海气耦合对MJO数值模拟的影响

采用一种基于降水异常追踪MJO（Madden–Julian Oscillation）东传的MJO识别方法（MJO Tracking）评估了参与MJOTF/GASS（MJO Task Force/Global Energy and Water Cycle Experiment Atmospheric System Study）全球模式比较计划的全海气耦合模式（CNRM-CM）、半海气耦合模式（CNRM-ACM）和大气模式（CNRM-AM）1991～2010年模拟MJO的能力，探究了海气耦合过程对模式模拟MJO能力的影响机理。CNRM-CM模式模拟的MJO结构更加接近观测，该模式不仅具有最高的MJO生成频率，也能够模拟较强的MJO强度以及较远的传播距离。海气耦合过程会造成CNRM-CM和CNRM-ACM模式中印度洋—太平洋暖池区域海温气候态的冷偏差。但是这种海温气候态的偏差基本没有改变模式模拟MJO的能力。CNRM-CM中MJO对流中心东（西）侧存在较强的季节内尺度海温暖（冷）异常，纬向梯度明显，而CNRM-ACM和CNRM-AM中不存在这样的海温东西不对称结构。结果表明在CNRM模式中海气耦合过程调控模式海温季节内尺度变率对模式MJO模拟能力的影响比调控模式海温气候态更加重要。

Impact of Air-Sea Coupling on MJO Simulation

The MJO (Madden-Julian Oscillation) simulation ability of the numerical experiments from CNRM GCMs (General Circulation Models) participated in the MJOTF/GASS (MJO Task Force/Global Energy and Water Cycle Experiment Atmospheric System Study) project was evaluated by tracking the eastward propagating positive equatorial precipitation anomalies. The GCMs included a fully coupled simulation (CNRM-CM), a half-coupled simulation (CNRM-ACM), and an uncoupled simulation (CNRM-AM) based on 1991-2010 data. The possible impacts of air-sea coupling on the MJO simulation were investigated. The CNRM-CM showed the highest ability in simulating the MJO characteristics, in terms of the occurrence frequency, amplitude, and propagation range. The climatological sea surface temperature (SST) of the CNRM-CM and CNRM-ACM showed a distinct cold bias over the Indo-Pacific warm pool region, compared with that of the CNRM-AM. The cold bias did not strongly impact the MJO simulation ability compared with the impacts on the MJO simulation abilities of the CNRM-ACM and CNRM-AM. The zonal gradient of the intraseasonal SST was significant in the CNRM-CM, with a strong positive intraseasonal SST anomaly to the east of the MJO convection center and a negative intraseasonal SST anomaly to the west of the MJO convection center when the MJO was over the Indian Ocean. In contrast, such a gradient was lost in the CNRM-ACM and CNRM-AM. The results indicate that the impact of the air-sea coupling on the MJO simulations by the CNRM GCMs was mainly through the influences on the intraseasonal SST variability.