Introduction

Groundwater is considered as a blue gold of vital social and economic importance. Its quality has significant effects on human health and agriculture. The sustainability of groundwater quality and quantity is essential for its domestic and agricultural uses, so that it can be utilized effectively for a long time in an optimal manner without damaging the environment (Singh 2010). As water is a good solvent, it is highly vulnerable to quality degrade action due to geological and anthropogenic sources of pollution and other developmental actions (Tiwari et al. 2017; Das et al. 2010). Therefore, hydrogeochemical processes that impact groundwater quality are a matter of growing interest. Vasanthavigar et al. (2010) considered hydrogeochemical parameters to ascertain groundwater suitability for human needs as well as for irrigation purposes. Li et al. (2016) reported that variations in regional hydrology and water resources are driven by human activities and natural environmental changes. When groundwater comes in contact with dissolved salts derived by weathering of mineral rocks, its salinity increases the concentrations of various cations and anions in it as well. The active management of agricultural landscapes affects the quality of groundwater in aquifers. The use of wide variety of fertilizers, manure, and excessive phosphorus and nitrogen substances during agricultural production causes degradation of groundwater quality (Getahun and Keefer 2016). Groundwater quality in arid to hyper-arid regions occurs due to over-exploitation of groundwater and exposure of groundwater to mineral deposits in aquifers.

Many researchers have carried out groundwater quality assessment in India (Kamra et al. 2002; Singh et al. 2002; Ravindra and Garg 2006; Gautam et al. 2015; Kumar et al. 2015). Certain studies have found that over-exploitation of groundwater results in degradation of its quality and gradual drop of groundwater table up to 30 m for many wells (Chintalapudi et al. 2017). These increase groundwater salinity, which increases the content of fluoride, nitrate, iron, and other heavy metals in the groundwater (Kamra et al. 2002; Singh et al. 2002; Gautam et al. 2015). Once contamination of groundwater occurs from agricultural residues, fertilizers, and over-exploitation, it can persist for decades because groundwater movement in aquifer is very slow. Earlier studies of groundwater quality reveal that its chemistry is governed by typical correlations and interactions among a wide range of physicochemical variables (Praveena et al. 2010; Singh and Mukherjee 2015). Singh and Mukherjee (2015) assessed groundwater geochemistry in the western part of India, whereas Praveena et al. (2010) assessed groundwater quality in unconfined aquifers using numerical and hydro-chemical approaches. Groundwater chemistry is a major criterion for its use for drinking and agricultural needs. Jasrotia et al. (2018) evaluated groundwater quality parameters using geochemical plots and various other hydro-chemical analytical methods in order to assess groundwater geochemistry.

Rajasthan is the largest state in India that stands at very critical juncture due to its alarming decrease in groundwater levels. Because of inadequate surface water potential and meager rainfall, there is an increased dependence on groundwater for meeting almost all types of water requirements. Contamination of groundwater in Rajasthan is prone to increase due to inefficient water pumps and irrigation systems. Excess infiltration of groundwater is due to its low prices and believed abundant availability in aquifers, which are not true for every location (Schmoll et al. 2006).

The present study deals with the Bikaner block of Rajasthan, which belongs to the category of hyper-arid zone because it receives 100–350 mm of precipitation every year. In recent years, the prevailing groundwater situation in the block has become alarming due to incessant falling of water levels in the wells. The number of pumping wells for drinking and irrigation purposes has increased drastically in the block in the last decade. The block is facing an acute shortage of groundwater resources with the stage of groundwater development reaching as high as ~ 147%. Most of the block comprises alluvial aquifer system and is underlain by highly permeable and well-drained coarse sandy to sandy loam soils. The alluvial aquifers become vulnerable to contamination due to their high permeability and shallow characteristics (Singh et al. 2005a, b). Along with this, anthropogenic sources and hydrogeochemical processes may be accountable for contaminating the groundwater. The aquifer chemistry is severely affected by the anthropogenic and natural activities, and so attention is required to focus on these activities for sustainable management of the ground water (Subba Rao et al. 2006). Thus, there is a need to carry out extensive investigation on the issues to assess different processes involved in groundwater contamination in the block (Helena et al. 1999).

