Stable Isotope and Hydrochemical Evolution of Groundwater in Mining Area of the Changzhi Basin, Northern China

The Changzhi Basin of China is an economically and ecologically important area with intensive human activities. To foster the sustainable development of groundwater resources and the economy, a total of 117 groundwater samples were collected in shallow and deep aquifers, including 91 2 H and 18 O isotope samples, to improved understanding of the natural geochemical processes and the impacts of anthropogenic activities on the groundwater chemistry. Synthetical application of the stable isotopes, Piper diagram, Gibbs diagram, ionic ratios and saturation indices to data analysis led to identication of hydrochemical zones for both aquifers from west to east of the basin. Isotopic analyses suggested that the groundwater recharge mainly comes from inltration of rain water, hydraulic interaction between surface water and shallow groundwater, and lateral recharge from ssure water at the edge of the basin. The predominant natural geochemical processes include mineral dissolution in conjunction with the cation exchange. The excess deuterium method revealed that mineral dissolution contributed 81%–98% to the salinity of shallow groundwater and 84%–98% to the salinity of deep groundwater. Anthropogenic activities are secondary contributions to the hydrochemical evolution with fertilizer application, human waste and sewage discharges causing an increase in NO 3 -N content and coal mining activities affecting the ion content of Na + , Cl - , SO 42- , and HCO 3- in the groundwater.


Page 3/13
A total of 117 groundwater samples were collected in April and November, 2018. The samples comprised 76 shallow groundwater samples, 36 deep groundwater samples, and 5 mine drainage samples. The sampling locations are shown in Fig. 1. Mine drainage samples were taken from the water drained from the mines to enable the exploitation of coal-bed methane and coal. Additionally, 24 rainwater samples were collected from central, eastern, and western areas of the basin from July, 2017 to July, 2018 for stable isotope tests; whereas, three of these samples were also used for hydrochemical tests. The samples were ltered through 0.45 µm membranes on site and then stored at 4°C. The bottles were rinsed twice with deionized water before sampling. For cation analysis, water samples were acidi ed using analytically pure nitric acid to pH < 2. Samples for stable isotope analysis (δ 18 O and δ 2 H) were collected in 50 mL glass bottles, which were sealed with airtight caps.

Measurement methods
Water temperature, pH, and electrical conductivity were directly measured on-site using a HANNA HI 991301 multi-parameter instrument. Major anions, cations and minor elements were analyzed by a Thermo Scienti c Dionex ICS-4000 (precision = ± 1%) and PerkinElmer Optima 8300 inductively coupled plasmaoptical emission spectrometer (precision = ± 1%) at the Groundwater Mineral Water and Environmental Monitoring Center in the Institute of Hydrogeology and Environmental Geology at the Chinese Academy of Geological Sciences. The analytical precision and electrical balance error of the hydrochemical data were within ± 5%.
Stable isotope ratios were expressed in δ (‰) notation and calculated with respect to Vienna Standard Mean Ocean Water (VSMOW). The δ 18 O and δD values in water samples were obtained using a Picarro L2130-i Analyzer at the Institute of Hydrogeology and Environmental Geology at the Chinese Academy of Geological Sciences. The analytical precision for δD was ± 1‰ and for δ 18 O was ± 0.1‰.

Hydrochemical characteristics
The physicochemical parameters of the water samples are shown in Table 1; all water sample data were plotted on Piper diagrams (Fig. 3a-

Rain water, surface water, and mine drainage
The pH values of rain water ranged from 6.39 to 6.63, indicating weakly acidic conditions. Surface water (pH 7.03-7.94) and mine drainage (pH 7.85-8.90) were generally neutral to weakly alkaline. The concentrations of the chemical components in rain water were, in general, low. Most of the chemical components in surface water and rain water exhibited a weak to medium variability, but a wide variability in mine drainage. The maximum concentrations of total dissolved solids (TDS), Na + , Cl − , and SO 4 2− for mine drainage were 2,901, 1,176, 1,078, and 1,218 mg·L − 1 , respectively, and were signi cantly greater than the maximum concentrations for surface water and rain water.

