The Sargodha Division is situated in central Punjab, Pakistan and lies between latitude 31.3°–32.6° N and longitude 71.8°–73.4° E. It is comprised of four districts which are Sargodha, Khushab, Mianwali and Bhakkar. The area is located in the fertile plains of the Punjab basin and is primarily dependent on surface as well as groundwater resources for agricultural, livestock and domestic use. The study area has a semi-arid climate with warm to hot summers (45 °C) and mild winters. The annual mean precipitation varies from 300 to 500 mm and concentrated mainly during the monsoon months. The region is lithological controlled by alluvium deposits of silt, clay, and sand that control groundwater replenishment and quality. The aquifer is of the depth from 40 m to 100 m and is mainly accessed by using hand pumps and tube wells. The Sargodha is an urban mixed rural area and its population is over 8 million. Most live-in hinterland communities on untreated groundwater for daily use. These characteristics make Sargodha a suitable model to evaluate water quality risks in the areas where underground water is the main source of water. Profile of the study area has been depicted in Figure 3.

The groundwater samples were obtained from the Sargodha city, in the Punjab province of Pakistan, at sixteen different points (Figure 3). The sites were selected using a non-probability purposive sampling approach to cover generalized properties with respect to urban situations and population densities, as well as residential and commercial water usage. A total of 30 ground water samples were collected from the depth of 23–67 m. Samples were collected in 1.5 L plastic bottles, after prewashing them with de-ionized water and rinsing with sample water before collecting the samples. After the pumping of hand pump for 2–3 min to stabilize physical parameters, samples were collected. For nitrate measurement, 100 mL of water was collected in the sterilized bottles and 1 mL of boric acid solution by the sterile syringe was injected through it after to stop the reaction further. All samples were analyzed in Pakistan Council of Research in Water Resources.

The pH and total dissolved solids (TDS) of the groundwater samples collected (n = 30) were determined with a glass electrode (pH meter, Adwa AD 111) and an electrical conductivity (EC) meter (Adwa AD 330) both were conducted TDS was calculated. A flame photometer (JENWAY PFP7) was used to determine sodium and potassium concentrations. The sulfate content was determined gravimetrically, and the bicarbonate level was obtained by titration. The argentometric titration method was used for determining chloride concentration. Calcium concentration was determined by the standard EDTA titration method (1992), and magnesium concentration was calculated by deriving magnesium content from the hardness (total hardness calcium) using a standard formula. For the determination of nitrate concentration, groundwater samples were preserved with boric acid and analyzed with the cadmium reduction method (HACH-8171) using a spectrophotometer (HACH, Germany). The pH meter (Adwa AD 111) and electrical conductivity (EC) meter (Adwa AD 330) were supplied by Adwa Instruments, Szeged, Hungary. The flame photometer (JENWAY PFP7) was manufactured by Jenway Staffordshire, United Kingdom. The determination of nitrate was carried out in a spectrophotometer (HACH) model with HACH Company, Düsseldorf, Germany. All reagents of chemical analysis in the present study were of analytical grade and obtained from accredited suppliers. For analytical quality control and data verification, all instruments were calibrated as recommended by the manufacturers before use. Method detection limits (MDLs) for each analyte were calculated from the repeated analysis of low-level standards such that the thresholds of detection were below WHO guideline levels.

Sulfate (SO₄ ^2-) concentration was analyzed by turbidimetric method in APHA standard methods (4500-SO₄ ^2- E). Here, in this method, the sulfate ions in the sample reacts with barium chloride to give a turbidity which is proportional to the sulfate concentration. The turbidity was determined by the UV–Vis spectrophotometry at 420 nm with a UV–Visible spectrophotometer (HACH, DR 3900 TGK, Germany). Calibration curves were obtained using certified sulfate standards (R² > 0.99). The analysis’ method detection limits were 0.5 mg/L and precision was validated by performing duplicate analyses (recoveries 95%–102%). Nitrate were determined by the cadmium reduction method (APHA 4500-NO₃ ^- E) with a UV–Visible spectrophotometer (HACH DR 3900, Germany). In this process the nitrate is reduced to nitrite by passing the sample through a reduction column (copper-cadmium). The nitrite produced reacts with sulfanilamide and N-(1-naphthyl)-ethylenediamine dihydrochloride to produce a pink azo dye, which was measured at 543 nm. Before analysis, the groundwater samples were preserved using 1 mL of boric acid per 100 mL to inhibit the microbial activity, and stored at 4 °C Calibration was made using potassium nitrate standards in the range 0.1–20 mg/L Analytical quality control comprised method blanks, duplicates and spiked recoveries (90%–104%).

