Traceability of Phreatic Groundwater Contaminants and the Threat to Human Health: A Case Study in the Tabu River Basin, North China

Abstract

Groundwater is the main clean water resource in northern China, and its quality is critical for both human health and social sustainable development. Due to complex anthropogenic and/or geogenic processes, the sources of groundwater contaminants are not easy to determine. The Tabu River Basin, located in northern China, is an agriculture and pasture interlaced area in which phreatic groundwater is the predominant water resource for domestic and agricultural purposes. Groundwater with abnormally high levels of NO₃⁻, F⁻, and TDS was observed here based on 87 groundwater samples collected from the phreatic aquifer in 2022. In this study, hydrogeochemical and isotopic methods were used to trace groundwater contaminants in the phreatic aquifer, and a risk assessment was conducted to analyze their threat to human health. The results indicated that NO₃⁻ in the phreatic groundwater primarily originated from manure, the high concentration of TDS was highly associated with irrigation, and the enrichment of F⁻ was mainly controlled by geogenic factors, including alkaline condition, competitive adsorption, the dissolution of fluorine-bearing minerals, and cation exchange. A principal component analysis (PCA) showed that both anthropogenic (PC1, 50.7%) and geogenic (PC2, 19.9%) factors determined the quality of the phreatic groundwater in the study area. The human health risk assessment demonstrated that 98.9%, 92.0%, and 80.5% of the groundwater samples exceeded the permissible limit of the total noncarcinogenic risk for children, adult females, and adult males, respectively. The monitoring results from 2022 to 2023 suggested that phreatic groundwater contamination could not be mitigated through natural attenuation under the existing external pressures. Measures need to be taken to decrease the contamination of phreatic groundwater and enhance the groundwater sustainability in the Tabu River Basin. The findings of this study can provide a reference for sustainable groundwater development in the Tabu River Basin and other arid and semi-arid regions worldwide.

1. Introduction

Clean groundwater is a crucial resource for industrial, agricultural, and domestic purposes in arid and semi-arid regions, influencing human health and social sustainable development [1,2]. Moreover, it serves as a significant supply source for rivers, lakes, and wetlands, thereby maintaining ecological balance [3]. However, the deterioration of groundwater quality has considerably reduced the availability of water resources, resulting in a global sustainability concern [4]. A comprehensive understanding of hydrogeochemical characteristics and contamination status is crucial for the protection of groundwater resources [5,6].

Anthropogenic activities, such as agriculture and industry, significantly contribute to the contamination of groundwater on a global scale, increasing the levels of various contaminating factors in groundwater including salinity, nitrogen, etc. [4]. There has been a growing focus on geogenic-contaminated groundwater in recent years [7,8,9]. Geogenic-contaminated groundwater refers to groundwater that contains abnormally high concentrations of hazardous substances derived from hostrocks or aquifer matrices. These hazardous substances are enriched through complex geological and hydrogeochemical processes [9,10]. For instance, geogenic factors, such as fluoride-bearing minerals dissolution, cation exchange, and alkaline conditions, can lead to the occurrence of high F⁻ groundwater [11,12]. Moreover, the co-occurrence of geogenic and anthropogenic contaminants in groundwater has been reported in various locations, such as in Rajasthan in India [13], the Bist-Doab region of Punjab in India [14], and the Pearl River Delta in China [15]. However, knowledge regarding the driving forces that control the multiple contaminations of groundwater is still limited due to complex hydrogeochemical conditions and anthropogenic activities.

It is crucial that we distinguish the effects of geogenic and anthropogenic factors on hydrogeochemical composition in order to develop a comprehensive understanding of the geochemistry and quality of groundwater [16]. Generally, contaminants from various sources exhibit significant variations in the hydrogeochemical parameters and isotopic composition, which can serve as reliable indicators for tracing their origins [17,18]. For example, chemical fertilizers exhibit high NO₃⁻/Cl⁻ ratios, a low Cl⁻ content, and high δ¹⁸O-NO₃⁻ values, whereas manure and sewage are characterized by low NO₃⁻/Cl⁻ ratios, a high Cl⁻ content, and low δ¹⁸O-NO₃⁻ values [6]. Additionally, multivariate statistical analyses, such as principal component analysis (PCA) and hierarchical cluster analysis (HCA), are robust methods for analyzing the sources of contaminants in groundwater [19,20]. HCA can assess the similarities and differences among the samples [21]. PCA is commonly employed to reduce data dimensionality and identify the primary factors influencing water quality [22]. Therefore, utilizing both HCA and PCA could enhance the interpretation of hydrogeochemical parameters and isotopic composition.

The primary concerns regarding contaminated groundwater are the potential health risks to individuals associated with long-term exposure [5,6]. For example, the consumption of water with elevated NO₃⁻ levels can lead to a reduced oxygen-carrying capacity, resulting in health issues, such as blue baby syndrome [23] and thyroid disease [24]. At lower F⁻ concentrations, water consumers are susceptible to dental caries, while, at concentrations higher than the stipulated levels, it can lead to dental fluorosis, skeletal fluorosis, crippling fluorosis, and adverse effects on the kidneys [25,26]. Exceeding the pollution limit for a specific contaminant in groundwater does not necessarily pose a threat to human health, as the potential risk depends on various factors such as exposure pathways and target population [27]. In addition, the concentration of contaminants in groundwater below the pollution limit does not guarantee the absence of threats to human health, as the assessment of health risks should consider all hazardous substances and multiple exposure pathways [28]. Therefore, conducting a comprehensive human health risk assessment associated with groundwater quality is crucial for the management of water resources and the protection of human health.

