Multivariate Statistics and Hydrochemistry Combined to Reveal the Factors Affecting Shallow Groundwater Evolution in a Typical Area of the Huaibei Plain, China

Abstract

Understanding the characteristics of groundwater chemistry is essential for water resource development and utilization. However, few studies have focused on the chemical evolution processes of shallow groundwater in typical areas of the Huaibei Plain. We analyzed 28 water samples from the study area using hydrogeochemical mapping, multivariate statistical analysis, and other approaches. The study found that the hydrogeochemical facies of groundwater are mainly HCO₃-Ca·Mg (64.3%), mixed SO₄·Cl-Ca·Mg, and SO₄·Cl-Na. The hydrochemical composition is primarily controlled by natural water–rock interactions, including carbonate weathering and cation exchange processes. Correlation analysis and principal component analysis (PCA) revealed that mineral dissolution was the predominant source of Na⁺, Mg²⁺, Cl⁻, and SO₄²⁻ in shallow groundwater, significantly contributing to total dissolved solids (TDS) accumulation. Hierarchical cluster analysis (HCA) identified three characteristic zones: (1) agricultural/urban-influenced areas, (2) high-F⁻/low-hardness zones, and (3) nitrate-contaminated regions. These findings provide critical insights for assessing the geochemical status of groundwater in the Huaibei Plain and formulating targeted resource management strategies.

1. Introduction

Groundwater has a complex composition and transports elements through migration, dispersal, and enrichment processes across geological strata [1,2]. Groundwater undergoes chemical exchanges during interactions with various strata; therefore, hydrochemical studies enable a scientific assessment of groundwater resource dynamics and the rational development of groundwater resources [3]. When compared to medium and deep groundwater, shallow groundwater is not only easier to extract and primarily used for agricultural irrigation and decentralized rural water supply but also serves as a renewable resource for sustainable utilization due to its rapid participation in the hydrological cycle and is recharged primarily through atmospheric precipitation [1].

The Huaibei Plain is a resource-rich area with shallow groundwater serving as the primary water source for agriculture. Additionally, the Huaibei mining area is a major coal energy base in North China, with the earliest developed and most extensively exploited coal resources. Previous research on groundwater in the Huaibei Plain has primarily focused on five key areas. First, based on the monthly groundwater depth data of the observation wells and the monthly rainfall data of the meteorological station in the Huaibei Plain, the temporal and spatial variation of the shallow groundwater depth was studied [4,5]. Similar studies have been conducted in comparable regions, such as the Sanjiang Plain in Heilongjiang Province, China, and the arid and semi-arid plains of India [6,7]. Second, the interaction between agricultural activities and groundwater changes was investigated, such as the influence of groundwater depth on soil moisture in the shallow burial area of Huaibei Plain during the maize growth period [8], the impact of groundwater level fluctuations on drought propagation probability, growing cycles, and drought resistance capacities of winter wheat and summer maize [9], and the effects of intensive vegetable cultivation on the main ionic characteristics, dissolved organic matter and microbial community structure of groundwater [10]. Third, the impact of mining activities on groundwater evolution and microbial community structure has been explored [11,12]. Globally, studies on groundwater quality in coal mining areas have been conducted in India [13,14], while Ghana has focused on the effects of mining activities on groundwater characteristics [15,16]. Fourth, comprehensive analyses of the sources of fluorine and nitrate contamination in groundwater and their associated health risks were systematically evaluated [17,18,19,20]. Related studies have also been conducted in the Weibei Plain of Shandong Province, China [21], the Noyyal Basin in India [22], the coastal areas of Bangladesh [23], and southwestern Ghana [16]. Fifth, various statistical methods are used to determine the source, natural background levels, and threshold values of major ions in shallow groundwater [24]. However, systematic investigations into the controlling factors of shallow groundwater evolution in the Huaibei Plain remain inadequate. These knowledge gaps highlight the need for a systematic investigation of hydrochemical drivers in this region.