Strict quality control procedures were performed to guarantee the quality of the results. Daily calibration of all equipment was performed according to manufacturer instructions. CRM (High Purity Standards, United States) was analyzed together with the samples for further validation of accuracy, the recovery percentage efficiency for all major ions was observed to be 95%–103%. Calibration curves of the parameters were generated by standard solutions of five concentration levels, and the correlation coefficients (R²) were more than 0.99. The Limit of Detection (LOD) was 0.01 mg/L for nitrate, 0.5 mg/L for sulfate, and below WHO guideline thresholds for other ions. Blank measurements were performed using ultrapure deionized water (Milli-Q system, Millipore, United States), and no significant contamination was detected. Replicate analyses (n = 3) of randomly selected samples showed a relative standard deviation (RSD) below 5%. Spiked recoveries were within 96%–104%, confirming the robustness of the analytical procedures. These validation parameters demonstrate that the reported values are accurate, reproducible, and suitable for hydrogeochemical assessment.

The Water Quality Index (WQI) in this study was calculated using the weighted arithmetic index method (Brown et al., 1972), which is a refinement of the original approach by Horton (1965). While Horton’s method provided the foundational concept, the present study applies updated permissible limits and parameter weights following WHO standards. The classification of water quality was adopted from Brown et al. (1972), adapted to our dataset to define categories such as excellent, good, fair, poor, very poor, and unfit for drinking. Following many studies done after 2017, the WQI methodology is one that has been successfully performed to study and interpret water quality problems (Kumar et al., 2019). Approaches using WQI are well-known and well-accepted methods of determining water quality in the scientific literature, including (Patel et al., 2023), as examples. Which paves the way for an in-depth analysis of various physical, chemical, and biological parameters required to assess the groundwater quality in the Sargodha Division. Studies that highlight the versatility and use of WQI in assessing water quality especially (Gupta et al., 2021). Therefore, WQI is suitable for our study. We will use the WQI methodology to generate a numerical water quality quantifier which can be used for evidence-based decision-making. This will help prioritize actions and resources needed to ensure the sustainable management of water resources and safeguarding of public health in Sargodha Division (Krishan et al., 2023). The water quality index is computed by the following expression as shown in Equations 1–3. W Q I = ∑ Q i × W i / ∑ W i (1) Were, WQI = Water Quality Index.

Qi = Quality rating for the ith parameter.

Wi = Unit weight for the ith parameter.

The quality rating (Qi) is calculated using; Q i = [ ( V i − V o S i − V o ] × 100 (2) Were, Vi = Observed concentration of parameter i (e.g., mg/L)

Vo = Ideal value (0 for all except pH = 7).

Si = Standard permissible value of parameter i (e.g., mg/L).

The unit weight (Wi) is determined by; W i = K / S i (3)

Where, K = Proportionality constant.

Si = WHO standard permissible value for parameter i (mg/L or unitless for pH) K = 1 / ∑ V s (4)

Calculating the WQI of groundwater parameters comprises multiple steps apart from the estimation of quality rating of each parameter using a general formula. A quality score of 0 means there are no pollutants and a quality score in the 0 to 100 range means the level of pollutants is within acceptable standards. But if the score is greater than 100, it means that pollutants are above the acceptable limits (Gungoa, 2016). To calculate the unit weight for all physicochemical parameters using specific formula is another task to do in the following step. These unitary weights are used to normalize parameters that exist in different dimensions and scales on a common scale (Bora and Goswami, 2017).

