1,343.43	98.97	21.38	Gastrointestinal conditions
TH	313.75 ± 452.60	24.48	3.79	Problems with the stomach and kidneys, as well as artery calcification
Ca2+	49.87 ± 53.28	10.69	0.34	Kidney stones, hypercalcaemia, improper brain, and heart function
Mg2+	51.43 ± 110.15	52.07	17.24	The effects of laxatives and hypermagnesemia
Na+	97.87 ± 146.28	13.45	0.34	Blood pressure elevation, coronary heart disease, and hypertension
K+	7.77 ± 22.54	5.52	82.76	Hyperkaleamia can lead to a heart attack, as well as problems with the digestive system and brain systems
	322.41 ± 176.12	68.97	2.41	Respiratory and metabolic acidosis
Cl−	115.83 ± 325.21	6.90	1.03	Cause heart disease, asthma, eczema, dry, itchy skin, and even cancer
	73.39 ± 197.36	7.24	0	The risk of dehydration from diarrhoea, as well as the laxative effect
F−	0.63 ± 0.65	4.48	4.48	Fluorosis in the skeleton and teeth

View Large

Table 3

Weight and relative weight of the physiochemical variables

Variable .	Drinking water .			Irrigation water .
	Standards [29,30,20] (BIS/ICMR/WHO) .	Weight (wi) .	Relative weight (Rwi) .	Standards (FAO) .	Weight (wi) .	Relative weight (Rwi) .
pH	6.5–8.5	1	0.04	6.5–8.5	1	0.0625
EC	300	5	0.20	1,000	2	0.1250
TH	300	3	0.12	712	1	0.0625
Ca2+	75	2	0.08	400	1	0.0625
Mg2+	30	2	0.08	60	1	0.0625
Na+	150	2	0.08	920	1	0.0625
K+	12	3	0.12	2	5	0.3125
	244	4	0.16	610	1	0.0625
Cl−	250	1	0.04	1,065	1	0.0625
	200	1	0.04	1,920	1	0.0625
F−	1.5	1	0.04	1.5	1	0.0625
		∑ = 25			∑ = 16

Variable .	Drinking water .			Irrigation water .
	Standards [29,30,20] (BIS/ICMR/WHO) .	Weight (wi) .	Relative weight (Rwi) .	Standards (FAO) .	Weight (wi) .	Relative weight (Rwi) .
pH	6.5–8.5	1	0.04	6.5–8.5	1	0.0625
EC	300	5	0.20	1,000	2	0.1250
TH	300	3	0.12	712	1	0.0625
Ca2+	75	2	0.08	400	1	0.0625
Mg2+	30	2	0.08	60	1	0.0625
Na+	150	2	0.08	920	1	0.0625
K+	12	3	0.12	2	5	0.3125
	244	4	0.16	610	1	0.0625
Cl−	250	1	0.04	1,065	1	0.0625
	200	1	0.04	1,920	1	0.0625
F−	1.5	1	0.04	1.5	1	0.0625
		∑ = 25			∑ = 16

Note: Standards are in mg/L and EC in μS/cm.

View Large

The WQI is a crucial factor in determining the quality of groundwater and its appropriateness for drinking and irrigation. It has been extensively employed to assess the suitability of water resources for domestic purposes (Varol & Davraz 2015). Long-term consumption of hard to very hard water in the diet may result in higher rates of health issues, such as anencephaly, parental mortality, and cardiovascular illnesses (Segun & Raimi 2021). In Table S1, the hardness of the study region is classified, in which 84.83% of samples fell in the ‘hard’ to ‘very hard’ category. Out of 290 sampling sites, 4 sites (1.38%) had excellent water quality while 21 sampling sites (7.24%) were unsuitable for drinking (Table S2).

At present, the groundwater of the UP plays a significant role in supporting approximately 70% of irrigated agriculture in the state, in addition to meeting around 90% of rural domestic requirements and over 75% of urban water consumption. Furthermore, it satisfies 95% of industrial, infrastructural, and commercial water demands (State of Groundwater in Uttar Pradesh 2021). The feasibility of using groundwater for irrigation in the study area has been assessed using the WQI. In addition, a variety of indicators are employed with the Wilcox diagram to determine if irrigation is suitable, including the SAR, sodium hazard (Na%), RSBC, potential salinity (PS), MAR, KR, etc. The classification of these indicators and the percentage of water samples that fall into each category are calculated in Table 4.