Multivariate methods, such as hierarchical cluster analysis (HCA) and principal component analysis (PCA), have been applied for understanding multifaceted data in order to derive a clearer understanding of water quality in an area under consideration (Srinivas et al. 2015). Probable factors that are liable for variations in groundwater quality can be determined effectively by applying the above-said methods. These methods are important tools for formulating suitable policies for real-time and sustainable management of groundwater assets (Singh and Ghosh 2003; Singh et al. 2005a, b; Singh 2008). Bhakar and Singh (2018) assessed environmental impacts of groundwater supply system in a hyper-arid region of India and found that life cycle assessment is also a possible tool for identifying major hot spots for sustainable groundwater management.

In the present study, groundwater samples have been collected during both pre- and post-monsoon periods in the year 2016. The samples were analyzed using charge balance error method, and statistical z-scores were computed. Finally, 14 out of 43 samples were found appropriate for further analysis. Different multivariate statistical techniques were applied to extract information about the similarities or dissimilarities between the sampling sites for the identification of (a) water quality variables responsible for spatial and temporal variations in groundwater quality, (b) hidden factors explaining the structure of the database, and (c) the influence of possible factors (natural and anthropogenic) on groundwater quality parameters.

The major issues addressed in this study are (1) evaluation of the suitability of groundwater in the Bikaner block of Rajasthan, (2) assessment of prevailing association among the contributing parameters and/or groundwater occurrence, and (3) determination of the utility of multivariate statistical methods in order to achieve sustainable groundwater quality management.

Figure 1 depicts the graphical representation of the research work discussed in this paper. It explains the major problem and need of the study, research methodology adopted, and findings of the present study. This study meticulously contributes to the limited number of groundwater quality studies in the Bikaner block, which faces water scarcity, and discusses its applicability to other similar regions. The current study presents a novel approach using multi-criteria decision-making tools and hydrogeochemical plotting to understand the interrelationship among various groundwater quality parameters in a hyper-arid region of western Rajasthan in order to support decision-makers in devising a suitable policy for sustainable groundwater management. This study discusses critical issues of groundwater quality and identifies critical physicochemical parameters regulating the groundwater quality in the block, which will hopefully create mass awareness among the various stakeholders and policymakers as well. Hence, it is expected to yield direct benefits to society and to create awareness among people in a groundwater-dependent ecosystem.

Figure 1
figure 1

Summary of graphical representation of the present study

Full size image

Literature Review

Table 1 depicts the major research work carried out within the knowledge field in which thus study falls in terms of contribution, location, parameters, methodology, and regional aspects. It is observed from the contemporary literature that a groundwater quality assessment in terms of hydrogeochemical parameters is a common approach to understand the hydrology and status of groundwater. The studies reviewed by this study are found close to our research aim, which is toward assessment of groundwater quality in an arid to hyper-arid region. The studies reviewed aim to promote research in groundwater quality, which is rapidly deteriorating due to intensified human activities and fresh water needs (Li et al. 2016). A study by Li et al. (2017) discussed contemporary research on groundwater quality level and its deterioration due to fast economic and anthropogenic activities in an arid region in Western China.

Table 1 Summary of the existing literature on groundwater/surface water quality analysis
Full size table

Important observations made from the literature review are the following:

  • Most of the existing studies found that groundwater quality is deteriorating due to fast economic and anthropogenic growth.

  • In many of the studied regions, groundwater is not drinkable, especially in arid and semiarid regions.

  • The existing studies assessed the major factors of groundwater quality deterioration.

  • The research community believes that both anthropogenic and natural processes affect groundwater quality.

  • The research community argues that groundwater recharge and groundwater contamination are two major problems.

Study Area

Location and Extent

The Bikaner block is located in the northwestern part of the state of Rajasthan, and it has an international border with Pakistan (Fig. 2). It occupies 30,381.75 sq. km and lies between 27°11′ and 29°03′ north latitudes and 71°54′ and 74°12′ east longitudes.

Figure 2
figure 2

Location of the study area

Full size image

Climate and Rainfall

The climate of the Bikaner block ranges from arid in the east to extremely arid/hyper-arid in the west and is characterized by extreme high temperature, erratic rainfall, and high evaporation. Being situated on the western side of the Aravalli hill ranges, the area is characterized as a typical rain shadow region resulting in low precipitation. The normal annual average rainfall of the block is 262.11 mm during the last 108 years (1901–2008). The temperature in winter season is fairly low, and there is a probability of frost occurrence once in three years. Wind speeds during winter season are low with an average of 4.5 km/h, the main direction of wind being NE. High temperatures in the block start from April onwards, whereas May and June are the hottest months of the year.

