Risk prediction and driver tracking based on watershed resilience dynamics: A case study in Dahei River Basin, China

The watersheds face increasing pressure from both human activities and natural factors, which exacerbate potential risks and pose significant challenges to integrated watershed management. This study developed an effective methodology to evaluate watershed sustainability and predict potential risks based on watershed resilience dynamics by combining catastrophe theory, adaptive cycle theory, and the Copula-Bayesian Network. Taking the Dahei River Basin (DRB) in China as a case study, we systematically evaluated the resilience dynamics, diagnosed risks, and tracked key driving factors to propose management strategies. During 2013 to 2022, the resilience of the DRB generally improved from 0.52 to 0.88, primarily influenced by the water resources subsystem, with hilly areas demonstrating superior resilience in comparison to the agricultural and urban areas. It has exhibited two adaptive cycles, with 2015 and 2017 marking as exploitation phases, indicating periods of rising sustainability. Probability distributions of future resilience levels were quantified by using the Copula-Bayesian Network, enabling a comprehensive forward-looking risk assessment for the watershed. Projections indicated moderate sustainability of the holistic watershed, with 36.6 % being non- or low-resilience and only 15 % being high-resilience. These results suggest that the DRB is likely to face considerable risks associated with low-resilience in the future. Key drivers, including water reuse rate, ammonia nitrogen emissions, and environmental governance measures, were identified for different functional areas to improve watershed resilience and mitigate basin risks. Site-specific measures, such as improving agricultural irrigation and promoting urban water-saving technologies, tailored to local conditions, were proposed to promote sustainable development of the basin.