Hydrochemical Characteristics, Mechanisms of Formation, and Sources of Different Water Bodies in the Northwest Coal–Electricity Agglomeration Area

Abstract

Water resources are relatively scarce in Northwest China. Therefore, this study aimed to identify the hydrochemical characteristics and sources of different water bodies in the Northwest Coal–Electricity Agglomeration area, and the utilization of water resources in the region. Hydrochemical diagrams and correlation analysis were applied to data obtained through the collection of 40, 14, and 42 surface water, shallow groundwater, and deep groundwater samples, respectively. The Positive Definite Matrix Factor Decomposition (PMF) model was used to explore the origins of ions in different water bodies. The results show the following: (1) The rank of anions in surface water, shallow groundwater, and deep groundwater in water bodies of the Bulianta mining area during the wet period according to concentration was as follows: SO₄²⁻ > Cl⁻ > HCO₃⁻ > NO₃⁻; that of cations was as follows: Na⁺ > Ca²⁺ > Mg²⁺ > K⁺; (2) The chemical composition of surface water is mainly regulated by the dissolution of evaporites; that of shallow groundwater was regulated by silicates; that of deep groundwater was mainly regulated by the hydrolysis of silicates and the dissolution of evaporites; (3) Four main sources of ions in different water bodies were identified, namely agricultural activities, rock weathering, primary geology, and unknown sources. Two natural factors, namely rock weathering and primary geology, and human activities contributed to 35.2% and 38.8% of ions in shallow groundwater, respectively. Rock weathering and human activities contributed to 20.6% and 63.9% of ions of deeper groundwater, respectively. This study can provide a basis for the conservation and rational planning and utilization of water resources in the Northwest Coal–Electricity Agglomeration area.

1. Introduction

Water is essential for sustaining ecosystems and for human life and social development [1]. However, accelerated industrialization and the associated increases in human population and human activities since the 20th century have led to the pollution of freshwater sources [2,3,4,5]. Ordos Basin falls in the arid and semi-arid northwestern China, characterized by large coal and petroleum resources [6]. While the exploitation of coal resources has facilitated economic development, it has also resulted in several environmental challenges, including environmental pollution, land subsidence, aquifer destruction, and ecological damage.

Many recent studies have examined the impact of coal mining on regional aquifers, surface water, groundwater, and water conservation to understand the current status and impact of pollution on the water environment [7,8,9,10,11]. Han et al. (2020) used hydrochemical methods, including mathematical statistics, the Piper trilinear diagram, and principal component analysis to characterize the hydrochemistry of groundwater and the associated mechanisms in Shendong Daliuta coal mine [12]; Li et al. (2022) analyzed the characteristics and influencing factors of surface water and groundwater hydrochemistry in the Shendong Mining Area through the application of Piper and Gibbs diagrams. These past studies have shown that the hydrochemical composition of natural freshwater sources is mainly regulated by silicate rock and carbonate rock weathering and evaporite salts [13]. Yang et al. (2017) examined factors regulating the hydrochemistry of groundwater in the Shendong mining area [14]; Su et al. (2023) examined the characteristics of ion migration in groundwater of the Shendong mine under different water chemistry conditions [15]. The above two studies demonstrate the applicability of hydrochemical methods for accurately analyzing the hydrochemistry of mining sites, whereas the combined application of topical analyses with quantitative resolution and other methods can allow the comprehensive characterization of natural and anthropogenic mechanisms regulating the evolution of hydrochemical ions [16,17,18].

The present study aimed to identify the hydrochemical characteristics and sources of ions in different water bodies in the Northwest Coal–Electricity Agglomeration area. The study focused on a mine in Yijinholo, Ordos Basin, where preliminary analysis was conducted on the sources of elements in surface water and groundwater in the area using methods such as Piper’s three line diagram, Gibbs diagram, correlation analysis, and hydrogeochemical simulation, in order to provide a scientific basis for water safety, ecological environment protection, and sustainable economic development in the Bulianta mining area.

2. Materials and Methods

2.1. Study Area

The Bulianta Mine (109°30′00″–110°10′00″ E, 39°00′00″–39°50′00″ N) is in Yijinholo Banner, Ordos City, Inner Mongolia Autonomous Region (Figure 1). The study area has a hilly landform type with elevation ranging from 1200 to 1324 m above sea level, gradually decreasing from east to west. The study area falls into an arid and semi-arid continental monsoon climate zone, with annual rainfall and annual potential evaporation from 194.7 to 531.6 mm (average of 357.3 mm) and from 2297.4 to 2833.7 mm (average of 2457.4 mm), respectively. The region has a developed surface water system, with the Wulanmulun River, a direct tributary of the Yellow River, flowing through the mining area from northwest to southeast. Regional groundwater is mainly recharged by atmospheric precipitation.

