Hydrochemical Characteristics and Drinking Water Quality Assessment of Phreatic Groundwater in the Northwest of the Sichuan Basin, SW China

Abstract

In this study, a total of 26 groundwater samples were collected from the northwest of the Sichuan Basin. Statistical analysis revealed that Ca²⁺ was the predominant cation, followed by Na⁺, Mg²⁺, and K⁺. The anion concentrations followed the order HCO₃⁻ > SO₄²⁻ > NO₃⁻ > F⁻ > Cl⁻. Consequently, Ca-HCO₃ was identified as the dominant hydrochemical type in the study area. Geochemical modeling results indicated that silicate weathering and cation exchange processes were the primary factors influencing groundwater hydrochemistry. To provide an accurate assessment of water quality, a Comprehensive Water Quality Index (CWQI) was applied in this study. This novel method combined factor analysis and the entropy-weighted approach to derive integrated weights for water quality calculation. The CWQI results showed that 73.08% of the samples were classified as excellent for drinking, while 26.92% were classified as good. Sensitivity analysis further demonstrated the robustness of the drinking water quality model. The findings of this study could contribute to the enhancement of water quality evaluation in the Sichuan Basin.

1. Introduction

Water and sanitation are fundamental to sustainable development, and “Clean Water and Sanitation” has been established as one of the Sustainable Development Goals by the United Nations. However, in recent decades, overexploitation, pollution, and climate change have resulted in severe groundwater stress in regions around the world [1,2,3]. As a primary source of water for domestic and agricultural use, deteriorating groundwater quality has significantly impacted human health [4,5,6]. Consequently, understanding the driving factors behind groundwater chemistry and developing effective models for water quality assessment have become critical for managing groundwater resources.

The geochemical model is an effective tool for distinguishing the hydrochemical processes of groundwater [7,8]. This model typically includes hydrochemical diagrams, saturation indices, and geostatistical analysis tools [9,10]. Hydrochemical diagrams consist of the Gibbs diagram, Piper diagram, and scatter plots of specific ions, while geostatistical tools include interpolation analysis. Qin [11] highlighted both natural and anthropogenic factors affecting geochemistry in Chongqing using specific hydrochemical diagrams. Gao [12] successfully identified chlorine and nitrate pollution in Xi’an City through interpolation analysis in GIS. Thus, the geochemical model serves as an efficient tool for identifying the driving factors that control groundwater chemistry.

The Water Quality Index (WQI), proposed by Horton [13], is a traditional method widely used for assessing water suitability for various purposes [14,15,16]. This method is valued for its ability to integrate numerous individual hydrochemical parameters into a single score [17,18,19]. The core of WQI calculation lies in the weight assigned to each parameter [20,21]. Common methods for calculating weights in water quality assessment include the order relation analysis method (G1), analytic hierarchy process (AHP), criteria importance through inter-criteria correlation (CRITIC), entropy-weighted method, and factor analysis [12,22,23,24]. Among these, the entropy-weighted method is widely employed due to its reliance solely on dataset characteristics, thus minimizing the impact of subjective human evaluations [25,26]. Moreover, several studies have explored the integration of multiple weight calculation methods to more accurately reflect water quality [27,28]. For instance, Gao [12] used an integrated weight combining the G1 and entropy-weight methods to evaluate groundwater suitability for drinking purposes. Yan [22] developed a combined-weight WQI using AHP and CRITIC methods to assess drinking water quality in the Sichuan Basin. This novel-integrated approach offers a new perspective for accurately assessing drinking water suitability.

The Sichuan Basin is a major population center in western China, where groundwater plays a crucial role in domestic, industrial, and agricultural activities [10,26]. Numerous studies have been conducted to evaluate groundwater suitability in the metropolitan areas of the Sichuan Basin [7,29]. However, the hydrochemical processes and water quality suitability in surrounding areas remain poorly understood, hindering economic development in the region. To address these gaps, 26 groundwater samples were collected from the northwest of the Sichuan Basin. This study aims to (1) explore the dominant factors controlling groundwater chemistry and (2) evaluate drinking water quality using a chemical water quality index, while identifying the key factors influencing it. The findings of this research are expected to provide a fresh perspective on water quality evaluation.