Figure 4, shows the presence of two principal water types; (Ca–Mg–HCO₃) bicarbonate–alkaline earth and (Na–Cl) These Ca–Mg–HCO₃ waters are characteristic of young recharge in contact with carbonate lithologies (dissolution of calcite/dolomite) and are only slightly anthropogenically modified. Samples plotting near the Na–Cl corner are indicative of higher salinity and mineralization, typical of evaporite dissolution, higher residence times, cation exchange (Na+from clays being released and Ca ²+/Mg ²+ being taken up), and anthropogenic impact from irrigation return flow and domestic wastewater. Combined positions indicate evolutionary transition between these.

The Durov projection supports the Piper facies and emphasizes ion-exchange contributions in samples trended towards higher Na⁺ for the same alkalinity, suggesting the exchange of Ca²⁺/Mg²⁺ for Na⁺ on aquifer surfaces. The two aforementioned points as well as the other points that tend towards the Cl⁻–SO₄ ^2- end, combined with high TDS, are indicative of contributions from fertilizer residues, sewer leakage or evaporative concentration in irrigated tracts. See Figure 5, the clustering suggests spatially varying impact by the agricultural activity and local wastewater pollution on top of a lithological control.

Figure 6 indicates that the majority of samples are characterized in the rock water interaction dominance field, supporting that carbonate/silicate weathering governs base flow chemistry. A minor part migrates towards the evaporation–crystallization field, where they are explained by semi-arid conditions, shallow water tables in the impact of irrigation and solute accumulation by return flow. The trend towards higher Na/(Na + Ca) and Cl/(Cl + HCO₃) ratios with increasing TDS and the positive association with high TDS additionally indicates the influence of evaporative concentration and secondary processes (i.e., ion exchange and anthropogenic loading) as significant modifiers of the natural signal.

Relationship between water quality variables was investigated using a correlation matrix. There are very strong positive correlations between TDS and Na (r = 0.79), Cl and Na (r = 0.75) and Mg and hardness (r = 0.81), which may indicate common anthropogenic or geogenic sources (such as rock dissolution, sewage contamination). Figure 7 gave the correlation matrix of groundwater matrix.

The PCA transformed the dataset into three principal components (PCs) explaining 84% of total variance of the groundwater chemistry of Sargodha (Figure 8). PC1 (52.4% variance) presented marked positive loadings for TDS, Na, K, and Cl. This factor mainly represents mineral dissolution and salinity induced processes, such as halite dissolution and cation exchange of clay minerals with ground water. The closer relationship of Na with Cl indicates that both geogenic processes (e.g., rock-water interaction) and anthropogenic activities, e.g., the use of saline groundwater for irrigation and wastewater infiltration. The second component (PC2, 21.7% variance) was mostly characterized by nitrate and sulfate, and associated with agriculture runoff and domestic sewage inputs. High NO₃ ⁻ concentrations in multiple samples suggest nitrogen fertilizer leaching, whereas SO₄ ^2- can arise from both fertilizers and wastewater. This part reinforces the contribution of anthropogenic activities in spatial distribution of groundwater quality in Sargodha, especially in regions where there is excessive use of groundwater for agriculture. PC3 (9.9% of the variance) was dominated by high loading of Ca and Mg, which indicated carbonate and dolomitic lithologies in the aquifer system. These correlations show that dissolution of calcite and dolomite minerals is a geogenic control to groundwater chemistry as per the geological settings of the Punjab plains.

The combined above three factors show that the groundwater characteristics in Sargodha are largely controlled both by natural (carbonate weathering, minerals dissolution, ion exchange) and anthropogenic (fertilizer leaching, wastewater intrusion, and irrigation return flow) processes. Equivalent PCA interpretations have been found (Nasir et al., 2016; Rehman et al., 2024) in other parts of Pakistan and South Asia whereby Na–Cl enrichment and nitrate contamination were attributed to mix geogenic–anthropogenic sources. Therefore, the PCA not only confirms the correlation results, but also gives a good separation of the pollution sources and represents a powerful diagnostic method for groundwater management.

Levels of Na, Cl, and nitrate in general are comparable to or higher than the studies related to groundwater in Rawalpindi (Farooq et al., 2008), Faisalabad (Nasir et al., 2016) and Kasur (Arshad and Imran, 2017) as were reported in the case of contaminated groundwater. This is consistent with the results of Daud et al. (2017), who found that groundwater contamination in Pakistan was associated with untreated wastewater and agrochemical wash off. This cross-validation increases the confidence of our results, and underscores the necessity of local monitoring plans of groundwater quality.

