【摘要】 沿海城市极易受到极端降水和风暴增水引发的复合洪涝灾害影响。定量分析极端降水和风暴增水之间的结构依赖，准确评估雨潮复合洪涝灾害风险，对沿海城市防洪除涝及应对措施的制定具有重要指导意义。本研究以上海市为例，利用研究区1979-2014(36a)年日累积降水与吴淞口风暴增水数据，通过K-S、AIC和BIC检验方法优化边缘分布函数，并采用Copula函数定量评估了不同联合重现期下降水和风暴增水组合的复合洪涝风险。研究表明：（1）上海市36a最大日降水量和吴淞口相应风暴增水均适合运用GEV分布进行拟合，Frank Copula函数对雨潮遭遇联合分布特征拟合效果最好；（2）上海市在5、10、20、50和100a雨潮联合重现期下，降水风暴增水遭遇的联合概率是同现概率的4.12、7.51、14.21、34.27、67.72倍；（3）100a联合重现期下的降水与风暴增水分别约为276 mm和3.5m，说明上海市沿海地区要预防百年一遇的强降水或风暴增水，须在考虑天文大潮的基础上，设计构筑至少3.5 m的防汛墙。此研究表明二元Copula函数能够较为准确地计算出不同联合重现期下的设计降水和风暴增水，为优化防汛工程的设计标准、设计科学复合洪涝灾害情景提供了可能。更多还原

【Abstract】 Coastal cities are extremely vulnerable to compound flood disasters caused by extreme rainfall and high storm surge. In order to accurately assess the risk of compound flood disasters in coastal cities,it is essential to estimate the dependence between extreme rainfall and storm surge. In this study,we use the maximum daily accumulated rainfall in Shanghai from 1979 to 2014(36 years)and the maximum storm surge at Wusongkou station on the day of the maximum precipitation and 3 days after the maximum rainfall. The marginal distribution function is optimized by KS,AIC and BIC testing methods. The function quantitatively evaluates the compound flood risk of the combination of rainfall and storm surge under different joint return period in Shanghai. The results show that the36-year maximum daily rainfall and the corresponding maximum storm surge at Wusongkou are both suitable for fitting with the GEV distribution;the Frank Copula function has the best fitting for the joint distribution of rainfall-storm surge encounter in Shanghai;the joint probability of rainfall and storm surge is 4.12,7.51,14.21,34.27,67.72 times of the co-occurrence probability under different joint return period;Under the 100-year joint return period,the rainfall and storm surge are 276 mm and 3.5 m,which shows that coastal regions should prevent strong occurrences in 100 years. Beside this,an additional design and construction of a flood wall of at least 3.5 m shall be made in consideration of the astronomical tide. This study presents that the Copula function can more accurately calculate the design rainfall and storm surge under different joint return period,which provides the possibility for optimizing the design standards of flood control projects and designing scientific compound flood disaster scenarios.更多还原