Overview of the Study Area

Qianyingzi Coal Mine is located approximately 15 km southwest of Suzhou City’s center, spanning the administrative regions of Suzhou City and Suixi County, Huaibei City. Geographically, it lies between longitudes 116°51′00″–117°00′00″ E and latitudes 33°27′00″– 33°32′30″ N. The terrain is predominantly flat, with elevations ranging from +18.885 m to +24.720 m (averaging around +23 m), showing a gradual slope from higher elevations in the northwest to lower elevations in the southeast. Hydrologically, the Hui River – a tributary of the Huai River – traverses the central part of the mine. Specifically, the annual average water levels are +17.22 m upstream of Qixian Gate and +16.07 m downstream, with an average annual flow rate of 7.85 m³/s at Linhuan upstream. Additionally, the area is intersected by an extensive network of artificial channels and dense agricultural irrigation systems.

Climatically, the region experiences a monsoon-influenced warm temperate semi-humid climate, characterized by distinct seasonal variations: mild temperatures with variable spring warmth, hot and rainy summers, crisp autumns, and cold, dry winters. Notably, it benefits from abundant sunlight and a long frost-free period.

The groundwater within the study area is stratified into four distinct aquifers (groups) from the uppermost to the lowermost layers: the Cenozoic loose layer, the Permian main coal mining seam, the Taiyuan Formation limestone karstic fissured aquifer, and the Ordovician limestone karstic fissured aquifer. All samples were collected from the sandstone fractured aquifer associated with the top and bottom plates of the 32 coal seams. The burial depth of the floor varies between 187.95 and 1,206.83 meters, with an average of 592.09 meters. The thickness of the sandstone ranges from 0 to 130.60 meters, with a mean value of 21.08 meters. In this particular layer (section), the sandstone fractures are relatively well-developed yet exhibit a certain degree of heterogeneity. Specifically, the unified conversion values for q range from 0.0002 to 0.06445 L/(s•m), indicating weak water-bearing properties, and for k, they range from 0.00025 to 0.06157 m/d. Fig. 1 illustrates the geological relationship map of the study area along with the sample collection points.

Sample Collection and Processing

A total of 19 sandstone fissure water samples were collected from the study area. The 19 water samples were taken from various leak points at the mine face, with four collected from surface boreholes and seven and eight samples from rock workings on the east and west flanks, respectively.

During sampling, the following standardized procedures were implemented: Pre-rinsing: Sampling buckets were rinsed three times with deionized water. Field rinsing: Before collection, each bucket was rinsed three times with the sampled water to minimize contamination. Transport and storage: Samples were transported to the laboratory within 24 hours, filtered through 0.45 μm membranes, and stored at 4ºC for subsequent analysis.

Analytical methods included: pH and TDS: Measured using a portable OHAUS instrument. Cations (Na⁺, K⁺, Ca²⁺, Mg²⁺): Analyzed via an ISC-600 ion chromatograph, with 1.3 mL methanesulfonic acid added to a 1000 mL volumetric flask as the cation eluent. Anions (Cl^-, SO₄^2-): Tested using an ICS-900 ion chromatograph with ultrapure water eluent. Hardness and HCO₃^-: Determined through acid-base titration. Quality control measures included verifying instrument stability with standard samples and analyzing duplicate samples to ensure relative deviations remained below 5%.

Fig. 1.

Geological relationship map of the study area and the sample collection point.

Water Quality Evaluation Treatment Methods

Data Analysis Methods

Hydrochemical characteristics were analyzed using mathematical statistics, Piper trilinear diagrams, Gibbs diagrams, and ionic ratios. Descriptive statistics were performed in Excel, while diagrams were generated using Origin 2024 and CorelDRAW.

Hydrochemical Characteristics

Statistical results of major ion concentrations, total dissolved solids (TDS), and average pH values for sandstone fissure water samples from Qianyingzi Coal Mine are summarized in Table 1.

The pH ranges from 7.95 to 10.27 (average 8.68), indicating weakly alkaline conditions. TDS values ranged from 935.42 mg/L to 4,645.00 mg/L (mean 2,999.88 mg/L), which is characteristic of saline groundwater [19]. Among cations, alkali metal ions dominate, with concentrations following the order Na⁺>K⁺. Specifically, sodium concentrations range from 0.40 mg/L to 53.59 mg/L, while potassium ranges from 1.42 mg/L to 24.60 mg/L. The study area belongs to the coal measure strata, which are often rich in pyrite (FeS₂). The oxidation of sulfides is that the S that is insoluble in water enters the water in the form of SO₄^2- in large quantities, so the SO₄^2- in the groundwater in the study area comes from the dissolution of sedimentary rocks containing gypsum or other sulfates. The reaction is shown in Equation (9). The Na⁺ comes from the dissolution of rock salt and other sodium salts in sedimentary rocks.

