Evaluation and comparison of CMIP6 and CMIP5 model performance in simulating the seasonal extreme precipitation in the Western North Pacific and East Asia

This study evaluates the performance of the Coupled Model Intercomparison Project Phases 6 models (CMIP6) in simulating the seasonal evolution and extreme precipitation indices in the western North Pacific and East Asia region (WNP-EA), and compare the results with those from CMIP Phases 5 (CMIP5). In the ensemble of CMIP6 models, the seasonal evolution simulation demonstrates improvements in seasonal northward migration of the rain band from spring to summer and more intense precipitation, resulting in a higher skill score in the CMIP6 ensemble than the CMIP5 one. In general, the skill scores for the spatial pattern of simple daily intensity (SDII), total rainfall occurrence (Totfq), and consecutive dry days (CDD) are higher than those of extreme precipitation intensity (R99p). The CMIP6 ensemble mostly gains higher skill scores for extreme precipitation indices in the wet season. However, the improvement is limited in extreme indices during spring and fall. The probability distributions for maximum 1-day precipitation (RX1day), maximum 5-day precipitation (RX5day), and R99p in the CMIP6 models demonstrate a more realistic shape and stronger intensity, indicating the improvement over the CMIP5 models. However, the biased distributions of the overestimated (underestimated) occurrence for lighter (heavier) SDII and shorter (longer) duration CDD cases remain as a problem in CMIP6 model simulations. The higher skill scores for the spatial pattern of SDII, Totfq, and CDD are likely due to the compensation between the biased distributions mentioned above, implying further improvements are needed for correcting the deficiency in simulating the precipitation occurrence.

Automated summary

This study evaluates the performance of CMIP6 models in simulating the seasonal evolution and extreme precipitation indices in the WNP-EA region, finding improvements in seasonal evolution but limited improvement in extreme indices during spring and fall.