Impacts of watershed and meteorological characteristics on stream water quality resilience

In response to pollutant inflows caused by anthropogenic and meteorological influences within watersheds, effective hydrological management of stream ecosystems and water quality is crucial. This study employs a random forest regression model to explore the intricate, non-linear relationships between watershed characteristics and meteorological variability, and their impacts on stream water quality resilience. We used biological oxygen demand (BOD), total nitrogen (TN), and total phosphorus (TP) data from 270 water quality measuring sites in the Han River watershed in Korea from 2017 to 2019 to derive stream water quality resilience. The analysis identifies critical watershed attributes-including urbanization, water temperature, slope gradient, elevation, vegetative cover, soil drainage, evapotranspiration rates, and precipitation patterns-that influence stream water quality resilience. Our findings revealed that variables determining the hydrological cycle and the type of pollutant flowing into the stream exert a significant effect on resilience. Urban expansion and rising water temperatures were found to decrease the rate of recovery in water quality by increasing the influx of pollutants into streams or inhibiting the decomposition of pollutants within the water body. Conversely, steep slopes, high elevation, forest, and well-drained area expansion, increased evapotranspiration, precipitation, and the 6-month standardized precipitation index increased the rate of recovery in water quality. Furthermore, we examined the impact of regional differences on resilience increase by comparing the spatial distribution of predicted resilience values and resilience at the local level. Our findings indicate the need for targeted management strategies that leverage specific watershed and climatic characteristics to optimize stream water quality resilience and emphasize a spatially nuanced approach to hydrological management.