The chemical composition of shallow groundwater, which serves as water for domestic, agricultural, and industrial use, is influenced by a mixture of natural factors (e.g., recharge-runoff environments, geologic processes, and biochemical processes in the aquifer) and anthropogenic factors (including infiltration of contaminants from domestic, industrial, agricultural, and mining sources), posing challenges in interpreting hydrochemical signatures within extensive monitoring datasets [24]. Methods such as hydrogeochemical diagrams (e.g., Piper diagrams, Gibbs diagrams, and ion ratio analyses) and multivariate statistical analyses, which qualitatively reveal the factors influencing the hydrochemical characterization of groundwater and the probable origins of its chemical constituents, have been widely employed to evaluate hydrogeochemical processes and the geochemical evolution of complex systems at the plains scale [25]. Ion ratio analysis is one of the most widely used methods for determining the evolution of groundwater chemistry. Carbonate and sulfate minerals dissolve or precipitate, significantly altering groundwater chemical composition. In addition to mineral dissolution, groundwater components undergo ionic exchange processes, which dramatically affect the levels of key cations [11]. PCA, a multivariate statistical technique, enables dimensionality reduction to determine the source of parameter differences, and HCA can be used to determine whether samples can be classified into different populations (water chemistry groups), which may be important in a geologic or hydrologic context as well as from a statistical point of view [26].

Therefore, based on a regional hydrogeological investigation and sampling test work, combined with the multivariate statistical methods (e.g., PCA and HCA) and traditional hydrochemical analyses (including Gibbs diagrams and ion ratio analyses), this paper conducted research on the hydrochemical characteristics and evolution of shallow groundwater in typical areas of the Huaibei Plain, providing fundamental data for the study of hydrogeochemical evolution in the study area and offering critical insights for the sustainable management of groundwater in the study area.

2. Study Area

The study area spans from 116°40′00″ E to 117°00′00″ E longitude and 33°40′00″ N to 34°00′00″ N latitude, located on the alluvial plain region of the northeastern part of Huaibei City, Anhui Province, within the transitional zone of the Huang–Huai alluvial plain. Characterized by a warm temperate semi-humid monsoon climate, the region has a mean annual temperature of 11–14 °C. Annual precipitation ranges from 600–1000 mm, with 50–70% concentrated during July–September [19].

The area is dominated by Quaternary unconsolidated sediments, primarily composed of alluvial–pluvial sands, silts, and gravel layers with thickness increasing from northwest to southeast. A typical binary structure is observed: upper strata predominantly consist of silty clay and fine sands, while lower strata feature medium-coarse sands and gravel layers, forming a multi-layered porous aquifer system. The Quaternary aquifer basement, controlled by Paleogene basement structures, exhibits widespread carbonate and silicate weathering crusts. Localized evaporite remnants are observed at this interface, potentially influencing shallow groundwater chemistry.

The semi-closed groundwater system receives recharge primarily through vertical precipitation infiltration and lateral surface runoff. Seasonal rivers (e.g., the Sui River system) provide pulse-like recharge via channel leakage during flood seasons [4]. Groundwater flow direction, jointly controlled by regional topography and basement structures, displays an NW–SE pattern. Mineral–water interactions critically govern hydrochemical evolution: carbonate dissolution (calcite, dolomite) and silicate weathering (feldspar, quartz) dominate solute sources, while localized evaporite dissolution (gypsum, halite) elevates sulfate and chloride concentrations in groundwater stagnation zones.

3. Sample Collection and Analysis

The study area is characterized by a central urban zone in the northwest, where tap water has replaced groundwater, while hilly and mountainous regions lie to the north and east of the study area. Meanwhile, the plain features dense river networks and irrigation channels. Accordingly, 28 shallow groundwater samples (all from loose rock-type pore water) were collected, as shown in Figure 1. Sampling was conducted during March−April 2023 under non-pumping conditions. The sampling sites were primarily located in extensive farmland areas and farmers’ homes. Well depths showed a bimodal distribution: 14 wells at ~10 m (eastern sector) and 14 wells at ~20 m (western sector). The wells comprised two types: agricultural irrigation wells (50%) and domestic small-diameter wells (50%), exhibiting spatial intermingling. Functional classification showed 43% exclusive agricultural use (wheat/orchards), 32% domestic use (drinking/washing), 21% dual-purpose (vegetable irrigation/household), and one well (GW07) combining industrial sand washing with domestic use. Purification protocols included 20 min submersible pumping for irrigation wells and 5 min electric motor flushing for domestic wells. All samples were collected in triple-rinsed polyethylene bottles.

