Overview of Comprehensive statistical study of groundwater quality parameter
Table 1 illustrates the statistical analysis of the 18 groundwater samples obtained from May 2021 to April 2022 within the vicinity of the open dumpsite. Table 1 displays the maximum, mean, and lowest values of the groundwater parameters. This study included a total of 16 factors to assess the level of pollution in the surrounding area of the landfill site. Table 1 presents the pH values ranging from 6.8 to 7.8, with a mean value of 7.3, suggesting an alkaline nature. The alkaline groundwater at the Saduperi dumping site is attributed to the elevated pH resulting from the presence of heavy metals, as reported by (El Fadili et al. 2022). The electrical conductivity (EC) of the studied area exhibited significant variation, ranging from 1210 µS/cm-8420 µS/cm, with a mean value of 3750 µS/cm. Similarly, the total dissolved solids (TDS) had a range of 640 mg/L to 4610 mg/L, with a mean value of 1982 mg/L. The majority of sample wells within the study area are above the allowable thresholds as stipulated by the (BIS 2012; WHO 2022). The aforementioned observation suggests the existence of salt concentration, which may be used to infer the direction of water circulation and the penetration of ions (Masood et al. 2022).
Table 1
Statistics data of groundwater physicochemical parameters in the study area
S.No | Parameters | Minimum | Mean | Maximum | BIS Standards | WHO Standards |
1. | pH | 6.8 | 7.3 | 7.8 | 6.5–8.5 | 6.5–8.5 |
2. | EC (µS/cm) | 1210 | 3705 | 8420 | - | 400–2000 |
3. | TDS (mg/L) | 640 | 1982 | 4610 | 500–2000 | 500–1000 |
4. | F− (mg/L) | 0 | 1.16 | 2.47 | 0.5-1.0 | 0.5–1.5 |
5. | Cl− (mg/L) | 130 | 647 | 1849 | 250–1000 | 250–600 |
6. | SO42− (mg/L) | 5 | 156 | 259 | 200–400 | 25–250 |
7. | CO₃²⁻ (mg/L) | 0 | 73 | 280 | - | - |
8. | HCO3− (mg/L) | 80 | 406 | 650 | 200–600 | - |
9. | NO3− (mg/L) | 0 | 19 | 106 | 45-No relaxation | 45-No relaxation |
10. | Na+ (mg/L) | 50 | 256 | 601 | 20–200 | 20–200 |
11. | K+ (mg/L) | < 1 | 23 | 257 | 12–50 | 10–12 |
12. | Ca2+ (mg/L) | 17 | 177 | 601 | 75–200 | 100–200 |
13. | Mg2+ (mg/L) | 35 | 124 | 413 | 30–100 | 30–50 |
14. | Ni (mg/L) | < 1 | 0.04 | 0.19 | 0.02-No relaxation | 0.07-No relaxation |
15. | Ca (mg/L) | < 1 | 0.005 | 0.06 | 0.003-No relaxation | 0.003-No relaxation |
16. | Cr (mg/L) | 0.01 | 0.13 | 1.10 | 0.05-No relaxation | 0.05-No relaxation |
The presence of cations including Na+, K+, Ca2+, and Mg2+ as well as anions such as HCO3−, Cl−, SO42−, CO3−, F− and NO3− were identified. In general, the presence of cations followed the sequence Na+> Mg2+ > Ca2+>K+ being the most prevalent. The concentration of Na+ ranged from 50 to 601 mg/L, with a mean value of 256 mg/L, which is above the established guidelines set by both (BIS 2012; WHO 2022). Elevated levels of Na+ concentration result in the development of cardiovascular and circulatory diseases as well as renal disorders (Siddiqi et al. 2022). In a similar vein, the levels of Mg2+ and Ca2+ exhibit a range of 35 to 413 mg/L and 17 to 601 mg/L, respectively. The mean concentration is 124 mg/L and 177 mg/L, which exceeds the allowable limit as stated by the (BIS 2012; WHO 2022). Both Mg2+ and Ca2+ are known to contribute to the phenomenon of water hardness. It has been shown that prolonged consumption of groundwater with elevated concentrations of Ca2+ and Mg2+ may lead to adverse effects such as infertility and impaired development (Fatoba et al. 2017). The K+ exhibits relatively lower levels of prevalence in comparison to other cations, ranging from less than 1 to 257 mg/L. The average concentration of K+ is measured at 23 mg/L, which is above the acceptable threshold as defined by the (BIS 2012; WHO 2022).
