Hydrogeochemical Formation Mechanisms and Water Supply Suitability of Groundwater in the Northern Grassland Region on Inner Mongolia Plateau, China

ABSTRACT

Groundwater sustains the development of arid grasslands worldwide, yet its availability is increasingly threatened by human activities and climate change. This study takes the northeastern grassland on Inner Mongolian Plateau to get insights into the hydrochemical characteristics, formation mechanisms, and quality of groundwater resource in arid grasslands. The results show that groundwater in the study area is weakly alkaline and predominantly composed of HCO₃‐Ca, Cl‐Na, mixed HCO₃‐Na·Ca, and mixed Cl‐Mg·Ca facies. The hydrochemical evolution of the groundwater is primarily controlled by the dissolution of silicate minerals through water–rock interaction. Cation exchange also plays a crucial role in regulating the composition of major ions. Groundwater in the priority exploitation zones (the central area, eastern margin, and riparian zones of the two rivers) has relatively good quality, with 54.2% of samples had water quality suitable for direct consumption. The greatest susceptibility to noncarcinogenic risks was observed for infants, with nearly 30% of sampling sites falling into the high‐risk category for this subgroup. Meanwhile, NO₃ ⁻, F⁻, and Mn pose nonnegligible potential noncarcinogenic risks across all age‐gender groups. Although most samples were deemed suitable for agricultural irrigation, a subset from the northern hilly area was classified as unsuitable for long‐term use due to high sodium hazard levels. This research offers new understanding and a scientific basis for advancing sustainable groundwater management and drinking water safety protocols in similar agro‐pastoral ecotones.

Summary

• This study focused on a typical agro‐pastoral transitional zone of the arid grassland, providing systematic groundwater quality assessment for this ecologically fragile area.

• This study identified key pollutants (NO₃ ⁻, F⁻, and Mn) contributing to health risks, with infants being the most vulnerable subgroup.

• The assessment confirms that most of the groundwater is suitable for irrigation, while also warning of the potential risk of soil salinization in the northern hilly area.

• The results can serve the sustainable management of local groundwater and provide targeted scientific reference for drinking water safety in similar ecologically vulnerable regions.

This study constructs a systematic framework for investigating the hydrogeochemical formation mechanisms and water supply suitability of groundwater: hydrogeochemical formation mechanism identification via multiple models, drinking water quality assessment with the EWQI, health risk quantification of key contaminants, and irrigation suitability evaluation.

1. Introduction

Water resources represent a fundamental element that supports the functioning of the global life‐support system, and their sustainable use is essential for socioeconomic development and ecological stability (Karandish et al. 2025; Wang et al. 2025; Xiao et al. 2024). Accelerated aridification in semiarid regions, driven by global warming, disrupts the equilibrium of the hydrological system, creating a critical mismatch between water provision and need (Jia et al. 2026; Liu et al. 2025; Mo et al. 2025; Sardans et al. 2024). Ecologically fragile agro‐pastoral ecotones are particularly sensitive to such conditions because precipitation is highly variable and surface water dynamics are extremely unstable in these regions. In arid and semiarid regions, groundwater functions as a critical and growing resource for daily human consumption and irrigation in agro‐pastoral systems. This prominence is largely attributed to its widespread availability, considerable storage, and relative ease of abstraction (Karandish et al. 2025; Xiao, Zhang, et al. 2025; Zhu et al. 2025). Regional agricultural water demand in the agro‐pastoral ecotone of northern China has been increasingly amplified by cropland expansion and intensified land use, resulting in greater dependence on groundwater under conditions of variable precipitation and unstable surface water availability (Chen et al. 2023; Gao et al. 2025; Zhang et al. 2025). The attainment of regional sustainable development goals necessitates systematic assessment of groundwater security as an indispensable prerequisite.

Groundwater exploitation and management are constrained not only by water availability but also by hydrochemical composition and water quality safety. For drinking‐water purposes, safety is largely determined by the physical, chemical, and biological characteristics of water, which are directly linked to public health protection (Hao, Xiao, Cui, et al. 2025; Mukherjee et al. 2024). Global groundwater systems have been subjected to ongoing stress from rapid urbanization, population growth, and intensified agricultural production over recent decades, frequently leading to incidents of pollution by nutrients such as nitrate and ammonium nitrogen, as well as heavy metals (Hao, Xiao, Liu, et al. 2025; Hu et al. 2024; Lu et al. 2026; Qin et al. 2026). In addition to anthropogenic inputs, the chemical composition of groundwater is profoundly influenced by natural geochemical processes. Prolonged water–rock interactions in the subsurface environment can facilitate the mobilization and enrichment of hazardous constituents, including fluoride (F⁻), arsenic (As), and iodine (I), potentially leading to noncarcinogenic health risks among sensitive groups such as infants and children (Jiang et al. 2024; Oni et al. 2022). Accordingly, effective groundwater management and sustainable resource utilization depend on a comprehensive assessment of hydrogeochemical characteristics, genetic mechanisms, water quality conditions, and related health implications.

During the past decades, increasing attention has been directed toward groundwater quality and related health risks in arid and semiarid agro‐pastoral regions. The research focus has evolved from traditional hydrochemical analysis toward a broader framework involving contamination source apportionment, drinking‐water safety, irrigation suitability, and human health risk assessment (Ayangbenro and Babalola 2021; Feng et al. 2025; Xiao, Hu, et al. 2025; Zhang et al. 2024). Previous studies have shown that groundwater quality deterioration is jointly controlled by hydrogeological conditions, external contaminant inputs, and intrinsic geochemical processes. Shallow aquifers are generally more vulnerable to anthropogenic contamination associated with agricultural production and domestic wastewater, whereas confined aquifers are more likely to accumulate geogenic harmful elements under relatively isolated hydrogeochemical environments (Jia and Qian 2025). These aquifer‐specific differences demonstrate that groundwater risks depend heavily on both the hydrogeological setting and the water‐use purpose, necessitating an integrated and spatially explicit assessment approach. Nevertheless, irrigation suitability and drinking‐water health risks are still frequently assessed as separate issues, limiting the ability of existing evaluations to support differentiated water‐use planning and management (Garg and Arora 2025; Guo et al. 2026; Sharma et al. 2025). Moreover, the predominance of generalized regional‐scale assessments has often obscured the spatial heterogeneity of groundwater quality and risk distribution, thereby weakening the scientific basis for high‐risk area identification and targeted control measures (Jia et al. 2026).

Regarding the agro‐pastoral ecotone in northern China, existing studies have primarily focused on eco‐hydrological processes and total water resource assessments at large watershed scales (Chen et al. 2026; Pei et al. 2021; Xiao et al. 2026; Yang et al. 2025). Although localized hydrogeological surveys have provided some basic hydrochemical data, systematic and comprehensive evaluations of water quality and health risks that address the dual core needs of safe drinking water and agricultural irrigation water supply remain inadequate. Research on groundwater systems is particularly lacking in representative and ecologically critical areas such as the Inner Mongolia Plateau. This knowledge gap severely constrains the development of effective groundwater protection strategies and sustainable water supply management measures for this region and similar ecologically fragile zones.