The Tabu River Basin is situated in an agriculture and pasture interlaced area of the North Yinshan Mountain, which serves as a national ecological barrier for northern China [29]. Phreatic groundwater is the primary water source for domestic and agricultural purpose in the Tabu River Basin, playing a crucial role in social sustainable development [30]. This phreatic groundwater is primarily found in quaternary deposits with good permeability, such as sandy gravels, pebbles, and sands, and is buried at a shallow depth ranging from 1 to 5 m. The long-term cultivation of crops and livestock poses a significant threat to the groundwater quality in the Tabu River Basin due to the shallow burial depth and high permeability of phreatic aquifers. In addition, the co-occurrence of arsenic, fluoride, and TDS in groundwater due to geogenic factors has also been observed in the Yinshan Mountain region [8]. Therefore, it can be inferred that both geogenic and anthropogenic factors can affect the groundwater quality in the Tabu River Basin. Investigating the hydrogeochemical characteristics, quality, and human health risks of groundwater is crucial for the exploitation and utilization of groundwater resources in the Tabu River Basin, which, to the best of our knowledge, has not received attention yet.

From the perspectives of geological conditions and external factors, the Tabu River Basin is an ideal case study for investigating hydrochemistry and groundwater pollution under the influences of anthropogenic and geogenic factors. Therefore, hydrogeochemical and isotopic analyses coupled with multivariate analytical methods were used to elucidate the factors influencing phreatic groundwater contamination in the Tabu River Basin. The objectives of this study were to: (1) investigate the hydrogeochemical characteristics of phreatic groundwater in the Tabu River Basin; (2) identify the spatial patterns of phreatic groundwater contaminants; (3) clarify the geogenic and anthropogenic factors regulating phreatic groundwater contamination; and (4) assess the human health risks associated with contaminated phreatic groundwater through multiple exposure pathways.

2. Materials and Methods

2.1. Study Area

The Tabu River Basin is located in the northern foothills of the Yin Mountains in the Inner Mongolia Autonomous Region (Figure 1a). The terrain slopes from the south to the north, with elevations ranging from 900 to 2100 m. The Tabu River flows from the southwest to the north. The study area has an arid and semi-arid climate with low rainfall and high evaporation rates. The average annual rainfall is 311.2 mm, and the majority of this occurs between July and September, accounting for approximately 80% of the total annual rainfall. The average annual evaporation is 1365.2 mm, and the average annual temperature ranges between 1 and 6 °C. The concentrated rainfall and extensive evaporation have led to prolonged drought conditions for most surface water. Meanwhile, the majority of surface water in the research area is mainly replenished by groundwater, as it remains dry for long periods except during the rainy season. The primary land use types in the study area are grassland and farmland, and industrial activities are limited in this area (Figure 1b). The grassland is primarily located downstream of the Tabu River, whereas farmland is mainly found in the middle and upper reaches (Figure 1b). Livestock (cattle and sheep) and crop production are integrated at the household level in the Tabu River Basin. In this region, livestock manure is commonly used as a plant fertilizer after qualified preparation. Due to the lack of surface water, groundwater acts as the primary water source for agricultural activities in the grassland and farmland of the Tabu River Basin. Therefore, the management of groundwater resources is crucial for sustainable development in the study area.

The precipitation, evaporation, and temperature data were acquired from the National Data Center for Meteorological Sciences (http://data.cma.cn/, accessed on 1 December 2023). The land use data for 2020 were obtained from the Data Center for Resources and Environmental Sciences at the Chinese Academy of Sciences (http://www.resdc.cn/, accessed on 1 December 2023).

A detailed hydrogeological description can be found in our previous study [30]. Briefly, phreatic groundwater is mainly located in quaternary deposits with good permeability. The burial depth of the phreatic groundwater approximately ranges from 1 to 5 m, and the thickness of the phreatic aquafer approximately ranges from 10 to 20 m. The quaternary deposits mainly consist of sandy gravels, pebbles, and sands. Groundwater flows from the south to the north. Rivers serve as the primary discharge pathway for groundwater, while precipitation acts as the main source of groundwater recharge in the study area [30]. However, the discharge from river to groundwater is negligible, as the Tabu River is generally dry, aside from the flood season.

2.2. Sampling and Analysis

A total of 87 samples of phreatic groundwater (with a burial depth ranging from 0.75 to 22.15 m;

Table 1. Statistical summary of hydrogeochemical and isotopic parameters of groundwater samples and the saturation index of the selected minerals.

