Research on the hydrochemical characteristics of groundwater and dynamic changes in water level depth in Jinchang City

Introduction Jinchang City, an arid inland city in Northwest China, is among the country’s 110 most water-stressed cities with a per capita water resource of 1,173 m 3 —far below provincial and national averages. Rapid agricultural land expansion and large-scale groundwater exploitation have led to continuous groundwater level decline, threatening regional ecology, production, and livelihoods. However, the impact of groundwater dynamics on hydrochemistry and the key drivers of water level decline in this region remain insufficiently elucidated. This study aims to address these gaps to provide insights for targeted groundwater management. Methods We utilized two core datasets: hydrochemical data from 13 national monitoring wells (collected in September 2023) and monthly groundwater level data spanning 2017–2023. Multiple analytical tools were integrated, including statistical analysis (SPSS 20.0), hydrochemical graphing (Origin 2021), spatial interpolation (Surfer), and an optimized random forest model (R software, ntree = 50, mtry = 2) with model performance metrics of R 2 = 0.83 and RMSE = 0.19 m. These methods were combined to analyze hydrochemical characteristics and identify drivers of groundwater level changes. Results (1) Groundwater in the study area is slightly alkaline, with a pH range of 7.50–7.90 (average 7.73). Total hardness (TH) varies from 229 to 1720 mg/L (average 569.62 mg/L) and total dissolved solids (TDS) from 334 to 2,420 mg/L (average 905.92 mg/L). Dominant cations follow the order Ca 2+ > Na + > Mg 2+ > K + , and dominant anions SO 4 2− > HCO 3 − > Cl − > NO 3 − , with Ca-SO 4 as the main hydrochemical type. (2) Hydrochemical evolution is jointly controlled by silicate weathering, carbonate dissolution, and evaporite dissolution: Ca 2+ and Mg 2+ derive from carbonates and silicates; Na + mainly comes from halite dissolution (with minor silicate contribution); SO 4 2− and Cl − primarily originate from carbonates supplemented by evaporites. (3) Groundwater level dropped by 2.26 m from 2017 to 2023, with an average annual decline of 0.38 m. Random forest analysis identified over-exploitation (importance index = 1.2) and reduced runoff (0.67) as the primary driving factors. (4) Level decline was associated with increased ion concentrations (with SO 4 2− and NO 3 − exceeding standards), exacerbated pollution accumulation, and destabilized hydrochemical types. Discussion These findings clarify the coupling relationship between groundwater dynamics and hydrochemistry in arid inland areas, filling the local research gap in systematic hydrogeochemical analysis. The identification of over-exploitation and reduced runoff as key drivers aligns with the region’s agricultural development and water resource management context. Practically, the results provide a scientific basis for groundwater sustainability in Jinchang City: they support targeted measures for managers (e.g., controlling over-exploitation in central agricultural areas, optimizing upstream water diversion, and promoting water-saving irrigation) and help raise community awareness of water scarcity for rational water use. Future research could expand monitoring density to refine driver analysis across smaller spatial scales.