In this study, the concentration of fluoride was observed as 49% sites below 0.5 mg/L and 46% sites 0.6–1.5 mg/L, which is safe for drinking and does not cause fluorosis while 5% were not suitable for drinking and caused dental fluorosis [1% (1.6–2 mg/L), 3% (2.1–3 mg/L), & 1% (>3 mg/L)]. Out of 290 sampling sites in UP, it was observed that 13 sites had more than 1.5 mg/L which is above the permissible limit and caused dental to skeletal fluorosis such as Varanasi (2.2 mg/L) (Chaurasia et al. 2018), Chhatta, in Mathura district (2.08 mg/L) (Rawat et al. 2012), Dhankaur in Gautam Buddha Nagar district (3.14 mg/L), Dhata in Fatehpur district (2.14 mg/L), Lalganj-2 in Raebareli district (1.9 mg/L) (Sahu et al. 2019), etc. Maximum concentration was observed at Baldeo in Mathura district (6.53 mg/L) and minimum concentration was observed as Bhuta in Bareilly district, Patharrdewa in Deoria district and Chandraprabha in Chandauli district (0.15 mg/L).

The amount of fluoride in groundwater is influenced by factors such as pH, solubility, anion exchange capacity, and geological formation (Raju et al. 2012). Shallow groundwater tends to have higher fluoride concentrations due to increased weathering, while deeper water levels have lower fluoride concentrations (Pandey et al. 2016). It is essential to understand the possible consequences of exceeding allowable limits for fluoride in groundwater to make well-informed decisions on sustainable resource use, public health, and water management.

The Chadha diagram is a further representation of hydro-chemical characteristics. This categorization utilizes a modified Piper diagram. The general characteristics of water are described by the square or rectangular field. Each of the eight sub-fields in the rectangular field represents a different type of water and defines the basic characteristics of water (Balaji et al. 2017). The results of the analyses were plotted on the suggested diagram to determine its applicability to study the hydro-chemical processes and geochemical groundwater classification (Figure 5(b)). The findings of this analysis indicated that 75.15% of the samples were of CaHCO₃ water type, which signifies samples with recharge capacity, and 17.25% of the samples were of Ca–Mg–Cl water type, indicating reverse ion exchange. Furthermore, 3.11% of the samples were of the seawater category (NaCl water type), while 4.49% of the samples were categorized as NaHCO₃ type, representing base ion exchange water type (Sutradhar & Mondal 2021). It indicates that the groundwater in the study area was characterized mostly by high calcium and bicarbonate concentrations (Umar & Alam 2012). Most of the terrain of UP is made up of calcium carbonate-rich shale, dolomite, and limestone formations (CaCO₃). Rainwater passes through these rocks and dissolves the CaCO₃, which causes the groundwater to contain significant amounts of calcium and bicarbonate (HCO₃).

To assess the quality of the water, correlations between a sampling site and physicochemical variables of water have been identified using a Heatmap analysis (Figure 5(c)). Heatmap analysis revealed links between water quality variables for all districts of UP in which EC concentrations were observed very high at all the sampling sites compared to other variables. The variables, such as F⁻, pH, K⁺, Ca²⁺, Mg²⁺, ⁠, Na⁺, and Cl⁻ followed similar trends at all the sampling sites. TH and have higher concentrations due to carbonate and silicate rocks being dominant in all the sampling sites so weathering of these rocks showed a higher concentration of TH and ⁠. In the heat map graphs, we plotted the dendrogram (vertical and horizontal both) for different variables and dendrograms were divided into two clusters, i.e., Cluster I had six districts (sampling sites) which included Mahamaya Nagar, Aligarh, Mathura, Chitrakoot, and Jyotiba Phule Nagar. The remaining 63 districts were in Cluster II based on pollution sources and concentrations.

Irrigation water quality directly affects crop quality and repeated irrigation in the long term affects soil physicochemical properties. Using water full of all necessary nutrients and free of any pathogenic contamination is a requirement for optimal crop performance (Das et al. 2019). It has been evident that toxic and inadequate nutrient levels in the water have shown a negative impact on crop yield. As a result, several variables are mentioned below that have been considered to assess the suitability of water for irrigation (measurements in square brackets represent the appropriate element's respective ionic concentration levels in meq/L).