The anions were seen to exhibit the following sequence: Cl− followed by HCO3−, F−, SO42−, NO3 and CO3−. The Cl− serves as the prevailing anion, exhibiting a concentration range of 130 to 1849 mg/L and an average value of 647 mg/L. Elevated concentrations of Cl− are a diagnostic tool and an indication of environmental degradation in the context of polluted groundwater. The disposal of various substances such as domestic wastewater, livestock waste, inorganic fertilisers, animal feeds, paper, plastic garbage, and septic tank effluents in landfills has been identified as a significant factor in the anthropogenic contamination of chloride (Mor et al. 2006). HCO3− is a common chemical that occurs naturally and functions as an indicator of water of superior quality (Kumar et al. 2013). The concentrations of bicarbonate exhibit a range of 80 to 650 mg/L, with a mean value of 406 mg/L. The study identified the minimum and maximum concentrations of fluoride in several well sites are 0 and 2.47 mg/l respectively. Several studies (Adimalla and Wu 2019; Adimalla et al. 2019) have shown that the excessive presence of fluoride may lead to various adverse health effects, such as discoloured teeth, dental and skeletal fluorosis, deformities in the back, hands, and legs, as well as visual impairment (Ayoob and Gupta 2006). The sulphate content exhibits a range of 5 to 259 mg/L, with an average value of 156 mg/L. According to the (CPCB 2007), elevated levels of sulphate in groundwater have been shown to induce physiological effects in humans such as purgation and gastrointestinal discomfort. Nitrate is a crucial ion for the developmental processes of plants. According to (Kumar et al. 2013), the excessive use of nitrate fertilisers has the potential to enter groundwater and result in contamination. According to a study conducted by (Kumari et al. 2019), it was noted that anthropogenic sources, including household sewage, animal or human waste, agricultural runoff, and leachate from unlined landfills, contribute significantly to the elevated levels of nitrate. The concentration of NO3− spans from 0 to 106 mg/L, with an average value of 19 mg/L. The concentration of CO3− ranges from 0 to 280 mg/L, with an average of 73 mg/L, indicating the presence of a geological characteristic in the research area.
The groundwater sample exhibits the presence of elevated quantities of heavy metals, including nickel, cadmium, and chromium. The acceptable threshold for nickel concentration in potable water needs to be below 0.02 mg/L. The majority of samples have concentrations that are above the permissible limit. It was observed that the nickel has a high concentration of 0.19 mg/L. The standard threshold for cadmium concentration is 0.003 mg/L. The highest observed concentration was determined to be 0.06 mg/L, while the average concentration was calculated to be 0.005 mg/L. The allowed limit of chromium in drinking water is 0.05mg/L. The greatest value recorded is 1.10 mg/L, while the smallest value is 0.01mg/L. These measurements indicate the influence of leachate pollution in the groundwater around the dump site.
Hydro geochemical faces
Piper diagrams (Piper 1944) are used to determine the various types of hydrogeochemical characteristics of groundwater. Hydrogeochemical facies analysis is adequate to identify the kind of water. It exhibits the results of chemical interactions between groundwater and the lithologic framework of minerals (Sajil Kumar et al. 2013). The nature of water varies significantly between the southwest monsoon, the northeast monsoon, winter and summer seasons as shown in Fig. 2. This diagram demonstrates the four categories into which the groundwater samples may be categorized. Type 1 represents the natural water quality, which is not affected by any anthropogenic impacts, and it is represented by Ca- HCO3. Type 2 is mixed Ca-Mg-Cl, and it is created by the combination of polluted acidic water with natural water. Type 3 is Na- Cl- SO4, which displays the anthropogenic impact. Mixed Ca-Na-HCO3 represents type 4 and it highlights the effects of weathering and the ion exchange between groundwater-derived Ca and rock-derived Na. In the southwest monsoon, about 55% of the samples fall in type 3 indicating the contamination of groundwater due to the leachate migration of open dumpsite. The northeast monsoon indicates the mixed types of 2, 3, and 4. This shows that sample degradation would have occurred due to anthropogenic and rock-water interaction. In winter, the water type represents type 3 which is due to anthropogenic activity i.e., the influence of leachate from the dumpsite. In summer about 50% of the samples were of the Ca-HCO3 type, indicating that geogenic forces influenced the ions of natural origin.
