资源环境科技发展态势分析平台

Forced by two Coupled Model Intercomparison Project phase 5 (CMIP5) global models, EC-EARTH and IPSL-CM5A, both regional climate models (RCMs) of Regional Climate Model system version 4 (RegCM4) and Weather Research and Forecasting (WRF) are used to perform regional climate simulations over China for 1980-2000. In general, the simulations in Institute Pierre Simon Laplace-Climate Model version 5A (IPSL-CM5A) are improved by the downscaling of many aspects of both precipitation and surface air temperature, with improvements including the significant enhancement of the resolution and presentation of regional processes in the RCMs. However, the improvement over EC-EARTH is limited. In addition, the added values also depend on the seasons, geophysical locations over the country and the variables presented. For the mean climates, both RCMs are able to exhibit better annual and winter precipitation measurements when downscaling IPSL-CM5A over most areas. The superiority of the RCMs over IPSL-CM5A for the surface air temperature is only obtained in summer. Seasonal cycles of both precipitation and surface air temperature are well simulated by all the models. The RCMs succeed in improving the seasonal cycles of precipitation over both GCMs in most sub-regions, but the seasonal cycles of temperature are only improved over IPSL-CM5A. In addition, the downscaling demonstrates more consistent probability distribution functions (PDFs) of precipitation with those of the observations in most sub-regions. For extreme precipitation, the RCMs provide outstanding convective wet days (CWD) with reduced biases and enhanced presentations of spatial patterns. For the other three extreme precipitation indices, the improvement of the RCMs is relatively weak, with superior over the IPSL-CM5A in several areas. The extreme temperature indices are better modelled by the RCMs, except for the minimum values of the daily minimum temperatures (TNn), with greater added values over IPSL-CM5A than those over EC-EARTH. Comparing two RCMs, WRF shows advantages in the downscaling precipitation, while RegCM4 works better for the surface air temperature.