Water quality analysis was performed at the Laboratory of China Geological Survey, Nanjing Center. Cations (K⁺, Na⁺, Ca²⁺, Mg²⁺) and anions (Cl⁻, SO₄²⁻, F⁻, NO₃⁻) were quantified by inductively coupled plasma atomic emission spectrometry (ICAP 6300Duo, Thermo Fisher Scientific, Waltham, MA, USA) and ion chromatography (Dionex-2500, Thermo Fisher Scientific, USA), respectively [27,28]. HCO₃⁻ was measured using dual-indicator titration (phenolphthalein–methyl orange). All analyses met charge balance error criteria (±5%). The spatial distribution of hydrochemical parameters was mapped using ArcGIS 10.2 with the inverse distance weighting method [29]. In addition to the application of hydrogeochemical diagrams (Piper, Gibbs, ion-ratio), multivariate statistical analyses (correlation analysis, PCA, HCA) were also applied to reveal the hydrochemical evolution of groundwater. Correlation analysis was utilized to investigate potential linear correlations between different groundwater parameters [30]. PCA was subsequently performed to determine the possible factors that contribute to the overall chemical variance [31,32]. HCA was also conducted to link to groups of similar water chemistry [33]. Data processing utilized Origin (2021) and IBM SPSS 26.

4. Results and Discussion

4.1. Descriptive Analysis

Table 1. Statistical summary of hydrochemical parameters in shallow groundwater.

Parameters	Minimum Value	Maximum Value	Mean Value	Standard Deviation	Coefficient of Variation
K+	0.1	47.3	3.3	9.6	2.9
Na+	17.3	1170.0	180.7	238.7	1.3
Ca2+	27.8	286.8	126.1	64.5	0.5
Mg2+	24.1	262.7	69.9	54.0	0.8
Cl−	16.0	683.0	122.7	131.7	1.1
SO42−	38.1	2092.4	314.9	477.3	1.5
HCO3−	128.6	1043.1	610.7	204.9	0.3
F−	0.2	2.5	0.8	0.6	0.7
NO3−	0.0	219.0	30.9	57.0	1.8
TDS	512.8	4784.5	1149.0	898.6	0.8
TH	184.9	1533.8	604.0	267.3	0.4

Note: Concentrations of hydrochemical parameters are in mg/L.

presents the statistical summary of hydrochemical parameters in shallow groundwater from the study area. The average concentrations of the main cations follow the order: Na⁺ (180.7 mg/L) > Ca²⁺ (126.1 mg/L) > Mg²⁺ (69.9 mg/L) > K⁺ (3.3 mg/L), while major anions exhibit a descending order: HCO₃⁻ (610.7 mg/L) > SO₄²⁻ (314.9 mg/L) > Cl⁻ (122.7 mg/L) > NO₃⁻ (30.9 mg/L) > F⁻ (0.8 mg/L). Coefficients of variation exceeding 100% for K⁺, Na⁺, Cl⁻, SO₄²⁻, and NO₃⁻ (Table 1) indicate pronounced spatial heterogeneity. This variability likely originates from the semi-open nature of shallow groundwater systems, influenced by hydrometeorological conditions, industrial/mining activities, and agricultural practices [34]. The shallow groundwater in the study area occurs within loose Quaternary sediments, where significant regional variations in permeability and mineralogical compositions exist among different sediment types (e.g., sandy soils, silts, clays). The heterogeneous distribution of salt-bearing minerals (e.g., gypsum, halite) and K- and Na-bearing silicate minerals (e.g., feldspar) results in distinct weathering and dissolution capacities for releasing ions such as K⁺, Na⁺, Cl⁻, and SO₄²⁻. Highly permeable sandy layers facilitate ion migration and diffusion, whereas low-permeability clay layers hinder ion transport, leading to localized ion accumulation. NO₃⁻ primarily originates from agricultural fertilizer leaching, with groundwater concentrations in intensive agricultural zones (e.g., vegetable production bases) significantly higher than those in other areas. Industrial wastewater discharges (e.g., chemical manufacturing, paper production) contribute Cl⁻, SO₄²⁻, and Na⁺ to groundwater systems. Unregulated emissions of domestic sewage and landfill leachate further exacerbate localized enrichment of Cl⁻ and NO₃⁻.