Groundwater
The pH values of the shallow groundwater ranged from 7.13 to 8.07, and had a mean value of 7.53, which indicates near neutral to weakly alkaline conditions.  (Fig. 3c). The Cv values of Cl − , SO 4 2− , and NO 3 − were greater than 1.0, indicating that these hydrochemical components had a wide variation in spatial distribution. The TDS of the deep groundwater gradually increased (from 240.2 to 2,160 mg·L − 1 ) from the west to the east of the basin. In the west and central of the basin, the groundwater is mainly of the HCO 3 -Ca type. However, in the east of the basin, as a result of low discharge and weak self-puri cation capacity, the SO 4 ·HCO 3 -Ca/Mg, HCO 3 ·SO 4 ·Cl-Ca, and HCO 3 ·Cl·SO 4 -Ca types were all identi ed for the deep groundwater, which had a high TDS of approximately 800-2,160 mg·L − 1 .
The concentrations of minor and trace elements, such as F, Cr, As, Fe, Mn, and Pb, were generally low, and most were not detected. In terms of the Groundwater

Groundwater
The δ 18 O and δ 2 H of groundwater are plotted in Fig. 4b. The δ 18 O and δ 2 H compositions of groundwater are located near the LMWL, which indicates that the groundwater is mainly recharged by rain water. The slope of the shallow groundwater line (k = 6.19) is smaller than that of the deep groundwater line (k = 6.41), which means that the groundwater has undergone evaporation during the recharge process, especially in the case of shallow groundwater.  (Gibbs, 1970), was carried out to identify the dominant processes affecting evolution. The Gibbs diagram depicts the relative dominance of precipitation, rock weathering, and evaporation in semi-arid and arid regions. The diagrams show the weight ratios of Na + /(Na + +Ca 2+ ) and Cl − /(Cl − +HCO 3 − ) against TDS, as shown in Fig. 5. Figure 5b shows that the ratios of Na + /(Na + +Ca 2+ ) are mostly less than 0.5 and that the TDS is mostly low to medium, which indicates that rock weathering is the dominant mechanism in the geochemical evolution of groundwater for both shallow and deep groundwater. The ratios of Na + /(Na + +Ca 2+ ), however, show As shown in Fig. 6a, the average total d-excess and TDS for shallow groundwater and deep groundwater are 4.63‰ and 744.02 mg·L − 1 , 4.72‰ and 510.86 mg·L − 1 , respectively. The contribution ratios of mineral dissolution in shallow groundwater and deep groundwater are 81-98% and 84-98%, while the contribution ratio of evaporation is 0.2-4.7% and 0-2.4%. In the study area, most groundwater depths are greater than 5 m, and the groundwater is mainly recharged by rain water and bedrock ssure water, which favors the dissolution of minerals. Most of the water samples indicate no evaporation effects regardless of shallow groundwater or deep groundwater. Figure 6b and 6c show that there is an almost exponential positive correlation between TDS and the contribution of mineral dissolution, and an almost exponential negative correlation between TDS and the contribution of evaporation, which indicates that mineral dissolution is the main contributor to the total salinity of groundwater.

Effects of geochemical processes on hydrochemistry
The relationship between (Ca 2+ + Mg 2+ ) and (HCO 3 − + SO 4 2− ) concentrations in groundwater samples is close to the carbonate and gypsum dissolution line   The slope and correlation coe cients of the equations for the shallow and deep groundwater (Fig. 8b) are − 1.88 (R 2 = 0.86) and − 0.34 (R 2 = 0.88), indicating that Ca 2+ , Mg 2+ , and Na + participate in ion exchange. However, signi cant differences were observed between the theoretical and actual values, implying that cation exchange is not the sole process affecting the concentration of the three ions. The other processes affecting the ion content include the discharge of mine drainage, and the interaction between groundwater and surface water. These processes affect ion content because of the high Na + content of mine drainage and surface water (Table 1).
To better understand the hydrogeochemical processes in the aquifers, PHREEQC (Parkhurst and Appelo, 1999) was used to calculate the saturation indices of the major minerals. The saturation indices of minerals varied between − 0.5 and + 0.5, which indicates that groundwater is saturated (or in equilibrium) or near saturation with respect to these minerals. As shown in Fig. 9a and b, most of the groundwater samples are in a state of saturation or over-saturation with respect to calcite and dolomite. Almost all the groundwater samples are in a state of under-saturation with respect to gypsum and are highly unsaturated in terms of halite ( Fig. 9c and d). Precipitation is the main source of groundwater in the study area. During the percolation of weakly acidic rain, carbonate minerals dissolve quickly and it is easy for groundwater to reach a dissolution equilibrium with calcite and dolomite. No signi cant correlation is observed between TDS and the SI values of calcite and dolomite. SI values of gypsum and halite, however, tend to increase with TDS, which indicates that the dissolution of gypsum and halite is one of the main processes involved in the increase in groundwater salinity.

