基于Copula函数分析了由思贤滘连通的广东西江水文站-马口和北江水文站-三水构成的两个样本的洪水联合概率分布特征,获得如下结论:经过优选的马口、三水洪水边缘分布可分别由P-III型和GEV表示;拟合优度检验指标表明二者的最优连接函数均为Archimedean copula类的Gumbel-Hougaard Copula;重现期介于10~500 a之间的马口、三水洪水边缘分布与联合分布的洪水设计值相对差值大约介于0.6%~1.5%之间;基于条件概率计算的两站相同设计频率洪水的遭遇概率都大于88%。
The characteristics of joint probability distribution of flood were analyzed between the Makou Station(MS) in Xijiang River and the Sanshui Station(SS) in Beijiang River which are connected by the Sixianjiao Channal based on Copula function.Some conclusions were abtained as following:(1) Optimized marginal distributions of flood of Makou and Sanshui can be represended by the Pearson pattern Ⅲand GEV(generalized extreme value) distribution,respectively;(2) Gumbel-Hougaard Copula was the optimal copula selected by the results of goodness-of-fit test;(3) The relative differences of the special frequency design values between the marginal distribution of Makou and Sanshui floods and the joint distribution were from 0.6% to 1.5% for the return periods between 10 years and 500 years;(4) The probabilities of Makou and Sanshui meeting floods with each other were higher than 89% for the same design frequencies of two stations based on the conditional probability.