Figure 2 illustrates the regional distribution of hydrochemical parameters. The northeastern sector exhibits the highest ion concentrations (except for F⁻ and NO₃⁻). Additional small-scale high-concentration zones for K⁺, Mg²⁺, and Cl⁻ occur in the southwest, central, and western regions, respectively. Ca²⁺ and HCO₃⁻ demonstrate more extensive spatial distributions, with Ca²⁺ enrichment observed in the northwest and southeast (besides the northeast), and HCO₃⁻ concentrations predominating in the western and central areas (besides the northeast). Elevated F⁻ levels cluster in the northern, central, and southwestern regions, while NO₃⁻ hotspots are localized in the northwestern and central sectors. The permitted threshold for TDS (1000 mg/L) is met by 60.7% of samples, predominantly located in the southern sector. As shown in Figure 3, TDS and total hardness (TH) values range from 512.8 to 4784.5 mg/L and from 184.9 to 1533.8 mg/L, respectively. The shallow groundwater is classified as fresh to brackish, with hardness categories primarily classified as hard (300–450 mg/L) and very hard (>450 mg/L) [1,35,36,37].

4.2. Hydrochemical Types and Controlling Factors

The Piper diagram, a graphical tool consisting of a diamond and two triangles, is widely employed for water chemistry classification [36]. Water quality data from the collected samples were plotted on the Piper diagram (Figure 4), enabling the interpretation of hydrogeochemical phenomena through spatial distribution analysis [38]. Cation triangle analysis reveals that shallow groundwater samples predominantly cluster in zone B, followed by zone A, classified as “non-dominant” and “calcium” types. The anion triangle indicates that most samples fall within zone E, categorized as the “bicarbonate” type. Synthesizing these distributions, 64.3% of sample points are concentrated in Zone 3, where the hydrochemical type is HCO₃-Ca·Mg, primarily distributed in the central part of the study area. This finding aligns with Qiu et al. (2023), who reported HCO₃-Ca·Mg dominance in groundwater chemistry in the adjacent Huaibei area [18]. Secondary types (SO₄·Cl-Ca·Mg and SO₄·Cl-Na) collectively account for 28.6%, predominantly located in the northeast and southwest regions. The observed diversity in hydrochemical types suggests significant groundwater vulnerability to anthropogenic influences [34].

The hydrochemical facies in the study area are predominantly characterized by HCO₃⁻, which is primarily controlled by five key factors: geological background, climatic conditions, hydrological cycle, geochemical environment, and limited anthropogenic influence. First, the study area contains significant limestone (CaCO₃) and dolomite (CaMg(CO₃)₂) in its strata. During groundwater flow through these carbonate rocks, dissolution processes release considerable amounts of HCO₃⁻. Second, the warm temperate semi-humid monsoon climate provides sufficient precipitation, while elevated temperatures and humidity enhance chemical weathering rates, thereby continuously replenishing HCO₃⁻ concentrations. Third, the groundwater system is characterized by slow runoff due to its relatively closed nature. This prolonged water–rock interaction in carbonate rock pores facilitates effective mineral dissolution while limiting the accumulation of competing anions (e.g., SO₄²⁻ and Cl⁻). Fourth, carbonate dissolution maintains groundwater at neutral to slightly alkaline conditions (pH 6.8–7.5, avg. 7.2), as documented by Zhang et al. (2021) [11], favoring HCO₃⁻ stability. Additionally, the limited availability of other anion sources (e.g., the absence of evaporite deposits) limits SO₄²⁻ concentrations. Fifth, remoteness from pollution sources or the presence of effective agricultural/industrial controls minimizes anthropogenic inputs of SO₄²⁻, Cl⁻, and NO₃⁻, enabling HCO₃⁻ to predominate through natural leaching processes.

