Water quality is a pressing issue affecting the sustainable development of lakes. To elucidate the spatial and temporal characteristics of water quality in Bosten Lake, China, this study constructed a comprehensive water quality index (CWQI) based on key water quality indicators, utilizing water quality data collected from 17 sampling sites spaning from 2011 to 2019. Key water quality indicators were determined using factor analysis, and the spatial and temporal characteristics of key water quality indicators and the CWQI were examined using multivariate statistical analysis. The key water quality indicators included pH, chemical oxygen demand (COD), water transparency (SD), \({\rm NO}_{{3}^-}\) , total dissolved solids (TDS), Cl−, SO42−, and electrical conductivity (EC). Furthermore, the contribution rates of all water quality indicators to the water quality were quantitatively elucidated using the SHapley Additive exPlanations (SHAP) values, thereby validating the factor analysis outcomes. Among the eight key water quality indicators, the COD had the most significant influence on the water quality of Bosten Lake. The water quality condition of Bosten Lake has remained at Class III from 2011 to 2019 (CWQI ranging from 3.19 to 3.90). The water quality of Bosten Lake was characterized by distinct regional differences that arose from hydrodynamic processes within the lake and upstream water quality. The southwestern region exhibited the best water quality (mean CWQI of 3.47), whereas the northwestern region exhibited the worst (mean CWQI of 3.58). It is crucial to acknowledge that alongside the increase in industrial and agricultural effluent discharge monitoring, a series of ecological restoration projects for the lake basin have been initiated. Over time, the water quality of Bosten Lake showed gradual improvement (improvement rate of CWQI at 0.05/a). This study provides a critical scientific basis for enhancing the understanding and effective management of water quality in the Bosten Lake Basin through a comprehensive analysis of its spatial and temporal evolution and driving mechanisms.