Parameter	Min	Max	Median	SD
K+ (mg/L)	0.8	272.6	2.6	29.2
Na+ (mg/L)	8.9	1130.0	76.0	166.1
Ca2+ (mg/L)	23.5	565.9	85.3	108.8
Mg2+ (mg/L)	11.5	292.4	39.2	51.6
Cl− (mg/L)	17.7	1156.0	98.4	190.9
SO42− (mg/L)	17.9	974.7	100.5	155.5
HCO3− (mg/L)	174.0	716.0	293.1	128.9
F− (mg/L)	0.2	4.6	0.8	1.0
NO3− (mg/L)	0.0	1274.0	81.3	308.0
pH	7.37	8.66	8.10	0.29
TDS (mg/L)	278.0	5398.5	712.0	882.4
Burial depth (m)	0.75	22.15	5.82	3.91
δ2H-H2O (‰)	−111.6	−32.6	−75.4	10.1
δ18O-H2O (‰)	−15.1	−4.0	−10.2	1.5
δ15N-NO3− (‰)	−4.7	31.2	14.0	4.7
δ18O-NO3− (‰)	−7.6	10.8	−1.1	2.9
SIcalcite	0.26	1.6	0.91	0.25
SIdolomite	0.01	3.12	1.71	0.62
SIfluorite	−2.39	0.05	−1.11	0.5
SIgypsum	−2.41	−0.74	−1.66	0.4

) from the overall the Tabu River Basin groundwater was collected in May 2022. All groundwater samples were taken from wells that served for domestic and agricultural irrigation. Prior to sampling, the groundwater wells were purged for approximately 5 min and the sample bottles were rinsed with abstracted water three times. Each water sample was filtered using a 0.45 μm membrane filter (PES) [31]. The samples for anion and cation analysis were stored in 100 mL poly-ethylene bottles. The samples for δ²H-H₂O and δ¹⁸O-H₂O analyses were collected and sealed in 10 mL Eppendorf plastic tubes. The samples for δ¹⁵N- NO₃⁻ and δ¹⁸O-NO₃⁻ analyses were collected and sealed in 10 mL Eppendorf tubes. All samples were kept at 4 °C before analysis. In addition, five samples of surface water were collected in May 2022 (Figure 1). The sampling processes and parameter analyses of the surface water were the same as those of the groundwater.

Temperature, pH, and ORP were analyzed using a multiparameter instrument (Pro Plus, YSI, Yellow Springs, OH, USA). Na⁺, K⁺, Ca²⁺, and Mg²⁺ concentrations were measured using inductively coupled plasma optical emission spectrometry (Optima 8300, PerkinElmer, Waltham, MA, USA) according to a Chinese standard (water quality—determination of 32 elements—inductively coupled plasma optical emission spectrometry) [32]. SO₄²⁻, Cl⁻, NO₃⁻, and F⁻ concentrations were measured using ion chromatography (850 Professional IC, Metrohm, Switzerland) with Na₂CO₃ and NaHCO₃ as the eluent [33]. The HCO₃⁻ concentration was measured using acid–base titration (HCl and NaOH) within 24 h after sampling with phenolphthalein and methyl orange as indicators. Total dissolved solids (TDS) were determined using gravimetric analysis (the difference in weight between the water sample before and after drying at 105 °C [34]).

δ²H-H₂O and δ¹⁸O-H₂O were measured using a high-precision isotopic water analyzer (L2130-i, Picarro, Santa Clara, CA, USA). The values of δ²H-H₂O and δ¹⁸O-H₂O were reported as per mille (‰) relative to the Vienna Standard Mean Ocean Water (VSMOW). The reproducibility of the triplicate measurements was less than 0.2‰ and ± 0.1 ‰ for δ²H-H₂O and δ¹⁸O-H₂O, respectively. The denitrification method was used to analyze δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻ with Pseudomonas aureofaciens as the denitrifying bacteria [35]. Briefly, NO₃⁻ in the groundwater sample was fully converted to N₂O gas by Pseudomonas aureofaciens in reaction bottles using a constant temperature shaker (25 °C and 180 rpm) after 12 h of incubation. After purification and extraction, N₂O was introduced to an isotope ratio mass spectrometer (MAT 253, Thermo Fisher Scientific, Waltham, MA, USA) to measure δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻. δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻ were reported as per mille (‰) relative to air and VSMOW, respectively. The δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻ values were calibrated through four international standards (USGS32, USGS34, and USGS35, as well as IAEA-N3). The reproducibility of the triplicate measurements was less than ±0.3 ‰ and ±0.5 ‰ for δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻, respectively.

2.3. Data Handling

The ionic charge balance error percentage was calculated to ensure the accuracy of the hydrogeochemical analysis (Equation (1)). The relationships between the hydrogeochemical parameters and isotopic composition, including pH, Na⁺, K⁺, Ca²⁺, Mg²⁺, HCO₃⁻, SO₄²⁻, Cl⁻, NO₃⁻, F⁻, TDS, and δ¹⁸O-H₂O, were analyzed using Spearman correlation (SC) [36]. These hydrogeochemical parameters were also used for HCA and PCA to gain additional insights [22]. The statistical analyses, including SC (‘corrplot’ package), HCA (‘factoextra’ package), and PCA (‘factoextra’ package), were conducted using R software (R version 4.3.0) [37]. Prior to HCA and PCA, the data were standardized. Ward’s method using Euclidean distance was employed to separate different clusters in the HCA. Wald’s method is an algorithm for data classification and clustering, which is a distance-metric-based approach that aims to cluster similar data points together and separate dissimilar data points (for more details about the Wald’s method, see Nourani et al., 2022 [38]). The variation in hydrochemistry parameters between 2022 and 2023 was analyzed by the Wilcoxon test using R software [20].

