Hydrogeochemical Characteristics and Sulfate Source of Groundwater in Sangu Spring Basin, China

Abstract

The Sangu Spring Basin is located in an important economic area, and groundwater is the main source of water for local life and industry. Understanding the sources of chemical components in groundwater is important for the development and utilization of groundwater. In this paper, we analyzed the origin of the chemical components of groundwater and their evolution in the Sangu Spring Basin using statistical analysis, Piper diagrams, Gibbs diagrams, ion ratios, and combined hydrochemistry–isotope analyses. The results show that the groundwater in the Sangu Spring Basin is mainly derived from atmospheric precipitation, that the groundwater in stagnant and confined environment zones was formed under colder climatic conditions, and that the surface water (SW) has a close hydraulic relation with the groundwater. Water–rock interaction is the main factor controlling the composition of groundwater. The compositions of groundwater are mainly derived from carbonate weathering, silicate weathering, and dissolution of gypsum. Na⁺ and K⁺ in groundwater mainly come from the dissolution of albite and potassium feldspar, rather than rock salt. Ion exchange occurs in karst groundwater (KGW) and fissure groundwater (FGW), and ion exchange is dominated by the exchange of Mg²⁺ and Ca²⁺ in the groundwater with Na⁺ and K⁺ in the rock or soil. Sulfate in groundwater is derived from dissolution of gypsum, infiltration of atmospheric precipitation, and leakage of SW. Groundwaters with the highest sulfate content are located in the vicinity of SW, as a result of receiving recharge from SW seepage. Groundwaters with higher sulfate contents are located in the stagnant and deeply buried zones, where sulfate is mainly derived from the dissolution of gypsum. SW seepage recharges groundwater, resulting in increased levels of Cl⁻, NO₃⁻ and SO₄²⁻ in groundwater. These insights can provide assistance in the protection and effective management of groundwater.

1. Introduction

The Sangu Spring Basin is one of the important coal-producing areas in China and has an important economic status. Coal mining and chemical production are the main economic activities in the study area. The Sangu Spring Basin is located in arid and semi-arid areas, and, coupled with the lack of SW resources and poor SW quality, this means that groundwater has become the main source of local industrial water and domestic water [1,2]. Groundwater is an essential element of the hydrological cycle and plays a key role in the base flow of rivers and the maintenance of terrestrial aquatic ecosystems [3,4]. There is a close hydrological connection between groundwater and SW in the Sangu Spring Basin, so it is of great significance to study the evolution of groundwater hydrochemistry for the management of groundwater.

Hydrochemical components and isotopes of groundwater are often used to analyze problems in hydrogeology [5,6,7,8]. Gibbs diagrams, Piper diagrams, deuterium–oxygen relationship diagrams, ion ratios, and combined hydrochemistry–isotope analyses are commonly used to analyze the origin and evolution of the chemical composition of groundwater [9,10]. Stable isotopes are the tools for studying the origin of groundwater, recharge elevations, recharge mechanisms, and hydraulic relations, and for calculating the mixing ratio of different water sources [11,12,13]. Combined use of δ³⁴S-SO₄²⁻, δ¹⁸O-SO₄²⁻, δ¹⁸O-H₂O, δD-H₂O, and hydrochemistry–isotopes can be used to study the sources of sulfate in groundwater or SW and calculate the mixing ratios between different water sources [14,15,16]. Therefore, the method of combining hydrochemistry with isotopes can determine the geochemical factors and mechanisms that control the chemical composition of groundwater.

Previous studies on groundwater in the Sangu Spring Basin have mainly focused on the determination of spring boundaries [17,18,19], evaluation of groundwater quality [20,21], evaluation of water resources [17,22], decline of the groundwater level [23,24], protection of aquifers [2,25,26], and the impact of coal mining on the groundwater environment [27,28]. Only Zhang et al. [29] have studied the hydrochemical characteristics and evolution mechanism of KGW in the Sangu Spring Basin. At present, there is no systematic and comprehensive investigation on the hydrogeochemical behavior of groundwater in the Sangu Spring Basin. This paper studies the sources of hydrochemical components and their evolution of pore groundwater (PGW), FGW, and KGW in the Sangu Spring Basin by using statistical analysis, Gibbs diagrams, Piper diagrams, ion ratios, and joint hydrochemical–isotope analysis. The research results provide a scientific basis for the protection, pollution prevention, exploitation, and management of groundwater in the spring area.

2. Study Area

2.1. General Setting

The Sangu Spring Basin is located in the southeast of Shanxi province (Figure 1), China, with an area of 3214 km². The Sangu Spring is exposed in the valley of the Dan River, which has now been flooded by the Qingtianhe Reservoir. The Sangu Spring Basin is surrounded by high mountains, with undulating intermountain basins and hills in the middle. The main rivers in Sangu Spring Basin are the Dan River, Baishui River, and Dongdan River, which belong to the Yellow River system. There are two medium-sized reservoirs, Renzhuang Reservoir and Qingtianhe Reservoir, as well as numerous small reservoirs in the Sangu Spring Basin. The Sangu Spring Basin belongs to the warm, temperate, semi-humid continental monsoon climate with four distinct seasons. The average annual temperature is 13.7 °C. The average annual precipitation is 593.66 mm and the average annual evaporation is 1633.95 mm [1,30].