Physiography and Geomorphology

The general regional slope of the block is from SSE to NNW, and regional elevation above MSL is about 152 m in the western part and about 275 m in the eastern part. The block has no major river. Alluvium, Tertiary, and Paleozoic sandstone are the main water-bearing formations in the block. The depth to water in the block ranges between 5 and 136 m below ground level. Based on historical evolution, slope, erosion and depositional characteristics, size and nature of sediments, drainage system and salinity hazards, the existing landforms of the block can be described as flat and gravelly aggraded older alluvial plains.

Land Use

About 35 percent of the total area of the block is cultivated. However, the area sown varies up to 45 percent depending upon rainfall occurrence in a particular season of the year. The percentage of uncultivable land is about 55 percent depending on annual rainfall characteristics, nature of geomorphology, and non-availability of water sources. The forest and pasture lands account for 2.59 percent and 1.77 percent of the total area of the block, respectively.

Geology

The rock formations in the block are mainly concealed beneath a thick cover of dune sand. However, a very few isolated rocks are exposed to the surface. Based on the available literature and lithology of existing dug wells and tube wells in the region, the rock units in the block exist from Paleozoic to Quaternary periods. A generalized geologic succession of the rock formations occurring in the block along with their lithological characters is given in Table 2.

Table 2 Geological succession in the Bikaner block
Full size table

Materials and Methods

Data Collection and Standardization

Groundwater samples were collected from 43 tube wells located in various villages in the Bikaner block during the pre- and post-monsoon periods in the year 2016. Sampling locations were identified using a global positioning system (GPS) as given in Table 3.

Table 3 GPS coordinates (longitude, latitude) and average mean sea level of villages
Full size table

Groundwater samples were collected in polyethylene bottles. The samples were filtered through 0.45-μm membrane filter and split into two portions: one acidified with concentrated HNO3− for dissolved cation estimation and the other for rest of the estimations. Electrical conductivity, pH, and total dissolved solids were measured in the field using portable meters. The samples were stored at 4 °C in the laboratory, and alkalinity, chloride, phosphate, nitrate, and sulfate were estimated as per the standard methods (Clesceri et al. 1998).

The accuracy of estimations was verified by analyzing standard reference materials. The data thus obtained were first of all tested for charge imbalance error, and only 14 samples were found suitable for further analysis. As far as PCA is concerned, Bartlett’s sphericity test and Kaiser–Meyer–Olkin (KMO) test were conducted to justify the groundwater sample data and their adequacy. The Bartlett’s sphericity test carried out on the correlation matrix shows a calculated χ2 value of 452.7, which is greater than the critical value as specified in the literature, χ2 = 146.6 (with Pearson’s coefficient (r) = 0.05% and degrees of freedom 91°). Thus, PCA can be applied to the data to achieve significant reduction in the dimensionality of the original dataset.

The initial step to carry out PCA and HCA is to standardize the dataset of all physicochemical water quality parameters, of all water samples. The hydro-chemical data were analyzed in the following steps. If xi,…,xp denotes the P variables, with N number of observations, the jth observation of the ith variable is Xij, where i = 1, …, P and j = 1, …, N. If S and Xm denote the standard deviation and mean, respectively, estimated from N observations of the ith variable, then the jth observation of the ith variable is expressed in the standardized forms as:

$$ Z_{ij} = \frac{{X_{ij} - X_{m} }}{{S_{i} }}, $$
(1)

where Zij is the jth value of the standardized variable Zi. The value of Zi is called as z-score (Liu et al. 2003). For PCA and HCA in this study, the data were first standardized by computing z-scores from the non-normal condition and ascertaining the normal distribution of the data. The mean and variance of z-score were found to be ‘zero’ and ‘one,’ respectively, for all variables. Standardization of data enhances the impact of variables having small variance and decreases the impact of variables with large variance. Further, standardization of data nullifies the impact of units of measurement and makes the data dimensionless.

Principal Component Analysis

In this study, PCA was applied to extract significant principal components (PCs) and the values thus obtained were subjected to varimax rotation analysis for generating varifactors (VFs). Principal components are linear combinations of original variables. The aim of PCA is to dimensionally reduce the contribution of the variable of less importance as compared to another variable in the water quality dataset. The variance of data obtained in the analysis helps to identify the most contributing variable in the dataset (Shrestha and Kazama 2007). The PCs so obtained lie along the directions of maximum variance. Three principal components were retained, following the criteria which suggest using all the PCs up to and including the first one after the break, with eigenvalues more than unity and describing percent of the variance in the original dataset (Cattell and Jaspers 1967). If the eigenvalue of a PC is more than unity, then it provides more useful information about the underlying facts and impact of contribution, thus ensuring the reduction in dimensionality (Cattell and Jaspers 1967). Along with obtaining PCs, PCA provides a correlation matrix that describes the relationship of each parameter with one another.

