Uncertainty analysis of impact of climate change on hydrology and water resources

A procedure for the uncertainty analysis of climate change impact assessment is proposed and analysed by application to two river basins in China. A monthly water balance model was chosen to simulate soil moisture and runoff. Monte Carlo simulation was used to generate different parameter sets, and probability density functions of runoff series were estimated by a nonparametric method. The results of the case study indicate that runoff is more sensitive to variation in precipitation than to increase in temperature; the smaller the runoff coefficient, the larger the uncertainty. At 5% significance level and the most likely climatic scenario (temperature increase of 1 degrees C and rainfall increase by 10%), the future peak flood discharge at Huayuan and Nantang basins in the south of China may increase by 47.41% and 38.16% respectively, which may seriously affect flood protection works and water resources systems.