2.2. Geological and Hydrogeological Setting

The outcrop strata in the Sangu Spring Basin include Cambrian, Ordovician, Carboniferous, Permian, and Quaternary strata. The Cambrian strata are distributed in a small area and only appear in deep gullies in the southeast of the spring region. The Ordovician and Carboniferous strata are widely distributed in the area, with large areas exposed in the western, central, eastern, and southwestern parts of the study area. The Quaternary strata are mainly exposed in the hills and basins between Gaoping City and Jincheng City. The Taiyuan Formation of the Carboniferous and the Shanxi Formation of the Permian systems are coal-bearing strata, which are widely distributed in the central and northern parts of the spring region.

The main aquifers (generalized) from bottom to top in the Sangu Spring Basin are Cambrian limestone aquifers, Ordovician limestone aquifers, Carboniferous thin limestone and sandstone aquifers, Permian sandstone aquifers, and Quaternary loose rock aquifers (Figure 2). KGW is mainly contained in the Cambrian and Ordovician strata, with the Zhangxia Formation of the Middle Cambrian and the Majiagou Formation of the Middle Ordovician as the main aquifers. The upper Cambrian and lower Ordovician Sanshanzi formations are a water-resisting layer in the region. FGW is stored in the Carboniferous and Permian strata, with the Taiyuan Formation limestone of the Carboniferous, the sandstone of the Shanxi Formation, and the Sandstone of the Shihezi Formation of the Permian as the main water-bearing layers. The Carboniferous Benxi Formation bauxite mudstone and Permian mudstone are water-resisting layers. PGW is stored in the Quaternary strata, and the aquifers are sand and gravel layers.

KGW is the most widely distributed and best water-rich groundwater in the Sangu Spring Basin, and it is also the most important source of water supply. FGW is the groundwater that is most affected by coal mining, and most of the springs’ flow rates become smaller or dry, and the water supply function becomes weaker gradually. PGW is mainly located in the mountain basins and river terrace in the middle of the spring area, with the smallest distribution area, thin aquifer thickness, and uneven water-richness.

Characteristics of the supplement, runoff, and draining of KGW are affected by various factors such as geological structure, stratigraphic lithology, distribution of modern water networks, and human activities [1,17,30].

The recharge sources of KGW include infiltration of atmospheric precipitation, leakage of surface water, and leaking recharge of the overlying aquifer. It is recharged by direct infiltration of atmospheric precipitation in the exposed areas of limestone, by leakage of surface water in the valleys and reservoirs of the limestone area, and by leaking recharge from the overlying aquifers in the overlying and buried areas.

There are three strong-flow zones in the KGW of the Sangu Spring Basin (Figure 1): one along the middle and lower valleys of the Dan River, one along the Gaoping–Bagong–Beishidian–Jincheng City line, and another along the valley of the Dongdan River [1,17,18].

The drainage methods of KGW include springs and artificially extracted groundwater. Drainage of spring water is mainly concentrated in the Guobi Spring (flow rate 0.4–0.66 m³/s) and the Sangu Spring (flow rate 3.5–7.0 m³/s). KGW is the main source of water for industrial production and domestic use in the Sangu Spring area, and groundwater exploitation is mainly distributed in the north–central part of the spring area, with an extraction volume of about 1.1568 × 10⁸ m³/a.

3. Materials and Methods

3.1. Sample Collection and Treatment

From 2014 to 2015, 234 groundwater samples, 27 SW samples, and 12 atmospheric precipitation samples were collected, with the sampling point locations shown in Figure 1. Prior to the collection of well-water samples, the wells were first pumped until the electrical conductivity (EC) of the well water was stabilized. The samples were filtered through a 0.45 μm membrane filter and then placed into 1500 mL and 550 mL polyethylene bottles for analysis of major ionic constituents and trace elements. Nitric acid was added to the 550 mL samples until the pH was below 2, and the 1500 L samples did not need to be treated. The samples were stored below 4 °C until they were analyzed. Samples for stable isotope analysis (δ¹⁸O-H₂O and δD-H₂O) were collected into 50 mL glass bottles with a sealed mouth.

From 234 groundwater samples, 152 samples were selected for δ¹⁸O-H₂O and δD-H₂O analysis. A total of 12 atmospheric precipitation samples were all analyzed for δ¹⁸O-H₂O and δD-H₂O. Sixteen of the 27 SW samples were selected for δ¹⁸O-H₂O and δD-H₂O analysis.

A total of 33 samples were selected from 234 groundwater samples for δ¹⁸O-SO₄²⁻ and δ³⁴S-SO₄²⁻ analysis. Two samples were selected from 27 SW samples for δ¹⁸O-SO₄²⁻ and δ³⁴S-SO₄²⁻ analysis. The δ¹⁸O-SO₄²⁻ and δ³⁴S-SO₄²⁻ analyses were performed on 5 samples from 12 atmospheric precipitation samples. Samples of δ¹⁸O-SO₄²⁻ and δ³⁴S-SO₄²⁻ isotopes were collected in 5 L polyethylene containers and acidified with HCl to pH 2–3, and then saturated BaCl₂ solution was added to precipitate BaSO₄.

3.2. Analytical Methods

The hydrochemical components were determined by the Groundwater-Mineral Water and Environmental Monitoring Center, Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences. Major cations and trace elements were analyzed using a PerkinElmer Inductively Coupled Plasma Emission Spectrometer Optima Model 8300 (accuracy ±1%). The reliability of the water chemistry data was assessed by examining the ionic balance, with the ion charge imbalances within ±5%.

