Compound flooding potential in the Yangtze Estuary based on Bayesian inference

In the context of escalating global warming, the compound effects of typhoons, intense rainfall, riverine flooding, and storm surges have led to a heightened frequency of compound flood events, posing considerable threats to the flood control security of coastal cities. Using observational data and numerical simulations, a comprehensive dataset encompassing compound flood hazard factors from 1959 to 2020 has been reconstructed. A Bayesian inference-based Copula approach is employed to evaluate the risks and uncertainties associated with double and triple compound events of storm surge in the Yangtze Estuary. The results indicate that, from 1988 to 2020, the annual extreme high water level at the Xuliujing station in the Yangtze Estuary exhibited a significant downward trend, primarily due to the weakening intensity of storm surges. The coincidence risk between extreme storm tide levels and heavy rainfall in the Yangtze Estuary is relatively high, with a return period of 89 years. Conversely, the Datong runoff is statistically independent of typhoons and their associated events. Neglecting the dependency between variables would lead to an underestimation of compound flood disaster risks, which will become more pronounced as the return period of hazard factors increases. For instance, the risk underestimation of double compound events of storm surge reaches up to 77.7%. The uncertainty in the return period of double compound events of storm surge due to natural variability ranges from − 12.6% to 16.3%. Compared to extreme storm tide levels and dependence, precipitation is the major contributor to the uncertainty of compound flooding.