Hierarchical Cluster Analysis

In the present study, HCA was applied to identify the relatively homogenous groups of cases or variables based on their intrinsic properties. In HCA, clusters are generated successively starting with the most alike pairs of variables and generating greater clusters progressively. Cluster analysis is the task of combining a set of variables in a way that the variables in the group known as cluster show much more similarity with one another than those variables in other clusters. The outcomes of HCA help in understanding the data and designating the patterns (Singh et al. 2005a, b). The HCA was applied to the standardized dataset by deploying Ward’s method using Euclidean distance as a degree of resemblance.

Results

Principal Component Analysis

The PCA was carried out to assess water quality by evaluating the chemical associations as described by principal component loadings and using a dataset consisting 14 groundwater samples from Bikaner block. The first three PCs were selected by adopting the correlation criteria as described in Table 4, to represent the prevailing hydrogeochemical phenomenon described in Figure 3, which helped in the formation of the present groundwater chemistry without losing the information of interest. The output of PCA reveals that the eigenvalues of the first three PCs together account for over 87% of the total variability of the combined population for pre-monsoon and over 85% for post-monsoon (Tables 5 and 6, respectively).

Table 4 Correlation criteria
Full size table
Figure 3
figure 3

Scree plots for (a) pre-monsoon data and (b) post-monsoon data

Full size image
Table 5 Total variance explained by the pre-monsoon data
Full size table
Table 6 Total variance explained by the post-monsoon data
Full size table

For the pre-monsoon data, the first principal component (PC1) after varimax rotation accounts for more than 63% of the total variance (Table 7a) and has very high loadings on TDS, Na+, K+, and NO3− and significant/high loadings on EC, Mg++, F−, SO4−, and TH. For the post-monsoon data, the PC1 has very high loadings on EC, TDS, Na+, Mg++, Cl−, and TH and significant/high loadings on Ca++ and SO4− (Table 7b). For the pre-monsoon data, the second principal component (PC2) after varimax rotation, which accounts for more than 16% of the total variance, has very high loading on pH and significant/high loading on Ca++. For the post-monsoon data, PC2 has significant/high loadings on K. For the pre-monsoon data, the third principal component (PC3) after varimax rotation, which accounts for more than 7% of the total variance, has significant/high loadings on HCO3−, whereas for the post-monsoon data, the PC3 has very high loading on CO3− and significant/high loading on HCO3−.

Table 7 Rotated components for (a) pre-monsoon data and (b) post-monsoon data
Full size table

The major contributor of Ca++ seems to be gypsum, dolomite, and limestone, which occur in the sedimentary basin of the area. Water in contact with gypsum can attain higher calcium contents, and solubility of gypsum increases in saline waters. Principal sources of Na+ could be the precipitate of sodium salts impregnating the soil in shallow water tracts, particularly in arid and semiarid regions. Certain clay minerals and zeolites may contribute to the sodium content in groundwater (Karanth 1987). The primary source of carbonate and bicarbonates in the present samples could be due to higher pH values of the samples ranging in between 4.5 and 8.2 and above 8.2. The presence of bicarbonate is indicated when pH is between 4.5 and 8.2 and carbonate if pH above 8.2 (Karanth 1987). The important contributor of sulfate in present samples seems to be gypsum and anhydrite found in the sedimentary rocks of the reason. Local abnormal concentrations of sulfate may be due to traversing of groundwater through lignite, coal, and gypsiferous beds. In sedimentary rocks, dolomite and limestone contain magnesium carbonates that seem to be the major contributor of magnesium in the water samples (Karanth 1987). A solution of halite and other evaporite deposits in the aquifer material are the primary sources of chloride in the present case along with atmospheric sources. Fluoride in the water samples seems to be derived from certain amphiboles, mica, and complex fluoride-bearing silicates (Karanth 1987). By far the greatest contribution of nitrate in groundwater may be from the excessive use of fertilizers and dung of field grazing cattle such as cow, sheep, and goat in the study area. As the area is a hyper-arid zone, mineralization and concentration of ions by evaporation (Jalali 2012; Marghade et al. 2015; Yang et al. 2016; Singh et al. 2017) may result in higher values of TDS in the samples.