The concentration of HCO₃⁻ was determined by phenolphthalein titration. Except for HCO₃⁻, major anions were analyzed using a Thermo Scientific Dionex ICS-4000 (Dionex, Sunnyvale, CA, USA; accuracy: ±1%). The reliability of the water chemistry data was assessed by checking that the ionic balances were within ±5%.

The δ¹⁸O-H₂O and δD-H₂O isotopes were determined by the Picarro L 2130-i analyzer (Picarro, Santa Clara, CA, USA) of the Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences. The data are expressed in δ-symbols (δ) and have a precision of ±0.1‰ for δ¹⁸O-H₂O and ±1‰ for δD-H₂O relative to Vienna Standard Mean Ocean Water (V-SMOW).

δ¹⁸O-SO₄²⁻ and δ³⁴S-SO₄²⁻ were analyzed in the Analytical Laboratory Beijing Research Institute of Uranium Geology (ALBRIUG). δ¹⁸O-SO₄²⁻ was determined by coupling CO production with a Finnigan MAT 253 on a Thermal Conversion Elemental Analyzer (TCEA, Thermo Fisher Scientific, Bremen, Germany) at a temperature of 1450 °C. δ³⁴S-SO₄²⁻ was determined by coupling SO₂ production with a Finnigan MAT 253 Isotope Ratio Mass Spectrometer (IRMS) on a TCEA (Thermo Fisher Scientific, Bremen, Germany). The working standards for δ¹⁸O-SO₄²⁻ were GBW 04409, GBW 04410, and NBS 28. The working standards for δ³⁴S-SO₄²⁻ measurements were GBW 04414 and GBW 04415. δ³⁴S-SO₄²⁻ values corresponded to the Vienna Canyon Diablo Troilite (VCDT) standard, and δ¹⁸O-SO₄²⁻ values correspond to the V-SMOW standard. The accuracy of the δ¹⁸O-SO₄²⁻ and δ³⁴S-SO₄²⁻ analyses was better than ±0.5‰ and ±0.2‰, respectively, based on repeated analyses of internal standards.

Maximum, minimum, mean, and standard deviation of water chemistry components and deuterium–oxygen isotopes were calculated using IBM SPSS Statistics 26 statistical analysis software.

4. Results

4.1. Hydrochemical Characterization of Groundwater

The basic characteristics of the groundwater are shown in

Table 1. Hydrochemical components of groundwater in the Sangu Spring Basin.

	K+	Na+	Ca2+	Mg2+	Cl−	SO42−	HCO3−	NO3−	TDS	pH
KGW (n = 137)
Min	0.26	3.54	49.37	12.59	0.36	21.86	159.40	0.00	218.70	7.20
Max	20.40	198.00	487.70	130.00	232.20	1413.00	441.10	378.50	2411.00	8.57
Mean	1.50	43.61	126.23	33.42	24.18	230.89	301.21	23.72	653.18	7.73
Std	2.45	52.88	60.64	15.71	26.07	207.25	39.64	38.04	337.31	0.25
FGW (n = 74)
Min	0.07	4.62	54.12	8.43	4.58	23.6	10.80	1.98	245.80	6.71
Max	7.78	282.60	307.00	51.40	230.10	846.00	465	286.00	2424.50	8.51
Mean	0.96	67.49	136.60	23.32	30.85	253.30	293.53	35.75	776.23	7.79
Std	1.09	74.44	56.57	10.33	35.11	213.34	85.12	1.56	408.26	0.33
PGW (n = 23)
Min	0.28	10.80	58.30	12.04	8.81	40.38	250.00	1.63	330.50	7.35
Max	9.45	196.00	286.90	60.38	101.00	521.00	478.60	227.50	1226.00	8.28
Mean	1.47	57.00	145.73	26.39	37.49	233.21	324.10	45.52	796.10	7.78
Std	1.92	57.81	54.32	11.26	25.24	135.76	50.89	47.37	231.40	0.26

Note on units: concentrations are expressed in milligram per liter (mg/L), except for pH.

and Figure 3.

A total of 137 samples of KGW were collected in the Sangu Spring Basin. The pH of the KGW ranged from 7.20 to 8.57, and the TDS ranged from 218.7 to 2411.0 mg/L. The hydrochemical types were dominated by HCO₃-Ca, HCO₃-Ca·Mg, HCO₃·SO₄-Ca·Mg, SO₄·HCO₃-Ca·Mg, SO₄·HCO₃-Na·Ca, and HCO₃·SO₄-CaHCO3-Ca. HCO₃⁻ and SO₄²⁻ were the dominant anions, and Ca²⁺, Mg²⁺, and Na⁺ were the dominant cations.

A total of 74 samples of FGW were collected in the Sangu Spring Basin. The pH of the FGW ranged from 6.71 to 8.51, and the TDS ranged from 245.80 to 2424.50 mg/L. The hydrochemical types were dominated by HCO₃-Ca, HCO₃·SO₄-Ca, SO₄·HCO₃-Ca, and SO₄·HCO₃-Ca·Na. HCO₃⁻ and SO₄²⁻ were the dominant anions, and Ca²⁺ and Na⁺ were the dominant cations.