2. Materials and Methods

2.1. Study Area

The study area is located in the northwest of the Sichuan Basin, covering an area of 2719 km² (Figure 1). The terrain slopes from northwest to southeast, with the dominant landforms being plains and hills. The climate of the study area is characterized as a subtropical humid monsoon climate, with four distinct seasons (hot summers and mild winters). The annual average temperature is 8 °C, and the annual precipitation is 788.2 mm. Precipitation is concentrated between June and September and exhibits spatial variation, decreasing from northwest to southeast. Croplands are the predominant land use in the area, followed by forests and urban built areas (Figure 1).

The strata exposed in the study area belong to the Yangzi stratigraphic zone, including the Longmenshan stratigraphic subdivision and the Sichuan Basin stratigraphic subdivision. The Paleozoic and Mesozoic strata are the main strata, followed by the Cenozoic strata. The strata covered seven periods: Silurian periods (S), Devonian periods (D), Permian periods (P), Triassic periods (T), Jurassic periods (J), Cretaceous periods (K), and Quaternary periods (Q). The carbonates were extensively developed in the Silurian period to Permian period, while the Permian period to Cretaceous was characterized by the accumulation of giant thick layers of red clastic sediments. The type of groundwater was mainly Quaternary pore water, interstitial water, fissure water, and karst water. Quaternary pore water is shallowly buried and distributed along the river. The typical spring flow rates of the fissure water aquifer in the red clastic rock area are 0.1 to 1 L/s. Spring flow rates of karst water aquifer in the carbonate area are less than 100 L/s. Groundwater in the study area is mainly recharged by precipitation, followed by agricultural irrigation and infiltration of surface water bodies. The gully and plain areas serve as both groundwater storage and discharge zones.

2.2. Data Collection

A total of 26 groundwater samples were collected across the study area in April. The precise locations of each sample site were recorded using the Global Positioning System (GPS). Prior to sampling, groundwater was pumped for more than 30 min to eliminate standing water. Additionally, sample bottles were thoroughly cleaned by rinsing them more than five times after the removal of stagnant water. The pH and total dissolved solids (TDS) were measured on-site using a WTW Multi 340i/SET (Munich, Germany). Then, groundwater was filtered through 0.22 μm membrane filters, bolted with wax in the polyethylene bottles, and transported to the laboratory under 3 °C. Inductively coupled plasma mass spectrometry (Agilent 75000ce ICP-MS) (Santa Clara, CA, USA) and ion chromatography (LC-10Advp) (Kyoto, Japan) were used for the significant cations (Na⁺, Ca²⁺, Mg²⁺, and K⁺) and anions (Cl⁻, SO₄²⁻, NO₃⁻, and F⁻) determination. Acid-base titration methods were used to determine the HCO₃⁻ concentrations. Moreover, ions charge balance error (%ICBE) was applied to ensure the reliability and accuracy of the study [30]. ICBEs of less than 5% were used for further analysis in this study.

$$\%\mathrm{I}\mathrm{CBE}={\displaystyle \frac{\sum \mathrm{C}\mathrm{ations}-\sum \mathrm{A}\mathrm{nions}}{\sum \mathrm{C}\mathrm{ations}+ \sum \mathrm{A}\mathrm{nions}}}$$

(1)

2.3. Methods of Results Analysis

2.3.1. Geochemical Modeling

Geochemical models can effectively distinguish the hydrochemical process of groundwater [7,8]. This integrated model generally comprises hydrochemical diagrams, saturation index, and geostatistical analyst tools [9,10]. Hydrochemical diagrams include Gibbs diagram, Piper diagram, and scatter plot of specific ions, which can be drawn with Python 3.12.4 (The Netherlands). The saturation index (SI) is a thermodynamic indicator that is widely used in recognizing the sources of ions in the water-rock process (Equation (2)). This unique index can be calculated using PHREEQC Version 3 (USA). Geostatistical analyst tools like the inverse distance weighted (IDW) interpolation, the Kringing interpolation method, and the Empirical Bayesian interpolation method can be achieved with the software ArcGIS Pro 3.1 (USA).

$$\mathrm{SI}=\lg {\displaystyle \frac{\mathit{IAP}}{K}}$$

(2)

where IAP donates the ion activity products, K is the equilibrium constant. When SI > 0, the minerals in groundwater are in a saturated status, whereas the value is less than zero signifying the unsaturated status for specific minerals.