The chemical evolution of groundwater is governed by three predominant factors: rock weathering, atmospheric precipitation, and evaporation [36,38]. Gibbs diagrams utilizing TDS vs. Na⁺/(Na⁺ + Ca²⁺) and TDS vs. Cl⁻/(Cl⁻ + HCO₃⁻) ratios were constructed to elucidate the principal mechanisms regulating groundwater chemical composition [39]. As shown in Figure 5, the hydrochemical characteristics of shallow groundwater in the study area are predominantly controlled by rock weathering processes (particularly silicate and carbonate dissolution), with secondary influences from evaporite dissolution (e.g., halite and gypsum) and minimal anthropogenic impact. The groundwater sample points cluster in the upper-middle section of the Gibbs diagram (Figure 5a), exhibiting a Na⁺/(Na⁺ + Ca²⁺) ratio ranging from 0.6 to 0.8. This pronounced sodium dominance over calcium likely originates from sodium-bearing mineral dissolution coupled with cation exchange processes [3,38]. Figure 5b reveals that only a limited number of samples display Cl⁻/(Cl⁻ + HCO₃⁻) ratios exceeding 0.6, with generally low chloride concentrations, corroborating the relatively insignificant influence of anthropogenic contamination on the shallow aquifer system [3].

To better understand hydrochemical processes and identify groundwater evolution, binary scatter plots of major ion ratios were analyzed. Carbonate weathering exerts a stronger control on shallow groundwater chemistry than silicate weathering, whereas the latter dominates over sulfate dissolution. End-member plots of Mg²⁺/Na⁺ vs. Ca²⁺/Na⁺ molar ratios were constructed to assess water–rock interactions of silicate rocks, carbonate rocks, and evaporite (e.g., sulfate minerals) influencing the hydrochemical characteristics of the study area [35,36]. Figure 6a demonstrates that over 40% of sampling points cluster between silicate and carbonate end-members, as well as near the end-members of silicate rocks, while fewer are proximal to evaporite sources (e.g., sulfate minerals dissolution), collectively indicating carbonate-silicate weathering dominance. To differentiate weathering dominance, a (Ca²⁺ + Mg²⁺)/(HCO₃⁻ + SO₄²⁻) ratio plot was generated [30,34]. Ratios > 1 suggest reverse cation exchange coupled with silicate weathering, whereas ratios < 1 indicate cation exchange coupled with carbonate weathering [38,40]. As shown in Figure 6b, more than 85% of samples cluster left of the 1:1 line, demonstrating that shallow groundwater chemistry is primarily controlled by carbonate rock weathering, with cation exchange significantly influencing cation content. This observation is reinforced by the Ca²⁺/SO₄²⁻ plot (Figure 6c), where more than 64% of samples plot in the lower right quadrant relative to the y = x line, indicating limited sulfate dissolution contribution. The observed Ca²⁺ excess likely originates from carbonate mineral dissolution [30,38]. This interpretation aligns with Zhang et al. (2021), who reported that the Ca²⁺ concentration in groundwater samples from the Huaibei Suixiao district exceeds SO₄²⁻ concentrations, further confirming carbonate mineral dissolution as the dominant hydrogeochemical process in this region [11].