The mining area has a monoclinic geological structure, with no major fold development and a relatively flat topography. The major exposed aquifers in the area, ordered in decreasing age, are the Triassic Upper Yongping Formation (T_3y), Jurassic Lower and Middle Yan’an Formation (J_2y), Middle Zhiluo Formation (J_2z), Anding Formation (J_2a), Lower Cretaceous to Upper Jurassic Shidan Group (J₃-K_1zh), Tertiary, and Quaternary. Two main types of aquifers occur in the study area: (1) the sandy mudstone aquifer between the Quaternary aquifer and the Zhiluo Formation aquifer; (2) the sandy mudstone aquifer between the Zhiluo Formation aquifer and the Yan’an Formation aquifer.

2.2. Sample Collection and Measurements

The present study collected water samples from the study area in April, June, September, and November 2021–2023, with Figure 1 showing their spatial distribution. Water sampling was conducted according to the Technical Specification for Groundwater Environmental Monitoring (HJ/T164-2004) [19]. Water sampling points were established in an area bounded by the Ulaanmulun River and its tributaries. The present study collected 40, 14, and 42 surface water, shallow groundwater, and deep groundwater samples, respectively, from straight sections of the river with steady flow, rural water supply wells (depth of 20–70 m), and from coal mines and centralized water supply wells in villages and towns (depth of 100–300 m), respectively. Groundwater samples were collected using well pumps, with the pump activated for 3 min prior to sampling to purge stagnant water, and the groundwater samples collected in 500 mL brown polyethylene slim neck bottles, which were sealed with parafilm and stored on ice in an insulated box.

Portable field meters were used to measure total dissolved solids (TDS), electrical conductivity (EC), and the pH of water samples. Laboratory testing determined the cations (K⁺, Ca²⁺, Na⁺, and Mg²⁺) of water samples using Aptar 940 high-pressure analytical ion chromatography; anions (SO₄²⁻, NO₃⁻, Cl⁻) were determined by Agilent ICP-MS 7800 inductively coupled plasma mass spectrometry; HCO₃⁻ and CO₃²⁻ were determined by acid–base titration. The water sample data were analyzed using correlation analysis [20,21], Piper trilinear plot analysis, Gibbs plot analysis [22,23], saturation index (SI), chloride alkalinity index (CAI-I and CAI-II), and Positive Definite Matrix Factor Decomposition (PMF).

The Gibbs plot can isolate water samples according to the three major processes influencing water hydrochemistry, namely evaporative-concentration-dominated, rock-weathering-dominated, and atmospheric-precipitation-dominated. The Gibbs plot consists of two sets of scatter plots with semi-logarithmic coordinates that plot the ratio of Cl⁻/(Cl⁻ + HCO₃⁻) or Na⁺/(Na⁺ + Ca²⁺) in the horizontal coordinates and TDS in both vertical coordinates.

The chloride alkalinity index (CAI-I and CAI-II) proposed by Schoeller can be used to identify cation exchange in groundwater [24].

$$\mathit{CAI}\text{-}I=\frac{\mathit{Cl} -(N{\mathrm{a}}^{+}{+K}^{+})}{\mathit{Cl}}$$

(1)

$$\mathit{CAI}\text{-}\mathit{II}=\frac{\mathit{Cl} -(N{\mathrm{a}}^{+}{+K}^{+})}{{(\mathit{HCO}}_3^{-}{+\mathit{SO}}_4^{2-}{+\mathit{NO}}_3^{-}{+\mathit{CO}}_3^{2-})}$$

(2)

Both CAI-I and CAI-II exceeding 0 indicates positive cation exchange adsorption defined by Equation (3) during the formation of water chemical components; their negative values indicate reverse cation exchange adsorption defined by Equation (4) [25,26].

$$2N{\mathrm{a}}^{+}+(C\mathrm{a}{,\mathit{Mg})X}_2\iff {(C\mathrm{a},\mathit{Mg})}^{2+}+2N\mathrm{a}X$$