Hierarchical Cluster Analysis

On the basis of similar hydrogeochemical features and sources of natural background, HCA effectively reproduced three clusters in pre-monsoon water samples and four clusters in post-monsoon water samples. The details of the produced clusters are given in Table 8. The nearby locations of the water sample sites primarily fall in the same cluster due to the orderly and appropriate involvement of sampling sites in the development of cluster. The clustering of water samples in both pre- and post-monsoon shows that groundwater quality varies smoothly with a few gradual changes over the entire block. This variation may be due to the equally prevailing hydrogeological environment in the area. The dendrograms for both pre- and post-monsoon (Figs. 4 and 5, respectively) show a clear picture of the spatial similarity occurring among the water samples in the area under consideration.

Table 8 Distribution of groundwater sample sites in different cluster
Full size table
Figure 4
figure 4

Dendrogram using Ward’s method for pre-monsoon data

Full size image
Figure 5
figure 5

Dendrogram using Ward’s method for post-monsoon data

Full size image

The branching-type nature of the dendrograms allows to trace backward or forward to any individual case or cluster at any level. In our case, if we examine the dendrogram for pre-monsoon, it is very clear that samples from Bikaner, Malasar, Molaniya, Shivbari, Gersar, Ranisar, Runia Bas, Surdhana, Bambloo, Deshnoke, and Naurangdesar villages are very similar, whereas in another cluster, samples from Himtasar and Jaimalsar villages are very similar. Further, we can extract sub-clusters from the picture shown in Figure 4. From the dendrogram for post-monsoon, it is very clear that samples from Chattargarh and Kavani villages are very similar, whereas in another cluster samples from Tejrasar, Gigasar, Gadwala, Kesardesar Jatan, Bambloo, Molaniya, Deshnoke, Ladera, Deli Talai, and Lalamdesar villages are very similar. Further, we can extract sub-clusters from the picture shown in Figure 5. To develop an effective groundwater quality monitoring system in the block, HCA offers a reliable classification of groundwater by selecting representative wells within a cluster that will help in formulating well-suited policy to spatially monitor groundwater quality. This will also help to significantly reduce number of sampling sites as well as monitoring cost.

Hydrogeochemical Plotting of Water Samples

The Bikaner block’s groundwater is the only source of water for domestic and agricultural purposes; groundwater quality of the block has also been evaluated by the use of geochemical plots, such as Piper diagram (Piper 1944) and Wilcox plot (Wilcox 1955). These graphical representations of groundwater quality data help to visualize the status of the prevailing groundwater quality in the block. A Piper diagram basically helps to make several conclusions about water type, precipitation or solution, mixing, and ion exchange. This diagram developed by Piper (1944) is a combination of anion and cation triangles with an intervening diamond. In the lower left triangle, values of three cations (Ca++, Mg++, Na+) are plotted, whereas in the lower right triangle three anions (HCO3−, SO4−, Cl−) are plotted, and the central diamond is used to show the overall chemical character of water. On the basis of placement of samples near the four corners of the diamond, water can be classified into four basic categories. Water samples that plot near the top of the diamond are high in Ca++ + Mg++ and Cl− + SO4− and are characterized by permanent hardness. Water samples that plot near the left corner are rich in Ca++ + Mg++ and HCO3− and are characterized by temporary hardness. In the lower corner, water samples are basically comprised of alkali carbonates (Na+ + K+ and HCO3– + CO3–), whereas in the right side of the diamond, water samples are categorized as saline water types (Na+ + K+ and Cl− + SO4−).

The prime use of Wilcox plots is to categorize water samples for irrigational use and to classify water quality, so that the seasonal effects on water quality could be visualized. The Wilcox diagram is plotted with EC as abscissa and sodium absorption ratio (SAR) as ordinate. The Wilcox diagram explicitly depicts the two important aspects of water quality named as salinity hazard and sodium (alkali) hazard. The diagram in terms of salinity deduces the water samples as low-salinity (C1), medium-salinity (C2), high-salinity (C3), and very-high-salinity water (C4). In terms of alkali hazard, the water samples can be deduced as low-sodium (S1), medium-sodium (S2), high-sodium (S3), and very-high-sodium water (S4).