2.3.2. Factor Analysis Method

Factor analysis is a widely used dimensionality reduction technique for handling comprehensive hydrochemical datasets. It is commonly applied in source identification and water quality assessments [31,32]. To characterize the data in this study, factor analysis was conducted. The calculation process for this method involves five steps:

Step 1: Calculate the KMO (Kaiser-Meyer-Olkin) value and conduct the Bartlett test to determine the feasibility of factor analysis. A KMO value greater than 0.6 and a p-value less than 0.05 indicate that the dataset is suitable for factor analysis.

Step 2: Standardize the data set (matrix X) with the Z-score method.

$${\mathrm{R}}_{\mathit{mn}}={\displaystyle \frac{{\mathrm{X}}_{\mathit{mn}}-{ \mathrm{\mu }}_n}{{\mathrm{\sigma }}_n}}$$

(3)

where ${\mathrm{X}}_{\mathit{mn}}$ is the hydrochemical dataset; m is the number of samples; n is the hydrochemical parameter of each sample; ${ \mathrm{\mu }}_n$ and ${\mathrm{\sigma }}_n$ donate the mean values and the standard deviation of the jth hydrochemical parameter.

Step 3: Find the eigenvalues and eigenvectors of the matrix R.

Step 4: Determine the number of factors and calculate the loading matrix A.

Step 5: Calculate the weight ($w_f$) of each hydrochemical parameter with loading coefficients, eigenroots and variance, explained.

$$w_f={\displaystyle \frac{A_j{ \lambda }_j}{{\sum }_{i=1}^k{ \lambda }_i}}$$

(4)

where $A_j$ is the loading coefficients of jth hydrochemical parameter; ${\lambda }_j$ is the eigenvalue of jth hydrochemical parameter; ${\sum }_{i=1}^k{ \lambda }_i$ donates the sum of all factors. $w_f$ represents the weight of each parameter based on the factor analysis method.

2.3.3. Entropy-Weighted Method

The entropy-weighted method is based on the concept of entropy in information theory and determines the weights of indicators according to the degree of dispersion of each indicator [33]. In the context of multi-indicator weighting, this method helps avoid the bias introduced by subjective assignments, thereby enhancing the objectivity of the evaluation results. The calculation process of the entropy-weighted method includes three steps:

Step 1: Normalize the eigenvalue matrix.

$$\mathrm{X}=\left[\begin{array}{cccc}{\mathrm{x}}_{11}& {\mathrm{x}}_{12}& \cdots & {\mathrm{x}}_{1\mathrm{n}}\\ {\mathrm{x}}_{21}& {\mathrm{x}}_{22}& \cdots & {\mathrm{x}}_{2\mathrm{n}}\\ ⋮& ⋮& \ddots & ⋮\\ {\mathrm{x}}_{\mathrm{m}1}& {\mathrm{x}}_{\mathrm{m}2}& \cdots & {\mathrm{x}}_{\mathrm{mn}}\end{array}\right]$$

(5)

$${\mathrm{y}}_{\mathrm{ij}}={\displaystyle \frac{{\mathrm{x}}_{\mathrm{ij}}-({\mathrm{x}}_{\mathrm{ij}}{)}_{\min }}{({\mathrm{x}}_{\mathrm{ij}}{)}_{\max }– ({\mathrm{x}}_{\mathrm{ij}}{)}_{\min }}}$$