(3)

$${\left(C\mathrm{a},\mathit{Mg}\right)}^{2+}+2N\mathrm{a}X\iff 2N{\mathrm{a}}^{+}+(C\mathrm{a}{,\mathit{Mg})X}_2$$

(4)

The mineral saturation index (SI) plays an important role in assessing the degree of dissolution–precipitation equilibrium between the groundwater body and minerals [27]. The SI can be applied to identify the major water–rock reactions occurring during water chemistry formation and is calculated as

$$\mathit{SI}=\mathit{Ig}\frac{\mathit{IPA}}{K},$$

(5)

where IPA is the product of ion activity in the water body and K is the equilibrium constant. An SI > 0, =0, and <0 indicates a saturated and mineral-precipitation-dominated water body, mineral dissolution, and precipitation in equilibrium at equal rates, and an unsaturated water body, respectively. The dissolution rate of the water body exceeds the sedimentation rate, and the absolute value of SI is proportional to the dissolution capacity of the water body [28].

Positive Definite Matrix Factor Decomposition (PMF) is widely used for identifying the sources of pollutants in the atmosphere, aquatic environments, and sediments. PMF identifies classes of pollution sources by introducing data uncertainty estimates [29]. The PMF model was plotted using EPA PMF version 5.0 [30], which decomposes the sample concentration data matrix, Xin, into two factor matrices, g_ik and f_kj, and a residual matrix, e_ij, calculated as

$$X_{\mathit{ij}}={\displaystyle \sum _{k=1}^p{g}_{\mathit{ik}}}{f}_{\mathit{kj}}{+e}_{\mathit{ij}},$$

(6)

where x_ij is the concentration of the jth element in the ith sample; g_ik is the contribution of source k to the ith sample; f_kj is the concentration of the jth element in source k; and e_ij is the residual matrix, i.e., the part of the PMF model that does not account for the sample concentration matrix x_ij.

The PMF method is based on least-squares qualification and performs iterative calculations to minimize the objective function Q, calculated as

$$Q={{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^m\left(\frac{e_{\mathit{ij}}}{u_{\mathit{ij}}}\right)}}}^2,$$

(7)

where μij is the uncertainty of the concentration of the jth element in the ith sample. When an element concentration exceeds the metal detection limit (MDL), the uncertainty (Unc) can be calculated as

$$\mathit{Unc}=\sqrt{{\left(\delta \times c\right)}^2+{\left(0.5\times \mathit{MDL}\right)}^2} (c>\mathit{MDL})$$

(8)

where δ is the relative standard deviation; C is the elemental content (mg·kg⁻¹), and MDL is the heavy metal method detection limit (mg·kg⁻¹).

If the element concentration is less than the MDL, the magnitude of the uncertainty is calculated as

$$\mathit{Unc}=\frac{5}{6}\times \mathit{MDL} (c\le \mathit{MDL})$$

(9)

3. Results

3.1. Characteristics of Chemical Components of Different Water Bodies

Figure 2 visually characterizes the hydrochemistry of the various water bodies in the Pelenta mine area. The pH of surface water during the wet and dry periods ranged from 7.46 to 10.04 (mean of 8.28) and 7.55 to 8.65 (mean of 8.14), respectively; that of shallow groundwater ranged from 7.02 to 8.08 (mean of 7.56) and 7.62 to 7.89 (mean of 7.77), respectively; that of deep groundwater ranged from 6.61 to 10.21 (mean of 7.89) and 7.48 to 8.77 (mean of 8.02), respectively. The TDS of surface water, shallow groundwater, and deep groundwater during the wet period ranged from 112 to 4116 mg/L (mean of 853.85 mg/L), 141.0 to 806.0 mg/L (mean of 156.36 mg/L), and 114.92 to 2361.05 mg/L (mean of 405.89 mg/L), respectively; during the dry period they were 166 to 2810 mg/L (mean of 499.34 mg/L), 213 to 498 mg/L (mean of 110.67 mg/L), and 176 to 3289.95 mg/L (mean 666.08 mg/L), respectively.