Mechanism governing groundwater chemistry
The facts of the research region show that both natural and artificial activities have an impact on groundwater. Plotting a Gibbs diagram (Gibbs 1970) will provide insight view of the three main sources of solute in groundwater: precipitation, rock-water interaction, and evaporation. As shown in Fig. 3a during southwest Monsoon few of the sample location falls on the rock dominance indicating the rock water interaction on the location. It was also observed that many samples fall outside the Gibbs field, indicating the possible input from anthropogenic sources. In the northeast monsoon it was found that a sign of the dominance of rock water interaction and human activities such as dumping of waste in landfills, industrial pollution, etc. leads to a high concentration of Na-Cl in groundwater are shown in Fig. 3b. During the winter season, few samples fall in the rock dominance and more samples fall outside of Gibbs field due to the anthropogenic sources and are shown in Fig. 3c. In the summer season, about 30% of the sample falls under rock water dominance and the remaining falls outside the Gibbs field indicating the anthropogenic activity of leachate contamination from the dumpsite shown in Fig. 3d.
Pollution indices
Leachate Pollution Index (LPI)
The assessment of leachate pollution potential originating from landfills has considerable significance and is determined using a quantitative technique referred to as the Leachate Pollution Index (LPI) (Kumar and Alappat 2005a). In this research, based on a theoretical range of 5 to 100, the LPI values were utilised to evaluate the polluting potential of the leachate. The leachate facilitates the transportation of a substantial pollution burden, mostly consisting of heavy metals, organic substances, and a substantial population of harmful bacteria. The LPI calculation was also conducted under the norms outlined in the Municipal Solid Waste Management and Handling Rules of 2016. The permissible limit of LPI value was assigned as 7.88 and 6.44 for surface and subsurface water disposal. The LPI values were found to be 42.5, 35.65, 40.2 and 37.5 in the Summer, SWM, NEM and Winter were shown in Table 2 respectively. It was observed the LPI value was found to be higher during summer than other seasons indicating the higher concentration of leachate. On the other hand, dissolution of leachate due to precipitation indicates a comparatively lesser value during other seasons. The computed LPI values indicate that the toxicity of leachate reaches its peak during all the seasons. The calculated LPI > 35 in all seasons indicates the prevailing poor environmental condition. These LPI values indicate the groundwater is highly contaminated around the dump site. The groundwater is used as the main source for domestic purposes around the dumpsite hence it leads to several health issues. It was found the leachate originating from the Saduperi dump site has a substantial contamination potential, as shown by its LPI values. Hence, it is highly recommended that the leachate be subjected to treatment and meticulous monitoring in order to prevent any adverse effects on the environment and to promptly initiate necessary corrective measures.
Table 2
Leachate Pollution Index (LPI) values of Saduperi dumpsite
S.No | Season | LPI value |
1. | Summer | 42.5 |
2. | SWM | 35.65 |
3. | NEM | 40.2 |
4. | Winter | 37.5 |
Heavy Metal Pollution Index (HPI)
The HPI is a robust method used to evaluate water quality by analysing the levels of heavy metals. The HPI was calculated according to the established international standard set by (Edet and Offiong 2002). The resulting index values are denoted as HPI. In this study, the parameters such as Cd, Cr, and Ni are considered for the calculation of HPI. The study examines the mean concentrations of heavy metals throughout the year. This methodology allows for the assessment of water quality at each designated sample site, facilitating the comparison of the indices associated with each individual specimen. The HPI has been calculated for four seasons Summer, SWM, NEM and Winter were shown in Table 3. It was observed that about 56% of the sampling site was affected by heavy metal concentration. The HPI value was found to be more than the critical value of 100 in the wells such as S1, S5, S6, S7, S10, S11, S12, S13, S14 and S18 for all seasons. HPI ranges from 184.51 to 677.80, 138.29 to 637.59, 97.07 to 683.17, 93.78 to 590.73 during the summer, SWM, NEM and Winter respectively. Based on the findings of the HPI, it is evident that all water samples, with the exception of samples such as S2, S3, S4, S8, S9, S15, S16, and S17 obtained from the monitoring wells, exhibited considerably higher values compared to the critical threshold. This critical threshold represents the point at which the level of water pollution becomes deemed unacceptable. If appropriate measures are not implemented to mitigate the accumulation of heavy metals within a given area, it is plausible that the levels of these contaminants might increase in subsequent periods. The fluctuation of HPI and its subsequent elevation of groundwater level throughout the rainy season serves as an indicator of the influence of water sources on the quality of water in the given geographical area. Moreover, the observed elevations in HPI at the sites of observation wells might potentially be attributed to the presence of elevated amounts of Cd and Cr. Due to the relatively low-weight units assigned to other metal Ni, its impact on the evaluation of the groundwater's HPI surrounding the dump site was minimal. However, Cd and Cr exhibited large weight units and made a substantial contribution to the overall assessment. Heavy metal concentration in the sampling site is mainly due to the mitigation of leachate from the Saduperi dumpsite. Hence it is imperative to monitor the heavy metal concentration.