Only three components, representing 77.92% of the total variance, were significant for groundwater datasets (eigenvalues >1; see Figure 9(a)). According to Wu et al. (2020), component loading values classified as strong (more than 0.75), minor (0.75–0.50), and weak (below 0.50) are significant and can be employed (0.50–0.30). With 53.88% of the total variance, PC1 in the datasets is the most significant one. It also exhibits positive loading with EC, TH, Ca²⁺, Mg²⁺, Na⁺, Cl⁻, and SO₄. PC2 shows positive loading with pH and F and accounts for 13.74% of the overall variance. PC3 shows positive loading with K⁺ and and it explains 10.31% of the overall variation. The loading plot (Figure 9(b)) displays the interactions between variables in the area covered by the first two components. We can see from the loading plot that EC, TH, Ca²⁺, Mg²⁺, Na⁺, Cl⁻ and have similar heavy loadings for principal component 1 (PC1). However, pH and fluoride have identical high loadings of principal component 2 (PC2). PC1 has strong positive loadings for EC, TH, Ca²⁺, Mg²⁺, Na⁺, Cl⁻ and ⁠. PC2 has strong positive loadings for pH and F, which may reflect dolomitization. A minor positive correlation between F and pH is observed (Figure 8), showing that the greater alkaline quality of the water accelerates fluoride enrichment and thus normally impacts fluoride content in groundwater (Adimalla et al. 2019). PC3 is dominated by large positive loadings for K⁺ and ⁠, a strong relation between potassium and bicarbonate, indicating a common source of contamination. Similar to Selvam (2015), the discharge of industrial waste, fertilizer application, and waste materials disposal can all raise the pH and fluoride levels in groundwater. K⁺ and are both significant constituents in many fertilizers, particularly synthetic fertilizers that contain potassium salts and ammonium bicarbonate. If these fertilizers are carried away from agricultural or landscaped areas into groundwater, the levels of both ions may rise at the same time.

In this study, basic hydro-chemical methods were used to determine whether UP's groundwater quality was suitable for consumption and irrigation. The main conclusion of the study is summarized as follows.

Hydrochemistry indicated the average ionic dominance sequence in the following order: Na⁺ has the highest dominance, followed by Ca²⁺, Mg²⁺, and K⁺, while has the highest dominance among anions, followed by ⁠, Cl⁻, and F⁻.

The WQI has been used to assess the water's quality for drinking and agricultural purposes according to standards set by various agencies. Less than 50% of groundwater samples nearly 45.52% had good drinking water quality, whereas the other samples have rather poor water quality. Due to this, they must be treated carefully before consumption. According to the WQI for irrigation, the majority of samples, nearly 84%, have very good quality water for agriculture purposes. The two most important variables, sodium and salinity hazards, indicate agricultural productivity. Hydro-chemical facies through Chadha, USSL, and Wilcox diagram indicated that most of the waters were of Ca type, with low to medium sodium hazards, and were under the excellent category in this region.

Based on similarities and dissimilarities in water quality variables, the CA result was derived using a dendrogram. Grouping all districts into two clusters (among the most polluted districts, Cluster I contained Mahamaya Nagar, Aligarh, Mathura, Chitrakoot, Jyotiba Phule Nagar, and Agra) and water quality variables into two clusters (because of the higher concentration, only EC was characterized in a separate cluster, while other physicochemical variables were characterized in another). PCA factors are the key source, accounting for 77.92% of the changes or variations. The first component, which accounts for 53.88% of the variance and is cited as the main reason for changes in water quality, consists of EC, TH, Ca²⁺, Mg²⁺, Na⁺, Cl, and SO₄. The second component, which comprises pH and F, has a variance of 13.74%, and the third component consists of K and with 10.31% of the variation.

This study provides important background information on the physicochemical variables, potential sources, and governing variables of groundwater quality and its spatial variation in the studied area. As a result, while using groundwater for irrigation and drinking, the results may be considered by policy-makers and government organizations. This study provides significant insight into the status of groundwater quality and geochemistry at a particular point in time. But for gaining a holistic understanding of these critical aspects requires more serious studies which will facilitate a clear understanding of the factors governing the groundwater quality. Utilizing data from multiple years would enable us to capture not only seasonal fluctuations but also potential long-term trends, thereby providing a more balanced picture of groundwater management.

The authors are highly thankful to the Central Water Commission (CWC) and Central Pollution Control Board (CPCB) of the Government of India, for providing the physicochemical water quality data in the online website ‘India Water Resources Information System’ of groundwater quality of UP to carry out this study.

The first author Ms. Supriya Chaudhary performed planning, data collection & analysis, and writing of the manuscript, the second author Mr. Gurudatta Singh contributed to theanalysis of data, the third author Mr. Deepak Gupta performed the collection and analysis of data fourth author Ms. Suruchi Singh Maunas helped in manuscript writing; corresponding author Prof. Virendra Kumar Mishra conceptualize the research, provided mentoring, correction of manuscript.

No funding was received for conducting this study.

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

The authors declare there is no conflict.