资源环境科技发展态势分析平台

This study aims to evaluate the precipitation bias in the grey zone simulation (15km) using the Central Weather Bureau Global Forecast System (CWBGFS). We develop a new evaluation method using the object-based precipitation system (OPS) to examine the bias associated with the degree of convection organization. The 2016 South China Sea (SCS) Summer Monsoon onset is selected to evaluate the model's performance due to its sharp transition of large-scale circulation, which contributes to the complexity of precipitation pattern. The results based on OPS show that the observed precipitation tends to aggregate toward the central part of SCS during the post-onset period, while the precipitation in the model distributes more sparsely over the ocean. The observed precipitation intensity increases with the size of OPS especially for the extremes; however, the model underrepresents the relationship between the precipitation spectrum and the size of OPS. Moreover, the model simulates earlier diurnal peak time of precipitation over land in the organized systems than observation. The results also suggest that the convection scheme is insensitive to column moisture during the pre-onset period which seems to be one of the key factors to the excessive precipitation in the model. Using high horizontal resolution, however, does not improve the simulation of precipitation much in the model. The current study suggests that the precipitation bias related to aggregation of the convective systems should be regarded as an essential objective of model evaluation and improvement.