In addition, binary scatter plots of (Ca²⁺ + Mg²⁺ − HCO₃⁻ − SO₄²⁻)/(Na⁺ + K⁺ − Cl⁻) reveal that cation exchange significantly influences groundwater chemistry, with the equivalent ratio of Na⁺ + K⁺ − Cl⁻ to Ca²⁺ + Mg²⁺ − HCO₃⁻ − SO₄²⁻ approaching −1 [34,36,38]. Figure 6d demonstrates a strong linear correlation (R = −0.97) between Na⁺ + K⁺ − Cl⁻ and Ca²⁺ + Mg²⁺ − HCO₃⁻ − SO₄²⁻, confirming cation exchange as a key regulator of shallow groundwater hydrochemical processes in the study area. The linear regression (y = − 1.06x − 0.97, slope ≈ −1) further supports the predominance of cation exchange in shaping groundwater hydrochemical characteristics [29]. The term Ca²⁺ + Mg²⁺ − HCO₃⁻ − SO₄²⁻ quantifies contributions of Ca²⁺ and Mg²⁺ from non-carbonate/sulfate sources, while Na⁺ + K⁺ − Cl⁻ represents Na⁺ inputs beyond atmospheric precipitation [37]. Figure 6d reveals elevated non-precipitation-derived Na⁺ in shallow groundwater, showing an inverse relationship between Na⁺ concentrations and Ca²⁺ + Mg²⁺ concentrations. Sample clustering below the 1:1 line suggests either Ca²⁺ + Mg²⁺ concentrations decrease or HCO₃⁻ + SO₄²⁻ concentrations increase during hydrochemical evolution. Integration with Figure 6b shows Ca²⁺ + Mg²⁺ concentrations exhibit a deficiency relative to SO₄²⁻ + HCO₃⁻. Given cation exchange affinity hierarchies where Ca²⁺ typically displaces Mg²⁺, our data indicate that Ca²⁺ is the principal exchange ion [11]. It is clear that Ca²⁺ in shallow groundwater replaces Na⁺ in soil particles via ion exchange, resulting in a decrease in Ca²⁺ and an increase in Na⁺ concentration. These mechanisms explain two key observations: despite (1) the broader spatial distribution of Ca²⁺ and (2) the predominance of calcium-type hydrochemical facies, the average Na⁺ concentration remains higher. These findings align with Zhang et al.’s (2021) work in the adjacent Huaibei area, which reported similar ion exchange dynamics [11].

The hydrochemical characteristics of shallow groundwater in the study area demonstrate comparatively limited anthropogenic influence, with industrial/mining activities exhibiting the most pronounced impacts, surpassing combined effects from domestic sewage and agricultural inputs. Samples significantly impacted by agricultural practices exhibit heightened NO₃⁻/Na⁺ and Cl⁻/Na⁺ molar ratios, whereas those influenced by domestic wastewater show depressed NO₃⁻/Na⁺ ratios [41]. As illustrated in Figure 6e, agricultural and domestic sources account for approximately 40% and 60% of anthropogenic impacts on shallow groundwater, respectively. Three key factors explain this observed equivalence: First, both sources introduce degradable pollutants (nitrates/pesticides vs. organics/pathogens) via comparable infiltration mechanisms (surface runoff/soil infiltration). Second, their diffuse spatial distributions hinder centralized remediation efforts—agricultural pollution shows rainfall/irrigation-dependent dispersion patterns, while domestic sewage in rural areas lacking drainage networks manifests decentralized discharge characteristics. Third, natural attenuation processes (soil adsorption vs. microbial degradation) produce similar magnitude groundwater impacts, despite differing pollutant compositions. The relationship between NO₃⁻/Ca²⁺ and SO₄²⁻/Ca²⁺ (milligram equivalent ratio) was employed to assess relative anthropogenic contributions to aqueous nitrate concentrations. When SO₄²⁻/Ca²⁺ exceeds NO₃⁻/Ca²⁺, industrial/mining influences predominate, whereas agricultural/domestic sources prevail under inverse conditions [37]. Notably, Figure 6f reveals that 64% of sampling sites are concentrated within the SO₄²⁻/Ca²⁺ < 1 and NO₃⁻/Ca²⁺ < 1 ranges, corroborating the finding of limited anthropogenic impacts. This distribution further substantiates that industrial/mining activities exert greater influence on shallow groundwater compared to agricultural/domestic sources. Three synergistic mechanisms underlie this predominance: (1) acid mine drainage from coal mining introduces sulfates, iron, and manganese, thereby lowering groundwater pH and triggering heavy metal dissolution; (2) mining-induced subsidence fractures aquifer aquitards, facilitating vertical contaminant transport; and (3) industrial and mining pollution sources exhibit distinct spatial clustering patterns, with wastewater discharge volumes substantially surpassing those from agricultural non-point sources.