WQI is employed to analyze groundwater quality in the Sargodha division. By representing water quality with a single item on a numeric scale, it streamlines conditional criteria for exposure assessment by condensing the concentrations of multiple parameters into one value for a given sample. The method generates a consolidated index across multifarious water quality parameters and offers a holistic interpretation of groundwater quality for various usages like drinking, irrigation and industrial (Rana and Ganguly, 2020). The First step for calculating groundwater’s WQI is to estimate each parameter’s quality rating using the formula Q n = 100 * V n − V i / V s − V i . If quality rating Q n = 0 means complete absence of pollutants, while 0 < Q n < 100 implies that the pollutants are within the prescribed standard and when Q n > 100 implies that the pollutants are above the standards (Gungoa, 2016). However, the explanation of collected groundwater samples showed that TDS (126.18), Na (104.68), K (102.78), and NO₃ (126) had a value of more than 100 (Table 2), which means these factors were critical in the decomposition of water quality. The unit weight (Wn) for each physicochemical parameter was then calculated using the formula (Table 2). The computed WQI, according to Brown et al. (1972), was 84.57 where groundwater quality was deemed poor, i.e. 76 to 100 (Table 3).

The WQI of 84.57 found for the overall index reflects the categorization of the groundwater quality as poor (76–100). TDS, Na, K, and NO₃ were major factors that controlled the high WQI value. pH, HCO3 and SO4 being within the WHO ranges had minimal impact on the index due to the variations observed. The computed values from the weighted arithmetic index of the WQI reveals that, the groundwater of the study area is unsuitable for drinking purpose but suitable for agricultural practices.

The results from this study highlight the necessity of region-specific, integrated approaches to control groundwater contamination in Sargodha. Pakistan has, however, developed various policy frameworks in order to regulate water quality including the National Water Policy (2018) the Punjab Drinking Water Policy (2007) and Pakistan Environmental Protection Act (PEPA) 1997, however, the deployment of such policies is restricted in Pakistan due to a variety of institutional and administrative bottlenecks. Among them fractured governance between the federal, provincial, and municipal governments, and among water, sanitation, and public health authorities; inadequate budgets and weak enforcement capacity at the district level. To fill this implementation gap, there is an urgent need to improve capacity of local water monitoring systems. Creating tehsil-level water quality monitoring centers within the PCWR could potentially enable on-going testing, data transparency and interventions in a timelier manner. Better co-ordination between departments, most notably health, environment and local government will be vital in ensuring that the management of water, fits within a wider public health and environmental rationale. Additionally, raising public awareness plays a key role in moderating behavior. This knowledge can help empower local communities to not only educate them in terms of hazards from untreated ground water, but to also promote household level filtration technologies and water conservation efforts, which all can reduce exposure to contaminated water sources. At the policy level, there is also a need for amending groundwater extraction regulations to make it mandatory for providing quality-based conditions for granting a license for digging borewells. That would be an important step towards reducing over-exploitation of groundwater and the degradation of water quality. Another important area is agriculture, because agricultural runoff is a proven source of nitrate and sulfate contamination of drinking water sources, water quality policies should encourage and train farmers to adopt more sustainable farming practices that minimize chemical usage. In conclusion, investment in small-scale water treatment infrastructure and sanitation systems, which are particularly needed in poorly-served rural communities, is necessary for the public health to be enjoyed by all population segments.

Although the study was conducted in several tehsils of Sargodha District including both urban and rural settings, but it is also constrained by the number of NH3 sampling sites (30 points) which may possibly not have covered micro-level spatial gradient. The study was also limited to physicochemical parameters, thus microbiological contaminants and heavy metals such as arsenic and lead were left out. Species specific dose-response models for hazard quotients with chronic daily intake models were also not developed given the full-scale of the health risk assessment was not conducted for this study. Future research work ought to take into account seasonality, larger sample sizes, microbiological and heavy metal analysis and sophisticated modeling methods for better understanding of the groundwater quality and its health impacts. A detailed health impact assessment is important to correctly measure the health risks associated with eating and drinking contaminated water.