Introduction

Groundwater resources are among most important sources of fresh water. Similar to surface water resources, water quality is an important environmental issue in groundwater worldwide which largely depends on a number of physicochemical parameters (Kumar Tiwari et al. 2017). Characterization of water quality for groundwater resources should be carried out before making any decisions as how to allocate water for different purposes (drinking, irrigation, industry, etc.). In many regions of the world, groundwater chemistry has been utilized as a measure to determine water quality for drinking and irrigation purposes (Edmunds et al. 2002).

Complexity of geological formations and flow domains are widely observed in groundwater resources. Hydrogeological and geochemical studies are, therefore, warranted for groundwater quality assessment and groundwater resource management, particularly in an event when groundwater quality is potentially degraded by non-natural activities (Colak 2012; Tang et al. 2013; Jabbari et al. 2016). Groundwater sampling and monitoring are also required on an ongoing basis to back up hydrogeological studies and to more accurately assess water quality and propose sustainable management strategies.

Heavy metals are carcinogenic compounds that are among the most common environmental pollutants with their occurrence in water and biota reflecting presence of natural or anthropogenic sources (Adaikpoh et al. 2005; Akoto et al. 2008). Heavy metals exist in water in colloidal, particulate and dissolved phases (Adepoju-Bello et al. 2009) with their occurrence in water bodies being either of natural origin (e.g. eroded minerals within sediments, leaching of ore deposits and volcanism extruded products) or of anthropogenic origin (i.e. solid waste disposal, industrial or domestic effluents) (Marcovecchio et al. 2007). Variations in natural and human activities reflect spatial variations of the aquifer and the hydrochemical parameters of the groundwater. Elevated heavy metal concentrations in water lead to environmental problems which make it necessary to center studies around potential water contamination from metals sources. Industrial, agricultural, and solid waste disposal practices are famous examples of human activities playing role in increasing levels of heavy metals in water, soil, and air (Adekunle et al. 2007; Nagajyoti et al. 2010; Jaishankar et al. 2014).

When studying heavy metals, pollution parameters are generally monitored for assessment of water quality and are then compared with allowable heavy metal concentrations recommended by water standards. Several water quality indices and corresponding applications of index methods have been proposed for estimation of water quality (Edet and Offiong 2002; Bhuiyan et al. 2010). Pollution indices are useful tools for water quality executives, environmental managers, and decision makers as measures which have combined influences of all parameters. In recent years, much attention has been given to evaluation of heavy metal pollution in groundwater using the heavy metal pollution index (HPI) (Rajkumar et al. 2015). The spatial study of heavy metals HPI can be helpful in identifying and quantifying trends in water quality (Mishra et al. 2017). Another index is the general metal index (MI) for drinking water, which takes into account possible additive effects of heavy metals on human health (Enaam Abdullah 2013). MI helps to quickly evaluate the overall quality of drinking water (Enaam Abdullah 2013).

Metals indices can be coupled with statistical methods to obtain more robust results when analyzing water quality data. Multivariate statistical techniques, such as correlation matrix and principal component analysis (PCA), are examples of statistical methods that have been broadly used by researchers around the world for the assessment of water quality (Arora and Mehra 2009; Vieira et al. 2012). Application of multivariate statistical techniques facilitates interpretation of complex data matrices for a better understanding of water quality and a variety of environmental factors.

In this study, we are focused on assessing the prevailing water quality condition to identify the pollution status and probable sources of pollutants in the study area. This work is primarily aimed to assess physicochemical parameters of groundwater [i.e. pH, electrical conductivity (EC), temperature, and total dissolved solids (TDS)] and calculate HPI and MI indices. The indices are used as quantitative criteria to pinpoint areas with high degree of heavy metal pollution. Spatial frequencies and distributions of heavy metal concentration are then evaluated by applying multivariate statistical methods. Heavy metals of concern include arsenic (As), chromium (Cr), copper (Cu), nickel (Ni), lead (Pb), and zinc (Zn). The result of this research can help identifying segments of groundwater with suitable quality for drinking water purposes.