The percentages of anions by total anion mass varied between the different water bodies and between the wet and dry periods (Figure 2). The rank of anions in surface water, shallow groundwater, and deep groundwater during the wet period according to the relative proportion of total mass was SO₄²⁻ > Cl⁻ > HCO₃⁻ > NO₃⁻, with SO₄²⁻ accounting for 51.27%, 59.40%, and 62.42% of the total anions in shallow groundwater, deep groundwater, and surface water, respectively. This result suggests that anions in all three water types may be regulated by accelerated oxidation of pyrite introduced by coal mining. The rank of cations in surface water and deep groundwater according to the relative proportion of total mass was Na⁺ > Ca²⁺ > Mg²⁺ > K⁺; that in shallow groundwater was Ca²⁺ > Na⁺ > Mg²⁺ > K⁺. Among the cations, Na⁺ accounted for 80.49% and 70.92% of total cations in surface water and deep groundwater, respectively; Ca²⁺ accounted for 60.21% of total cations in shallow groundwater. The rank of anions of surface water during the dry period according to relative proportion of total mass was SO₄²⁻ > HCO₃⁻ > Cl⁻ > NO₃⁻; that of shallow groundwater was SO₄²⁻ > Cl⁻ > NO₃⁻ > HCO₃⁻. Given the limited depth of shallow groundwater, chemical constituents of shallow groundwater were strongly influenced by anthropogenic activities, with high concentrations of NO₃⁻ detected in a small number of samples. The rank of anions of deep groundwater according to the relative proportion of total mass was SO₄²⁻ > Cl⁻ > HCO₃⁻ > NO₃⁻; that of cations of surface water and deep groundwater was Na⁺ > Ca²⁺ > Mg²⁺ > K⁺; that of cations of shallow groundwater was Ca²⁺ > Mg²⁺ > Na⁺ > K⁺. Na⁺ accounted for 78.47% and 55.13% of the total cations in surface water and deep groundwater, respectively; Ca²⁺ accounted for 43.72% of the total cations in shallow groundwater, mainly originating from the dissolution of carbonate minerals [31].

3.2. Analysis of the Sources of Ions in Different Water Bodies

3.2.1. Differences in Hydrochemistry among Different Water Sources

The present study applied Piper trilinear diagrams to analyze differences in water chemistry among different water sources in the study area and between the wet and dry periods. The results of the Piper trilinear diagram (Figure 3) in combination with the Shukarev method showed that surface water during the wet period was dominated by the SO₄-Cl-Na type water, accounting for 74% of water samples, with the calcium sulphate type accounting for a few samples; the SO₄-Cl-Ca and SO₄-Cl-Na-Ca types dominated shallow groundwater; the SO₄-Cl-Na and SO₄-Na types dominated deep groundwater. Surface water during the dry period was mainly dominated by the HCO₃-SO₄-Na and SO₄-Cl-Na types, accounting for 50% and 43.75% of water samples, respectively; the HCO₃-Ca type dominated shallow groundwater; the HCO3-SO₄-Na and HCO₃-SO₄-Ca types dominated deeper groundwater. The high concentrations of SO₄²⁻ in the various water bodies could be mainly attributed to oxidative dissolution of sulfide iron ore, linked to the effects of coal mining, and resulting in the formation of SO₄²⁻. The spatial distribution of the results of the Piper trilinear plots showed higher ion concentrations in shallow and deep groundwater. The chemical types of deep groundwater were regulated by a higher diversity of ions, mainly due to its sensitivity to anthropogenic activities, such as coal mining, as well as to complex sources of recharge.

3.2.2. Identification of Correlations between Water Hydrochemistry and Different Sources of Ions

Correlation analysis (CA) is a statistical method used to measure the degree of correlation between different variables. CA allows a visual representation of the degree of correlation between different water chemistry indicators. Pearson’s correlation coefficients (r) were calculated between 12 variables (TDS, EC, pH, F⁻, Cl⁻, NO₃⁻, SO₄²⁻, HCO₃⁻, Ca²⁺, K⁺, Mg²⁺, and Na⁺) for the 40, 14, and 42 surface water, shallow groundwater, and deep groundwater samples from the study area, respectively (Figure 4).