Effects of anthropogenic activities on hydrochemistry
The Changzhi Basin has a long history of agricultural development, and the main crops grown are corn and wheat. Furthermore, the Changzhi Basin is located in southeastern Shanxi Province, near the location of the Jindong coal-based industries. Human activities associated with these socio-economic developments impose extensive impacts on the groundwater environment. These impacts stem from the use of fertilizers in the agricultural areas, water drainage during coal mining, and sewage discharge from urban areas.
Groundwater pollution caused by anthropogenic activities is a world-wide issue (Li et al., 2019). According to the monitoring results from 195 cities in China, 97% of urban groundwater has been polluted (Zhang, 2015). From Table 1 Mine drainage during coal mining is also a cause for concern because of the relatively high Na + , Cl − , SO 4 2− , HCO 3 − , and TDS concentrations. When compared with the combined average values for shallow groundwater, deep groundwater, and surface water, the mine water has the following characteristics: Mean Na + content of mine drainage was 7.69-23.87 times that of the combined average. The amount of mine drainage was 0.39 million m 3 ·d − 1 in the Changzhi mining area, and was directly discharged into surface water, which interacts with groundwater, especially shallow groundwater, thus leading to an increase in Na + , Cl − , SO 4 2− , and HCO 3 − content of groundwater.

Conclusions
In this study, the integrated approach consisting of Piper diagram, stable isotopes, Gibbs diagrams and ionic ratios provided an e cient way for analyzing the groundwater origin and hydrochemical processes that affected water chemistry. The Piper diagram and coe cient of variation were used to characterize the groundwater hydrochemistry and the stability of ions content; stable isotopes was a useful tool for analysis the origin and transformation of groundwater; Gibbs diagrams were used to establish the dominant effects of precipitation, rock weathering, or evaporation on geochemical evolution of groundwater, and deuterium excess was a capable way to quantify the contribution of evaporation to groundwater salinity; ionic ratios and saturation indices were used to depict the effects of mineral dissolution or precipitation on groundwater salinity. These methods were complementary of and verify each other. The main conclusions drawn are summarized as follows: The groundwater chemistry type of both shallow and deep groundwater demonstrates zonational characteristics from the west to east of the basin; the TDS content gradually increases (from 208.8 to 2,559 mg·L − 1 ) and the types of hydrochemistry tend to be complex. In both shallow and deep groundwater, the hydrochemistry types are mainly HCO 3 -Ca and HCO 3 ·SO 4 -Ca·Mg. The stable isotope compositions suggest that rain water is main recharge source for both shallow and deep groundwater. The interactions between shallow and deep groundwater, surface water and shallow groundwater and the lateral recharge from ssure groundwater at the edge of the basin have affected the isotopic composition of groundwater.
The hydrochemical and isotopic interpretation showed that the hydrochemical composition of the groundwater was controlled by geochemical processes. Gibbs diagrams suggested that water-rock interaction was the main mechanism controlling groundwater chemistry. The deuterium excess method revealed that that mineral dissolution accounts for 81-98% of the salinity of shallow groundwater and 84-98% of deep groundwater. The dissolution of gypsum and halite makes a signi cant contribution to the increase of groundwater salinity. Overall, rock weathering in conjunction with the cation exchange absolutely predominated in the geochemical evolution of groundwater.
The hydrochemical composition of groundwater in the study area is also affected by anthropogenic activities. NO 3 -N pollution occurs in the central area, the northeastern agricultural area, and the urban area in the east of the basin, and is more serious for shallow groundwater than for deep groundwater. The main sources of NO 3 -N are fertilizer application, human waste, and sewage. Mine drainage has relatively high concentrations of Na + , Cl − , SO 4 2− , HCO 3 − , and TDS when compared with shallow groundwater, deep groundwater, and surface water. Mine drainage is directly discharged into surface water and consequently interacts with groundwater, thus leading to an increase of Na + , Cl − , SO 4 2− , and HCO 3 − content of groundwater. The results of the present study provide a deeper insight into the water quality situation and geochemical evolution of groundwater, and will assist decision-makers to formulate sustainable groundwater management strategies for the study area.

Con icts of interest
The authors declare no con icts of interest.

Figure 2
Hydrogeological cross-section (along line A-A' in Fig. 1) of the study area   Plots of δ18O versus δ2H for all water bodies