4.3. Multivariate Statistical Analysis

The Pearson correlation analysis provides valuable insights into hydrogeochemical processes within the aquifer system, as significant correlations between parameter pairs may arise from mineral dissolution or precipitation processes [30]. Figure 7 shows the Pearson correlation matrix of the main chemical parameters of the shallow groundwater of the study area. Strong correlations were observed between Na⁺ and SO₄²⁻ (r = 0.95), Mg²⁺ and Cl⁻ (r = 0.67), and Mg²⁺ and SO₄²⁻ (r = 0.60), indicating a common source likely derived from mineral dissolution. Consistent with Wang et al. (2023), the Na⁺ and SO₄²⁻ concentrations are attributed to the dissolution of mirabilite (Na₂SO₄·10H₂O) [30]. The weak Ca²⁺-SO₄²⁻ correlation (r = 0.47), combined with corresponding binary scatter plot analysis (Figure 6c), implies dominant carbonate mineral dissolution over sulfate mineral dissolution. TDS exhibited significant positive correlations with Na⁺ (r = 0.97), Mg²⁺ (r = 0.71), Cl⁻ (r = 0.95), and SO₄²⁻ (r = 0.97), with weak associations observed for K⁺ (r = 0.36), HCO₃⁻ (r = 0.14), and F⁻ (r = 0.02), which demonstrates that ionic dissolution predominantly governs TDS enrichment [30].

PCA was performed to examine the interrelationships among groundwater hydrochemical parameters and identify their potential sources (

Table 2. Principal component and varimax rotated component matrix values of chemical components in shallow groundwater.

Variables	Component Matrix				Rotated Component Matrix
	PC1	PC2	PC3	PC4	PC1	PC2	PC3	PC4
K+	0.32	−0.68	0.48	−0.06	0.31	0.82	0.11	0.09
Na+	0.94	0.09	0.24	−0.02	0.96	0.10	−0.11	−0.11
Ca2+	0.47	−0.73	−0.37	−0.09	0.37	0.33	0.81	0.07
Mg2+	0.75	0.55	−0.12	0.22	0.79	−0.52	−0.19	0.02
Cl−	0.96	0.00	−0.02	−0.01	0.94	0.00	0.14	−0.08
SO42−	0.96	−0.14	0.20	−0.07	0.95	0.26	0.07	−0.10
HCO3−	0.18	0.82	−0.36	0.08	0.21	−0.85	−0.22	−0.15
F−	−0.02	0.67	0.66	0.12	0.10	−0.10	−0.94	−0.05
NO3−	−0.17	−0.43	0.04	0.88	−0.11	0.17	0.08	0.97
TDS	0.99	−0.01	0.10	0.02	0.99	0.08	0.06	−0.04
TH	0.91	0.02	−0.32	0.13	0.88	−0.23	0.33	0.06
Eigenvalues	5.47	2.64	1.15	0.88	5.38	1.96	1.78	1.01
Variance(%)	49.72	23.98	10.49	7.98	48.93	17.84	16.19	9.21
Cumulative(%)	49.72	73.70	84.19	92.17	48.93	66.77	82.96	92.17

) [42]. Four principal components were extracted based on eigenvalues > 1 and a cumulative variance percentage exceeding 85%. These four principal components accounted for 92.17% of the variance in the 11 groundwater variables. The first principal component, PC1, accounted for 49.72% of the variance and was positively correlated with Na⁺, Mg²⁺, Cl⁻, SO₄²⁻, TDS, and TH. These ions exhibited high loadings in PC1 and are typically associated with natural minerals, geomorphic components, and anthropogenic activities, which influence groundwater complexity [31,33]. PC1 demonstrated strong positive loadings for Na⁺ (0.96), Cl⁻ (0.94), SO₄²⁻ (0.95), and TDS (0.99), suggesting that their dissolution primarily governs TDS increases in groundwater. The elevated concentrations of these ions may be attributed to mineral dissolution and anthropogenic inputs. Elevated Mg²⁺ concentrations were associated with increased groundwater TH [31,32,33,42]. PC2, PC3, and PC4 accounted for significantly less variation in the chemical composition of the groundwater samples compared to PC1. PC2 (17.84% variance) showed a strong positive loading for K⁺ (0.82) and a negative loading for HCO₃⁻ (−0.85), attributable to silicate mineral weathering processes [32]. Ca²⁺ showed a significant positive correlation with PC3, while F⁻ had a negative association with PC3 and a strong negative correlation with Ca²⁺. This suggests that F⁻ enrichment in groundwater may have a facilitating or inhibitory effect [32]. Agricultural fertilizer use (e.g., urea and NPK fertilizers) on flat farmland could enhance the leaching of K⁺ and NO₃⁻ into groundwater, explaining the notable positive association of K⁺ with PC2 and NO₃⁻ with PC4 [31,33].

