摘要In the context of climate change,coastal cit-ies are at increased risk of extreme precipitation and sea level rise,and their interaction will aggravate coastal floods.Understanding the potential change of compound floods is valuable for flood risk reduction.In this study,an integrated approach coupling the hydrological model and copula-based design of precipitation and storm tides was proposed to assess the compound flood risk in a coastal city—Hai-kou,China.The copula model,most-likely weight function,and varying parameter distribution were used to obtain the combined design values of precipitation and storm tides under the nonstationary scenario,which were applied to the boundary conditions of the 1D-2D hydrological model.Sub-sequently,the change of the bivariate return periods,design values,and compound flood risks of precipitation and storm tides were investigated.The results show that the bivariate return period of precipitation and storm tides was reduced by an average of 34％under the nonstationary scenario.The maximum inundation areas and volumes were increased by an average of 31.1％and 45.9％respectively in comparison with the stationary scenario.Furthermore,we identified that the compound effects of precipitation and storm tides would have a greater influence on the flood risk when the bivariate return period is more than 50 years,and the peak time lag had a significant influence on the compound flood risk.The proposed framework is effective in the evaluation and pre-diction of flood risk in coastal cities,and the results provide some guidance for urban disaster prevention and mitigation.