Study area

Similar to many other countries in arid and semi-arid regions of the world, groundwater is a major source of water supply for domestic, industrial and agricultural sectors in Iran (Rezaei and Hassani 2017). The area of study of this research is located in north of Isfahan province in Iran which is historically known for low annual precipitations and long droughts. More precisely, this area is located in northern part of the Sanandaj–Sirjan zone and southern part and middle division of central Iran tectonic zone. The study area lies between longitudes 52°15′00″ to 53°15′00″E and latitudes 33°20′05″ to 33°30′05″N, covering an estimated area of approximately 1500 km2 (Fig. 1). Groundwater is an important water resource especially for drinking and agricultural uses in the study area.

Fig. 1
figure 1

Location map of the study area and sampling stations. Green dashed line is used on top right figure to show sampling area extension with respect to jurisdictional boundaries of Isfahan province

Full size image

Geological features and information of the study area are discussed in this section. The main lithology consists of Eocene volcanic and pyroclastic rocks (Fig. 2). In the Oligocene–Miocene, marl, limestone, sandstone, shale and gypsum units are known as Qom Formation. In the Late Miocene, deposition of sandstone, marl, conglomerate and evaporate rocks known as the upper red formation developed in a molasses-type condition of a sedimentary environment as a result of slow positive movement of the basin (Babaahmadi et al. 2010). Natural sources of heavy metals include volcanism, bedrock erosion, and atmospheric transport. The main sources of potential pollution are deemed to be related to activities in mining, agricultural, municipal and also in part to natural sources.

Fig. 2
figure 2

Geological map of the study area (Amidi and Zahedi 1988)

Full size image

Materials and method

As the first step in investigation of heavy metals concentrations, samples were drawn from 35 monitoring wells in the area. Specifications and coordinates of sampling stations are listed in Table 1.

Table 1 Specifications and coordinates of sampling stations in the study area
Full size table

Water samples were collected from stations during April and May 2016. Location of sampling stations is demonstrated in Fig. 1 and Table 1. Physicochemical parameters including temperature, pH, electrical conductivity (EC), and total dissolved solids (TDS) were measured in situ, using a multi-parameter portable meter (HATCH, Germany). Groundwater sample collection procedure included drying, cleaning and sterilizing in polyethylene bottles. Samples were, then, labeled, sealed and transported to the laboratory and preserved in the refrigerator at a temperature of about 4 °C until analysis time. Water samples were analyzed for major cations and anions using standard methods within 48 h after sampling. Quality control and analysis procedures for obtaining accurate data were performed, including probe calibration for the in situ parameters. The concentration of heavy metals (i.e. As, Cr, Cu, Ni, Pb, and Zn) was determined using inductively coupled plasma mass spectrometry (ICP-MS) method at the Lab West Laboratories, Australia. The accuracy and precision of the measurements were checked using duplicate samples.

Indexing approach

In this research, two indices, namely, heavy metal pollution index (HPI) and metal index (MI), are applied as quantitative measures of water quality.

Heavy metal pollution index (HPI)

Heavy metal pollution index (HPI) is defined as a quantity reflecting the composite influence of different dissolved heavy metals (Sirajudeen et al. 2014). HPI is a powerful tool for ranking amalgamated effects of individual heavy metals on the overall water quality and determining suitability of groundwater for human consumption with values below 100 safe for consumption (Rizwan et al. 2011). The HPI is a method of assigning a rating or weighting (Wi) for each selected parameter. The rating is a value between zero and one, reflecting the relative importance of individual quality considerations and defined as inversely proposal to the recommended standard for each heavy metal (Mishra et al. 2017). The rating value can also be defined as inversely proportional to the recommended standard (Si) for each parameter (Mohan et al. 1996; Prasad and Kumari 2008). The highest tolerance value for drinking water (Si) refers to the maximum allowable concentration (MAC) in drinking water in the absence of any alternate water source. The desirable maximum value (Ii) indicates the standard limits for the same parameters in drinking water. The HPI index is calculated as follows (Mohan et al. 1996):

The first step involves computing the relative weight (Wi) of each parameter using Eq. 1.

$${W_i}\alpha \frac{1}{{{\text{MAC}}}} \to {W_i}=\frac{k}{{{\text{MAC}}}}$$
(1)

with k being the constant of proportionality.