Against this background, a typical semiarid agro‐pastoral ecotone on the northeastern Inner Mongolia Plateau was selected as the study area. The strong dependence of local agricultural and pastoral production on groundwater resources provides a suitable basis for investigating groundwater quality, resource suitability, and water supply security. This study aims to (1) characterize the hydrochemical characteristics of groundwater in the arid grassland area; (2) elucidate the controlling mechanisms of groundwater hydrochemical evolution; and (3) evaluate groundwater suitability for drinking and irrigation, and associated health risks. This work could provide scientific support and a valuable reference for the exploitation and sustainable management of groundwater resources in arid grasslands worldwide.

2. Study Area

The study area lies within a basin at the transition between the Hulunbeier Plain and the Greater Khingan Mountains on the northeastern Inner Mongolian Plateau (Figure 1), covering approximately 50 km² between 119°08′11″ to 120°17′04″ E and 49°06′40″ to 49°41′54″ N. Meadow steppe is the dominant vegetation type across the study area. Meadow steppe is the prevailing vegetation type in the study area, and the climate belongs to the midtemperate continental semiarid category.

FIGURE 1.

Location of (a) the Inner Mongolia Plateau, (b) the study area, and (c) groundwater sampling sites in the study area.

The central region corresponds to the basin that extends from the northeast to the southwest, whereas the southern part forms the Hailar River valley plain. Surrounding the area to the north, east, and west are hills composed of Mesozoic volcanic and Paleozoic metamorphic rocks. Inside the basin, the eastern part is distributed with high plains, and the western part is an alluvial plain. Elevation across the study area exhibits a general decline from north to south and from east to west, with a progressive decrease along the northeast‐to‐southwest gradient. Topographic and geomorphic conditions in this region are highly heterogeneous and complex.

The natural boundaries of the study area are defined by two perennial fluvial systems, namely, the Hailar River and the Morigele River. Both rivers share a common origin in the western piedmont of the Greater Khingan Mountains. The Hailar River flows westward across the study area's southern section, whereas the Morigele River extends along its western basin margin in a northeast‐southwest orientation. Collectively, the two rivers offer stable lateral recharge to groundwater and serve as the regional drainage base for the groundwater system.

The groundwater system exhibits high coupling with regional topography and geomorphology. This system is characterized by distinct horizontal zonation and features a layered water‐bearing structure developed across the basin interior. The Hailar River is one of the main runoff and enrichment zones of groundwater, hosting Quaternary loose rock pore phreatic water. Composed of alluvial, glaciofluvial, and alluvial‐lacustrine sandy gravel deposits, the study area aquifer has a thickness range of 30–50 m. Measured hydrogeological parameters show that single‐well water inflow is mostly above 1000 m³/day, indicating the aquifer has extremely high water richness. In addition, the water table depth is generally shallower than 3 m.

3. Research Methods and Materials

3.1. Sample Collection and Analytical Methods

Groundwater samples were collected from 24 boreholes across the study area in August 2025. The sampling network was spatially well distributed, providing broad coverage of the entire region while incorporating the principal hydrogeological units and areas potentially affected by anthropogenic activities. Standardized protocols were rigorously applied during sampling to guarantee both the representativeness of the samples and the reliability of the analytical results. Prior to sample collection, stagnant water in the borehole was pumped out until the effluent became clear with no visible impurities, and the high‐density polyethylene sample containers were rinsed at least three times with the in situ groundwater from each sampling site to eliminate cross‐contamination from external factors. All samples were preserved at 4°C in a portable constant‐temperature incubator with transport and analysis completed within 48 h to minimize compositional alterations from prolonged storage.

Groundwater physicochemical parameters were obtained through a combination of in situ measurements and laboratory analyses. Temperature (T), pH, and electrical conductivity (EC) were measured directly in the field using a portable multiparameter meter (Multi 350i/SET, Munich, Germany). Subsequent laboratory determinations were carried out at the laboratory. Major cations, including K⁺, Na⁺, Ca²⁺, and Mg²⁺, together with Mn, were analyzed by inductively coupled plasma mass spectrometry (ICP‐MS; Agilent 7500ce, Japan). Ammonium (NH₄ ⁺) and major anions, including Cl⁻, SO₄ ²⁻, NO₃ ⁻, NO₂ ⁻, and F⁻, were determined using ion chromatography (Shimadzu LC‐10ADvp, Kyoto, Japan). Bicarbonate (HCO₃ ⁻) and total dissolved solids (TDS) were quantified by acid–base titration and the gravimetric method, respectively.

Standard analytical procedures were strictly implemented throughout the testing process. All analyzed samples exhibited a percent charge balance error within the acceptable threshold of ± 5% (Adimalla 2021), ensuring the reliability of the hydrochemical data.

$$\%\mathrm{IBE}=\frac{\left(\sum \text{Cations}-\sum \text{Anions}\right)}{\left(\sum \text{Cations}+\sum \text{Anions}\right)}\times 100$$ (1)

3.2. Water Quality Index Assessment

By applying information entropy theory to assign objective weights to relevant parameters, the Entropy Weight Water Quality Index (EWQI) facilitates a comprehensive assessment of water quality (Su et al. 2018; Yang et al. 2023). Its core advantage lies in assigning weights based on the degree of data dispersion itself, effectively reducing the deviation caused by subjective weight setting (Adimalla et al. 2020). At the same time, it can integrate multiple types of indicators such as TH, TDS, and ion concentration, making it suitable for the comprehensive quality evaluation of groundwater in this study. The operational steps are as follows.

The methodological framework involves constructing the initial data matrix X (Formula 2), with dimensions m × n (sampling sites × water quality indicators), and its normalized counterpart Y (Formula 3):

$$X=\left[\begin{array}{ccc}\begin{array}{cc}{x}_{11}& x_{12}\\ x_{21}& x_{21}\end{array}& \cdots & \begin{array}{c}{x}_{1n}\\ x_{2n}\end{array}\\ ⋮& \ddots & ⋮\\ x_{m1}{x}_{m2}& \cdots & x_{\mathit{mn}}\end{array}\right]$$ (2)

$$Y=\left[\begin{array}{ccc}\begin{array}{cc}{y}_{11}& y_{12}\\ y_{21}& y_{22}\end{array}& \cdots & \begin{array}{c}{y}_{1n}\\ y_{2n}\end{array}\\ ⋮& \ddots & ⋮\\ \begin{array}{cc}{y}_{m1}& y_{m2}\end{array}& \cdots & y_{\mathit{mn}}\end{array}\right]$$ (3)

Here, the matrix element x _ij represents the measured concentration (or value) of indicator j obtained from sampling site i. And y _ij is the normalized value of x _ij to eliminate the influence of different indicator dimensions.