A total of 23 PGW samples were collected from the Sangu Spring Basin. The pH of PGW ranged from 7.35 to 8.28, and the TDS was 330.5–1226.0 mg/L. The hydrochemical types were dominated by HCO₃·SO₄-Ca, SO₄·HCO₃-Ca (Na), SO₄·HCO₃-Na (Ca), HCO₃-SO₄-Ca, SO₄-HCO₃-Ca (Na), and SO₄-HCO₃-Na (Ca). HCO₃⁻ and SO₄²⁻ were the dominant anions, and Ca²⁺ and Na⁺ were the dominant cations.

4.2. Isotope Characterization of Groundwater

The δD-H₂O and δ¹⁸O-H₂O values of the KGW ranged from −83.0‰ to −42.0‰ and from −11.2‰ to −4.5‰, respectively. The δD-H₂O and δ¹⁸O-H₂O values of the FGW ranged from −83.0‰ to −57.0‰ and from −11.5‰ to −8.1‰, respectively. The δD-H₂O and δ¹⁸O-H₂O values of the PGW ranged from −69.0‰ to −48.0‰ and from −9.6‰ to −5.8‰, respectively. The δD-H₂O and δ¹⁸O-H₂O values of the SW ranged from −70.0‰ to −42.0‰ and from −9.1‰ to −3.9‰, respectively (

Table 2. δD-H₂O and δ¹⁸O-H₂O isotope composition of all the water samples.

	KGW (n = 112)				FGW (n = 32)
	Min	Max	Mean	Std	Min	Max	Mean	Std
δD-H2O (‰)	−83.0	−42.0	−70.5	4.03	−83.0	−57.0	−67.7	4.68
δ18O-H2O (‰)	−11.2	−4.5	−9.6	0.57	−11.57	−8.1	−9.2	0.67
	PGW (n = 8)				SW (n = 16)
	Min	Max	Mean	Std	Min	Max	Mean	Std
δD-H2O (‰)	−69.0	−48.0	−63.0	6.74	−70.0	−42.0	−56.3	8.16
δ18O-H2O (‰)	−9.6	−5.8	−8.5	1.17	−9.1	−3.9	−7.4	1.38

). The δD-H₂O and δ¹⁸O-H₂O contents of SW and PGW were higher, while the δD-H₂O and δ¹⁸O-H₂O contents of some KGW and FGW were lower. The δD-H₂O and δ¹⁸O-H₂O contents of KGW and FGW had a wide range and were closely related to their hydrogeological conditions, and groundwaters that directly receive infiltration recharge from atmospheric precipitation or seepage recharge from SW had higher δD-H₂O and δ¹⁸O-H₂O contents.

5. Discussion

5.1. Groundwater Origins Indicated by δD-H₂O and δ¹⁸O-H₂O

The isotopes of δD-H₂O and δ¹⁸O-H₂O are commonly used to analyze the origin of groundwater [13]. The relationship between δD-H₂O and δ¹⁸O-H₂O in the samples is shown in Figure 4. The local meteoric water line (LMWL) is fitted to δD = 7.78δ¹⁸O-H₂O + 10.33 [1], and its slope is slightly lower than the global meteoric water line (GMWL:δD = 8δ¹⁸O-H₂O + 10) [31,32]. All groundwater samples were distributed along the atmospheric precipitation line (Figure 4), showing the origin of groundwater atmospheric precipitation.

SW is mainly distributed in Group 4, and some SW showed a drift of δ¹⁸O-H₂O (Figure 4). SW close to the LMWL is mainly formed by the recent atmospheric precipitation recharge or received groundwater recharge, such as the middle and lower reaches of the Baishui River and the lower reaches of the Dan River. SW with δ¹⁸O-H₂O drift is mainly affected by evaporation. For example, J53 is a sample of reservoir water from Shan’erdong Reservoir, and the water level of the reservoir was at the stagnant level when the sample was taken, with less recharge and strong evaporation, so there was a strong δ¹⁸O-H₂O drift.

There are one PGW sample and one KGW sample in Group 4 (Figure 4). P11 is shallow PGW, located in the first terrace of Dan River. Its recharge sources are atmospheric precipitation and river-water infiltration (the river water comes from the reservoir). It is assumed that the higher deuterium–oxygen isotope content is due to the combined effect of recharge from the river water and evaporation. P10-f is KGW, and it is assumed that its higher deuterium–oxygen isotope content is due to the recharge of SW from Renzhuang Reservoir. The staggered distribution of SW and groundwater indicates a close hydraulic relationship between SW and groundwater.

The mixed distribution of SW, PGW, FGW, and KGW in Group 3 (Figure 4) indicates that there is a close hydraulic relationship between them. For example, the higher deuterium and oxygen isotope content of Je35, D112, etc. in KGW is due to the fact that they receive recharge from overburdened PGW, which in turn receives seepage recharge from the Dan River, and the river water is affected by evaporation. For instance, the shallow KGW (XB90, XA165, E184) directly receives infiltration recharge from atmospheric precipitation, has a short cycle path, a shallow water level, and is affected by a certain degree of evapotranspiration, resulting in a high deuterium–oxygen isotope content.

Group 2 is dominated by KGW (Figure 4), located in a zone of weak runoff, with slow water turnover and tritium levels of less than 7 TU. The groundwater may contain groundwater recharged during colder climates.