For anions, sulfate (SO₄^2-) is the primary constituent, exceeding bicarbonate (HCO₃^-) in most samples. To assess data variability, the coefficient of variation (CV) was calculated, which is defined as the standard deviation divided by the mean. Higher CV values (>1) indicate greater hydrochemical instability, reflecting diverse formation processes. Notably, the TDS CV was <1, suggesting stable hydrochemical conditions [20]. Similarly, the pH CV is <0.1, implying minimal spatial variability [21]. However, exceptions with CV >1 highlight localized instability in certain components, likely influenced by environmental fluctuations [22].

The Piper trilinear diagram (Fig. 2) was employed to classify hydrochemical types and compare compositional trends [23]. Key observations include: Alkali metals (Na⁺ + K⁺) constitute >80% of cations, exceeding alkaline earth metals (Ca²⁺ + Mg²⁺). Dominant hydrochemical type: SO4·HCO₃-Na+K. Low Ca²⁺ and Mg²⁺ concentrations, except in rare cases. The water samples exhibit diverse hydrochemical properties. The reasons are as follows: These samples share the same recharge sources and climatic conditions; however, due to different infiltration methods of groundwater, varying burial depths of groundwater runoff, and discrepancies in the water environment and hydrochemical processes, different hydrochemical outcomes arise.

In Fig. 2a), all samples cluster in Zone 2 (alkali metals > alkaline earth metals). Fig. 2b) shows that strong acid anions (SO₄^2-, Cl^-) dominate most samples, with weak acids (HCO₃^-) prevalent in a few. Fig. 2c) categorizes samples into Zone 7 (non-carbonate hardness >50%, alkali metals + strong acids) and Zone 8 (carbonate hardness >50%, alkaline earth metals + weak acids).

Fig. 2.

Piper diagram of a sandstone fissure water sample.

Water-Rock Interaction Mechanisms

Gibbs diagrams (Fig. 3) were used to elucidate hydrochemical controls, classifying mechanisms into three domains: evaporation-crystallization, rock weathering, and atmospheric precipitation [24].

In Fig. 3a), TDS values (1,000-10,000 mg/L) and Na⁺/(Na⁺ + Ca²⁺) ratios (0.961-0.999) cluster within the evaporation-crystallization zone. Fig. 3b) shows that Cl^-/(Cl^- + HCO₃^-) ratios increase with TDS (<0.5), indicating a transition from rock weathering to evaporation dominance. These patterns collectively suggest that groundwater chemistry is primarily controlled by evaporation-driven water-rock interactions, with minimal influence from weathering or atmospheric precipitation [25].

Ion Ratio Analysis

Analysis of ionic ratios in groundwater provides insights into hydrogeochemical processes [26], as illustrated in Fig. 4.

Na⁺/Cl^- ratio of 1 indicates that the halite dissolution dominates groundwater chemistry. Ratios exceeding 1 suggest the presence of additional sodium sources (e.g., other sodium salts) [27].

Fig. 4a) shows that all samples fall below the Na⁺/Cl^- = 1 line, implying contributions from silicate dissolution or cation exchange adsorption, where Na⁺ replaces adsorbed Ca²⁺/Mg²⁺, elevating Na⁺ concentrations [28]. These may include silicate dissolution or cation exchange adsorption occurring in the fissure water. Specifically, cation exchange adsorption replaces Ca²⁺ and Mg²⁺ in the water with Na⁺ from rock-soil particles, thereby increasing the concentration of Na⁺ in the aqueous phase. It is therefore concluded that, in the study area, the Na⁺ in sandstone fissure water originates not only from rock salt dissolution but also from additional processes such as cation exchange adsorption.

Fig 3.

Gibbs plot of the study area.