(6)

$$\mathrm{Y}=\left[\begin{array}{cccc}{\mathrm{y}}_{11}& {\mathrm{y}}_{12}& \cdots & {\mathrm{y}}_{1\mathrm{n}}\\ {\mathrm{y}}_{21}& {\mathrm{y}}_{22}& \cdots & {\mathrm{y}}_{2\mathrm{n}}\\ ⋮& ⋮& \ddots & ⋮\\ {\mathrm{y}}_{\mathrm{m}1}& {\mathrm{y}}_{\mathrm{m}2}& \cdots & {\mathrm{y}}_{\mathrm{mn}}\end{array}\right]$$

(7)

where $({\mathrm{x}}_{\mathrm{ij}}{)}_{\max }$ and $({\mathrm{x}}_{\mathrm{ij}}{)}_{\min }$ represent the maximum and minimum values of jth hydrochemical parameters. Y represents the standardized water quality numerical matrix.

Step 2: Get the information entropy.

$${\mathrm{p}}_{\mathrm{ij}}={\displaystyle \frac{{\mathrm{y}}_{\mathrm{ij}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{y}}_{\mathrm{ij}}}}$$

(8)

$${\mathrm{e}}_{\mathrm{j}}=-{\displaystyle \frac{1}{\mathrm{lnm}}}{\sum }_{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{P}}_{\mathrm{ij}}\ln {\mathrm{P}}_{\mathrm{ij}}$$

(9)

where ${\mathrm{e}}_{\mathrm{j}}$ donates the entropy information of jth hydrochemical parameters

Step 3: Calculate the entropy weight.

$${\mathrm{w}}_{\mathrm{j}}={\displaystyle \frac{1-{\mathrm{e}}_{\mathrm{j}}}{{\sum }_{\mathrm{j}=1}^{\mathrm{n}}(1-{\mathrm{e}}_{\mathrm{j}})}}$$

(10)

where ${\mathrm{w}}_{\mathrm{j}}$ represents the weight of each parameter based on the entropy-weighted method.

2.3.4. Chemical Water Quality Index

Determining the weights for each hydrochemical parameter is a crucial step in water quality evaluation. To obtain integrated weights for each parameter, the additive model was applied in this study. This model incorporates a preference coefficient P, which minimizes the difference between the integrated and individual weights [12]. The calculation for integrated weight and preference coefficient P are as follows:

$${\mathrm{w}}_{\mathrm{c}}=\mathrm{p}{\mathrm{w}}_{\mathrm{f}}+(1-\mathrm{p}){\mathrm{w}}_{\mathrm{j}}$$

(11)

$${\mathrm{Z}}_{\min }=\sum _{\mathrm{j}=1}^{\mathrm{n}}[{\left({\mathrm{w}}_{\mathrm{c}}-{\mathrm{w}}_{\mathrm{f}}\right)}^2+{\left({\mathrm{w}}_{\mathrm{c}}-{\mathrm{w}}_{\mathrm{j}}\right)}^2]$$

(12)

After getting an integrated weight of each hydrochemical parameter, the chemical water quality index (CWQI) can be calculated by the following step:

$${\mathrm{q}}_{\mathrm{j}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{j}}}{{\mathrm{S}}_{\mathrm{j}}}} \times 100$$

(13)

$$\mathrm{CWQI}={\displaystyle \sum _{\mathrm{j}=1}^{\mathrm{n}}}{\mathrm{w}}_{ \mathrm{c}}{ \mathrm{q} }_{\mathrm{j}}$$

(14)

where ${\mathrm{C}}_{\mathrm{j}}$ donates the concentration of each hydrochemcial parameter; ${\mathrm{S}}_{\mathrm{j}}$ represents the Chinese Standard for Groundwater Quality (GB/T 14848 2017 Class III) [34] or the World Health Organization [35].