The groundwater in Group 1, which includes two KGW and two FGW samples, has the lowest deuterium–oxygen isotope content (Figure 4). The two KGW samples in Group 1 are located on the boundary of the Sangu Spring Basin, which is in the stagnant-flow area of KGW, the water alternation is extremely slow, and the groundwaters may have formed in a period of colder climate [6,7,32]. The two FGW samples in Group 1 are water discharged during the extraction of coalbed methane (ZC38, ZC112), and the groundwaters are in a confined environment, speculated to have been formed during a period of colder climate [33].

5.2. Causes of Groundwaters

The Gibbs diagram can be used to study the major diagenetic mechanisms of groundwater evolution. At present, this method is widely used to identify the sources of groundwater chemical composition [34,35]. The Gibbs diagram is generally divided into rock weathering dominant zones, evaporation dominant zones, and precipitation dominant zones, and the water–rock action dominant zones can be further divided into carbonate dissolution dominant zones and silicate dissolution dominant zones [36]. Groundwater samples from the Sangu Spring Basin were mainly distributed in areas where rock weathering is dominant (Figure 5), indicating that groundwater is mainly controlled by water–rock interaction. The Na⁺/(Na⁺ + Ca²⁺) of groundwater varies in a larger range, while TDS varies in a smaller range, indicating that cation exchange may influence the variation of Na⁺ and Ca²⁺ content in groundwater [37]. Rock weathering is the main factor affecting the hydrochemical components of groundwater and does not prove that groundwater formation mechanisms are not affected by human activities. The influence of human activities includes NO₃⁻, SO₄²⁻, and Cl⁻ directly or indirectly caused by acid rain, sewage, and chemical fertilizer running into groundwater.

The relationship diagrams for Mg²⁺/Na⁺ versus Ca²⁺/Na⁺ and HCO₃⁻/Na⁺ versus Ca²⁺/Na⁺ were plotted (Figure 6), through which the sources of dissolved substances in the groundwater can be further determined [38,39,40,41]. KGW, FGW, and PGW are located on the weathering line of silicates–carbonates and near the end element (Figure 6), indicating that carbonate weathering and silicate weathering are the main controlling factors of groundwater components in the Sangu Spring Basin. FGW is located on the weathering line of silicate–carbonate, which is consistent with the nature of the surrounding rocks of the aquifer. The aquifers of FGW are sandstone and limestone, especially in the Carboniferous strata, where limestone is the main aquifer. The PGW is also located in the weathering line of silicate–carbonate, which is consistent with the hydrogeological conditions. PGW is mainly located in intermountain basins and river terraces, and groundwater from the surrounding mountains is the main source of recharge.

5.3. Major Ion Ratios of Groundwater and Their Implications for Mineral Dissolution

Ion ratios are a common method for analyzing the source of groundwater components and have been widely used [7,35,42,43]. The chemical reaction equations for carbonate and silicate weathering are shown in Equation (1) through Equation (3). As a result of weathering, the anions increase HCO₃⁻, SO₄²⁻, and NO₃⁻ and the cations increase Ca²⁺, Mg²⁺, Na⁺, and K⁺ in the chemical fractions of groundwater.

Carbonate minerals weathering:

$$\begin{gathered} 4\left(C{a}_{1-\mathrm{x}}M{g}_x\right)C{O}_3\left(\mathrm{s}\right)+H_{2}C{O}_3+H_{2}S{O}_4+HN{O}_3\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em} \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\to 4(1-\mathrm{x}){\mathrm{C}a}^{2+}\left(\mathrm{aq}\right)+4\mathrm{xM}{g}^{2+}\left(aq\right)+5HC{O}_3^{-}\left(\mathrm{aq}\right)+N{O}_3^{-}\left(\mathrm{aq}\right)+S{O}_4^{2-}\left(aq\right) \end{gathered}$$

(1)

Silicate minerals weathering:

$$\begin{gathered} CaA{l}_{2}S{i}_2{O}_8\left(\mathrm{s}\right)+C{O}_2\left(\mathrm{g}\right)+2H_{2}O\left(l\right)+H_{2}S{O}_4+HN{O}_3\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em} \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\to 2\mathrm{C}{a}^{2+}\left(aq\right)+HC{O}_3^{-}\left(aq\right)+N{O}_3^{-}\left(aq\right)+S{O}_4^{2-}\left(aq\right)+4Si{O}_2+Al{(OH)}_3+3AlOOH \end{gathered}$$

(2)

$$\begin{gathered} 4Na\left(K\right)AlS{i}_3{O}_8\left(\mathrm{s}\right)+C{O}_2\left(\mathrm{g}\right)+2H_{2}O\left(l\right)+H_{2}S{O}_4+HN{O}_3\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em} \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\to 4\mathrm{N}{a}^{+}\left(K^{+}\right)\left(aq\right)+HC{O}_3^{-}\left(aq\right)+N{O}_3^{-}\left(aq\right)+S{O}_4^{2-}\left(aq\right)+12Si{O}_2+Al{(OH)}_3+2AlOOH \end{gathered}$$

(3)