(Ca²⁺ + Mg²⁺)/HCO₃^- ratio of 1 suggests carbonate (calcite, dolomite) and sulfate (gypsum, anhydrite) mineral dissolution [29]. Fig. 4b) reveals (Ca²⁺ + Mg²⁺)/ (HCO₃^- + SO₄^2-) ratios <1, indicating low Ca²⁺/Mg²⁺ levels due to: (1) limited carbonate sources, (2) oxidation-induced SO₄^2- enrichment, (3) silicate weathering, or (4)cation exchange (Na⁺ displacing adsorbed Ca²⁺/Mg²⁺), corroborating Fig. 4a).

Additionally, the dissolution of sulfate minerals serves as a significant source of ions in groundwater. Therefore, the (Ca²⁺ + Mg²⁺)/SO₄^2- ratio is commonly employed to investigate the potential sources of Ca²⁺, Mg²⁺, and SO₄^2- in groundwater [30]. As illustrated in Fig. 4c), most water samples plot above the (Ca²⁺ + Mg²⁺)/SO₄^{2 -}= 1 baseline, indicating that the primary source of SO₄^2- is not limited to sulfate dissolution but also involves additional contributions. (Ca²⁺ + Mg²⁺)/ HCO₃^- ratio of 1 signifies carbonate dissolution [31]. However, low Ca²⁺/Mg²⁺ ratios in the samples (Fig. 4d) highlight the contributions of silicate weathering.

Collectively, ionic ratios confirm that sulfate dissolution and silicate weathering are the primary ion sources in fissure water.

Evaluation Results

Nineteen groundwater samples were analyzed using single-parameter, F-value, and fuzzy comprehensive evaluation methods. As summarized in Table 2, both the single-parameter and F-value methods categorized 18 samples (94.7%) as Class V (extremely poor quality) and 1 sample (5.3%) as Class IV (poor quality). The primary pollutants – total dissolved solids (TDS), chloride (Cl^-), and sodium (Na⁺) – consistently exceeded the Class V thresholds specified by the Chinese Groundwater Quality Standard (GB/T 14848- 2017). For instance, Sample 1 exhibited TDS levels of 2000 mg/L (Class V limit: 1000 mg/L), with Cl^- and Na⁺ concentrations reaching 500 mg/L and 400 mg/L, respectively. Furthermore, pH fluctuations (Class III-V) and elevated sulfate levels indicated potential risks of acidification and sulfate contamination.

Fig. 4.

Ion ratio results in water samples.

The F-value method generated a composite pollution index (F) by aggregating weighted parameters, with all samples exceeding the threshold value of 7.0 (scores >7.45). However, this method’s reliance on expert-assigned weights (e.g., 30% for TDS) introduced subjectivity, potentially distorting the evaluation outcomes.

In contrast, the fuzzy evaluation method (Table 3) employed membership functions – mathematical tools that quantify uncertainties in classification – to achieve higher precision. For example, Sample 1 demonstrated a 98.42% membership probability in Class V, confirming severe contamination. Overall, 15 samples (79%) were unambiguously classified as Class V, while 4 samples (21%) were assigned to Class IV due to marginal improvements in parameters such as pH. These results aligned with the single-parameter classification but better captured subtle parameter variations through probabilistic modeling.

Comparative Analysis and Recommendations

Methodological disparities stem from distinct theoretical foundations, such as:

Single-parameter approach: Although conservative, this method neglects parameter synergies [32]. For example, Sample 7 was erroneously classified as Class IV under this method due to the neutralizing effect of pH (Class IV) and sulfate (Class I). In contrast, the fuzzy logic method correctly assigned Class IV by integrating multi-parameter weights.

F-value method: While linear weighting accounts for parameter interactions, it is prone to subjective weight allocation (e.g., 10% for Cl^-), which can potentially lead to underestimating pollutant impacts.

Fuzzy evaluation: This method utilizes membership functions and fuzzy matrices to address data uncertainty and nonlinear relationships. In Sample 3, TDS (Class IV) combined with Cl^- (Class V) yielded a 71.77% membership probability in Class V, consistent with observed contamination levels.

In accordance with WHO and EU guidelines, fuzzy methods demonstrate superior performance in multi-pollutant scenarios. Future research should focus on two priorities: (1) incorporating machine learning algorithms for dynamic weight optimization to reduce subjectivity and (2) enhancing real-time monitoring systems to improve data reliability.

The severe groundwater contamination identified in this study necessitates urgent remediation measures, including industrial emission controls and the implementation of constructed wetlands. These findings also underscore the need for transregional water resource management, particularly in mining or industrial zones with analogous hydrogeological conditions.