3.2. Hydrochemical type analysis
Based on the Piper trilinear diagrams, we studied the hydrochemical types of groundwater in the study area. The results show that the main hydrochemical type of groundwater in the study area is HCO
3•SO
4-Ca•Mg (
Figure 4), which is specifically manifested as: during the PUS, the hydrochemical types are mainly HCO
3•SO
4-Ca•Mg or HCO
3-Ca•Mg, accounting for 98.3%, with 1.70% of water samples being SO
4•HCO
3-Ca type. During the IUS, HCO
3•SO
4-Ca•Mg and HCO
3-Ca•Mg account for 86.0% of the total water samples, and there was a significant decrease compared to the previous stage, while 14.0% of water samples are SO
4•HCO
3-Ca•Mg or SO
4-Ca type. For the AUS, the proportion of HCO
3•SO
4-Ca•Mg or HCO
3-Ca•Mg decreases to 73.7%, SO
4•HCO
3-Ca•Mg accounts for 15.8%, and an additional 10.5% of Cl•HCO
3•SO
4-Ca•Mg type water appears. It can be seen that different patterns of hydrochemical types are presented at different levels of urbanization. This changing trend is mainly due to the fact that during the PSU, the hydrochemical types of groundwater in the study area are less affected by human activities and mainly controlled by the characteristics of the strata. However, in the AUS, with the socio-economic development accelerates, the water demand for industrial and agricultural was increased sharply, along with the shortage of surface water sources, thus leading to severe over-extraction of groundwater. In addition, the non-compliant discharge of domestic and industrial wastewater causes an increase in the concentration of major ions in the groundwater, and result in a change in hydrochemical types. It can be seen that the impact of human activities on the hydrochemistry of groundwater gradually increases with the rapid development of urbanization and industrialization, resulting in a diversified and complex trend of hydrochemical types.
3.4. Driving mechanism of water chemistry temporal evolution
Groundwater chemistry evolution is influenced by a combination of various factors. In this study, PCA was used to identify the driving factors for water chemistry evolution during the three urbanization stages in the study area. Ten water quality parameters (pH, K
+, Na
+, Ca
2+, Mg
2+, Cl
-, SO
42-, HCO
3-, NO
3-, TDS, and TH) were selected for PCA analysis. Before entering the data into the model, Kaiser-Meyer-Olkin and Barlett's sphericity tests were performed on the research data, with the results showing KMO values of 0.502, 0.454, and 0.621 for the three urbanization stages and Barlett's sphericity test values of 820, 1272, and 468 (
P<0.001), indicating that the water chemistry data met the requirements for PCA. Based on eigenvalues >1, two, two, and three main factors controlling water chemistry evolution in the PSU, ISU, and ASU were identified, explaining 74.4%, 69.9%, and 86.9% of all variables, respectively, representing all information from the ten water chemistry indicators (
Table 1).
In the PSU, PC1 explained 62.4% of the total variable information, and the indicators exhibiting strong positive correlations with PC1 were TDS, Ca2+, Mg2+, NO3-, TH, SO42-, and HCO3-, while Cl- had a moderate positive correlation with PC1. As proven in section 3.3.2, TDS, Ca2+, Mg2+, TH, HCO3-, and Cl- are mainly related to carbonate dissolution. Previous studies have shown that NO3- in the water environment mainly comes from chemical fertilizers, sewage, manure, soil nitrogen, etc. (Zhang and Wang, 2020), and at this stage, the amount of agricultural fertilization is relatively small, population density is low (230 people/km2), and the concentration of nitrate in groundwater is relatively low (13.7 mg/L), indicating that the impact of agricultural activities is not significant. But, the coal was main fuel for the industrial activities (mining activities, coking plants, and power plants, etc) in the phase. The large discharge of industrial wastewater will inevitably lead to an increase in SO42- concentration in groundwater. Therefore, SO42- of groundwater mainly originated from industrial activities. Thus, PC1 represents the control of groundwater chemistry by carbonate dissolution and industrial activities. PC2 explained 12.0% of the total variable information, with Na+ and K+ showing strong positive correlations with PC2 and pH displaying a moderate positive correlation. Based on previous analyses, Na+ in the study area mainly originates from halite dissolution and cation exchange. Groundwater pH is influenced by various factors and does not have directionality, and the pH during this period was relatively stable (7.20~8.15). Therefore, PC2 represents groundwater chemistry were controlled by halite dissolution and cation exchange.
During the ISU, PC1 explained 62.4% of the total variable information, and indicators exhibiting strong positive correlations with PC1 included TDS, SO42-, Ca2+, TH, while Cl- and K+ + Na+ showed moderate positive correlations with PC1. Among them, TDS, SO42-, Ca2+, and TH are mainly related to carbonate and gypsum dissolution. As proven in section 3.3.3, with the gradual implementation of government environmental protection policies, strict supervision of industrial non-compliant emissions has reduced the impact of industrial pollution on groundwater chemistry, and SO42- concentration (151 mg/L) is relatively stable. Thus, PC1 represents the groundwater chemistry were controlled by carbonate and gypsum dissolution. PC2 explained 21.5% of the total variable information, with Mg2+ and pH showing strong positive correlations with PC2 and HCO3- and NO3- displaying moderate positive correlations. As known from sections 3.3.2 and 3.3.3, this stage belongs to a period of rapid agricultural development, and increased groundwater extraction intensity and high fertilizer application (0.742 t/hm2), this not only affect water-rock interactions but also allow unutilized nitrogen fertilizers to infiltrate groundwater through irrigation return water and rainfall runoff. The combined action of these factors leads to an increase in Mg2+, HCO3-, and NO3- concentrations. Additionally, livestock breeding and domestic sewage also affect groundwater chemistry. Therefore, PC2 represents the groundwater chemistry were controlled by groundwater extraction intensity, agricultural fertilization, and domestic sewage.
In the ASU, PC1 explained 49.5% of the total variable information, and indicators showing a strong positive correlation with PC1 included Mg
2+, Cl
-, and TH, while Ca
2+ and K
++Na
+ had a moderate positive correlation with PC1. Mg
2+, TH, and Ca
2+ primarily originate from carbonate dissolution, while Cl
- and Na
+ mainly come from halite dissolution (as proven in section 3.3.2). Therefore, PC1 represents the groundwater chemistry were controlled by carbonate and halite dissolution. PC2 explained 24.8% of the total variable information, with indicators showing a strong positive correlation with PC2 including NO
3-, HCO
3-, and TDS, while pH had a moderate negative correlation.
Section 3.3.3 has proven that high concentrations of NO
3- primarily come from fertilizers and sewage during this stage. The negative correlation between HCO
3- and pH indicates decarbonation occurring in groundwater (Zhang et al.; 2011). Under intense human exploitation of water, pressure and temperature changes in groundwater occur, and increased extraction intensity also affects water-rock interactions, ultimately influencing TDS variations. Therefore, PC2 represents the groundwater chemistry were controlled by agricultural activities, sewage, and extraction intensity. The main factor PC3 explained 12.6% of the total variable information, and the indicator showing a strong positive correlation with PC3 was SO
42-. In this time period, the concentration of SO
42- was close to that of the previous period (180 mg/L and 190 mg/L, respectively), both originating from gypsum dissolution. Thus, PC3 represents the groundwater were controlled chemistry composition by gypsum dissolution.