2.3.5. Sensitivity Analysis

A sensitivity analysis was conducted in this study to further explore the dominant factor affecting drinking water quality and test the reasonability of the selected index system [36]. A lower sensitivity result could respond to the stability of the water quality model [37]. The sensitivity of CWQI can be calculated using Equation (15):

$${\mathrm{S}}_{\mathrm{i}}={\displaystyle \frac{|{\mathrm{V}}_{\mathrm{i}}/N-{\mathrm{v}}_{\mathrm{i}}/n|}{{\mathrm{V}}_{\mathrm{i}}}}\times 100\%$$

(15)

where ${\mathrm{V}}_{\mathrm{i}}$ represents the CWQI scores and N donates the hydrochemical parameters used in water quality calculation; In this study, N is 9; ${\mathrm{v}}_{\mathrm{i}}$ represents the CWQI after moving one hydrochemical parameter; Thus, n is 8 in this study.

3. Results

The descriptive statistical results of the physical and chemical parameters of the groundwater samples in the study area are presented in

Table A1. Statistical analysis of hydrochemical parameters of groundwater samples (Standards are in milligrams per liter).

Parameters	Min	Max	Mean	SD	CV	Limit	% of SEL
pH	7.10	7.96	7.49	0.26	0.35	6.5–8.5 *	0
TDS	205.54	486.80	347.98	74.45	100.88	1000.00 *	0
K+	0.92	4.02	2.34	0.91	1.23	-	0
Na+	8.30	62.94	32.73	14.51	19.66	200.00 *	0
Ca2+	35.80	93.22	62.34	18.17	24.62	400.00 *	0
Mg2+	8.67	25.10	13.40	3.44	4.66	150.00 *	0
Cl−	0.12	3.95	0.93	0.83	1.12	250.00 *	0
HCO3−	135.79	351.62	221.66	55.66	75.41	250.00 *	0
SO42−	43.83	221.39	89.03	39.62	53.69	450.00 *	0
NO3−	0.11	52.11	17.25	17.10	23.18	20 **	30.77
F−	0.23	2.47	0.95	0.67	0.90	1.00 **	38.46

Note: SD, standard deviation; CV (%), coefficient of variation; % of SEL, % of samples exceeding acceptable limit; *, WHO Guideline; **, Chinese Guideline.

. To better illustrate the characteristics of the statistical variables in the table, the study presented the violin plot for the aforementioned indicators in Figure 2. The groundwater samples in the study area exhibited a pH range of 7.10 to 7.96, with a mean of 7.49. The TDS (Total Dissolved Solids) values of groundwater varied from 205.54 to 486.80 mg/L, averaging 347.98 mg/L, classified as freshwater (Figure 3a). The TH (Total Hardness) values of groundwater ranged from 127.72 mg/L to 335.96 mg/L, with an average value of 212.29 mg/L. The groundwater in the study area was categorized as soft and moderate hard water (Figure 3b). The primary cation in groundwater was Ca²⁺, which varied between 35.80 and 93.22 mg/L, with an average concentration of 62.34 mg/L. The concentrations of K⁺, Na⁺, and Mg²⁺ in the groundwater were relatively low, ranging from 0.92 to 4.02, 8.30 to 62.94, and 8.67 to 25.10 mg/L, respectively. HCO₃₋ was the predominant anion among the major anions, with concentrations ranging from 135.79 to 351.62 mg/L, and an average value of 221.66 mg/L. SO₄²⁻ ranked second, with a concentration in the range of 43.83 to 221.39 mg/L and a mean value of 89.03 mg/L. The concentration of Cl⁻ was relatively low, ranging from 0.12 to 3.95 mg/L, with an average value of 0.93 mg/L. Ca²⁺ and HCO₃⁻ were the dominant cation and anion in groundwater. Combined with Figure 3, it can be seen that the groundwater samples in the study area were dominated by HCO₃-Ca type. All the above parameters and ions were within the limits, reflecting good water quality. As for the concentrations of NO₃⁻ and F⁻ in the groundwater, samples exceeded the Chinese guideline (20 mg/L for NO₃⁻ and 1 mg/L for F⁻) by 30.77% and 38.46%, respectively, which may be attributed to the impact of agricultural activities.