The term P _ij (Formula 4) denotes the proportion of the ith sampling site for the jth indicator, the information entropy e _j (Formula 5) of the jth indicator, and the corresponding entropy weight ω _j (Formula 6) are calculated sequentially as follows:

$$P_{\mathit{ij}}=\frac{y_{\mathit{ij}}}{{\sum }_{i=1}^m{y}_{\mathit{ij}}}$$ (4)

$$e_j=-\frac{1}{\ln m}{\sum }_{i=1}^m\left(P_{\mathit{ij}}\times \ln P_{\mathit{ij}}\right)$$ (5)

$${\omega }_j=\frac{1-e_j}{{\sum }_{j=1}^n\left(1-e_j\right)}$$ (6)

Next, the quality score q _j (Formula 7) of the jth indicator is obtained by standardizing the measured value X _j with the national groundwater quality standard limit S _j :

$$q_j=\frac{X_j}{S_j}\times 100$$ (7)

Finally, based on the determined quality ratings (q _j ) and entropy weights (ω _j ), these values were combined to yield the comprehensive EWQI for groundwater (Formula 8):

$$\mathit{\text{EWQI}}={\sum }_{j=1}^n\left({\omega }_j\times q_j\right)$$ (8)

3.3. Assessment of Human Health Risks

The principal pathways of human exposure to groundwater contaminants include oral intake through drinking water, dermal absorption during water‐related activities, and inhalation of volatilized contaminants (Karunanidhi et al. 2020; Price 2023). Extensive studies have supported the finding that oral ingestion of drinking water represents the primary exposure pathway for most noncarcinogenic contaminants, and groundwater quality is therefore a critical factor influencing long‐term human health and quality of life. To quantify the combined risk of multiple contaminants, the US Environmental Protection Agency (USEPA 2004) developed a standardized assessment framework based on the hazard index (HI), which has been widely validated and applied in groundwater health risk evaluation worldwide. Employing the aforementioned assessment framework, an evaluation was performed to quantify potential noncarcinogenic risks from groundwater, focusing on the dominant exposure route of drinking‐water ingestion. The inclusion of four sensitive population groups (infants, children, females, and males) ensures that risk estimates reflected community‐wide variations in susceptibility. The calculation process follows the standard steps below.

First, for the ith pollutant, the chronic daily intake (CDI _i ) via the oral drinking water pathway is computed according to Formula (9):

$${\mathit{CDI}}_i=\frac{C_i\times \mathit{IR}\times \mathit{EF}\times \mathit{ED}}{\mathit{BW}\times \mathit{AT}}$$ (9)

where each parameter is defined as follows:

• C _i : measured concentration of the ith pollutant in the studied groundwater (mg/L).

• IR: daily ingestion rate of drinking water (L/day).

• EF: annual frequency of exposure (days/year).

• ED: total duration of exposure (years).

• BW: average body weight of the exposed population (kg).

• AT: average cumulative exposure time (days) (Formula 10):

$$\mathit{AT}=\mathit{ED}\times 365$$ (10)

Next, the hazard quotient for the ith pollutant (HQ _i ) represents the ratio of the actual daily intake (CDI _i ) dose to the reference dose of the target pollutant (Formula 11):

$${\mathit{HQ}}_i=\frac{{\mathit{CDI}}_i}{{\mathit{RfD}}_i}$$ (11)

Herein, RfD _i represents the reference dose for oral exposure to contaminant i (Duan et al. 2014). Finally, the HI serves to assess the combined noncarcinogenic risk from all target pollutants and is derived through the summation of their respective hazard quotients (Formula 12):

$$\mathit{HI}=\sum _{i=1}^i{\mathit{HQ}}_i$$ (12)

HQ and HI values greater than 1 indicate that groundwater in the study area may present potential noncarcinogenic health risks to the relevant population groups. The magnitude of risk increases with increasing assessment values. The relevant parameter values for different groups are shown in Table 1.

TABLE 1.

Metrics for human health risk evaluation in potable water.

Parameter	Males	Females	Children	Infants	Index	RfD (mg·kg/day)
DI (L)	2.75 a	2.5 a	1.5 b	0.65 b	NO3 −	1.6
EF (days/year)	365	365	365	365	NO2 −	0.1
ED (years)	30 b	30 b	16.5	0.5	NH4 +	0.97
BW (kg)	72.42	62.26	36.66	9.27	F−	0.04
BH (cm)	169.22	157.66	132.22	70.44	Mn	0.046
AT (days)	10,950	10,950	6022.5	182.5
SA (cm2)	17,900	16,000	11,400	4200
EV (times/day)	1 c	1 c	1 c	1 c
CF	0.001 c	0.001 c	0.001 c	0.001 c

^a From Zhai et al. (2017).

^b From Su et al. (2013).

^c From Yang et al. (2012).

3.4. Agricultural Irrigation Suitability Evaluation

The primary economic activity in the study area is centered on animal husbandry. Secondary economic activities include the cultivation of grain and cash crops, which together form a considerable part of the local economy. This region relies on groundwater as its central water source, for which it provides drinking water for grazing livestock and irrigation for cultivated agricultural products. Livestock health and farmland productivity are both directly affected by, and are key determinants for assessing, groundwater quality. And inappropriate groundwater utilization for irrigation and drinking will not only restrict crop growth and degrade soil quality but also threaten the safety of livestock drinking water, eventually acting as a significant barrier to the sustainable progression of agriculture and animal husbandry in the area (Houatmia et al. 2016). To support the enduring sustainability of regional agriculture and livestock production, a comprehensive assessment of groundwater irrigation suitability is imperative.

Key determinants of groundwater suitability for irrigation include total salinity and the sodium‐to‐other‐cations ratio (Gharaibeh et al. 2021). This assessment employed an integrated approach to evaluate both salinity and sodium hazards, through the application of six widely recognized indices: sodium adsorption ratio (SAR), soluble sodium percentage (%Na), residual sodium carbonate (RSC), magnesium hazard (MH), permeability index (PI), and electrical conductivity (EC). The five core indicators were computed via standard equations. The specific significance and interpretive implications of each index are described in the following section, providing a systematic framework for groundwater quality evaluation.