Cl⁻ in surface water showed significant positive correlations (p < 0.05) with SO₄²⁻, Ca²⁺, Mg²⁺, Na⁺, and K⁺, with correlation coefficients of 0.59, 0.63, 0.57, 0.75, and 0.62, respectively. Cl⁻ in shallow groundwater showed significant positive correlations with SO₄²⁻, Ca²⁺, Mg²⁺, Na⁺, and K⁺, with correlation coefficients of 0.68, 0.76, 0.55, 0.67, and 0.83, respectively. Cl⁻ in shallow groundwater showed strong negative correlations with F⁻ and pH, indicating that OH⁻ did not inhibit the adsorption of F⁻ in the clay minerals. Cl⁻ in deep groundwater showed significant positive correlations (p < 0.05) with SO₄²⁻, Ca²⁺, Mg²⁺, Na⁺, and K⁺, with correlation coefficients of 0.62, 0.39, 0.38, 0.65, and 0.43, respectively. These results indicated that several ions were strongly influenced by Cl⁻, resulting in their possible homology. There were relatively strong positive correlations (p > 0.05) between SO₄²⁻ and K⁺ and between Mg²⁺ and Na+ in surface water, as well as between SO₄²⁻ and K⁺ and between Ca²⁺ and Na⁺ in shallow groundwater. There were significant correlations (p < 0.05) between Na⁺ and Cl⁻ and between Na⁺ and SO₄²⁻ in deep groundwater. These results indicated that deep groundwater was influenced by the dissolution of rock salt and gypsum. There was a positive correlation (p > 0.05) between K⁺ and NO₃⁻ in surface water, whereas no such relationship occurred in shallow and deep groundwater. This could be attributed to the influence of nitrogen and potassium compound agricultural fertilizers on surface water [32].

3.2.3. Mechanisms Regulating the Hydrochemistry of Different Water Sources

(1)

Factors regulating the hydrochemistry of different water sources according to the Gibbs diagrams

As shown in Figure 5, the major factor influencing the hydrochemistry of almost all groundwater and surface water samples in the Shuilianta mine area was rock weathering, which is consistent with the conclusions of Zengfeng et al. [33]. As shown in Figure 5, the shift of most samples of shallow groundwater to the right could be attributed to the dominant effect of human activities on their hydrochemistry. The placement of some surface water and deep groundwater sampling points outside the areas of control of evaporative concentration and water–rock interactions indicated that they were influenced by other factors.

(2)

End-element plots of water bodies from different sources

The end-element plots provide three different end-elements based on rectangular diagonals, i.e., three different water–rock actions: (1) carbonate dissolution; (2) silicate weathering; (3) evaporite dissolution. The distance of a water sample point from an end-element is inversely related to the relative importance of the hydrochemical process represented by that end-element in the formation of hydrochemistry of the sample; the placement of a sample between the two end-element regions indicates that the hydrochemistry of the sample is regulated by both processes [34,35]. As shown in Figure 6, surface water was mainly regulated by evaporite dissolution, whereas shallow groundwater was regulated by silicate weathering, and silicate hydrolysis and evaporite dissolution were the main processes regulating the hydrochemistry of deep groundwater. The abundant F⁻ in groundwater could be attributed to the decomposition and release of fluorine-containing minerals, such as silicates (e.g., smectite, albite, quartz, etc.). In comparison to deep groundwater, factors regulating the hydrochemistry of shallow groundwater and surface water were more dispersed, indicating that their hydrochemistry is regulated by complex weathering, likely related to coal mining activities.

4. Discussion

4.1. Dissolved Filtration in Waters from Different Sources

Mineral solubility, the solute saturation of the water body, geotechnical porosity, dissolved gases, and the alternating rate of groundwater circulation regulates the intensity of mineral dissolution–precipitation. Calculating and plotting the ratio of major ions in groundwater and the mineral saturation index allows discrimination of the possible mineral dissolution–precipitation effects in different water bodies in the study area.

