Climate change affects the frequency and intensity of extreme hydrological events, e.g. extreme storm in a very direct way.Studying these effects caused by climate change will provide great support for disaster mitigation and engineering design. This paper applied the LARS-WG weather generator to simulate synthetic weather data under the A1B, A2 and B1 emission scenarios from Intergovernmental Panel on Climate Change (IPCC) using the results of General Circulation Model HadCM3. Based on the L-moments approach, design storms of different return periods were calculated for the Qiantangjiang River Basin in 2055s using the P-III distribution function. Thefinal results show that the LARS-WG weather generator has good capability in simulating synthetic weather data in the Qiantang RiverBasin and the maximum bias between observed and simulated design storms of 200a return period is 8.52%, occurring at the ShengxianStation. The design storms of different return periods tend to increase under the A1B and B1 scenarios, while decline under the A2 scenarios at the most stations. And design storm of 100a return period at the Hangzhou Station will be 209.14mm under the A1B scenarios, 11percent larger than that in the baseline period. The biggest increase and decrease of the design storm of 100a return period is 39.54% and42.79%, occurring at the Jinhua Station under the B1 scenarios and Tianmushan station under the A2 scenarios respectively.