SAR serves as an indicator of the relative capacity of sodium ions in irrigation water to exchange with soil cations. Higher values of SAR indicate a greater propensity for sodium to displace calcium and magnesium ions bound to soil colloids, increasing the risk of soil dispersion, structural degradation, and alkalization (Formula 13):

$$\mathit{SAR}=\frac{{\mathrm{Na}}^{+}}{\sqrt{\frac{{\mathrm{Ca}}^{2+}+{\mathrm{Mg}}^{2+}}{2}}}\times 100\%$$ (13)

%Na represents the relative abundance of sodium among the total cations in irrigation water. Elevated sodium concentrations reduce water infiltration rates by promoting soil particle dispersion and clay swelling, so %Na is a key indicator of potential sodium‐related structural degradation (Formula 14):

$$\%\mathit{Na}=\frac{{\mathrm{Na}}^{+}+{\mathrm{K}}^{+}}{{\mathrm{Ca}}^{2+}+{\mathrm{Mg}}^{2+}+{\mathrm{Na}}^{+}+{\mathrm{K}}^{+}}\times 100\%$$ (14)

RSC represents the surplus of CO₃ ²⁻ and HCO₃ ⁻ over Ca²⁺ and Mg²⁺ in irrigation water. Excess CO₃ ²⁻ and HCO₃ ⁻ precipitate Ca²⁺ and Mg²⁺ from solution, leaving sodium bicarbonate in soil solution, which alters soil structure and promotes soil hardening (Formula 15):

$$\mathit{RSC}=\left({\mathrm{HCO}}_3^{-}+{\mathrm{CO}}_3^{2-}\right)-\left({\mathrm{Ca}}^{2+}+{\mathrm{Mg}}^{2+}\right)$$ (15)

MH evaluates the negative effect of high relative magnesium concentrations on soil structure and crop growth. When magnesium accounts for a large proportion of exchangeable cations, it reduces soil aggregate stability, lowers infiltration rates, and increases soil pH and alkalinity risk (Formula 16):

$$\mathit{MH}=\frac{{\mathrm{Mg}}^{2+}}{{\mathrm{Ca}}^{2+}+{\mathrm{Mg}}^{2+}}\times 100\%$$ (16)

PI is defined by the collective, long‐term effects of Na⁺, Ca²⁺, Mg²⁺, and HCO₃ ⁻ on the permeability characteristics of soil. Chronic irrigation with water of unfavorable composition gradually alters soil exchangeable cation composition and reduces infiltration capacity over time, and PI captures this cumulative risk (Formula 17):

$$\mathit{PI}=\frac{{\mathrm{Na}}^{+}+\sqrt{{\mathrm{HCO}}_3^{-}}}{{\mathrm{Ca}}^{2+}+{\mathrm{Mg}}^{2+}+{\mathrm{Na}}^{+}}\times 100\%$$ (17)

4. Results and Discussion

4.1. Hydrochemical Characteristics of Groundwater

Statistical analysis evaluated the hydrochemical characteristics of groundwater, whereas violin plots (Figure 2) visualized the distributions of key water quality parameters. Subsequent comparative assessments referenced the allowable concentration limits of the Chinese National Standard for Groundwater Quality (GAQS 2017) and the World Health Organization guidelines (WHO 2022).

FIGURE 2.

Violin plots of physicochemical parameters of sampled groundwater in the study area: (a) pH, (b) EC, (c) TDS, (d) TH, (e) K⁺, (f) Na⁺, (g) Ca²⁺, (h) Mg²⁺, (i) Cl⁻, (j) SO₄ ²⁻, (k) HCO₃ ⁻, (l) NO₃ ⁻, (m) NO₂ ⁻, (n) NH₄ ⁺, (o) F⁻, (p) Mn.

The hydrochemical parameters of the sampled groundwater display distinct variations across the study area. pH values range from 6.4 to 8.0, averaging 7.2, indicative of a neutral to weakly alkaline condition (GAQS 2017). Electrical conductivity (EC) spans 225.0–3610.7 μS/cm, with a mean of 1207.9 μS/cm, reflecting significant variability in ionic content. Total dissolved solids (TDSs) vary between 144.0 and 3087.0 mg/L (average 936.2 mg/L), with 37.5% of samples exceeding the WHO (2022) limit of 1000 mg/L. Total hardness (TH) ranges from 59.2 to 1072.4 mg/L, averaging 328.6 mg/L; 20.8% of samples surpass the 450 mg/L threshold (WHO 2022), suggesting localized hardness concerns.

The major cations in groundwater are characterized by the following concentration ranges: Na⁺ (13.9–674.9 mg/L, mean 201.4 mg/L), Ca²⁺ (14.5–214.1 mg/L, mean 79.7 mg/L), Mg²⁺ (5.6–130.6 mg/L, mean 31.5 mg/L), and K⁺ (1.3–10.5 mg/L, mean 4.2 mg/L). Na⁺ is predominant, with 41.6% of samples exceeding the 200 mg/L limit. All Ca²⁺ values are below the 300 mg/L guideline, and only one sample surpasses the Mg²⁺ limit of 100 mg/L. Groundwater anion concentrations are characterized by a HCO₃ ⁻‐dominant composition (79.5–1164.2 mg/L, mean 355.7 mg/L), followed by Cl⁻ (3.0–8954.0 mg/L, mean 164.0 mg/L) and SO₄ ²⁻ (0.4–514.5 mg/L, mean 138.9 mg/L). Exceedances above the 250 mg/L guideline were observed for Cl⁻ and SO₄ ²⁻ at 16.7% of the sampling points; the overall anion abundance hierarchy is HCO₃ ⁻ > Cl⁻ > SO₄ ²⁻.

NO₃ ⁻ concentrations exhibited substantial variation, ranging from 0.1 to 130.7 mg/L (mean 20.5 mg/L). Only three samples (G7, G8, and G12) exceeded the safety threshold of 50 mg/L (GAQS 2017). NO₃ ⁻ concentrations in natural waters are generally considered to remain around 10 mg/L under background conditions (Xiao et al. 2026). However, elevated NO₃ ⁻ levels at several sampling sites (G3, G7, G8, G9, G12, G14, G19, and G24) indicate a clear influence of anthropogenic activities. Conversely, NO₂ ⁻ concentrations (0.0–2.0 mg/L; mean: 0.2 mg/L) exceeded the recommended limit of 0.02 mg/L at 66% of the sampling sites, indicating widespread contamination. NH₄ ⁺ concentrations range from 0.1 to 22.5 mg/L (mean 1.7 mg/L) and 58.3% of samples surpassing the 0.2 mg/L guideline. F⁻ and Mn also demonstrated notable exceedances, with three F⁻ samples (0.1–2.4 mg/L, mean 0.6 mg/L) above 1 mg/L and 33.3% of Mn samples (0.0–2.4 mg/L, mean 0.3 mg/L) exceeding the 0.1 mg/L limit.

