摘要
基于1991年-2008年香溪河流域降雨量和径流量资料,将最大熵-Copula(MEC)方法应用于流域降雨-径流相关性分析中。结果表明:MEC方法生成的边缘分布能够有效拟合降雨量和径流量的观测值,且在边缘分布确定的过程中无需对线型进行假设,以均值、标准差、偏度和峰度作为约束也能够很好的体现变量的统计特征;GumbelHougaard Copula函数对于降雨量和径流量联合分布的拟合效果较好;香溪河流域降雨量和径流量存在上尾相关性,这对防洪排涝有着重要的作用,同时也为流域工程规划设计工作提供了参考依据。
Based on the rainfall and runoff in Xiangxi River Watershed,this study applied a maximum entropy-copula（MEC）method to investigate the relation ship between the rainfall and runoff of watershed.Results show that：the generated marginal distributions of the rainfall and runoff can effectively fit the observed values without any hypothesis,and the statistical characteristics of random variables can be reflected well as the constraints of mean,standard deviation,skewness and kurtosis;Among all the Copula functions used,Gumbel hougaard copula showed the best effect of fitting generation the joint distribution of the rainfall and runoff;the upper tail dependence was observed between the rainfall and runoff in Xiangxi River Watershed,which not only can play agreat role in flood and drainage control in Xiangxi River Watershed,but also provide a reference basis for watershed project planning and design.
出处
《南水北调与水利科技》
CAS
CSCD
北大核心
2016年第1期84-89,共6页
South-to-North Water Transfers and Water Science & Technology
基金
福建省教育厅 A 类项目( JA15371)
厦门理工学院科学技术研究项目( YKJ3018R
XYK201419)