The spatial distribution maps of hydrochemical parameters are presented in Figure 4. The pH distribution showed a trend of low in the northeast and high in the southwest (Figure 4a). The distribution patterns of TDS, K⁺, Na⁺, Ca²⁺, and HCO₃⁻ were comparable, showing concentrated concentrations in the northeast and scattered concentrations in the central region (Figure 4b–e,g). The spatial distribution map of Mg²⁺ and SO₄²⁻ displayed similar characteristics (Figure 4f,h). The high concentration of NO₃⁻ may be distributed along the river where the land use type is cropland. This phenomenon indicated that agricultural activities were the main factor controlling NO₃⁻ in groundwater. The F⁻ concentrations were mainly distributed in the middle of the study area (Figure 4k).

4. Discussion

4.1. Driving Factors Controlling Groundwater Chemistry

Gibbs diagram is a powerful tool for evaluating the factors influencing the hydrochemical characteristics of groundwater, primarily including evaporation, water-rock interaction, and precipitation as the three key factors [38]. All samples were located within the rock weathering zone, suggesting that water-rock interactions are the primary factors influencing the hydrochemical composition (Figure 5a,b). The Gaillardet diagram further identifies the rock types involved in natural rock-water interactions based on the end-members of evaporation, silicate, and carbonate types [39]. The positions of groundwater samples close to the silicate end member indicated that both silicate weathering and carbonate dissolution were the primary sources of major ions (Figure 5c,d).

By analyzing the relationships among major ions, we can further elucidate the minerals involved in the water-rock interactions. The Cl⁻ vs. Na⁺ diagram can be utilized to trace the sources of Cl⁻ and Na⁺ in groundwater (Figure 6a). All samples were located below the y = x line, indicating that Na⁺ in groundwater in the study area was not only from halite dissolution (Equation (16)). The fact that most samples were plotted below the y = x line indicated that the concentrations of Ca²⁺ and SO₄²⁻ were not derived from gypsum dissolution (Equation (17)) (Figure 6b). The origins of Ca²⁺ and Mg²⁺ ions in groundwater can be identified through the calculation of the molar ratio between (Ca²⁺ + Mg²⁺) and (HCO₃⁻ + SO₄²⁻) (Equations (18) and (19)). The positioning of groundwater samples above the y = x line suggested that silicate dissolution was a significant hydrochemical process (Figure 6c). In Figure 6d, all samples were located near the y = x line and tilted towards the HCO₃⁻ axis, indicating that the hydrochemistry of groundwater was also affected by calcite dissolution. To determine the direction of cation exchange, the chlor-alkali index (CAI-I and CAI-II) was employed for analysis, as described in equations 20 and 21. The majority of water samples exhibited CAI-I and CAI-II values below zero, suggesting that the excess Na⁺ concentrations stemmed from cation exchange reactions in addition to silicate dissolution (Figure 6e). Calculation of the saturation index for various minerals revealed that the saturation index values of calcite, dolomite, and fluorite were close to zero, indicating that these three minerals were in a state of saturation in the groundwater (Figure 6f). Given the negative saturation indices, it can be inferred that gypsum and rock salt were in a soluble state in groundwater (Figure 6f).

In Figure 4e, elevated concentrations of Ca²⁺ were observed in the north and southwest of the study area. High concentrations of HCO₃⁻ were observed in the northern part of the study area (Figure 4g), where the geological strata are carbonates. The calcite dissolution led to the concentration of HCO₃⁻ and Ca²⁺ in the northern study area. Furthermore, high concentrations of SO₄²⁻ were identified in the southwest of the study area. The scatter plot of Ca²⁺ and SO₄²⁻ have revealed that gypsum dissolution was a source of calcium ions. Therefore, it can be inferred that the gypsum dissolution in the southwestern part of the study area contributed to the elevated concentrations of Ca²⁺ and SO₄²⁻. All in all, combined with the water-rock interactions, the spatial distribution of hydrochemical parameters, and the geological background, it is concluded that high concentrations of Ca²⁺ in the study area are the result of both gypsum and calcite dissolution.