In the second step, an individual quality rating (Qi) is computed for each parameter using Eq. 2.

$${Q_i}=\sum\limits_{{i - 1}}^{n} {\frac{{\left| {{M_i} - {I_i}} \right|}}{{{S_i} - {M_i}}}} \times 100$$
(2)

where Mi is the monitored value of heavy metal of ith parameter in the water sample, Ii is the ideal value of the ith parameter, and Si is the standard value of the ith parameter. Third, summing these sub-indices to compute the overall index:

$${\text{HPI}}=\frac{{\sum\nolimits_{{i=1}}^{n} {{W_i}{Q_i}} }}{{\sum\nolimits_{{i=1}}^{n} {{W_i}} }}$$
(3)

where, Qi is the sub-index of the ith parameter, Wi is the unit weight for the ith parameter, and n is the number of parameters. Generally, the critical value is 100 for drinking water.

Metal index (MI)

The metal index (MI) was preliminarily defined by Tamasi and Cini (2004). This index can be expressed by Eq. 4, where MI is the metal index, C is the concentration of each element in the solution, MAC is the maximum allowed concentration of each element, and the subscript i indicating the ith sample.

$${\text{MI}}=\sum\limits_{{i=1}}^{n} {\frac{{{C_i}}}{{{{({\text{MAC}})}_i}}}}$$
(4)

The higher the concentration of a metal compared to its respective MAC value, the worse the quality of the water. MI value > 1 is a threshold of warning (Bakan et al. 2010).

Statistical analyses

Multivariate statistical analyses are applied to identify the relationship between groundwater parameters in the study area. Correlation matrix of heavy metal concentrations and calculated indices was constructed to find relations among parameters. The bivariate correlation analysis method was applied to quantify the relation between hydrogeochemical parameters and heavy metals (As, Cr, Cu, Ni, Pb, and Zn). Also, Ca2+, Mg2+, K+, Na+ and HCO3− and SO42− were included in the matrix to capture their potential effect on redox and the metals mobility.

Spearman’s correlation coefficient was utilized for displaying relationships among variables using SPSS 18.0 statistical package (SPSS Inc. 2017). Because the distribution of ions in groundwater samples is non-parametric, the Spearman’s correlation analysis is appropriate to be considered here. A high correlation analysis [correlation coefficient (r) near + 1 or − 1] indicates a good relationship between two variables (either positively or negatively correlated) and a value around zero is reflective of no relationship. It is should be noted that, if the r value is greater than 0.7, two parameters are considered to be strongly correlated; whereas an r value between 0.5 and 0.7, indicates a moderate correlation at 95% significance level (Guey-Shin et al. 2011).

Principal component analysis (PCA) was next performed to reduce the number of variables with meaningful relations. PCA is one of the multivariate statistical methods that can be used to reduce the complexity of the variable analysis and also a better interpretation of large volumes of information and data (Noori et al. 2010). This method allows us to display most of the original variability in a smaller number of dimensions and has been widely used in geochemical and hydrochemical studies (Razo et al. 2004). In this study, PCA with varimax normalized rotation (VNR) was carried out using SPSS 18.0 (Kaiser 1958).

All of the hydrochemical variables measured consisting of T, TDS, EC, pH, anions and cations, heavy metals (As, Cr, Cu, Ni, Pb and Zn), HPI and a MI were used in this statistical studies.

Discussion

Physiochemical parameters of groundwater samples

The physiochemical parameters and statistical summary for all of the 35 water samples are shown in Table 2.

Table 2 Summary statistics of physicochemical parameters of groundwater samples of the study area
Full size table

The analytical results of physiochemical parameters were compared with the standard guideline values recommended by the World Health Organization (WHO) for drinking and public health purposes (WHO 2004). In the study area, pH values of groundwater samples vary from 7.05 to 8.76 with a mean of 7.78, indicating a neutral to slightly alkaline water (Fig. 3). TDS values range from 230 to 2200 mg/l, with a mean value of 640 mg/l. Groundwater with high TDS is not suitable for both drinking and irrigation purposes. The maximum allowable limit for the TDS in drinking water is 1000 mg/l (WHO 2004) (Fig. 4). EC and TDS are the most important parameters to demarcate salinity hazard and suitability of water for irrigation purposes. Water EC values vary from 480 to 4300 µS/cm with an average of 1280 µS/cm. It should be noted that maximum allowable limit for EC in drinking water is 1500 µS/cm according to WHO 2004 (Fig. 5). Also, water temperature (T) values vary from 17.2 to 30.1 °C with an average of 21.25 °C (Fig. 6).

