Spatio–temporal patterns of river water quality based on entropy-weighted monitoring data: A case study of the Tanjiang River Basin, China

Reliable river water quality assessment at the basin scale requires the joint interpretation of long-term monitoring data across time, space, and multiple indicators. However, most routine assessment approaches still analyze these dimensions separately, which limits their ability to identify priority pollution periods and critical river sections for management.In this study, a spatio--temporal--indicator assessment framework was developed to support integrated interpretation of monitoring data. Monthly observations from 82 monitoring sections in the Tanjiang River Basin (South China) during 2019--2023 were organised into a three-dimensional data matrix and analysed using three-dimensional principal component analysis (3D-PCA) combined with entropy-weighted water quality indices (WQI-3D and WQI\((^\ast)\)).The results show that seasonal hydrological variability and nutrient-related anthropogenic pollution represent the two dominant drivers of basin-scale water quality. Ammonia nitrogen (NH\((_3)\)–N) and total phosphorus (TP) were identified as the most influential indicators, and pollution pressure was significantly higher during the flood season (June–October). Spatial analysis revealed persistent water-quality hotspots in urban-industrial tributaries, including the Longwan, Mazongsha, and Tiansha Rivers. Compared with the conventional WQI-3D, the enhanced index WQI\((^\ast)\) more clearly highlighted spatio-temporally coincident pollution extremes, improving the identification of critical monitoring sites and periods.The proposed framework provides a practical and data-driven tool for basin-scale water-quality monitoring and management, enabling more targeted pollution control and prioritisation within existing monitoring networks.