$$\%\mathrm{ICBE}={\displaystyle \frac{\sum \left({\mathrm{Mg}}^{2+}+{\mathrm{Ca}}^{2+}+{\mathrm{Na}}^{+}+{\mathrm{K}}^{+}\right)\text{-}\sum \left({\mathrm{HCO}}_3^{\text{-}}+{\mathrm{SO}}_4^{2\text{-}}+{{\mathrm{NO}}_3^{\text{-}}+\mathrm{Cl}}^{\text{-}}+{\mathrm{F}}^{\text{-}}\right)}{\sum \left({\mathrm{Mg}}^{2+}+{\mathrm{Ca}}^{2+}+{\mathrm{Na}}^{+}+{\mathrm{K}}^{+}\right)+\sum \left({\mathrm{HCO}}_3^{\text{-}}+{\mathrm{SO}}_4^{2\text{-}}+{{\mathrm{NO}}_3^{\text{-}}+\mathrm{Cl}}^{\text{-}}+{\mathrm{F}}^{\text{-}}\right)}}\times 100$$

(1)

2.4. Human Health Risk Assessment (HHRA)

Generally, dermal and oral contact are the most common pathways of exposure to groundwater pollutants for human beings. Consequently, an assessment of human health risk was conducted by calculating the oral (HQ_oral) and dermal (HQ_dermal) hazard quotient and the resulting total non-carcinogenic health risk (HI_total) according to Equations (2)–(7) based on the model recommended by the United States Environmental Protection Agency [28]. The empirical parameters are summarized in Table S1.

$$\mathrm{H}{\mathrm{Q}}_{\mathrm{oral}}={\displaystyle \frac{{\mathrm{CD}\mathrm{I}}_{\mathrm{oral}}}{\mathrm{Rf}{\mathrm{D}}_{\mathrm{oral}}}}$$

(2)

$$\mathrm{CD}{\mathrm{I}}_{\mathrm{oral}}={\displaystyle \frac{\left({\mathrm{C}}_{\mathrm{i}}\times \mathrm{IR}\times \mathrm{EF}\times \mathrm{ED}\right)}{\mathrm{BW}\times \mathrm{AT}}}$$

(3)

$$\mathrm{H}{\mathrm{Q}}_{\mathrm{dermal}}={\displaystyle \frac{{\mathrm{CDI}}_{\mathrm{dermal}}}{\mathrm{Rf}{\mathrm{D}}_{\mathrm{oral}}\times \mathrm{AB}{\mathrm{S}}_{\mathrm{gi}}}}$$

(4)

$$\mathrm{CD}{\mathrm{I}}_{\mathrm{Dermal}}={\displaystyle \frac{\left({\mathrm{C}}_{\mathrm{i}}\times \mathrm{K}\times {\mathrm{S}}_{\mathrm{a}}\times \mathrm{T}\times \mathrm{EV}\times \mathrm{CF}\times \mathrm{EF}\times \mathrm{ED}\right)}{\mathrm{BW}\times \mathrm{AT}}}$$

(5)

$${\mathrm{HI}}_{\mathrm{total}}={\sum }_{\mathrm{i}=1}^{\mathrm{n}}\mathrm{H}{\mathrm{I}}_{\mathrm{i}}$$

(6)

$${\mathrm{HI}}_{\mathrm{i}}={\mathrm{HQ}}_{\mathrm{oral}}+{\mathrm{HQ}}_{\mathrm{dermal}}$$

(7)

3. Results and Discussion

3.1. General Hydrogeochemical Parameters of Groundwater Samples and the Spatial Distribution of Groundwater Pollution

A statistical summary of the hydrogeochemical and isotopic parameters of the phreatic groundwater in the Tabu River Basin is exhibited in Table 1. The pH values of the groundwater ranged from 7.37 to 8.66 (median: 8.10), indicating a slightly alkaline condition. The TDS ranged from 278.0 to 5398.5 mg/L, with 77.0% and 2.3% of samples classed as slightly saline (1000 < TDS < 3000 mg/L) [39] and moderately saline water (3000 < TDS < 10,000 mg/L), respectively. Only 20.7% of the samples were classified as freshwater (TDS < 1000 mg/L). The relative abundance of major anions followed the order of SO₄²⁻ > HCO₃⁻ > Cl⁻, while the order of major cations was Ca²⁺ > Na⁺ > Mg²⁺ ≫ K⁺. The groundwater samples were scattered near the local meteoric water line (LMWL) (Figure S1), indicating that precipitation is the primary source of recharge for groundwater in the study area.

The concentration of NO₃⁻ in the phreatic groundwater ranged from 0 to 1274.0 mg/L, with a median value of 81.3 mg/L (Table 1). A total of 67.8% of groundwater samples exceeded the acceptable limit of NO₃⁻ (50.0 mg/L) [40]. The high NO₃⁻ groundwater was mainly distributed in the upper and middle reaches of the Tabu River Basin (Figure 2a), which are characterized by extensive farmland and rural residential areas (Figure 1b). This result suggests the significant impact of anthropogenic activities. The concentration of F⁻ ranged from 0.2 to 4.6 mg/L (median: 0.8 mg/L; Table 1) in the phreatic groundwater. A total 31.0% of groundwater samples exceeded the acceptable limit of F⁻ (1.5 mg/L) [40]. The high F⁻ groundwater was mainly distributed in the downstream area of the Tabu River Basin (Figure 2b) with less anthropogenic activity (Figure 1b), indicating a geogenic origin.