Table 3
Heavy metal Pollution Index (HPI) values of Saduperi dumpsite
S.No | Well Number | HPI value |
Summer | SWM | NEM | Winter |
1. | S2 | 683.17 | 637.56 | 677.80 | 590.73 |
2. | S3 | 234.27 | 192.68 | 152.07 | 146.95 |
3. | S4 | 184.51 | 138.29 | 97.07 | 93.78 |
4. | S5 | 490.68 | 361.71 | 405.02 | 447.29 |
5. | S6 | 490.17 | 447.29 | 404.85 | 404.10 |
6. | S7 | 578.54 | 445.24 | 489.76 | 534.27 |
7. | S8 | 671.73 | 627.56 | 585.98 | 585.78 |
8. | S9 | 538.90 | 495.24 | 495.12 | 453.00 |
9. | S10 | 450.02 | 406.95 | 491.85 | 449.98 |
10. | S13 | 450.73 | 406.95 | 492.71 | 493.12 |
Partial least squares‑structural equation modelling (PLS‑SEM)
This work uses partial least square modelling as an alternative to the current covariance-based techniques such as LISREL (linear structural relations) and AMOS (analysis of moment structures). This study uses PLS-SEM due to its suitability for conducting both exploratory and confirmatory research. The two approaches used in SEM are covariance-based and Partial Least Squares-SEM (PLS-SEM). Unlike SEM, which is most suitable for hypothesis generation, PLS is often used for hypothesis validation. The PLS-SEM based approach has two components: weighing and measuring. A model that incorporates many orders and numerous variables is most suitable for PLS-SEM. PLS-SEM may get equivalent benefits from doing minimum data analysis. Performing parameter calculations with PLS-SEM is straightforward. This study uses the SmartPLS 4.0 version for the computation of the models. All the output models generated by SmartPLS were similar to the one shown in Fig. 4, serving as an illustrative example. The measurable variables (MV) are shown as yellow rectangles, while the latent variables (LVs) are represented as blue circles. The R2 value is shown inside the LVs. Each LV was constituted by one or more MVs. In this particular study, four LVs are considered such as “IOT Parameters”, “Leachate Parameters”, “Heavy Metal” and “Groundwater Quality”. LV “IOT Parameters” was established, consisting of four MV’s such as “Leachate Level”, “Moisture Content”, “Soil EC”, and “Soil Temperature”. The LV “Leachate Parameters” consisted of eight MV parameters such as “BOD”, “COD”, “Chloride”, “Chromium”, “Iron”, “Lead”, “pH-L”, and “TDS-L”. This LV is influenced by “IOT Parameters” as it has a connection with the detection and prediction of leachate from the Saduperi dumpsite. The LV “Heavy Metal” was composed of three MVs such as “Cd”, “Cr”, and “Ni” which have an effect on the LV “Leachate Parameter” as these leachate parameters are responsible for the contamination of heavy metals in the groundwater. The “Groundwater Quality” LV has combined impacts of other LVs such as “Leachate Parameters” and “Heavy Metal” which solely depend on the contamination of groundwater. The LV “Groundwater Quality” consist of eleven MVs pretraining the different kinds of parameters that are affecting the groundwater quality: “Alkalinity”, “Ca”, “Cl”, “F”, “K”, “Mg”, “Na”, “NO3”, “SO4”, “TDS”, “pH”.