The major ion ratios were plotted for the sampling points for the different water bodies in the study area during the wet and dry periods (Figure 6). The ion ratio relationship diagram could be used to analyze sources of major ions, with the source of Na⁺ + K⁺ and Cl⁻ distinguished by the (Na⁺ + K⁺)/Cl⁻ relationship in Figure 7a. The results showed that rock salt mineral dissolution occurred when the ratio of (Na⁺ + K⁺)/Cl⁻ was close to 1, whereas silicate mineral dissolution occurred when the ratio of (Na⁺ + K⁺)/Cl⁻ exceeded 1. As shown in Figure 7a, surface water and deep groundwater sampling points during the wet and dry periods mostly plotted above the y = x line. This result indicates that Na⁺ and K⁺ originate from the dissolution and filtration of evaporated rock salts and silicates as well as from other sources, and may also be affected by other processes, such as anion and cation exchange. Anion and cation exchange effects may be influenced by the dissolution and filtration of rock salts and silicates. Anion exchange can be identified by the (Na⁺ + K⁺ − Cl⁻)/((Ca²⁺ + Mg²⁺) − (SO₄²⁻ + HCO₃⁻)) relationship, with a ratio close to −1 indicating a cation exchange effect. As shown in Figure 7b, the samples of surface water, shallow groundwater, and deep groundwater showed negative (Na⁺ + K⁺ − Cl⁻)/((Ca²⁺ + Mg²⁺) − (SO₄²⁻ + HCO₃⁻)) relationships. In addition, this relationship in the water samples was more dispersed, suggesting the presence of the ion exchange effect. A (Ca²⁺ + Mg²⁺)/(HCO₃⁻ + SO₄²⁻) ratio close to 1 indicates that Ca²⁺ and Mg²⁺ originated from carbonate and silicate dissolution. As shown in Figure 7c, the (Ca²⁺ + Mg²⁺)/(HCO₃⁻ + SO₄²⁻) ratios for most of water sampling points of deep groundwater and shallow groundwater were = 1. Ca²⁺ and Mg²⁺ in the groundwater of the study area originate from gypsum dissolution. The ratio of (Ca²⁺ + Mg²⁺)/(HCO₃⁻ + SO₄²⁻) for surface water diverged away from 1, indicating weak exposure to gypsum dissolution, mainly from silicate dissolution. As shown in Figure 7d, most water sample points were plotted above the 1:1 line, indicating evaporite dissolution filtration. A small proportion of the water sample points plotted below the 1:1 line, suggesting regulation of water chemistry by dissolution filtration of carbonate minerals. As shown in Figure 7e, groundwater Ca²⁺ and Mg²⁺ were regulated by the joint effects of calcite, dolomite, and silicate dissolution. The Ca²⁺/Mg² ratio of a small proportion of water sample points plotted below 1, with almost no groundwater sample points falling between 1 and 2 and most falling above 2. The above results suggest that groundwater Ca²⁺ and Mg²⁺ mainly originate from silicate dissolution and calcite dissolution [36].

Figure 7f shows the ratios between the chlor-alkali indices CAI-I and CAI-II for surface water, shallow groundwater, and deep groundwater samples during the wet and dry periods. The CAI-I and CAI-II indicators were mostly negative, suggesting the occurrence of alternate cation adsorption in most of the water bodies in the study area [37].

The changes in SI can be used to identify the different stages of groundwater chemistry. The Phreeqc Interactive 3.7.3 simulation software was used to simulate the SI of calcite, dolomite, gypsum, rock salt, and other common soluble salt minerals in different water bodies in the study area during the wet and dry periods. As shown in Figure 8, small proportions of surface water, shallow groundwater, and deep groundwater during the dry period exhibited calcite and dolomite saturation, whereas most remained in an unsaturated gypsum and rock salt state. These results indicated complex chemical genesis of the water bodies in the study area, with large influences from anthropogenic factors.

4.2. Analysis of Sources of Chemical Ions in Water Bodies of the Study Area

Recent studies have widely applied PMF as a factor decomposition receptor model to calculate the contributions and sources of pollutants in the environment [38]. The present study applied the PMF model to quantitatively resolve the 10 water chemistry indicators (TDS, F⁻, Cl⁻, NO₃⁻, SO₄²⁻, HCO₃⁻, Ca²⁺, K⁺, Mg²⁺, and Na⁺) for the 40, 14, and 42 samples of surface water, shallow groundwater, and deep groundwater, respectively, in the Parenta mine area and to calculate the contributions of each source (Figure 9, Figure 10 and Figure 11). A calculated signal-to-noise ratio exceeding 2 indicates good suitability of the data for application to the model, with the ratio proportional to the likelihood of the sample being detected and its suitability for the model [34]. Subsequent to model preparation, four factors were determined, and twenty operation iterations were performed to select the best Q value. All parameter values were between −3 and 3, and the calculation results stabilized between the measured values and the predicted values.