3.2. Influences of Anthropogenic Factors on Groundwater Quality

Generally, the concentration of NO₃⁻ in groundwater is naturally below 10 mg/L [5]. Therefore, a high NO₃⁻ concentration in groundwater might be the result of anthropogenic input. Chemical fertilizers, manure, and sewage are common anthropogenic sources of NO₃⁻ in groundwater in agricultural areas [6,41,42]. Chemical fertilizers contain high concentrations of NO₃⁻ and low concentrations of Cl⁻, while manure and sewage have high concentrations of both NO₃⁻ and Cl⁻ [43,44]. In this study, NO₃⁻ exhibited a significant positive correlation with Cl⁻ (R = 0.69, p < 0.001; Figure 3), suggesting that manure and sewage are potential sources of the high concentration of NO₃⁻ in the phreatic groundwater.

Generally, contaminants from various sources exhibit significant variations in hydrogeochemical parameters and isotopic composition, which can serve as reliable indicators for tracing their origins [17,18]. Chemical fertilizers exhibit high NO₃⁻/Cl⁻ ratios, a low Cl⁻ content, and high δ¹⁸O-NO₃⁻ values, whereas manure and sewage are characterized by low NO₃⁻/Cl⁻ ratios, a high Cl⁻ content, and low δ¹⁸O-NO₃⁻ values [6]. The dual isotopes of NO₃⁻ (δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻) were used to identify the source of NO₃⁻ in phreatic groundwater. As shown in Figure 4a, the majority of the groundwater samples were scattered in the zone associated with manure and sewage, suggesting that manure and sewage were the primary sources of NO₃⁻. To further investigate the contribution of manure and sewage to the enrichment of NO₃⁻, a dual logarithmic diagram of Cl⁻/Na⁺ and NO₃⁻/Na⁺ was used. As shown in Figure 4b, the groundwater samples, and particularly those with a high NO₃⁻ concentration, were mainly located near the zone of agricultural activities rather than sewage, indicating that NO₃⁻ in the phreatic groundwater primarily originates from manure. Moreover, groundwater with a high NO₃⁻ concentration showed a substantial overlap with farmland areas in terms of spatial distribution (Figure 1b and Figure 2a), where crop–livestock integration was widely observed and manure served as a plant fertilizer after a qualified preparation. This result further demonstrated the primary contribution of manure to the enrichment of NO₃⁻. Only a small proportion of groundwater samples fell within the overlapping ranges of the NH₄⁺ fertilizer, soil NH₄⁺, and sewage (Figure 4a), suggesting a limited impact on NO₃⁻ enrichment in the phreatic groundwater.

In addition to NO₃⁻, the TDS was also found to be abnormally high in groundwater. A previous study reported that the long-term and intensive use of groundwater for irrigation in an arid environment, where water discharges only through evapotranspiration and infiltration, can lead to an accelerated accumulation of salt in the phreatic groundwater [47,48]. The Tabu River Basin is located in an arid and semi-arid region, and groundwater is the primary water resource for irrigation [30]. In this study, the TDS showed a significant positive correlation with NO₃⁻ (R = 0.71, p < 0.001). Moreover, the high NO₃⁻ area showed a substantial overlap with the high TDS area in terms of spatial distribution, (Figure 2a,c), where farmland was widely distributed (Figure 1b). Therefore, the use of groundwater for irrigation significantly contributed to the high TDS in the phreatic groundwater.

Anthropogenic activities can also contribute to the enrichment of F⁻, such as the discharge of sewage and an excessive use of fertilizers [6,49]. In this study, there was no significant correlation between F⁻ and NO₃⁻ (p > 0.05; Figure 3), indicating that anthropogenic activities had a negligible impact on the enrichment of F⁻ in phreatic groundwater. Additionally, areas of high F⁻ groundwater overlapped with the grassland in the Tabu River Basin, where anthropogenic activities were limited (Figure 1b and Figure 2b). This result further demonstrated that anthropogenic activity was not the primary factor controlling F⁻ enrichment in the phreatic groundwater.

3.3. Influences of Geogenic Factors on Groundwater Quality

3.3.1. Evaporation

Evaporation plays a crucial role in groundwater quality, particularly in arid and semi-arid regions [12,50]. Most of the groundwater samples fell on the lower right of the global meteoric water line (Figure S1), indicating the influence of evaporation. Theoretically, the concentrations of hazardous substances and δ¹⁸O-H₂O value in groundwater can increase simultaneously during the evaporation processes [6]. However, δ¹⁸O-H₂O exhibited no significant correlation with NO₃⁻, TDS, and F⁻ (p > 0.05; Figure 3). Therefore, the influence of evaporation on the phreatic groundwater quality was limited.

3.3.2. Interaction between Surface Water and Groundwater

Phreatic surface water can introduce contaminants into groundwater through hydrologic exchange [12]. Therefore, the surface water quality and its influence on groundwater were investigated. For the surface water, the NO₃⁻ content ranged from 1.9 to 38.2 mg/L (median: 32.0 mg/L; Table S3) in the Tabu River Basin. A total of 78.2% of the groundwater samples had a significantly higher NO₃⁻ content than 38.2 mg/L (Table S2). If the NO₃⁻ in groundwater mainly originates from surface water, groundwater with a higher NO₃⁻ content than the surface water should be formed through evaporative concentration. However, evaporation had no significant influence on groundwater quality (Figure 3). Moreover, our previous study demonstrated that surface water was recharged by groundwater in the Tabu River Basin rather than being discharged to groundwater [30]. These results indicate that the contamination of the phreatic groundwater cannot be attributed to surface water’s input.