4.2. Hydrogeochemical Types of Groundwater

As a classic graphical visualization method in the field of hydrogeochemistry, the Piper trilinear diagram (Piper 1944) has long been regarded as one of the most widely used tools for discriminating groundwater hydrochemical facies and tracking the evolution path of regional groundwater chemical composition. In this study, all sampling point data were projected onto the Piper trilinear diagram (Figure 3). In the left cationic triangular domain, Ca²⁺ and Na⁺ emerge as the dominant cationic constituents. In contrast, the right anionic triangular domain is predominantly characterized by HCO₃ ⁻‐dominated samples, with only a small fraction of data points extending into alternative anionic classification zones. Specifically, two samples occupy the SO₄ ²⁻‐dominated zone, and four samples reside within the Cl⁻‐dominated zone. The observed distribution pattern confirms a diversified anionic composition of groundwater.

FIGURE 3.

Piper plot illustrating the hydrochemical facies of sampled groundwater in the study area.

The integrated cationic and anionic information projected in the central diamond field indicates that the groundwater samples can be classified into four principal hydrochemical facies: HCO₃‐Ca, Cl‐Na, mixed HCO₃‐Na·Ca, and mixed Cl‐Mg·Ca. The predominance of HCO₃‐Ca (31.2%) and mixed HCO₃‐Na·Ca (27.1%) facies, which together account for 58.3% of the samples, suggests that most groundwater is characterized by a bicarbonate‐rich and weakly mineralized freshwater signature. This hydrochemical pattern reflects relatively active recharge conditions and is likely associated with carbonate dissolution, silicate weathering, and shallow water–rock interaction. The occurrence of Cl‐Na water in 22.9% of the samples indicates localized salinity enrichment in certain parts of the aquifer, potentially related to soluble salt dissolution, sluggish groundwater flow, and prolonged water–rock interaction. The mixed Cl‐Mg·Ca facies, comprising 18.8% of the samples, represents a transitional hydrochemical state between chloride‐enriched water and alkaline‐earth‐dominated water. Overall, this facies distribution highlights the combined control of recharge sources, flow‐path evolution, aquifer mineral composition, and water–rock interaction intensity on groundwater chemistry.

4.3. Mechanisms of Hydrogeochemical Evolution

Three core natural processes are widely recognized as the primary controls on groundwater chemical composition: evaporative concentration, water–rock interaction, and atmospheric precipitation recharge. To distinguish the dominant hydrochemical formation mechanism for a given region, the Gibbs (1970) diagram has been established as a standard graphical approach extensively applied in hydrogeochemical research. The key mechanisms governing regional water chemical characteristics were elucidated through the analysis of the groundwater Gibbs diagram (Figure 4a). The clustering of most samples within the rock dominance domain suggests that water–rock interaction represents the dominant control on groundwater hydrochemical evolution. The slight shift of several samples toward the evaporation dominance domain further indicates that evaporative concentration contributes a secondary and spatially localized effect.

FIGURE 4.

(a) Gibbs diagrams illustrating natural controlling mechanisms of groundwater hydrochemistry and (b) Mg²⁺/Na⁺ and Ca²⁺/Na⁺ end‐member diagram of sampled groundwater in the study area.

To further identify the types of parent rocks contributing ionic components during water–rock interactions, this study conducted an in‐depth analysis using the end‐member diagram of Mg²⁺/Na⁺ and Ca²⁺/Na⁺ ratios (Figure 4b). This approach facilitates evaluation of the relative contributions of carbonate, silicate, and evaporite weathering‐dissolution processes to groundwater composition, thereby clarifying their respective influences on groundwater chemistry (Gaillardet et al. 1999). The majority of groundwater samples fall within or adjacent to the silicate weathering field, demonstrating that silicate mineral weathering is the dominant natural process governing groundwater ion composition. However, several samples are positioned between the silicate and carbonate fields, implying that carbonate mineral dissolution may exert a secondary influence on the formation of groundwater chemistry. Additionally, a few samples are located within or near the evaporite dissolution field, indicating that local evaporite dissolution exerts a secondary superimposed influence on groundwater chemistry. The above‐mentioned spatial differentiation phenomenon of hydrochemistry is inferred to be closely related to the synergistic effects of multiple hydrogeochemical processes, such as the heterogeneous distribution of aquifer lithology, the multistage evolution of water–rock reaction paths, and cation exchange adsorption.

As a primary driver of water–rock interaction in aquifer systems, cation exchange plays a central regulatory role in both the genesis and the evolutionary modification of groundwater hydrochemical compositions (Xiao et al. 2018). Chloro‐alkaline indices (CAI) serve as quantitative markers for cation exchange processes within the study area, employing CAI‐1 and CAI‐2 (Formulas 18 and 19) to discern ion exchange reaction directions. As depicted in Figure 5a, approximately 87.5% of groundwater samples exhibit negative chloro‐alkaline index values, corroborating the extensive development of cation exchange between groundwater and aquifer media. Specifically, Ca²⁺/Mg²⁺ in groundwater displace Na⁺/K⁺ adsorbed on clay mineral surfaces, thereby enriching Na⁺/K⁺ and depleting Ca²⁺/Mg²⁺. These diagnostic patterns indicate that ion exchange plays an important role in regulating groundwater chemical composition and constitutes a key process in hydrochemical evolution.

$$\mathit{CAI}-1=\frac{{\mathrm{Cl}}^{-}-\left({\mathrm{Na}}^{+}+{\mathrm{K}}^{+}\right)}{{\mathrm{Cl}}^{-}}$$ (18)

$$\mathit{CAI}-2=\frac{{\mathrm{Cl}}^{-}-\left({\mathrm{Na}}^{+}+{\mathrm{K}}^{+}\right)}{{\mathrm{HCO}}_3^{-}+{\mathrm{SO}}_4^{2-}+{\mathrm{CO}}_3^{2-}+{\mathrm{NO}}_3^{-}}$$ (19)

FIGURE 5.

(a) Scatter plots of CAI‐1 versus CAI‐2 and (b) saturation states of major minerals in sampled groundwater in the study area.

The saturation index (SI) assesses the dissolution‐precipitation equilibrium state of major minerals and evaluates their potential impact on groundwater chemistry (Figure 5b). The SI analysis elucidates the distinct behaviors of different mineral groups. Among undersaturated minerals (SI < 0), gypsum, anhydrite, and fluorite exhibit SI values clustering near the equilibrium line (SI = 0), indicating dissolution is ongoing but approaching dynamic equilibrium. In contrast, halite shows SI values markedly below zero, reflecting a persistently undersaturated state and ongoing dissolution. Notably, rhodochrosite is predominantly undersaturated yet close to equilibrium, suggesting active dissolution nearing saturation in some waters.

For oversaturated minerals (SI > 0), the carbonate minerals aragonite, calcite, and dolomite commonly exhibit slight supersaturation (SI slightly > 0), indicative of potential limited precipitation or metastable equilibrium. Conversely, manganese oxides demonstrate a stronger precipitation tendency (hausmannite shows high SI values), whereas manganite and pyrolusite are distinctly supersaturated. Pyrochroite, though consistently supersaturated, remains close to the equilibrium line, implying its precipitation potential is limited by ambient hydrogeochemical conditions.

