Abstract: To study the runoff response of the upper Hanjiang River basin under future climate change,a downscaling model based on the RBF (Radial Basis Function) neural network and the SWAT (Soil and Water Assessment Tool) hydrological model are coupled. The RCP8.5 and RCP2.6 scenario data under the CanESM2 model from 2020 to 2099 are downscaled to each station in the basin to generate future climate elements(temperature and precipitation) and simulate the runoff response of the basin under future climate change. Results unveil that the runoff in the upper Hanjiang River will increase slightly in the future,with a slightly smaller increase in runoff in the RCP8.5 scenario than that in the RCP2.6 scenario. The annual runoff distribution in both scenarios is roughly the same with that in the base period. In both scenarios,runoff in flood season reduces slightly possibly because of a smaller maximum precipitation generated by the downscaling model. The research results provide a scientific basis for the comprehensive management in the Hanjiang River Basin.

Key words: climatic change, runoff change, SWAT(Soil and Water Assessment Tool) , RBF(Radial Basis Function), statistical downscaling model, upper Hanjiang River basin