Fig. 3
figure 3

Spatial distribution of pH values of the groundwater samples in the study area

Full size image
Fig. 4
figure 4

Spatial distribution of TDS values of the groundwater samples in the study area

Full size image
Fig. 5
figure 5

Spatial distribution of EC values of the groundwater samples in the study area

Full size image
Fig. 6
figure 6

Spatial distribution of T ºC values of the groundwater samples in the study area

Full size image

Figures 3, 4, 5 and 6 show the spatial distribution of pH, TDS, EC and T values in the study area. The concentration gradient is calculated using the inverse distance weighting interpolation method.

Hydrogeochemical facies and classification (water types)

The Piper (1944) diagram was used to identify water types in the study area. The chemical composition of groundwater is dependent on the geology, especially lithology, and solution kinetics, flow patterns of the aquifer and anthropogenic activities which take place within the aquifer system. To find the dominant hydrochemical type and characterization of waters in the study area, groundwater samples are plotted onto trilinear Piper’s diagrams based on major ion concentrations in meq/l (Piper 1944). The Piper plot (Fig. 7) shows that groundwater in the study area consists of four water types: Na–Cl, Na–SO4, Na–HCO3 and Ca–HCO3. This suggests that there might be a considerable number of different processes influencing the chemistry of groundwater in the study area and indicating the variable nature of the groundwater chemistry. Also, Fig. 8 shows the distribution of water types in groundwater of the study area. The results show that the majority of samples (17 samples, 49%) belong to the mixed Na–SO4 water type. Ca–HCO3 water type (one sample, 3%): this type indicates mineral dissolution (probably secondary carbonate and silicate minerals) with sufficient recharge from fresh water (Ako et al. 2012; Srinivasamoorthy et al. 2014). In Na–Cl water type (eight samples, 23%), the final stage of evolution of water types from recharge zone to discharge zone (i.e. bicarbonate → sulfate → chloride) can be observed in water samples of the study area at the discharge zone. In addition, water–evaporitic mineral interactions can be other potential reasons for the presence of chloride type in this zone (Todd and Mays 2005; Merkel and Planer-Friedrich 2008). Na–HCO3 (nine samples, 26%) is another water type which can be a result of ion exchange.

Fig. 7
figure 7

Piper diagram illustrating the main hydrochemical features of the groundwater in the study area

Full size image
Fig. 8
figure 8

Distribution of water types in the groundwater of the study area

Full size image

Classification of water

To classify water samples of the study area, the relationship between pH and metal load (defined as As + Cd + Cr + Cu + Fe + Mn + Ni + Pb + Zn in mg/l) was employed (Ficklin et al. 1992; Caboi et al. 1999). Figure 9 shows the relationship between total metal contents (mg/l) and pH of the analyzed samples. The results indicate that the majority of samples (86%, 30 samples) are classified as near neutral–extreme metal (stations 1–20, 22–23, 25–26, 28–31, 34–35), while five samples (14%) show near neutral–high metal (stations 21, 24, 27, 32, 33). The high metal contents of these stations can lead to serious threats for drinking water consumers.

Fig. 9
figure 9

Classification of water samples based on the plot of metal load and pH

Full size image

Heavy metals concentrations

Quality of water in the study area, with regard to the concentration of heavy metals (As, Cu, Cr, Ni, Pb, and Zn) is assessed herein. The concentrations of heavy metals in water samples are listed in Table 3 and are compared with heavy metals MACs from WHO (2004). The variation of heavy metals in the study area water at different locations is also presented in Fig. 10. The average heavy metal concentration in the groundwater samples follows this series (in decreasing order): lead > nickel > zinc > copper > arsenic > chrome (Table 3).