Furthermore, the Pearson correlation heatmap of groundwater hydrochemical indicators provides additional covariation evidence for the aforementioned hydrogeochemical processes (Figure 6). TDSs are extremely significantly positively correlated with EC at the 0.01 level and are also extremely significantly positively correlated with TH, Na⁺, Ca²⁺, Mg²⁺, Cl⁻, and SO₄ ²⁻, with correlation coefficients of 0.85, 0.91, 0.80, 0.84, 0.89, and 0.86, respectively.

FIGURE 6.

Correlation coefficients among hydrogeochemical indices of sampled groundwater in the study area.

Correlation analysis indicates that carbonate dissolution is an important contributor to groundwater chemistry, as reflected by the strong Ca²⁺ with Mg²⁺ correlation (r = 0.85) and the positive, although relatively weak, Ca²⁺ with HCO₃ ⁻ relationship (r = 0.37). The high Na⁺ and Cl⁻ correlation (r = 0.94) suggests a potential contribution from halite dissolution or evaporite‐related sources. Moreover, the moderate correlations between alkaline earth metals and Na⁺ indicate that cation exchange may have modified the relative distribution of major cations in groundwater. pH shows no significant strong correlation with most major ions, indicating its limited regulatory effect on the dominant mineral dissolution‐precipitation processes.

Overall, indicators of anthropogenic activity (including NO₃ ⁻, NO₂ ⁻, and NH₄ ⁺) show weak correlations with major ions. The weak and statistically insignificant correlation between NO₂ ⁻ and HCO₃ ⁻ indicates that nitrite is not closely associated with carbonate‐related geochemical processes. In contrast, the significant positive correlation between NO₂ ⁻ and NH₄ ⁺ at the 0.05 level suggests a potential linkage between reduced nitrogen species, possibly reflecting localized nitrogen transformation processes. Moreover, NH₄ ⁺ shows a significant positive correlation with Mn at the 0.05 level, implying that ammonium enrichment may be partly associated with reducing groundwater conditions. Overall, these relationships indicate that natural hydrogeochemical processes remain the dominant control on groundwater chemistry, whereas nitrogen‐related anthropogenic inputs appear to exert only localized or limited influence.

4.4. Health Implications of Drinking Water for Humans

Groundwater serves as an important source of domestic drinking water, and its quality is directly related to water supply security and public health. To comprehensively assess drinking‐water suitability and potential health impacts, TDS and TH are first used to classify overall groundwater quality. The EWQI is then applied to provide an integrated evaluation of groundwater quality status. On this basis, the HQ and HI framework is further introduced to quantify potential noncarcinogenic health risks associated with drinking‐water ingestion among different population groups.

The preliminary characterization of groundwater quality (Figure 7a) indicates that 37.5% of the samples belong to the soft‐fresh water category. The samples of other water quality categories collectively account for 62.5% of the total samples. Among them, five groundwater samples (G1, G3, G14, G20, and G22) constitute 20.8% of the total collection and are characterized as very hard brackish water.

FIGURE 7.

Scatter plots of (a) TDS versus TH and (b) EWQI versus TDS of sampled groundwater in the study area.

For the EWQI evaluation, 13 physicochemical indicators were selected after excluding K⁺ and HCO₃ ⁻. As shown in Figure 7b, EWQI values show substantial variation across the groundwater samples, with values ranging from 29.1 to 978.6. This wide range indicates highly heterogeneous groundwater quality conditions. Classified by the standard EWQI grading scheme, 29.2% of all sampling points fall into the excellent to good grades (Grades 1 and 2) with EWQI values below 100. Among the evaluated samples, 25% exhibit EWQI values between 100 and 150, thereby accounting for Grade 3 (medium water quality) classification. Collectively, water quality at 54.2% of sampling points meets the standard for domestic drinking water use. Notably, the proportion of poor and extremely poor‐quality groundwater (Grades 4 and 5, with EWQI values above 150) reaches 45.8%, meaning nearly half of the sampled groundwater is unsuitable for direct domestic use without proper treatment.

Notably, the EWQI and HQ/HI frameworks emphasize different aspects of drinking‐water safety. The EWQI provides an integrated evaluation of overall groundwater quality based on multiple physicochemical parameters, whereas the HQ/HI framework focuses on pollutant‐specific noncarcinogenic risks by considering toxicological thresholds and exposure parameters for different population groups. Therefore, the two assessments should be interpreted as complementary rather than interchangeable. Groundwater samples with relatively low EWQI values may still pose potential risks to infants if individual hazardous components such as NO₃ ⁻, NO₂ ⁻, NH₄ ⁺, F⁻, or Mn exceed risk‐relevant exposure thresholds. Conversely, samples with high EWQI values may mainly reflect elevated TDS or TH, which reduce overall drinking‐water suitability but do not necessarily correspond to high noncarcinogenic health risks.

All pollutants exceeding standard limits (NO₃ ⁻, NO₂ ⁻, and NH₄ ⁺), along with other detected target contaminants such as F⁻ and the heavy metal Mn, were incorporated into the assessment. The threshold for the HQ is set at 1, with values above this limit suggesting noncarcinogenic health risks and values below indicating negligible risk. From the overall HI results, the proportion of risk levels varies greatly among different age‐gender groups (Figure 8a). For infants, 29.1% of sampling points fall into the high‐risk category, 50% into the medium‐risk category, and only 20.9% into the low‐risk category. The risk distribution pattern across children, females, and males demonstrates a moderate predominance. The coverage analysis shows that high‐risk points account for 4.2% of the total, medium‐risk points for 54.2%, and low‐risk points for the remaining 41.6%.

FIGURE 8.

Box plots showing noncarcinogenic health risks: (a) total risk from multiple contaminants (HI) and hazard quotients for single contaminants including, (b) NO₃ ⁻, (c) NO₂ ⁻, (d) NH₄ ⁺, (e) F⁻, and (f) Mn in sampled groundwater in the study area.

When the HQ values of individual pollutants are analyzed, the risk characteristics of each component can be described as follows. For NO₃ ⁻ (Figure 8b), 12.5% of sampling points present high risk, and 20.8% present medium risk for infants, leaving 66.7% at low risk. For children and females, 4.2% of sites are high risk, and 12.5% are medium risk, with 83.3% of sites remaining at low risk. For males, 4.2% of sites are high risk, 8.3% are medium risk, and 87.5% are low risk, showing a gradual decrease in risk susceptibility from infants to adult males. For NO₂ ⁻ and NH₄ ⁺ (Figure 8c002Cd), only infants face 4.2% of medium‐risk sites, whereas all sampling points for all other population groups fall into the low‐risk category, indicating negligible noncarcinogenic risk from these two pollutants for most residents. For F⁻ (Figure 8e), infants face 4.2% high‐risk sites and 33.3% medium‐risk sites, with 62.5% of sites at low risk. The risk analysis encompassing children, females, and males indicate an absence of high‐risk sites, with 12.5% of locations exhibiting medium risk and 87.5% displaying low risk. For Mn (Figure 8f), 4.2% of sampling points are high risk, and 16.7% are medium risk for infants, whereas 79.1% remain low risk. For children and females, 16.7% of sites are at medium risk and 83.3% are at low risk. For males, 12.5% of sites are at medium risk, and 87.5% are at low risk.

