Attribution Analysis of Annual Precipitation Simulation Differences and Its Correction of CMIP5 Climate Models on the Chinese Mainland

Using the self-organizing maps (SOM) method, we ranked and compared simulation results of atmospheric circulation precipitation for 32 global climate models (GCMs) in Coupled Model Intercomparison Project Phase 5 (CMIP5) over China, found that ranking GCM’s ability to simulate frequency sea level pressure (SLP) weather patterns (WPs) was not correlated with its annual WPs. Then, attributed differences identified three main components multi-model results: internal variability, differences, combined term two, variability being largest components. These deviations depend ultimately on two factors: WPs corresponding average daily generated by these WPs, second factor playing a decisive role. address drawback model ensemble cannot be effectively improved when each single makes up is dry or wet, solution proposed correct differences: nodal WP were corrected. After correction capability all individual greatly improved, which increases our confidence using CMIP5 future forecasting.