The Ordovician strata in the Sangu Spring Basin contain gypsum intercalations, and the dissolution of gypsum can lead to the entry of Ca²⁺ and SO₄²⁻ into the groundwater (Equation (4)). If Ca²⁺ and SO₄²⁻ in KGW are mainly derived from gypsum dissolution, then the γCa²⁺/γSO₄²⁻ (γCa²⁺ represents the milligram equivalent of calcium ions in groundwater, and γSO₄²⁻ represents the milligram equivalent of sulfate ions in groundwater, in meq/L.) ratio is 1. KGW samples are not distributed along the line of gypsum dissolution (Figure 7a), which indicates that gypsum dissolution is not one of the main sources of Ca²⁺ and SO₄²⁻ in KGW.

$$C{a}_{2}S{O}_4\left(\mathrm{s}\right)\to \mathrm{C}{a}^{2+}\left(\mathrm{aq}\right)+S{O}_4^{2-}\left(aq\right)$$

(4)

If only carbonate weathering occurs, then γ(Ca²⁺ + Mg²⁺): γ(HCO₃⁻ + CO₃²⁻ + SO₄²⁻ + NO₃⁻) = 1:1, and the groundwater should be distributed along the carbonate weathering line (Figure 7c). If only silicate weathering occurs, then γ(Ca²⁺ + Na⁺ + K⁺): γ(HCO₃⁻ + CO₃²⁻ + SO₄²⁻ + NO₃⁻) = 1:1. If carbonate weathering, silicate weathering, and sulfate dissolution occur at the same time, then γ(Ca²⁺ + Mg²⁺ + Na⁺ + K⁺ − Cl⁻): γ(HCO₃⁻ + CO₃²⁻ + SO₄²⁻ + NO₃⁻) = 1:1. γ(Ca²⁺ + Mg²⁺ + Na⁺ + K⁺ − Cl⁻) represents the cationic component of carbonate weathering, silicate weathering, and sulfate dissolution into the groundwater. In the plot of γ(Ca²⁺ + Mg²⁺ + Na⁺ + K⁺ − Cl⁻) versus γ(HCO₃⁻ + CO₃²⁻ + SO₄²⁻ + NO₃⁻) (Figure 7b), the PGW, FGW, and KGW are all distributed along the line of 1:1, which indicates that the chemical fractions of the groundwater in the Sangu Spring Basin watershed mainly originated from carbonate weathering, silicate weatherig, and dissolution of sulfate.

Weathering of salt rocks causes increases in Na⁺, K⁺, and Cl⁻ content in groundwater (Equations (5) and (6)). The groundwater is not distributed along the dissolution line of the salt rocks (Figure 7d), which indicates that the dissolution of the salt rocks is not the main source of Na⁺ and K⁺ in the groundwater in the Sangu Spring Basin. Dissolution of carbonic, sulfuric, and nitric acids with sodium feldspar (NaAlSi₃O₈) and potassium feldspar (KAlSi₃O₈) (Equation (3)) is the source of Na⁺ and K⁺ content in groundwater.

$$NaCl\left(s\right)\to \mathrm{N}{a}^{+}\left(\mathrm{aq}\right)+\mathrm{C}{l}^{-}\left(aq\right)$$

(5)

$$KCl\left(s\right)\to K^{+}\left(\mathrm{aq}\right)+\mathrm{C}{l}^{-}\left(aq\right)$$

(6)

Precipitation contains SO₂ and NO_x, and their participation promotes the weathering of carbonate and silicate minerals (Equations (1) and (2)). The correlation coefficients between (Ca²⁺ + Mg²⁺) and (HCO₃⁻ + SO₄²⁻ + CO₃²⁻ + NO₃⁻) in KGW, FGW, and PGW are 0.85, 0.78, and 0.76, respectively, showing a significant positive correlation of level 0.01, indicating that the dissolution of minerals in groundwater was mainly controlled by carbonic acid, sulfuric acid, and nitric acid.

5.4. Ion Exchange

The exchange of Ca²⁺ and Mg²⁺ in groundwater with Na⁺ and K⁺ in soil or rock is a common cause of changes in ionic composition in groundwater. The equations for cation exchange occurring in groundwater are shown in Equations (7) and (8). In the Gibbs plots, the Na⁺/(Na⁺ + Ca²⁺) ratios of some groundwater samples are greater than 0.5, indicating that cation exchange has some influence on the change in cation content in groundwater [37,44]. Scherer (1977) [45] proposed a mathematical relationship that can better represent the ion exchange that occurs between groundwater and aquifer rocks, called the chlor–alkali index (CAI). The chlor–alkali index method provides a quantitative analysis of ion exchange and reverse ion exchange processes (Equations (7) and (8)) and is widely used [31,35,46,47].

$$C{a}^{2+}\left(M{g}^{2+}\right)\left(\mathrm{aq}\right)+2Na\left(K\right)X\left(s\right)\to 2\mathrm{N}{a}^{+}\left(K^{+}\right)\left(\mathrm{aq}\right)+Ca\left(Mg\right)X_2\left(s\right)$$

(7)

$$Ca\left(Mg\right)X_2\left(\mathrm{s}\right)+2N{a}^{+}\left(K^{+}\right)\left(\mathrm{aq}\right)\to \mathrm{C}{a}^{2+}\left(M{g}^{2+}\right)\left(\mathrm{aq}\right)+2Na\left(K\right)X\left(\mathrm{s}\right)$$

(8)

$$CAI1=(C{l}^{-}-\left(N{a}^{+}+K^{+}\right))/C{l}^{-}$$

(9)

$$CAI2=(C{l}^{-}-\left(N{a}^{+}+K^{+}\right))/\left(HC{O}_3^{-}+C{O}_3^{2-}+S{O}_4^{2-}+N{O}_3^{-}\right)$$