According to Wilcox (1955), water sample with SAR value up to 10 and EC up to 250 can be considered as S1 water. If the SAR value lies between 10 and 18 and EC value lies between 250 and 750, then it is considered as S2 water. Further, a SAR value up to 26 and EC value up to 2250 defines S3 water. Water with more than 26 SAR and more than 2600 EC can be defined as an S4 type. The various physicochemical parameters analyzed in this study were plotted in a Wilcox (Fig. 6) and Piper (Fig. 7) diagrams, and the results are tabulated in Table 9. The Wilcox plot clearly shows high salinity hazard in the block and medium-to-high alkali hazard in the majority of the block. Approximately 42% of the samples are under very high salinity hazard, 46% high salinity hazard, and rest moderate salinity hazard. In the same manner, 33% of the samples are under high alkali hazard, 38% medium alkali hazard, and the rest low alkali hazard.

Figure 6
figure 6

Wilcox plots for (a) pre-monsoon data and (b) post-monsoon data water samples

Full size image
Figure 7
figure 7

Piper plots for (a) pre-monsoon data and (b) post-monsoon data water samples

Full size image
Table 9 Water quality status of the study area
Full size table

Considering the Piper diagram, Na+ and Mg++ are the dominant cations, whereas Cl− and SO4− are the dominant anions in the respective left and right triangles of the diagram. This indicates that majority of the water samples falls under the category of saline water. Majority (37.5%) of groundwater samples are Na+–Cl− water type and ~ 33% of the samples represent alkali carbonates and remaining 29% are a mixed type (Table 9) for both pre- and post-monsoon samples.

Discussion

The methodology described in this study can assist the research community to include the hydrogeochemical information from the analysis of semiarid and arid region aquifer systems. The methodology used in this study allows the successful outcomes of each method accommodating all information to generate a robust interpretation by incorporating the strength of various geochemical, statistical, and spatial grouping tools. Therefore, optimal groundwater extraction, integrated groundwater management, and checked use of fertilizer and pesticides for crops are desired to ensure acceptable groundwater quality in the study area. The race toward livelihood resilience can explain why groundwater is overexploited by end users. In the last few decades, several problems have arisen from the evolution of groundwater for irrigation purposes in developing nations of Asia and Africa (Shah 2005). The contamination of groundwater and its quality degradation affect the farmers in terms of crop choice and availability of potable drinking water (Ranjan 2012). This calls for identification of alternate solutions for drinking water in problem areas. In addition, the filtration system commonly used in households of India is not efficient. They discard 75% of water supplied for purification and only 25% is used for drinking purpose (Bhakar et al. 2016). This makes fewer choices available to the end users for satisfying their daily needs of livelihood. Singh (2014) claimed that the conjunctive use of surface water and groundwater can be a better option for sustainable irrigation system. When it comes to hyper-arid regions where groundwater availability is much lower, groundwater quality assessment of aquifer systems can support decision-making for developing sustainable groundwater policies. The use of remote sensing in combination with multivariate statistical tools or artificial intelligence techniques can provide better possibilities to assess and monitor the quality of large samples of water. Such methodology can also work to remove uncertainties in the analysis of data (Swain and Sahoo 2017).

Conclusions

The various physicochemical parameters in the hydro-chemical datasets from the study area were analyzed using multivariate statistical methods to identify contributing variables and spatial similarity between the groundwater samples. By PCA, three significant principal components were extracted explaining 87% of the total variability of the combined population for pre-monsoon data and 85% for post-monsoon data. The major variations in the datasets are due to the solubility of gypsum, a precipitate of sodium salts impregnating the soil, higher pH values, excessive use of fertilizers, dung of field grazing cattle, concentration of ions by evaporation, and weathering of the rocks. On the other hand, HCA played a key role in identifying the spatial similarity between the groundwater samples from the study area. The clusters generated can be utilized in order to select a representative well from spatially similar clusters to achieve effective groundwater quality monitoring in the area under consideration. It can be concluded from the graphical plots of groundwater quality data that groundwater sources in the Bikaner block are affected by alkali hazards and salinity, and therefore are not suitable for drinking and irrigation purposes. The outcomes of the present study can be used by practitioners and policymakers for development and implementation of strict groundwater management policies for sustainable management of this natural resource. The study is limited to a small area for groundwater quality assessment; however, the tools used for the assessment have capabilities to assess large regions/areas as well. In future, time series analysis of data can be carried out to have a wider picture of groundwater quality. Along with multivariate statistical tools, remote sensing technique can also play a vital role in managing groundwater resources. Using remote sensing for assessment of groundwater quality can be a future scope of this study in the Bikaner block.