$$\mathrm{NaCl} \to \mathrm{N}{\mathrm{a}}^{+}+\mathrm{C}{\mathrm{l}}^{-}$$

(16)

$$\mathrm{CaS}{\mathrm{O}}_4·2{\mathrm{H}}_2\mathrm{O} \rightleftharpoons \mathrm{C}{\mathrm{a}}^{2+}+\mathrm{S}{\mathrm{O}}_4^{2-}+2{\mathrm{H}}_2\mathrm{O}$$

(17)

$$\mathrm{CaC}{\mathrm{O}}_3+\mathrm{C}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}\to \mathrm{C}{\mathrm{a}}^{2+}+2{\mathrm{HCO}}_3^{-}$$

(18)

$${\mathrm{CaMg}(\mathrm{C}{\mathrm{O}}_3)}_2+2\mathrm{C}{\mathrm{O}}_2+2{\mathrm{H}}_2\mathrm{O}\to {\mathrm{Mg}}^{2+}+{\mathrm{Ca}}^{2+}+4{\mathrm{HCO}}_3^{-}$$

(19)

$$\mathrm{CAI}-\mathrm{I}=\mathrm{C}{\mathrm{l}}^{-}-(\mathrm{N}{\mathrm{a}}^{+}+{\mathrm{K}}^{+})/\mathrm{C}{\mathrm{l}}^{-}$$

(20)

$$\mathrm{CAI}-\mathrm{II}=(\mathrm{C}{\mathrm{l}}^{-}-(\mathrm{N}{\mathrm{a}}^{+}+{\mathrm{K}}^{+}\left)\right)/(\mathrm{HC}{\mathrm{O}}_3^{-}+\mathrm{S}{\mathrm{O}}_4^{2-}+\mathrm{C}{\mathrm{O}}_3^{2-}+\mathrm{N}{\mathrm{O}}_3^{-})$$

(21)

4.2. Groundwater Suitability for Drinking Purposes

In this study, hydrochemical parameters (pH, TDS, Na⁺, Ca²⁺, Mg²⁺, Cl⁻, SO₄²⁻, NO₃^−, and F⁻) were selected for drinking water quality assessment. For factor analysis, the KMO value was 0.65 and the p-value was 0, demonstrating the suitability in weight calculations. The relative weight calculated by factor analysis was tabulated in

Table 1. The relative weight of each parameter for drinking water evaluation.

Parameter	pH	TDS	Na+	Ca2+	Mg2+	Cl−	SO42−	NO3−	F−
Wf	0.138	0.134	0.083	0.130	0.166	0.057	0.151	0.097	0.044
Wj	0.093	0.052	0.072	0.095	0.079	0.153	0.131	0.174	0.151
Wc	0.116	0.093	0.077	0.112	0.122	0.105	0.141	0.136	0.098

. Mg²⁺ showed the highest relative weight (0.166), followed by SO₄²⁻ (0.151), pH (0.138), TDS (0.134), Ca²⁺ (0.130), NO₃⁻ (0.097), Na⁺ (0.083), Cl⁻ (0.057), and F⁻ (0.044). Entropy weights were also determined by the entropy weight method. NO₃⁻ (0.174) had the largest weight as it provided the largest effective information entropy. Other large effective information values were obtained from Cl⁻ (0.153) and SO₄²⁻ (0.131), with relative weight values of 0.153 and 0.131. Both objective methods showed distinct results concerning selected parameters. In order to get reasonable and more accurate evaluation results, a weight integrated with W_f and W_j was calculated using an additive model (Table 1). The integrated weight emphasizes the impact of toxic elements on groundwater quality as the NO₃⁻ had the largest relative weight. SO₄²⁻ had a relative value of 0.141, followed by Mg²⁺ (0.122), pH (0.116), Ca²⁺ (0.112), Cl⁻ (0.105), F⁻ (0.098), TDS (0.093), and Na⁺ (0.077).