(10)

If the chemical reaction in Equation (7) occurs, CAI1 < 0, and CAI2 < 0, ion exchange occurs, and Mg²⁺ and Ca²⁺ in groundwater are exchanged with Na⁺ and K⁺ in rock or soil. If the chemical reaction in Equation (8) occurs, CAI1 > 0, and CAI2 > 0, the Na⁺ and K⁺ in groundwater are exchanged with Mg²⁺ and Ca²⁺ in rock or soil. The ranges of CAI1 and CAI2 in KGW were −30.12–5.86 and −0.58–0.08, respectively (Figure 8a). A total of 69.34% of KGW experienced ion exchange, which means that Mg²⁺and Ca²⁺ in KGW were exchanged with Na⁺ and K⁺ in rock or soil. A total of 8.03% of KGW experienced reverse ion exchange, which means that Na⁺ and K⁺ in KGW were exchanged with Mg²⁺and Ca²⁺ in rock or soil. The ranges of CAI1 and CAI2 in FGW were −40.41–0.46 and −0.66–0.10, respectively. A total of 85.14% of the FGW experienced ion exchange, which means that Na⁺ and K⁺ were exchanged with Mg²⁺ and Ca²⁺ in rocks or soils. A total of 12.16% of the FGW experienced reverse ion exchange, which means that Na⁺ and K⁺ were exchanged with Mg²⁺ and Ca²⁺ in rocks or soils. The ranges of CAI1 and CAI2 in PGW were −11.74–0.52 and −0.48–0.10, respectively. A total of 69.57% of the PGW had ion exchange, which means that Mg²⁺ and Ca²⁺ in the PGW were exchanged with Na⁺ and K⁺ in the rock or soil. A total of 26.09% of the PGW had reverse ion exchange, which means that Na⁺ and K⁺ were exchanged with Mg²⁺and Ca²⁺ in the rock or soil.

In recent years, the relationship plot of (K⁺ + Na⁺ − Cl⁻) versus ((Ca²⁺ + Mg²⁺) − (HCO₃⁻ + SO₄²⁻ + CO₃²⁻ + NO₃⁻) has often been used to determine whether cation exchange occurs in groundwater [31,48,49]. KGW, FGW, and PGW are basically distributed along the −1:1 line (Figure 8b), indicating that ion exchange occurs in groundwater. Ion exchange is dominated by the exchange of Mg²⁺ and Ca²⁺ in groundwater with Na⁺ and K⁺ in rock or soil. In a small portion of groundwater, Na⁺ and K⁺ are exchanged with Mg²⁺ and Ca²⁺ in rocks or soils. Although the PGW samples are distributed along the −1:1 line, the CAI1 and CAI2 of most of the samples are close to zero and the ion exchange is weak.

5.5. Sources of Sulfate

Sources of sulfate in groundwater include natural and anthropogenic sources [50,51,52]. Natural sources include atmospheric deposition, dissolution of sulfate minerals (e.g., gypsum), sulfide oxidation, seawater intrusion, and volcanic eruption [53,54]. Anthropogenic sources include domestic wastewater, industrial wastewater, mining activities, and the application of chemical pesticides, etc. [14,55,56,57,58,59,60]. The sulfur and oxygen isotopes of sulfates from various sources in groundwater have significant differences. Combined with the hydrogeological conditions, the sources of sulfate in groundwater can be identified by the comprehensive use of water chemistry and multi-isotope analysis [61,62,63,64].

In Figure 9a, no water samples fall within the experimental zone of sulfate generation from sulfide oxidation [65], indicating that sulfate in groundwater, SW, and atmospheric precipitation does not originate from sulfide oxidation. This is also illustrated by the absence of samples within the sulfate oxidation zone in Figure 9d.

KGW in Zone I in Figure 9b has higher SO₄²⁻content and the heaviest δ³⁴S-SO₄²⁻ [66]. These KGWs are located in the deeply buried groundwater zone in the northern part of the Sangu Spring Basin and in the stagnant-flow zone near the spring boundary. Due to the large circulation depth of the KGW, poor groundwater runoff conditions, and the long water–rock interaction time, the sulfate mainly comes from the dissolution of gypsum. The KGW located in Zone I in Figure 9b falls in the gypsum dissolution zone in Figure 9d, which also indicates that the sulfate in the KGW comes from the dissolution of gypsum. According to geological data, gypsum deposits are widespread in the Middle Ordovician strata in the Sangu Spring Basin. Gypsum has been dissolved in areas with good water circulation conditions, but is still present in deeply buried areas. Gypsum was found in the borehole where the ZK05 water sample was collected, in the Ordovician strata in the form of fissure filling, and the δ³⁴S-SO₄²⁻ content of the gypsum was 27.5. Figure 9c also proves that the KGW in the deeply buried zone and stagnant-flow zone has the heaviest δ³⁴S-SO₄²⁻ and the oldest age.

Groundwater in Zone II in Figure 9b has lower SO₄²⁻content and heavier δ³⁴S-SO₄²⁻. The two FGW samples are the groundwater discharged from the extraction of coalbed methane. The groundwater is in a reducing environment, and the reducing effect of sulfur bacteria leads to an increase in δ³⁴S-SO₄²⁻ and a simultaneous decrease in SO₄²⁻ content in the groundwater.

