13:45 〜 15:15
[ACG30-P10] A seasonal reproducibility and predictability of the boundary current synchronization in JMA/MRI-CPS3
キーワード:季節予測可能性、季節アンサンブル予報システム、境界流同期
A previous study showed that the Kuroshio Current and the Gulf Stream are tended to be synchronized on decadal time scales, and this synchronization was referred to as boundary current synchronization (BCS) by Kohyama et al. (2021). Their study suggested that the BCS is associated with extreme weather such as heat waves in the mid-latitude Northern Hemisphere, including Japan. In order to understand the predictability of seasonal forecasts, we investigate the seasonal reproducibility and predictability of the BCS by using hindcast data of JMA/MRI-CPS3 (CPS3), which is an operational seasonal prediction system of the Japan Meteorological Agency. It is found that the CPS3 can reproduce and predict the BCS more accurately than the previous version of the seasonal prediction system (CPS2). A correlation analysis of the BCS index shows that CPS3 has a high correlation even in the latter of the forecast period: the correlations exceed 0.5 up to 8 months forecast in all initial months. Furthermore more than half of forecast periods have higher correlation in CPS3 than that in CPS2. This means that CPS3 has high prediction accuracy. We also conduct a SVD analysis of the SST anomalies in the two current regions and a regression analysis for the surface temperature and zonal wind anomaly. These show that the BCS reproducibility in CPS3 is higher than that in CPS2. These results suggest that the high-resolution model such as CPS3 has an ability to reproduce and predict the BCS compared with low-resolution model such as CPS2, which are consistent with the result of Kohyama et al. (2021).