3.3.3. Water–Rock Interactions

In addition to evaporation, water–rock interactions are also crucial geogenic factors influencing groundwater quality [9]. Generally, water–rock interactions, such as dissolution/precipitation and ion exchange processes, generally have a limited effect on the NO₃⁻ concentration but have significant effects on the F⁻ concentration [23,51,52]. The dissolution of fluorine-bearing minerals, such as fluorite (CaF₂), can serve as a substantial source of F⁻ in groundwater (Equation (8)). This process can be facilitated by reducing the availability of Ca²⁺ through cation exchange (Equation (9)) or carbonate precipitation (Equation (10)) in aquifers.

CaF₂ → Ca²⁺ + 2F⁻

(8)

Ca²⁺ + 2NaX = 2Na⁺ + CaX₂

(9)

Ca²⁺ + 2HCO₃⁻ → CaCO₃ + H₂O + CO₂

(10)

A linear relationship between (Na⁺ + K⁺ − Cl⁻) and (Ca²⁺ + Mg²⁺) − (HCO₃⁻ + SO₄²⁻) with a slope close to −1 generally indicates cation exchange [12]. As shown in Figure 5a, the majority of the high F⁻ groundwater samples were distributed along the ’y = −x’ line, demonstrating the significant effect of cation exchange on the enrichment of F⁻ in the phreatic groundwater.

The saturation index was used to assess the influence of the carbonate precipitation on F⁻ enrichment [12]. In this study, the majority of groundwater samples were unsaturated with respect to fluorite (Figure 5b), facilitating the continuous dissolution of fluorite by the groundwater in the aquifers. Additionally, there was a significant positive correlation between F⁻ and SI_fluorite (Figure 5b), indicating the contribution of fluorite dissolution to F⁻ enrichment in the groundwater. As shown in Figure 5c, calcite was oversaturated in the groundwater, whereas fluorite was unsaturated. Carbonate precipitation continuously occurred, leading to a decrease in Ca²⁺ concentrations (Equation (10)). The decrease in Ca²⁺ concentrations make fluorite unsaturated in groundwater, thereby promoting fluorite dissolution (Equation (8)) [12,53]. Therefore, this result suggested the potential enhancement of fluorite dissolution due to calcite precipitation. Theoretically, if carbonatites precipitation dominates the dissolution of fluorine-bearing minerals, an increase in F⁻ concentration should be accompanied by a decrease in Ca²⁺ concentrations [12]. In this study, F⁻ exhibited a significant negative correlation with Ca²⁺ (R = −0.49, p < 0.01; Figure 3). Therefore, carbonate precipitation in the aquifers significantly contributed to the enrichment of F⁻ in the phreatic groundwater in the study area.

3.3.4. Alkaline Condition and Competitive Adsorption

Under an alkaline condition, an increase in OH⁻ can promote the reaction shown in Equation (11), promoting the release of F⁻ from fluorine-bearing minerals into groundwater [53]. Moreover, alkaline conditions can enhance the desorption of F⁻ that is adsorbed on silicate (Equations (12) and (13)) [54]. In this study, the groundwater was in a slightly alkaline condition (Table 1). There was a significant positive correlation between F⁻ and pH (R = 0.44, p < 0.001; Figure 3), confirming the contribution of the alkaline condition to the enrichment of F⁻ in the phreatic groundwater.

CaF₂ + 2OH⁻ → Ca(OH)₂ + 2F⁻

(11)

KAl₂[AlSi₃O₁₀]F₂ + 2OH⁻ → KAl₂[AlSi₃O₁₀][OH]₂ + 2F⁻

(12)

KMg₃[AlSi₃O₁₀]F₂ + 2OH⁻ → KMg₃[AlSi₃O₁₀][OH]₂ + 2F⁻

(13)

As shown in Figure 3, F⁻ exhibited a positive correlation with HCO₃⁻ (R = 0.40, p < 0.001) in this study, suggesting an association between the release of HCO₃⁻ and the enrichment of F⁻. HCO₃⁻ can reduce the adsorption sites on the sorbent, facilitating the release of F⁻ from the aquifer into the groundwater [6]. Additionally, the weathering of carbonates and silicates coexisting with fluoride can simultaneously elevate the levels of F⁻ and HCO₃⁻ in groundwater [6]. Therefore, the competitive sorption and weathering processes associated with HCO₃⁻ significantly contributed to the enrichment of F⁻ in the phreatic groundwater.

3.4. HCA and PCA Based on Hydrogeochemical Parameters

HCA was used to split the phreatic groundwater samples into four clusters (clusters I, II, III, and IV) (Figure S2). The hydrogeochemical parameters of the four groundwater clusters are exhibited in Table S2. As shown in Figure 6, cluster I was identified as seriously polluted groundwater with high concentrations of NO₃⁻ (median: 1232.5 mg/L) and F⁻ (median: 2.3 mg/L). Cluster II was identified as seriously polluted groundwater with high concentrations of NO₃⁻ (median: 570.2 mg/L) and low concentrations of F⁻ (median: 0.6 mg/L). Cluster III was identified as slightly polluted groundwater with low concentrations of NO₃⁻ (median: 62.3 mg/L) and F⁻ (median: 0.6 mg/L). Cluster IV was also identified as slightly polluted groundwater with low concentrations of NO₃⁻ (median: 50.0 mg/L) and high concentrations of F⁻ (median: 2.1 mg/L).

