GOALS/LASG模式对气候平均态的模拟

SIMULATION OF THE CLIMATE MEAN STATE IN THE GOALS/LASG MODEL

中文关键词:  全球海洋-大气-陆面耦合气候模式系统  气候平均态  模拟评估

英文关键词:Dynamical extended prediction  zonal flow  persistent prediction  climatic prediction  nonlinear regional prediction  correction

中文摘要:

      中国科学院大气物理研究所LASG最近发展了全球海洋大气陆面耦合气候模式系统(GOALS)的新版本,实现了全球大气环流谱模式(R42L9)与海洋环流模式(T63L30)在40°S～40°N之间的开洋面上海-气通量交换的完全耦合。该模式系统已积分了40a,基本上不存在明显的气候漂移。文中通过对所模拟的后30a平均的热带、副热带地区海温、海表风应力、洋面净通量和降水等的气候平均态与多种实测资料的对比分析,结果表明,GOALS模式基本上模拟再现了当今气候的一些主要特征,对热带气候平均态已具有一定的模拟能力,但也注意到,与观测相比,区域性差异是明显存在的,比如沿赤道西太平洋“暖池”区和靠近南美沿岸的东太平洋海域以及印度洋海表温度明显偏高约2℃,所模拟的赤道东太平洋海温冷舌西伸明显,造成赤道中太平洋海温明显偏冷等偏差。这些模拟误差,与模式中海表风应力和洋面所得到或释放的净热通量有密切的关系。SST的模拟误差反过来也影响到对降水的模拟。

英文摘要:

      Recently,the new version of the global ocean-atmosphere-land system(GOALS) has been developed at State Key Laboratory of Atmospheric Sciences and Geophysical Fluid Dynamics,Institute of Atmospheric Physics(LASG/IAP).It is a fully coupled model bet ween 40°S and 40°N latitudes where the ocean-atmosphere flux exchange has notany flux adjustment.A 40-year integration of the model shows that there is almost not climate drift.The climatic mean states of sea surface temperature(SST),wind stress,netenergy flux at surface, and precipitation in the tropic and subtropical Pacific,which is the mean averaged for the last 30-year simulations,were analyzed in comparison with observations.The results shown that the GOALS model reproduces some main characteristics of climate mean state in tropics.However,there still exist some large regional biases in the model if compared with observations.For examples,the simulated SST s in three regions of the equatorial western Pacific "warm pool",the eastern Pacific near the western coast of the South America,and the equatorial Indian Ocean are around 2℃ warmer than the observational data,and the cold "tong" in the equatorial eastern Pacific obviously spreads westward.The forming of those biases in the GOALS model is largely related to errors of wind stress and netenergy flux at the ocean surface.The SST bias,in turn,influences on the precipitation simulation.