The PLS-SEM model has been framed for all the sampling wells here four sampling wells such as S4, S5, S9 and S18 have been discussed in detail in the following section is shown in Fig. 5 as a yellow dot. The remaining sampling wells PLS-SEM models are referred to in the appendix.
PLS-SEM model for sampling well
The PLS-SEM model was partitioned into two sub-models, namely the inner and outer sub-models. The equations for the measured scores of each LV were computed using Equations (1) to (4) for the categories “IOT Parameters”, “Leachate Parameters”, “Heavy Metal” and “Groundwater Quality”. An equation representing the influence of MV on four LVs for the outer models for sample location S4 which is at a distance of 2 km from the Saduperi dumpsite which is shown in Fig. 6 is given below.
IOT Parameters Measured score = Leachate Level * (0.276) + Moisture Content * (0.305) + Soil EC * (0.262) + Soil Temperature * (− 0.299); ------ (3)
Leachate Parameters Measured score = BOD * (− 0.047) + COD * (− 0.023) + Chloride * (0.185) + Chromium * (0.157) + Iron * (0.180) + Lead * (0.183) + pH-L * (0.185) + TDS-L * (0.183); ------ (4)
Heavy Metal Measured score = Cd * (0.385) + Cr * (− 0.702) + Ni * (0.385); ----- (5)
Groundwater Quality Measured score = Alkalinity * (− 0.094) + Ca * (0.140) + Cl * (0.157) + F * (-0.053) + K * (0.143) + Mg * (0.125) + Na * (0.141) + NO3 * (0.128) + SO4 * (0.138) + TDS * (0.115) + pH * (0.092); ------ (6)
The inner model was composed of relations between latent variables, which are expressed by the equations (7) to (9) respectively;
Leachate Parameters Predicted score = IOT Parameters Measured score * (0.547); ------ (7)
Heavy Metal Predicted score = Leachate Parameters Measured score * (− 0.268); ----- (8)
Groundwater Quality Predicted score = Leachate Parameters Measured score * (0.760) + Heavy Metal Measured score * (0.396); ------ (9)
In order to comprehend the example model, it is necessary to analyse the weights and path coefficients concurrently, which may be seen in equations (7) to (9). It was observed that the positive value of the path coefficient has a higher impact and the negative value represents a lower impact on the LV. Based on these equations the measured value for “Leachate Parameters” is positive (0.547) which indicates that the “IOT Parameters” are responsible for the increase in contamination of the “Leachate Parameters”. The “Heavy Metal” is not much influenced by the leachate parameter as it has a negative coefficient (-0.268). On the other hand, the groundwater quality is influenced by both LV “Leachate Parameters” and “Heavy Metal” since it has a positive coefficient of 0.760 and 0.396 respectively. The R2 value of LV “Leachate Parameters” and “Heavy Metal” are 30%, 7.2% has a higher influence on the R2 value of LV “Groundwater Quality” which is represented as 50.5%. This indicates that the groundwater quality in the sampling well S4 is affected by both heavy metal and leachate as it is in the direction of the flow of groundwater.
The sampling location S5 shown in Fig. 7 is considered for the PLS-SEM model interpretation. The equations for the LVs are given below;
Heavy Metal Measured score = Cd * (− 0.244) + Cr * (0.452) + Ni * (0.544); ----- (10)
Groundwater Quality Measured score = Alkalinity * (− 0.176) + Ca * (0.183) + Cl * (0.205) + F * (-0.016) + K * (0.177) + Mg * (0.141) + Na * (0.192) + NO3 * (− 0.127) + SO4 * (− 0.015) + TDS * (0.060) + pH * (− 0.024); ------ (11)
The equation for inner model LVs is given below;
Heavy Metal Predicted score = Leachate Parameters Measured score * (− 0.153); ----- (12)
Groundwater Quality Predicted score = Leachate Parameters Measured score * (0.716) + Heavy Metal Measured score * (− 0.432); ------ (13)
It is observed that Eqs. (12) and (13) shows a negative (− 0.153) of “Leachate Parameters” LV measured value indicating the influence of leachate on the heavy not much. The measured value of LV “Leachate Parameters” in the “Groundwater Quality” LV is a positive value (0.716) which shows the influence indicating that the groundwater quality is been affected by the leachate of the Saduperi dumpsite. The R2 value of LV “Groundwater Quality” was found to be 61.3% indicating that the well is located in the flow direction of groundwater. It was observed the sampling well S4 was found to be more contaminated compared to S5. Since S5 is in close proximity to the dumpsite and it is also along the direction of flow of groundwater.
