Multimodel Coupled Ensemble Seasonal Predictions of the East Asian-Western North Pacific Summer Mons

• 作 者: 杨德剑 杨修群 张耀存
• 机 构: 南京大学, 南京 210023
• 出 处: 第一届中国大地测量和地球物理学学术大会
• 发表日期: 2014年10期
• 关键词: summer monsoon multi-model ensemble coupled model seasonal forecast skill

　　In this study,the seasonal forecast skill of the East Asian-Western North Pacific summer monsoon (EAWNPSM) was evaluated using state-of-the-art dynamic multi-model ensembles (MMEs) including the Ensemble-Based Predictions of Climate Changes and their Impacts (ENSEMBLES) ensemble,the Development of a European Multimodel Ensemble System for Seasonal-to-Interannual Prediction (DEMETER) ensemble,and the Canadian Historical Forecast Project (HFP2) ensemble.The analysis is focused on comparing the skill between MME and the contributing single model ensembles (SMEs) and on comparing the skill between coupled MME and uncoupled MME.It was found that coupled models are much more skillful than stand-alone atmospheric models forced by persisted sea surface temperature (SST) anomaly in predicting the EAWNPSM at the seasonal time scale.This advantage is mainly attributed to their better abilities in capturing the evolution of the underlying seasonal SST anomaly.In general,MME strategy is quite effective in improving the EAWNPSM prediction skill of SMEs.While MME only shows some improvement in correlation skill,it beats overwhelmingly the contributing SMEs in terms of Brier skill score (BSS),a probabilistic skill measure.In this study,the seasonal forecast skill of the East Asian-Western North Pacific summer monsoon (EAWNPSM) was evaluated using state-of-the-art dynamic multi-model ensembles (MMEs) including the Ensemble-Based Predictions of Climate Changes and their Impacts (ENSEMBLES) ensemble,the Development of a European Multimodel Ensemble System for Seasonal-to-Interannual Prediction (DEMETER) ensemble,and the Canadian Historical Forecast Project (HFP2) ensemble.The analysis is focused on comparing the skill between MME and the contributing single model ensembles (SMEs) and on comparing the skill between coupled MME and uncoupled MME.It was found that coupled models are much more skillful than stand-alone atmospheric models forced by persisted sea surface temperature (SST) anomaly in predicting the EAWNPSM at the seasonal time scale.This advantage is mainly attributed to their better abilities in capturing the evolution of the underlying seasonal SST anomaly.In general,MME strategy is quite effective in improving the EAWNPSM prediction skill of SMEs.While MME only shows some improvement in correlation skill,it beats overwhelmingly the contributing SMEs in terms of Brier skill score (BSS),a probabilistic skill measure.