Table 3 Heavy metal concentration (mg/l) in the study area
Full size table
Fig. 10
figure 10

Variation of heavy metals (As, Cu, Cr, Ni, Pb, and Zn) at groundwater samples of the study area

Full size image

Arsenic is a ubiquitous toxicant and a carcinogenic element associated with a wide range of adverse human health effects (WHO 2004; Tseng et al. 2002; Navas-Acien et al. 2006). Weathering and erosion of As-bearing minerals release As into surface water and groundwater resources (Wang and Mulligan 2006a, b; Armienta and Segovia 2008; Naidu and Bhattacharya 2009). The average Arsenic concentration in the study area was 0.023 mg/l, with a range of 0.0017–0.232 (Table 3). Higher concentrations of As at stations 1, 18, 19, 20 and 30 may be attributed to natural processes and anthropogenic sources. Natural processes in the study area consist of evaporation of water due to high temperature and low rate of rain falls in the last decade drought, which enhances a number of salts, trace, and toxic elements. Anthropogenic sources in the study area include the use of arsenical pesticides for protecting crops in the region, mainly. The As concentration at 17 stations is higher than the corresponding WHO values (Fig. 11). Also, transferring of As into the water system during water–acidic volcanic rock interactions are another source of arsenic in groundwater of the study area. The copper content in the water samples ranged from 0.0005 to 0.0709 mg/l with an average value of 0.0301 mg/l (Table 3). The higher level of Cu in some stations is attributed to runoff from farming areas (especially at stations 15 and 17) (Fig. 12). It is noteworthy to state that higher levels of Cu in some stations are due to existence of copper indices in the study area (natural processes). Nickel concentrations ranged from 0.011 to 0.078 mg/l with an average of 0.044 mg/l (Table 3). The Ni concentration at eight stations is higher than the corresponding WHO values (Fig. 13). Zinc concentrations ranged from 0.0005 to 0.035 mg/l with an average of 0.039 mg/l (Table 3). Zinc belongs to a group of trace metals, essential for growth in humans, animals, and plants and is potentially dangerous for the biosphere when present in high concentrations. High Zn concentrations were found at station 4, which can be attributed to the excessive use of zinc sulfate, an important constituent of fertilizers tremendously used in the region. Chromium is present in small quantities in nature. Chromium is extremely irritating and toxic to human body tissue owing to its oxidizing potential and the permeability of biological membranes (Anderson et al. 1996). Chromium levels in the study area ranged from 0.0005 to 0.040 mg/l, with an average of 0.0019 mg/l (Table 3). Lead concentrations ranged from 0.0005 to 0.27 mg/l with an average of 0.056 mg/l (Table 3). It has been observed that sampling stations 23 and 25 have lead levels above the WHO standard of 10 µg/l (Fig. 14). The Pb concentration at two stations is higher than the corresponding WHO values. Lead contamination of the groundwater in the study area can be the result of entry from agricultural run-off containing phosphate fertilizers, and human and animal excreta.

Fig. 11
figure 11

Spatial distribution of As values of groundwater samples of the study area

Full size image
Fig. 12
figure 12

Spatial distribution of Cu values of groundwater samples of the study area

Full size image
Fig. 13
figure 13

Spatial distribution of Ni values of groundwater samples of the study area

Full size image
Fig. 14
figure 14

Spatial distribution of Pb values of groundwater samples of the study area

Full size image

Heavy metals indices

The descriptive summary statistics of heavy metals and MACs are shown in Table 4. The concentration of Cr, Cu, and Zn were below the MAC in drinking water. The concentration of As (11% samples), Ni (66% samples) and Pb (54% samples) are more than the MAC (Table 4).

Table 4 Descriptive summary statistics for analyzed heavy metals in the study area (Adapted from Siegel 2002)
Full size table

For HPI calculation, the concentration limits (i.e. MAC, highest permissive value for drinking water (Si) and the maximum desirable value (Ii) for each heavy metal) were retrieved from the World Health Organization standard (WHO 2004).

The unit weight (Wi) for various water quality parameters is assumed to be inversely proportional to the maximum admissible concentration (MAC) for the corresponding parameter (Table 5).

Table 5 Standard values used for the indices computation
Full size table

\(\sum {{W_i}} {Q_i}=3.7808;\quad \sum {{W_i}} =0.1112\)

Mean HPI for all of the samples is calculated as 3.7808/0.1112 = 34 (Table 6).