The groundwater with the lowest SO₄²⁻ content and the lightest δ³⁴S-SO₄²⁻ is located in zone III of Figure 9b, and the atmospheric precipitation is also located in this zone. The sulfate in KGW and FGW in this area mainly comes from atmospheric precipitation.

The groundwater with the highest SO₄²⁻content and lighter δ³⁴S-SO₄²⁻ is located in zone IV of Figure 9b, and sample E141 from the Baishui River is also located in this zone. Due to the low SO₄²⁻ content in precipitation, atmospheric precipitation is unlikely to be the main source of SO₄²⁻ in these groundwater samples [67]. KGW in this area is affected by SW leakage, resulting in higher SO₄²⁻ content and lighter δ³⁴S-SO₄²⁻ in groundwater. Groundwater in Zone IV has a higher deuterium–oxygen isotope content and is in Group 3 with E141 in the δD-H₂O versus δ¹⁸O-H₂O relationship diagram.

Groundwater in the other areas in Figure 9b has higher SO₄²⁻ content and lighter δ³⁴S- SO₄²⁻; it is speculated that sulfate in groundwater comes from atmospheric precipitation, SW bodies, and gypsum dissolution. These samples in Figure 9d are also located between the atmospheric precipitation, SW, and gypsum end members, and are subject to the combined influence of atmospheric precipitation infiltration, SW body seepage, and gypsum dissolution.

In conclusion, the sulfate in the groundwater of the Sangu Spring Basin is mainly influenced by the dissolution of gypsum, the infiltration of atmospheric precipitation, and the seepage of SW bodies. Sulfate in the deeply buried and stagnant-flow areas of KGW is mainly derived from the dissolution of gypsum, and groundwater recharged by SW seepage tends to have a high sulfate content. FGW in reducing environments may have experienced the reducing effect of sulfur bacteria.

The hydrochemical components of groundwater in the Sangu Spring Basin are mainly affected by water–rock interaction and SW leakage. Further research on the water–rock interaction of groundwater and the hydraulic connection between surface water and groundwater is recommended. The protection and management of surface water quality should be strengthened, and already polluted surface water should be treated and monitored to avoid groundwater pollution caused by surface water seepage.

6. Conclusions

Using a combination of Piper diagrams, Gibbs diagrams, ion ratios, and hydrochemistry–isotope analyses is an effective means of determining the source of groundwater components and their influencing factors. The hydrochemical types of groundwater were mainly HCO₃-Ca, HCO₃·SO₄-Ca, HCO₃-Ca·Mg, HCO₃·SO₄-Ca·Mg, SO₄·HCO₃-Ca·Mg, SO₄·HCO₃-Ca (Na), and SO₄·HCO₃-Na·Ca. HCO₃⁻ and SO₄²⁻ were the dominant anions, and Ca²⁺, Mg²⁺ and Na⁺ were the dominant cations. All groundwater samples were distributed along the atmospheric precipitation line, suggesting an atmospheric precipitation origin for the groundwater. Groundwater in the stagnant zone and closed environment had the lightest δ¹⁸O-H₂O and δD-H₂O, indicating that the groundwater in this area was formed during a period of cooler climate. Groundwater with heavier δ¹⁸O-H₂O and δD-H₂O mainly receives recharge from SW.

The Gibbs diagram shows that the water–rock interaction is the main factor influencing the hydrochemical components of the groundwater, and the relationship diagrams for Mg²⁺/Na⁺ vs. Ca²⁺/Na⁺ and HCO₃⁻/Na⁺ vs. Ca²⁺/Na⁺ show that carbonate weathering and silicate weathering are the main sources of the chemical components of the groundwater. The ion ratio analysis also shows that the chemical components of groundwater mainly come from carbonate weathering and silicate weathering. The dissolution of minerals in groundwater is mainly controlled by carbolic acid, sulfuric acid, and nitric acid. Dissolution of salt rocks is not the main source of Na⁺ and K⁺ in groundwater in the Sangu Spring Basin, but dissolution of sodium feldspar (NaAlSi₃O₈) and potassium feldspar (KAlSi₃O₈) is the main source of Na⁺ and K⁺ in groundwater.

The ion exchange of groundwater is dominated by the exchange of Mg²⁺ and Ca²⁺ with Na⁺ and K⁺ in rocks or soil, and a small part of groundwater Na⁺ and K⁺ is exchanged with Mg²⁺ and Ca²⁺ in rocks or soil. Ion exchange mainly occurs in KGW and FGW, while ion exchange in PGW is weak.

Sulfate in the groundwater of the Sangu Spring Basin is affected by the combined effects of gypsum dissolution, infiltration of larger air precipitation, and leakage of SW. The sulfate in the KGW of the deeply buried and stagnant-flow zones is mainly derived from the dissolution of gypsum and has a higher sulfate content. Groundwater recharged by leakage from SW tends to have the highest sulfate content. The reducing effect of sulfur bacteria may have occurred in the FGW which is in a reducing environment.

In view of the environmental problems linked to groundwater in the Sangu Spring Basin, future work should strengthen the study of the hydraulic connection between KGW and SW, strengthen the protection of the quality of SW, and carry out monitoring of the hydrochemical anomaly zone of KGW, which is of great significance to the management of KGW resources in the Sangu Spring Basin.