变化环境下雅砻江流域水文极值演变及不确定性分析

为了探讨环境变化前后雅砻江流域的水文极值演变特征及不确定性,以雅砻江流域雅江水文站和洼里水文站为代表站,基于年极大值和超门限阈值样本,结合序列变异理论、多种极值统计模型(GEV、GPD、P-III)和轮廓似然函数参数估计及不确定性分析方法,分析了变化环境下雅砻江水文极值序列演变特征。结果表明:受气候变化和人类活动的影响,雅江水文站和洼里水文站环境变化前后的最优分布线型由P-III型分布转变为GEV分布。环境变化后的水文极值序列相应设计流量增大,重现期变小。设计值估算中,重现期越长,相应设计流量的不确定性也越大,其中参数的不确定性对于设计值的影响最为显著,但极值选样及分布函数的不确定性也不容忽视。轮廓似然函数法的使用可有效降低设计值不确定性,基于极值理论与轮廓似然函数估计法进行变化环境下水文极值演变分析可提高设计值的可靠性。研究成果可为雅砻江流域水库设计管理及防洪规划等提供参考。

Analysis on evolution and uncertainty of hydrological extreme value of Yalongjiang River Basin under changing environment

In order to discuss the evolution characteristics and uncertainty of the hydrological extreme value of Yalongjiang River Basin before and after the change of environment，Yajiang Hydrological Station and Wali Hydrological Station are taken as the representative stations，and then the evolution characteristics of the hydrological extreme value sequence of Yalongjiang River Basin under changing environment is analyzed herein on the basis of the samples of the annual maximum value and the superthreshold value therein in combination with the sequence variation theory，several extreme value statistical models ( GEV，GPD，P － III) and the parameter estimation of profile likelihood function as well as the relevant uncertainty analysis methods. The result shows that the optimal distribution linetypes of both Yajiang Hydrological Station and Wali Hydrological Station are transformedfrom P － III distribution to GEV distribution before and after the change of environment. The corresponding design flow of the hydrological extreme value sequence increases and the recurrence period shortens. In the estimation of the design value， the longer the recurrence period is， the greater the uncertainty of the corresponding design flow is to be; in which the influence from the parameter uncertainty on the design value is most significant，however， the uncertainties of the selection of extreme value sample and the distribution function cannot be ignored as well. The use of profile likelihood function method can effectively reduce the uncertainty of design value，while the extreme value theory and profile likelihood function estimation method-based analysis on the hydrological extreme value under changing environment can enhance the reliability of the design value. The study result can provide a reference for the reservoir design management and flood control planning within Yalongjiang River Basin.