Overall, infants are the most vulnerable population group to noncarcinogenic health risks associated with groundwater contaminants, as they account for the largest proportion of medium and high risk classifications. Among the assessed contaminants, NO₃ ⁻, F⁻, and Mn are the main contributors to noncarcinogenic health risks, with HQ values exceeding 1 at certain sampling sites and for specific population groups.

The HI exhibits a distinct spatial heterogeneity among the four exposed populations (Figure 9). In line with the risk classification framework: HI < 1 as low risk, HI between 1 and 4 as medium risk, and HI > 4 as high risk, the HI displays consistent regular spatial differentiation throughout the study area. Risk stratification across all population groups reveals that the central study area, eastern margin, and riparian zones along the two rivers consistently display lower HI values compared to other regions. The majority of low‐risk (HI < 1) sampling points are stratified and concentrated within these zones. High‐risk sampling points for infants (HI > 4) exhibit a primary concentration in the western study area. A small number of scattered high‐risk points also exist in the central and eastern regions, forming a triangular high‐risk distribution pattern. Within the midsouthwest quadrant of the research zone, a persistent risk nexus is evident for three distinct demographic groups (children, adult females, and adult males), suggesting that stable geological and/or anthropogenic influences drive the elevated noncarcinogenic groundwater risk. Consequently, prioritizing drinking‐water safety management in the western and central‐western regions is warranted.

FIGURE 9.

Spatial patterns of total noncarcinogenic health risk (HI) of groundwater for (a) infants, (b) children, (c) females, and (d) males in the study area.

4.5. Sustainability Assessment of Grassland Agriculture

Groundwater serves as the principal irrigation source for agriculture in the study area. Nevertheless, excessive salinity together with elevated concentrations of alkali ions such as sodium and magnesium can deteriorate soil physicochemical properties, impede crop nutrient acquisition, and ultimately constrain agricultural yields (Singh et al. 2024). For this reason, a systematic assessment of the irrigation suitability of local groundwater is of great practical guidance for the sustainable development of regional agriculture.

Evaluation of the six irrigation water quality indicators, according to the classification in Table 2. The majority of groundwater samples (91.7%) are rated as excellent in terms of SAR, with 8.3% classified as good and none falling into the doubtful or unsuitable categories, indicating the sodium hazard potential is predominantly low. For %Na, only 16.7% of samples are classified as excellent, 25% are good, and 20.8% are permissible, whereas 29.2% are doubtful, and 8.3% are unsuitable. Representing 37.5% of the total samples within the doubtful to unsuitable range, this finding underscores the elevated sodium‐alkali levels and the pronounced sodium hazard. The distribution of samples based on RSC criteria is predominated by the safe grade (37.5%), followed by good (29.2%), unsuitable (20.8%), and doubtful (8.3%), meaning more than 66% of samples are outside the safe range and pose potential harm to soil permeability. For MH, only 16.7% of samples are classified as desirable, whereas up to 83.3% of samples are classified as undesirable, indicating that magnesium content is generally high in groundwater of the study area, and long‐term irrigation easily causes soil magnesium alkalization, so most samples do not meet irrigation water quality requirements.

TABLE 2.

Assessment and classification of groundwater irrigation quality indices in the study area.

Indices	Range	Class	Percentage
SAR (meq/L)	< 10	Excellent	91.7%
	10–18	Good	8.3%
	18–26	Doubtful	\
	> 26	Unsuitable	\
%Na (%)	< 20	Excellent	16.7%
	20–40	Good	25%
	40–60	Permissible	20.8%
	60–80	Doubtful	29.2%
	> 80	Unsuitable	8.3%
RSC (meq/L)	< 0	Safe	37.5%
	0–1.25	Good	29.2%
	1.25–2.5	Doubtful	8.3%
	> 2.5	Unsuitable	20.8%
MH (%)	< 50	Desirable	83.3%
	> 50	Undesirable	16.7%
PI (%)	< 25	Unsuitable	\
	25–75	Moderately suitable	54.2%
	> 75	Suitable	45.8%
EC (μS/cm)	< 250	Excellent	4.2%
	250–750	Good	37.5%
	750–2250	Doubtful	45.8%
	> 2250	Unsuitable	12.5%

Based on the PI, all samples were classified as suitable for irrigation, with 54.2% moderately suitable and 45.8% fully suitable. These results suggest that the potential long‐term effects of groundwater on soil structural integrity within the study area are largely manageable. Analysis of groundwater samples for EC revealed that only 4.2% were excellent, whereas 37.5% were good. Consequently, only about 40% of the samples satisfy the salinity criteria required for high‐quality irrigation water; 45.8% of samples fall into the doubtful grade, and 12.5% are classified as the unsuitable grade, meaning more than half of samples exceed the salinity standard and have prominent salinity hazard risk. Comprehensive evaluation reveals that groundwater suitability is confined to satisfactory levels for SAR and PI. Conversely, parameters including %Na, RSC, MH, and EC show extensive noncompliance. The suitability for irrigation is primarily limited by sodium hazard, salinity hazard, and magnesium‐induced alkalization.

To integrate the information from all single indicators and generate a robust comprehensive assessment of irrigation suitability, the overall irrigation potential of groundwater in the study region was evaluated using three widely recognized graphical methods: the USSL diagram, the Wilcox diagram, and the Doneen diagram (Figure 10) (Yu et al. 2020).

FIGURE 10.

Groundwater irrigation suitability demonstrated by (a) USSL, (b) Wilcox, and (c) Doneen diagram in the study area.

As illustrated in Figure 10a, the USSL diagram classifies both salinity hazard and sodium hazard into four levels. Groundwater samples are distributed across C1–C4 salinity hazard classes, and the vast majority fall into the C2 (medium salinity) and C3 (high salinity) classes. The distribution of groundwater samples, with most clustered within the C2S1, C3S1, and C3S2 categories, reflects a prevailing medium to high salinity condition. Fewer samples were observed in the C1S1, C2S2, C3S3, C4S2, and C4S4 categories, indicating that extremely low or very high salinity‐sodium combinations are uncommon. This classification underscores that sodium hazards remain relatively low across the region, whereas only a minority of groundwater samples present conditions unsuitable for irrigation. The approach provides a clear framework for evaluating both salinity and sodium constraints in groundwater management.