HCA was applied to categorize water samples from similar monitoring sites based on their chemical composition (Figure 8) [33]. Group G1 samples, distributed in the central, northeastern, and southwestern parts of the study area, displayed the lowest average concentrations of Ca²⁺ (50.4 mg/L) and Mg²⁺ (68.8 mg/L). The corresponding average hardness (410.1 mg/L) was the lowest while containing the highest average F⁻ concentration (1.6 mg/L). Group G2 samples, located in the northeastern sector, demonstrated notably high TDS (mean > 2400 mg/L). Concentrations of Na⁺, Cl⁻, SO₄²⁻, F⁻, and TH exceeded permissible limits, suggesting significant impacts from agricultural activities and wastewater discharges [33]. Group G3 exhibited high NO₃⁻ (97.7 mg/L) and low Cl⁻ (56.7 mg/L) concentrations, primarily located in the southeastern region and along the Sui River. Group G1-G4 exhibited significant associations with SO₄²⁻, K⁺, F⁻, and NO₃⁻, which could be attributed to the excessive use of fertilizers and pesticides in agricultural practices, leading to their localized enrichment in groundwater. For example, Chorol et al. (2023) identified sulfate sources as compounds of agricultural fertilizers [33]. These findings reflect the extensive agricultural activities within the study area. Further analysis of fluoride and nitrate anomalies revealed the following: (1) Elevated (Na⁺ + K⁺)/Ca²⁺ ratios in fluoride-rich G1 samples (Figure 8c) suggest cation exchange processes favoring NaF solubility over CaF₂ precipitation [20]. (2) G3 cluster samples demonstrated significantly higher NO₃⁻/Cl⁻ molar ratios (Figure 8d), indicating predominant nitrate inputs from nonpoint sources such as soil nitrogen and fertilizer leaching [43,44,45].

5. Conclusions

The geochemical evolution of shallow groundwater in the Huaibei Plain remains a critical research priority. In this study, multivariate statistical analysis integrated with hydrochemical methods was employed to elucidate the dominant geochemical processes and ion sources governing groundwater evolution through analysis of 28 collected water samples.

The predominant hydrochemical type in the study area was HCO₃-Ca·Mg. Among cations, Ca²⁺ exhibited the widest spatial distribution, while Na⁺ emerged as the dominant cation overall. This phenomenon was attributed to cation exchange between Ca²⁺ in groundwater and Na⁺ adsorbed on clay minerals.

Carbonate weathering exerted a stronger influence on hydrochemical properties than silicate weathering, with both processes outweighing sulfate dissolution and anthropogenic inputs in intensity. Further analysis revealed that industrial and mining activities were more significant contributors to anthropogenic impacts than domestic sewage and agricultural activities, while the difference between the latter two was not statistically significant.

Correlation analysis and PCA indicated that mineral dissolution was the primary source of Na⁺, Mg²⁺, Cl⁻, and SO₄²⁻, directly elevating TDS. HCA identified the northeastern zone as highly susceptible to agricultural and domestic sewage influences. The central, northeastern, and southwestern areas exhibited elevated F⁻ concentrations coupled with low hardness. Elevated nitrate levels in the southeastern region and along the Sui River were attributed to non-point sources, including soil nitrogen and fertilizer leaching.

These findings provide policymakers with a scientific foundation for sustainable groundwater development, management, and protection.