As shown in Figure 9, the source of the main chemical indicator (factor 1) in surface water, NO₃⁻ (82.6%), was excessive application of nitrogen agricultural fertilizers, in general agreement with other studies [39]. In addition, factor 1 was strongly influenced by agricultural fertilizer, with an average contribution of 6.6%. The main ions contributing to factor 2 were F⁻, Cl⁻, SO₄²⁻, HCO₃⁻, K⁺, and Na⁺, with relative contributions of 83.2%, 62.5%, 100%, 70.8%, 48.2%, and 82.8%, respectively. Surface water was dominated by the SO₄-Cl-Na and HCO₃-SO₄-Na types. The dominance of Na⁺ among surface water cations could be attributed to the dissolution of silicate minerals and rock salt minerals. In addition, factor 2 represented primary geological action [40], with an average contribution of 8.2%. Since Ca²⁺ (48.6%) and Mg²⁺ (49.2%) were the main ions of factor 3, this factor represented rock weathering, with an average contribution of 13.2%. TDS (72%) and K⁺ (39.3%) were the main ions of factor 4, with this factor dominating the hydrochemistry of surface water (72%).

Figure 10 shows the results of the analysis of the sources of chemical indicators of shallow groundwater. F⁻ was the characteristic ion (67.5%) of factor 1. Most of the F⁻ in shallow groundwater originated from the decomposition and release of fluorine-containing minerals, such as silicates (e.g., black mica, white mica, quartz, etc.). Therefore, factor 1 was significantly (p < 0.05) influenced by primary geological effects, with an average contribution of 35.2%. Factor 2 of shallow groundwater was regulated by Cl⁻, SO₄²⁻, HCO₃⁻, Ca²⁺, K⁺, Mg²⁺, and Na⁺, whereas groundwater was regulated by the SO₄-Cl-Na, SO₄-Na, HCO₃-SO₄-Na, and HCO₃-SO₄-Ca water types. This result could be mainly attributed to the susceptibility of shallow groundwater to anthropogenic disturbance and sources of recharge. The average contribution of factor 2 was 38.8%. The average contribution of factor 3 in deep groundwater was 0.5%, and this factor was dominated by NO₃⁻. Factor 4 showed weak ionic contributions from TDS and F⁻, with an average contribution of 25.6%.

Figure 11 shows the results of source identification of chemical indicators in shallow groundwater. The main ions contributing to factor 1 were TDS, Cl⁻, SO₄²⁻, Ca²⁺, K⁺, Mg²⁺, and Na⁺. Among them, NO₃⁻ showed the highest contribution. Previous studies have indicated that frequent industrial activities lead to an increase in NO₃⁻ [39]. Factor 1 represented human activities, with the highest average contribution to the hydrochemistry of shallow groundwater of 63.9%. F⁻ was the main (54.8%) characteristic ion of factor 2 and originated from the dissolution of silicate. Therefore, factor 2 represents primary geological action, with an average contribution of 0.7%. HCO₃⁻ and Ca²⁺ are the main ions of factor 3, whereas at the same time, shallow groundwater was dominated by the HCO₃-Ca water type. In addition, the hydrochemistry of shallow groundwater was influenced by the dissolution of gypsum and silicate as well as calcite. These results suggested that factor 3 mainly represented rock weathering, with an average contribution of 20.6%. The characteristic ions of factor 4 were F⁻ and K⁺, with the average contribution of factor 4 of 14.8%.

5. Conclusions

(1)

The rank of anions in surface water, shallow groundwater, and deep groundwater in the water bodies of the Parenta mining area during the wet period according to relative proportion of total anion mass was SO₄²⁻ > Cl⁻ > HCO₃⁻ > NO₃⁻; that of cations in surface water and deep groundwater was Na⁺ > Ca²⁺ > Mg²⁺ > K⁺; that of cations in shallow groundwater was Ca²⁺ > Mg²⁺ > Na⁺ > K⁺.

(2)

The hydrochemistry of surface water was mainly regulated by evaporative dissolution; that of shallow groundwater was mainly regulated by silicate; that of deep groundwater was mainly regulated by silicate hydrolysis and evaporite dissolution. Most surface water, shallow groundwater, and deep groundwater in the study area remained in unsaturated gypsum and rock salt states during the dry season, indicating the complexity of factors regulating the hydrochemistry of water bodies in the Pelianta area and the considerable influences of anthropogenic factors.

(3)

The results of the PMF model indicated four main sources of ions in surface water, shallow groundwater, and deep groundwater, namely agricultural activities, rock weathering, primary geology, and unknown sources. The hydrochemistry of shallow groundwater was mainly affected by rock weathering and primary geology, collectively contributing to 35.2% of ions, and human activities, contributing to 38.8% of ions. Rock weathering and human activity contributed to 20.6% and 63.9% of the ions in deep groundwater, respectively.