Abstract:

Based on the observed runoff data of the Danjiangkou reservoir from 1969 to 2008 and the multicoupledmodels results of the 4th IPCC assessment report in the uppermiddle reaches of the Hanjiang River,1 day maximum flood volume series and 3 days maximum flood volume series of the upper valley of the Han River were respectively simulated by the generalized extreme value distribution model (GEV) and generalized Pareto distribution model (GPD).The optimal probability model which described the distribution of the 1 day maximum flood volume and 3 days maximum flood volume was selected,and then the corresponding design value of the return periods was calculated,finally the extreme runoff response to the climate change was analyzed.The results displayed that the 1 day maximum flood volume and 3 days maximum flood volume series conformed to the generalized extreme value distribution,and the generalized Pareto distribution,the corresponding design value of the generalized extreme value model were slightly larger than that of the generalized Pareto model under the same return periods.The simulated results of the generalized Pareto model were superior than that of the generalized extreme value model for the 1 day maximum flood volume and 3 days maximum flood volume series under the different scenarios of the future climate.The research indicates that the climate change has a direct impact on the flood extreme value