Cluster I was identified as Na-Cl·SO₄ groundwater (Figure 7a) with a high TDS (median: 4709.2 mg/L; Table S2). Cluster II was identified as mixed Ca·Mg-Cl groundwater with a high TDS (median: 1661.8 mg/L; Table S2). Both clusters I and II were seriously influenced by anthropogenic factors, but their major cations showed significant differences, which might be attributed to geogenic processes. As shown in Figure 7b, the Gibbs diagram indicated that cluster I was located in the zone of evaporation dominance, whereas cluster II was distributed in the zone of rock dominance. The precipitation of Ca²⁺ and Mg²⁺ occurs prior to Na⁺ under alkaline conditions [55]. Thus, the percentage of Na⁺ increased, whereas that of Ca²⁺ and Mg²⁺ decreased during the evaporation process. Therefore, the difference in major cations between clusters I and II was primarily attributed to evaporation.

Cluster III was identified as Ca-HCO₃ groundwater (Figure 7a) with a low TDS (median: 538.4 mg/L; Table S2). Cluster IV was identified as mixed Ca·Na-HCO₃ groundwater (Figure 7a) with a low TDS (median: 795.4 mg/L; Table S2). Both clusters III and IV were slightly polluted by anthropogenic inputs. The percentage of Na⁺ in the groundwater of cluster IV was higher than that of cluster III. The Gibbs diagram indicates that the cation-exchange process occurred in the groundwater of cluster III and IV, which might contribute to the difference in the major cation (Figure 7b). The chloro-alkaline indices (CAI-I and CAI-II) refer to ion exchange processes that take place when one ion is substituted by another ion at the solid material surface of soil or rock, such as clay minerals, organic matter, or metal oxyhydroxides [56]. Therefore, CAIs were used to assess the impact of the cation-exchange process in this study (Figure 8). Higher absolute values of these indices indicate a stronger cation exchange in the groundwater environment [57]. The cluster IV groundwater was primarily located in the cation-exchange zone (Figure 8), indicating the substitution of Ca²⁺ by Na⁺ (Equation (14)) [11]. Conversely, the groundwater of cluster III was mainly located in the reverse cation-exchange zone (Figure 8), indicating the substitution of Na⁺ by Ca²⁺. Therefore, the process of cation exchange led to a higher proportion of Na⁺ in the cluster IV samples in comparison to the cluster III samples.

2Na − clay + Ca²⁺ − water → Ca²⁺ − clay + 2Na⁺ − water

(14)

Under natural conditions, HCO₃⁻ is the major anion of the phreatic groundwater in the Tabu River Basin [30]. The major anion of clusters III and IV with slight contamination was HCO₃⁻. However, the major anions of cluster I and II with serious contamination were Cl⁻ and SO₄²⁻. The groundwater in this study area was significantly polluted by manure from agricultural activities, which can elevate the levels of Cl⁻ and SO₄²⁻ in the groundwater [5,43,44]. In this study, NO₃⁻ exhibited a significant positive correlation with Cl⁻ (R = 0.69, p < 0.001) and SO₄²⁻ (R = 0.52, p < 0.001). Therefore, agricultural activities significantly affected the hydrochemical composition of the groundwater in clusters I and II.

As shown in Figure 9, PCA based on the hydrogeochemical parameters and δ¹⁸O-H₂O explained 70.6% of the variance between the groundwater samples using the first two principal components (PC1 and PC2). PC1, which accounted for 50.7% of the variance, exhibited a negative correlation with NO₃⁻, TDS, Mg²⁺, Cl⁻, SO₄²⁻, HCO₃⁻, and Na⁺ (Table S4). As agricultural activity enhanced the inputs of NO₃⁻, SO₄²⁻, Cl⁻, and TDS, PC1 can be interpreted as those anthropogenic factors influencing groundwater quality. PC2 explained 19.9% of the variance, which was negatively correlated with pH, F⁻, Na⁺, and HCO₃⁻, but positively correlated with Ca²⁺ (Table S4). In the negative direction of PC2, F⁻ was positively correlated with pH and HCO₃⁻ (Figure 9), indicating the influences of alkaline conditions and competitive adsorption on the enrichment of F⁻ [6,10]. In the positive direction of PC2, Ca²⁺ was negatively correlated with F⁻ and Na⁺, suggesting the influences of cation exchange and fluorine-bearing mineral dissolution on the enrichment of F⁻ [12]. Therefore, PC2 could be interpreted as geogenic factors regulating the enrichment of F⁻ in groundwater. The results of the PCA further demonstrated that the enrichment of NO₃⁻ and F⁻ was dominated by anthropogenic and geogenic factors, respectively.

3.5. Assessment of the Risk to Human Health

The long-term consumption of groundwater contaminated with NO₃⁻ and F⁻ may pose health risks to humans, particularly when used as a source of daily drinking and bathing water without proper precautions being taken [5]. Thus, NO₃⁻ and F⁻ were considered in the assessment of non-carcinogenic risk for children and adults (females and males) [58]. As exhibited in Table S5, the HI_total varied between 0.97 and 39.50 (median: 4.79) for children, 0.57 and 23.09 for adult females (median: 2.80), and 0.42 and 16.96 for adult males (median: 2.06). A total of 86.2%, 60.9%, and 51.7% of the groundwater samples were found to exceed the permissible limit (HI_total = 1) for children, adult females, and adult males, respectively. Moreover, the spatial distribution of HI_total indicated that groundwater posed a potential risk to children, adult females, and adult males in almost the whole Tabu River Basin (Figure 10a–c). Therefore, the phreatic groundwater environment of the Tabu River Basin is facing challenges. The health threats of NO₃⁻ and F⁻ to humans through both oral and dermal contact followed the order of children > adult females > adult males.