Table 6 Mean HPI calculation for groundwater samples of study area
Full size table

Similarly, the HPI values for 35 samples in the study area, As, Cu, Cr, Ni, Pb, and Zn, are calculated and the results are given in Table 7 (Fig. 16). Also, bar diagram of HPI values at sampling stations is shown in Fig. 15.

Table 7 Heavy metal pollution index of study area at various sampling locations and median deviation
Full size table
Fig. 15
figure 15

Bar diagram of HPI values in the study area

Full size image
Fig. 16
figure 16

HPI values of groundwater samples in the study area

Full size image

Maximum value of HPI is 129.75, which is associated to station 20 and is potentially due to leaching of heavy metals from industries, such as fertilizers, pigments etc., located within the study area. This sampling site is adjacent to the Ardestan city (7 km distance). The HPI in this station is higher than the threshold index value of 100, indicating critical contamination with heavy metals. The median deviation for the HPI indices was computed for each sampling station (Table 7). The median deviation on the more negative side indicates a slightly better quality with respect to heavy metals. To apply these HPI indices in the present study, the scales were slightly modified using multiples of the median as a criterion. For this purpose, the values are divided into three classes to demarcate the different levels of contamination as low, medium, and high (Table 8). Using scaled HPI values, 51% of samples were classified as having low pollution levels, 46% with medium and 3% with high pollution levels (Fig. 16). Figure 17 shows the spatial distribution of MI values in the study area based on the inverse distance weighting interpolation method.

Table 8 Classification of groundwater water quality based on modified categories of HPI
Full size table
Fig. 17
figure 17

Spatial distribution of HPI values of groundwater samples in the study area

Full size image

Computed MI values for groundwater samples are presented in Table 9. The mean MI value is 3.88. Water quality classification using MI has been conducted by Lyulko et al. (2001) and Caerio et al. (2005) (Table 10).

Table 9 The mean MI calculation for groundwater samples in the study area
Full size table
Table 10 Water quality classification using MI
Full size table

Water quality classification using MI (Table 11) was carried out (Fig. 18).

Table 11 Results of water quality classification using MI in the study area
Full size table
Fig. 18
figure 18

MI values of groundwater samples in the study area

Full size image

The MI results show that two samples in the study area are significantly contaminated with heavy metals (sample numbers: 1 and 25). These sampling sites are located on Late Eocene (granodiorite, diorite, and gabbro) rock unit. The contamination may be due to the phenomenon of volcanism and bedrock erosion in the study area. Nineteen samples were observed strongly polluted with heavy metals (sample numbers 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 20, 23 and 30). This can be attributed to the phenomenon of volcanism, bedrock erosion and mining activities (often copper). Six samples were moderately polluted with heavy metals (sample numbers 18, 22, 26, 28, 29 and 31). Bedrock erosion in the study area can be a potential reason. Seven samples were slightly polluted with heavy metals (sample numbers 19, 24, 27, 32, 33, 34 and 35). This can be a result of agricultural activities in the region. One sample was clean with respect to heavy metal (sample number 21). This sampling site is near the Ardestan city and is located on a hillside with andesite–basalt rock unit. Figure 19 shows the spatial distribution of MI values in the study area based on the inverse distance weighting interpolation method.

Fig. 19
figure 19

Spatial distribution of MI values of groundwater samples in the study area

Full size image

Statistical analyses

The correlation coefficient matrix for the hydrochemical parameters, anions and cations, and heavy metals are presented in Table 12.