After obtaining the integrated weight of each hydrochemical parameter, drinking water quality was calculated (Figure 7a). The classic method categorized water quality into five ranks: excellent, good, medium, poor, and extremely poor, with water quality values of 0–50, 50–100, 100–150, 150–200, and >200, respectively. A total of 23 groundwater samples (73.08%) were classified as excellent for drinking, and 26.92% of samples were good for drinking. Zones of good drinking rank were similar to areas of high concentrations of NO₃⁻ and Ca²⁺, indicating both agricultural activities and the geological process could have an effect on water quality (Figure 7b). Overall, groundwater across the study area was suitable for drinking.

By calculating integrated weights of removing one hydrochemical parameter, the $v_i$ was computed and the sensitivity of each parameter was drawn in Figure 8. The effects of removing Cl⁻ (3.420%) are the most positive to the CWQI, followed by NO₃⁻ with an average sensitivity of 2.649%. TDS showed the smallest impact on CWQI with a mean value of 1.426%. Based on the results of sensitivity analysis, all selected parameters were not particularly sensitive to the CWQI values, demonstrating the suitability and robustness of the drinking water quality evaluation model. Moreover, the sensitivity analysis also proved that the hydrochemical process in the study area is controlled by both natural factors and human activities. The driving factors controlling groundwater chemistry showed that Cl⁻ in groundwater may come from the dissolution of halite. The concentration of NO₃⁻ could respond to agricultural activities.

4.3. Protection and Management Measures for Groundwater Resources

Based on a comprehensive analysis of the groundwater samples in the study area, the overall water quality was found to be generally good. However, there was a notable tendency for NO₃⁻ concentrations to exceed the standards in certain locations, indicating a potential deterioration of water quality. To ensure the continued sustainability and safety of the groundwater resources, several specific protection and management measures have been proposed:

• Implementing stricter regulations to address the elevated levels of NO₃⁻ in groundwater, which is primarily caused by agricultural activities. This could include enforcing guidelines on the use of fertilizers and pesticides to minimize the infiltration of nutrients and pollutants into the groundwater.
• Raising public awareness about the importance of groundwater protection and the harmful effects of excessive groundwater extraction. Educational campaigns would encourage individuals and businesses to adopt water-saving practices, such as the development of efficient irrigation systems to safeguard groundwater resources.
• Supporting the development of advanced technologies for groundwater monitoring, management, and protection. This includes establishing a network of monitoring stations, regularly analyzing groundwater samples to assess water quality, and closely tracking water quality trends in order to implement timely and effective protective measures.

5. Conclusions

This study provides a comprehensive assessment of the hydrochemical characteristics and drinking water quality of groundwater in the northwest of the Sichuan Basin. The analysis employed a geochemical model and a chemical water quality index (CWQI), offering valuable insights into the hydrochemical processes, groundwater quality, and its suitability for drinking purposes:

• The groundwater in the region was characterized by the HCO₃-Ca type, attributed to silicate weathering and calcite dissolution. Agricultural activities significantly influenced the concentrations of NO₃⁻, with 30.77% of samples exceeding the Chinese guidelines for drinking water. Cation exchange reaction played a crucial role in the enrichment of Na⁺ concentrations. The mineral saturation index of the groundwater indicated that calcite, dolomite, and fluorite were in a saturated state, with the gypsum and halite being soluble.
• The integrated weight analysis highlighted the significant impact of toxic elements on groundwater quality, with NO₃⁻ having the highest relative weight. Overall, the chemical water quality index (CWQI) revealed that the groundwater in the study area was deemed suitable for drinking, with 73.08% of the samples classified as excellent and 26.92% categorized as good. The good water quality grades in regions with elevated concentrations of NO₃⁻ and Ca²⁺ suggested that agricultural activities and geological processes played a significant role in shaping water quality.
• The sensitivity analysis results indicated that the removal of Cl⁻ had the most significant effect on the CWQI (average value of 3.420%), while TDS had the least impact (mean value of 1.426%). This result not only demonstrated the applicability and robustness of the drinking water quality assessment model but also highlighted that the hydrochemical processes were influenced by both natural factors and human activities.