汉江上游气象-水文干旱特征变量响应概率研究

Study on meteorological-hydrological drought characteristic variable response probability in upper reaches of the Hanjiang River Basin

为揭示汉江上游气象水文干旱特征变量响应关系,分别选用标准化降水指数、标准化径流指数表征气象干旱和水文干旱,采用线性、非线性函数模型构建气象水文干旱特征变量响应模型,进而构建了基于Copula函数耦合贝叶斯网络概率模型的气象水文干旱特征变量响应概率曲线。结果表明:气象水文干旱历时和烈度的最优响应模型分别为线性函数模型和基于Gumbel Copula函数的非线性模型;干旱历时和烈度的最优联合分布函数分别为Frank Copula函数和Gumbel Copula函数;水文干旱历时、烈度的响应概率随对应的气象干旱特征变量的增加而增加,水文干旱历时响应概率变化速率逐渐减小,最后趋于平缓,水文干旱烈度响应概率变化速率较小。

In order to reveal the response relationship between meteorological drought and hydrological drought characteristic variables in upper reaches of the Hanjiang River Basin,the standardized precipitation index and standardized runoff index were chosen to reflect the meteorological drought and hydrological drought,respectively,and the linear and nonlinear function models were used to construct the meteorological-hydrological drought characteristic variable response probabilistic model.Then,meteorological-hydrological drought characteristic variable probability curves were established based on the Copula function coupled with the Bayesian network probabilistic model.The results show that the optimal response models of meteorological-hydrological drought duration and intensity are the linear function model and the Gumbel Copula function-based nonlinear model,respectively;the optimal joint distribution functions of drought duration and intensity are the Frank Copula function and Gumbel Copula function,respectively;the response probability of hydrological drought duration and intensity increases with the meteorological drought characteristic variables,and the rate of change of response probability of hydrological drought decreases gradually,and finally tends to level off,and the rate of change of response probability of hydrological drought intensity is low.