Table 12 Correlation matrix for physicochemical parameters, anions and cations, and heavy metals in the study area (Spearman’s correlation)
Full size table

The highest correlation exists between EC and TDS (r = 0.99) because all of the dissolved components cause increased ionic concentration, as well as increased EC concentration. EC and TDS are moderately related to temperature (r = 0.678 and 0.672, respectively). A strong positive correlation was observed between TDS with Mg2+, Na+, and SO42− (r = 0.710, 0.772, and 0.942, respectively) and also between HCO3− with Na+ (r = 0.667). Also, the strong positive correlation observed between K+ with Ca2+ and Mg2+ (r = 0.818 and 0.759, respectively). This implies that these ions were derived from the common source. Ca2+ and Mg2+ presented a strong positive correlation (r = 0.855), indicating a common source. The pH was negatively correlated with all metals with the exception of lead. The positive correlation between pH and Pb in this research is remarkable and consistent with the redox potential of the metal relative to hydrogen. The correlation relations show that there is a significant positive relation between nickel and copper (r = 0.757, p < 0.01), and also between nickel with lead and MI (r = 0.695 and 0.714, respectively). The correlation studies show that the relation between copper and nickel with arsenic is negative (r = − 0.550 and r = − 0.506, respectively) and indicate one vaiable increases as the other decreases, and vice versa (Figs. 11, 12, 13). The correlation between HPI and MI is significant (r = 0.850) and the results show the similar trends at various sampling locations. MI shows high positive correlation with Ni (r = 0.714) and Pb (0.758). HPI has moderate negative relation with Zn (r = -0.521).

The rotated factor pattern after extraction using varimax rotation is given in Table 13. According to the results of the initial eigen values, five principal components extracted by scree plot explained 85.3% of the total variance (Fig. 20), which is quite good and can be relied upon to identify the main sources of variation in the hydrochemistry. Principal component analysis of different parameters is presented in Table 13.

Table 13 Principal component analysis (varimax with Kaiser normalization) for physicochemical parameters, heavy metals, HPI and MI in the study area
Full size table
Fig. 20
figure 20

PCA scree plot of the eigenvalues in the study area

Full size image

The first component (PC1), with 41.93% of the total variance, indicates that EC and TDS have high loading and reflecting the physicochemical characteristics of water quality while Temperature and Cr have moderate loading. Also, there are high positive loadings of Ca2+, Mg2+, Na+, SO42− and Cl−, and probably shows the result of mineral water reactions in the study area.

The second component (PC2), with 19.39% of the total variance, contains high loadings for HCO3−, Cu and Ni. PC1 and PC2 are assumed to be indicative of the natural processes and water–rock interaction.

The third component (PC3), with 10.1% of the total variance, indicates that K+ and As have high loading while pH and HPI have moderate loading and can be ascribed to the natural hydrogeochemical evolution of groundwater by groundwater–geological interaction. Also, it reflects that the source of As in the groundwater is the weathering process of minerals and anthropogenic sources (arsenical pesticides for protecting crops).

The fourth component (PC4) which contributes 7.8% of the total variance, indicates that Pb has high loading while MI has moderate loading and can be the result of entry from agricultural run-off.

The five component (PC5) contributes 6.1% of the total variance and is only associated with Zn. It can be attributed to the excessive use of zinc sulfate, which is an important constituent of fertilizers used in the region.

Conclusion

In this research, water quality and hydrogeochemistry characteristics of groundwater in north Isfahan of Iran have been investigated through analyzing samples from 35 stations. The quality of water in the study area, with regard to heavy metals (As, Cu, Cr, Ni, Pb, and Zn) has also been assessed. The Piper diagram showed that the groundwater in the study area has four types: Na–Cl, Na–SO4, Na–HCO3 and Ca– HCO3. This result suggests that there might be a considerable number of different processes influencing the chemistry of groundwater in the study area and indicating the variable nature of the groundwater chemistry. A maximum HPI value of 129.75 was obtained for a sampling site adjacent to Ardestan City. The HPI in this location is higher than the critical index value of 100, indicating critical contamination with heavy metals. Using HPI index metric, 51% of samples were classified as having low pollution levels, 46% medium pollution levels and 3% high pollution levels. Also, the MI results showed that two samples in the study area are significantly contaminated with a heavy metal.

Findings of the correlation analysis and PCA method show a strong impact of anthropogenic and agricultural sources for pollution load in the study area. Results indicate that there are water samples in the study area with the best quality for irrigation and drinking purposes in the future (e.g., stations 21, 22, 24, 26, 27, 31, 32, 33, 34 and 35). Using the methodology proposed in this work, one could perform similar analyses and nominate groundwater wells with suitable quality for different beneficial uses (drinking, industrial, agricultural, etc.).