The sampling location S9 shown in Fig. 8 is considered for the interpretation is at a distance of 150 m behind the Saduperi dumpsite by the following equations;
Heavy Metal Measured score = Cd * (0.5) + Cr * (− 0.210) + Ni * (0.5); ----- (14)
Groundwater Quality Measured score = Alkalinity * (− 0.113) + Ca * (0.135) + Cl * (0.127) + F * (-0.008) + K * (0.179) + Mg * (0.179) + Na * (0.172) + NO3 * (− 0.069) + SO4 * (− 0.095) + TDS * (− 0.172) + pH * (− 0.071); ------ (15)
The inner model of the sampling location S9 is given by the equations (16) and (17) respectively;
Heavy Metal Predicted score = Leachate Parameters Measured score * (0.107); ----- (16)
Groundwater Quality Predicted score = Leachate Parameters Measured score * (0.858) + Heavy Metal Measured score * (− 0.079); ------ (17)
Based on the measured value of equations (16) and (17) the “Leachate Parameters” is positive (0.107) as it shows the leachate from the Saduperi dumpsite shows the contamination of the heavy metals in the groundwater as the sampling location is located adjacent to the Saduperi dumpsite. The “Groundwater Quality” is highly affected by the “Leachate Parameters” which has a positive value of (0.858) rather than the “Heavy Metal” which is negative (− 0.079) as it is not influenced. The R2 value of LV “Groundwater Quality” is 72.8%. this indicates that the groundwater quality is affected by the leachate parameter and it is not affected by the other LVs as the sampling location was located adjacent to the Saduperi dumpsite.
The sampling location S18 which is far from the Saduperi dumpsite at a distance of 4 km shown in Fig. 9 has been considered for the interpretation of the PLS-SEM model. The equation below shows the measured values of LVs;
Heavy Metal Measured score = Cd * (0.396) + Cr * (− 0.631) + Ni * (0.491); ----- (18)
Groundwater Quality Measured score = Alkalinity * (− 0.121) + Ca * (0.220) + Cl * (0.198) + F * (0.033) + K * (0.208) + Mg * (− 0.094) + Na * (0.202) + NO3 * (− 0.139) + SO4 * (− 0.096) + TDS * (− 0.009) + pH * (− 0.068); ------ (19)
The inner model is shown by the following equations;
Heavy Metal Predicted score = Leachate Parameters Measured score * (− 0.134); ----- (20)
Groundwater Quality Predicted score = Leachate Parameters Measured score * (0.730) + Heavy Metal Measured score * (− 0.086); ------ (21)
The R2 value of LV “Groundwater Quality” is 55.6%. as the R2 value is less when compared to the wells on the flow direction indicating that the well is not much affected by the leachate.
Hence analysing 18 sampling wells using the PLS-SEM model shows that the wells located on the Eastern and North Eastern sides of the Saduperi dumpsite show more contamination in the groundwater quality compared to the wells located on the western side of the Saduperi dumpsite. This is due to the present groundwater flow direction being towards the eastern direction of the Saduperi dumpsite as the elevation on the eastern side is less when compared to the western side. It is observed that the “leachate parameter” LV shows a higher influence on groundwater contamination due to the presence of MVs such as “BOD”, “COD”, “Chloride”, “Chromium”, “Iron”, “Lead”, “pH-L”, and “TDS-L”. The heavy metals present in the leachate are responsible for the presence of heavy metal concentration in the sampling wells. These heavy metals are carcinogenic and cause human illness as the groundwater is utilized for domestic purposes.
In general, PLS-SEM offers a realistic approach to statistically analyse the interconnection between driving variables and water quality at the site level size. The major benefit of using the PLS-SEM approach is its capability to concurrently model, estimate, and test intricate ideas using empirical data. In the present research, it introduces a novel theoretical framework together with its components within the environmental setting.