Based on the integration of EC and %Na, the Wilcox diagram (Figure 10b) classifies irrigation suitability, providing clarification on overall groundwater feasibility. The classification of irrigation suitability reveals a predominance of excellent‐to‐good and good‐to‐permissible categories (87.5% of samples), contrasting with a minority (12.5%) in the doubtful‐to‐unsuitable range. This distribution indicates that, notwithstanding localized salinity and sodium concerns, the groundwater largely satisfies fundamental irrigation criteria.

For assessing the sustained impact of irrigation water on soil permeability, the Doneen diagram (Figure 10c) was employed as a more integrated method compared to single‐indicator analyses. Results show that every groundwater sample falls within Class I or II, corresponding to excellent and good long‐term irrigation suitability. Specifically, 83.3% of the samples are Class I, whereas 16.7% are Class II. Virtually, all groundwater samples meet the standard for sustainable agricultural use, confirming the satisfactory long‐term irrigation quality of the groundwater resources.

In general, the groundwater demonstrates moderate to good irrigation suitability across the study area, but it cannot be regarded as uniformly ideal. Despite the favorable performance of SAR and PI and the generally acceptable results from the USSL, Wilcox, and Doneen diagrams, localized limitations associated with salinity, sodium accumulation, and magnesium hazard remain evident.

4.6. Management Strategy Recommendations

Elevated concentrations of NO₃ ⁻, NO₂ ⁻, NH₄ ⁻, F⁻, and Mn were detected in groundwater samples, posing heightened health risks to infant populations. Concurrently, irrigation use is significantly constrained by salinity hazards, sodium hazards, and magnesium‐induced alkalinity risks. To address these dual challenges, implementation of a targeted management strategy (comprising zonal regulation, water‐use classification, source mitigation, and continuous monitoring) is recommended.

Delineation of the western and central‐western sections as high‐priority management zones is warranted given their pronounced health risks and drinking‐water safety challenges. Initial assessments should target self‐supplied wells and scattered water points in villages. Sources showing exceedances in NO₃ ⁻, F⁻, Mn, or composite risk indices should be limited, closed, or substituted, with safe drinking water guaranteed for sensitive subpopulations like infants. Pollution control efforts must prioritize domestic sewage, livestock effluents, nitrogenous fertilizers, and nonpoint contamination near riverine areas. Delineating safeguard zones around wells close to villages, farmlands, and rivers can prevent direct waste disposal, uncontrolled manure storage, and heavy fertilization. Recommended agronomic measures include optimized fertilizer dosing, staged application, and reduced preprecipitation inputs to minimize nitrate and ammonium infiltration.

Drinking water surveillance should mandate regular testing of TDS, TH, NO₃ ⁻, NO₂ ⁻, NH₄ ⁻, F⁻, and Mn. Noncompliant supplies warrant either decentralized treatment or integration into regulated distribution networks. In agriculture, regions with high EC, %Na, RSC, or MH are advised to adopt blended irrigation, crop‐water rotation, water‐efficient methods, and soil reclamation where needed. Wells with favorable chemistry should be preferentially used for irrigation of salt‐vulnerable crops such as vegetables and forage grasses. Implementation of a graded monitoring network (quarterly in critical areas and semiannually elsewhere) is essential to track water‐level and quality trends. Refocusing management on defining acceptable uses per water quality and spatial priority will advance groundwater safety and long‐term resource efficacy in the region.

5. Conclusions

A hydrochemical investigation was carried out on groundwater in the northeastern Inner Mongolia Plateau to systematically characterize its hydrochemical properties and water quality and further clarify its formation mechanisms in semiarid grasslands. The key findings are as below:

1. Groundwater is weakly alkaline and consists mainly of HCO₃‐Ca, Cl‐Na, mixed HCO₃‐Na·Ca, and mixed Cl‐Mg·Ca facies. HCO₃‐bearing waters, characteristic of bicarbonate‐rich freshwater, dominate the study area. The hydrochemical evolution of groundwater is primarily controlled by the dissolution of silicate minerals through water–rock interaction. Carbonate dissolution and localized evaporite dissolution exert secondary influences on groundwater chemistry. Cation exchange also plays a crucial role in regulating the composition of major ions. The saturation index analysis further reveals that evaporite minerals continue to dissolve, whereas the precipitation of carbonate minerals and manganese oxides is constrained.

2. Groundwater quality exhibits strong spatial variation. Approximately 54.2% of samples are suitable for direct use, whereas 45.8% require treatment before consumption. Infants show the highest noncarcinogenic health risk, whereas children and adults mainly experience moderate risk. NO₃ ⁻, F⁻, and Mn are the principal contributors. High‐risk zones are mainly located in the western part of the study area and at the northern and southern ends of the central region. Groundwater irrigation suitability is generally good, with more than 85% of samples classified as good to excellent. However, elevated sodium hazard occurs locally in the northern hilly region.

3. Elevated concentrations of NO₃ ⁻, NO₂ ⁻, NH₄ ⁻, F⁻, and Mn in groundwater pose significant health risks, particularly to infants. Additionally, high salinity and sodium levels severely limit the use of groundwater for irrigation. Consequently, a targeted management strategy is required, focusing on regional regulation, source control, and continuous monitoring. High‐priority zones should be established in the western and central‐western regions, with pollution control measures aimed at domestic sewage and agricultural runoff. Regular water quality monitoring and agronomic practices, including optimized fertilizer usage, are crucial for risk reduction. A graded monitoring network will help ensure long‐term groundwater safety and resource sustainability.

Author Contributions

Jiahao Liu: investigation, data curation, formal analysis, writing – original draft. Junting Liu: methodology, data curation, formal analysis. Yanlong Li: data curation, conceptualization, methodology, formal analysis. Yanxia Li: data curation, methodology, formal analysis. Qiulan Zhang: conceptualization, formal analysis, writing – review and editing. Jing Zhang: data curation, methodology, formal analysis. Jie Wang: investigation, formal analysis. Yong Xiao: conceptualization, formal analysis, writing – review and editing. Songtao Ma: investigation, formal analysis. Huizhu Chen: data curation, writing – review and editing. Wenxu Hu: investigation, data curation, formal analysis.

Funding

This research was funded by the Inner Mongolia Water Resources Basic Survey Project (33412024038); the MOE Key Laboratory of Groundwater Circulation and Environmental Evolution, China University of Geosciences (Beijing) (2023‐004); the National Natural Science Foundation of China (42477059); the New Interdisciplinary Cultivation Fund Project of Southwest Jiaotong University (2682026QZ032); the Innovative Practice Bases of Geological Engineering and Surveying Engineering of Southwest Jiaotong University (YJG‐2022‐JD04); and the Student Research Training Program of Southwest Jiaotong University (2025169).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data used to support the findings of the present study will be provided upon request by the corresponding author.