The HQ_total values of NO₃⁻ varied between 0 and 37.30 for children (median: 2.38), 0 and 21.81 for adult females (median: 1.39), and 0 and 16.00 for adult males (median: 1.02) (Table S5). A total of 86.2%, 58.6%, and 51.7% of the groundwater samples were found to exceed the permissible limit of 1 for children, adult females, and adult males, respectively. The HQ_total values of F⁻ varied between 0.21 and 5.36 (median: 0.90) for children, 0.12 and 3.14 for adult females (median: 0.52), and 0.09 and 2.30 for adult males (median: 0.38) (Table S5). A total of 44.8%, 31.0%, and 18.4% of the groundwater samples for children, adult females, and adult males were observed to have HQ_total values exceeding 1, respectively. Therefore, the contamination of phreatic groundwater in the Tabu River Basin poses potential risks to human health due to both anthropogenic and geogenic factors, with anthropogenic contamination (NO₃⁻) being more serious than geogenic contamination (F⁻).

For the dermal contact pathway, the HI_dermal values for both NO₃⁻ and F⁻ were less than 1 for children, adult females, and adult males (Table S5), suggesting that NO₃⁻ and F⁻ in the groundwater do not pose a non-carcinogenic risk to humans through the dermal contact pathway. For the oral contact pathway, HI_oral values for both NO₃⁻ and F⁻ varied between 0.97 and 39.35 (median: 4.78) for children, 0.57 and 23.00 for adult females (median: 2.79), and 0.42 and 16.87 for adult males (median: 2.05), respectively. The HI_oral values were close to HI_total, highlighting the potential risks of the oral contact pathway to human health. Therefore, most of the phreatic groundwater in the Tabu River Basin can be utilized for domestic rather than drinking purposes. However, high-quality groundwater for drinking purposes can be obtained by drilling wells downstream of the Tabu River Basin, where the human health risks are at an acceptable level (Figure 10).

From the perspective of the spatial distribution, the health risks associated with the groundwater in the middle and upper reaches are primarily influenced by NO₃⁻, whereas downstream, they are mainly affected by F⁻ (Figure S3). Therefore, the risks for the population of these different regions are also different due to different contamination conditions.

3.6. Development Trends and Countermeasures for Groundwater Contamination

In order to obtain the development trends of the contamination in the Tabu River Basin, 37 samples of phreatic groundwater were collected from some of the 2022 (Figure S4; Table S6) sites in May 2023. A comparison of NO₃⁻ and F⁻ in the groundwater samples between 2022 and 2023 was conducted. As shown in Figure 11, the NO₃⁻ and F⁻ content in phreatic groundwater remained stable from 2022 to 2023 (Wilcoxon test, p > 0.05), suggesting that the contamination of phreatic groundwater could not be mitigated through natural attenuation under existing external pressures. Therefore, measures need to be taken to decrease the impact of groundwater contamination and enhance groundwater sustainability in the Tabu River Basin.

It is important to take specific measures for areas with different contamination conditions in the Tabu River Basin. For the middle and upper reaches of the Tabu River Basin, NO₃⁻ needs to be controlled (Figure 2). Manure should be centrally disposed of after collection rather than being discharged in a disorderly manner. Moreover, the government should assist local residents with optimizing fertilizer usage to enhance efficiency and decrease overall nitrogen fertilizer consumption. Additionally, exploring the use of novel materials like polymer adsorbent [59] and composite adsorbent [60] presents potential solutions for addressing groundwater pollution. For the high-F groundwater downstream of the Tabu River Basin (Figure 2), it is impracticable to decrease the F⁻ content in the phreatic groundwater through artificial measurement, as F⁻ is a geogenic contaminant. Therefore, no measures need to be taken against geogenic high F⁻ groundwater unless it needs to be used for drinking purposes in the future. If so, strong fluorine removal engineering technology can be used after high-F groundwater has been extracted to waterworks [54]. In addition, Cl⁻ and other hazardous substances from chemical fertilizers, manure, and wastewater should be considered during water treatment if groundwater is used for drinking purpose.

4. Conclusions

Groundwater quality is critical for human health and sustainable development in northern China, as it is the main clean water resource. In addition to anthropogenic effects, geogenic processes may play an important role in groundwater quality variation. In this study, a study area in the Tabu River Basin was selected to investigate the contributions of anthropogenic and geogenic effects in phreatic groundwater quality evolution. Additionally, the effects of phreatic groundwater quality on human health were also evaluated. The main conclusions are as follows:

(1)

Agricultural activity is the primary anthropogenic factor influencing the quality of phreatic groundwater. NO₃⁻ pollution in the groundwater primarily originates from manure, and the high level of TDS is highly associated with irrigation.

(2)

The enrichment of F⁻ in the phreatic groundwater is dominated by geogenic factors, including alkaline conditions, competitive adsorption, the dissolution of fluorine-bearing minerals, and cation exchange.

(3)

Phreatic groundwater with high NO₃⁻ and F⁻ contents poses significant threats to human health through the oral contact pathway, especially for children. Most of the phreatic groundwater in the Tabu River Basin can be utilized for domestic purposes, with the exception of drinking water.

(4)

The contamination of phreatic groundwater cannot be mitigated through natural attenuation under existing external pressures. Measures need to be taken to decrease contamination of phreatic groundwater and enhance groundwater sustainability in the Tabu River Basin.