返回
Causes for Simulation Biases of ENSO Periods in Coupled Models
陈林
南京信息工程大学
To investigate the root causes leading to simulation biases of ENSO period in coupled models, four interesting coupled models are used to analyze their ENSO periods. Specifically, these four models are constructed by cross-coupling the atmospheric components and oceanic components of two commonly used coupled models. By comparing the ENSO period in the observation with their performances in model simulations, the four parent models are separated into two groups, rightly associated with their atmospheric components: Grid-point Atmospheric Model of IAP LASG Version 2 (GAMIL2)-based models whose ENSO periods are shorter than observation, and Community Atmosphere Model Version 4.0 (CAM4) -based models whose ENSO amplitudes are longer. Wyrtki index analysis on ENSO period shows that meridional structure of zonal wind stress is the most critical factor to determine the period simulation among these coupled models. The longer ENSO periods in CAM4-based models are attributed to broader meridional width of zonal wind stress, while short period in GAMIL2-based models are caused by the narrower meridional width. The meridional structures of anomalous zonal wind stress are mainly affected by the anomalous precipitation responses to ENSO, and the biases of climate mean SST simulation are the root causes which modulate the structures of precipitation anomalies.