Impacts of Initialization Schemes of Oceanic States on the Predictive Skills of the IAP Near-Term Climate Prediction System

Based on the near-term climate prediction system of the Institute of Atmospheric Physics (hereafter IAP-DecPreS), we developed two distinct initialization schemes for the Coupled Global Climate Models (CGCM), FGOALS-s2. The first scheme used the Incremental Analysis Update (IAU) to assimilate gridded oceanic temperature and salinity data derived from the EN3 dataset. The second scheme used the merge of the ensemble optimal interpolation (EnOI) and IAU scheme (hereafter EnOI-IAU) to assimilate raw observational oceanic temperature and salinity profiles. The predictive skills of the decadal prediction experiments based on the two schemes were compared. Several metrics including temporal correlation and root mean square skills score indicate that the experiment based on the EnOI-IAU shows significantly higher predictive skills in the Sea Surface Temperature (SST) anomalies in the North Pacific associated with the Pacific Decadal Oscillation (PDO), than the experiment based on the IAU. In contrast, for the Atlantic Multi-Decadal Oscillation (AMO), the predictive skills of the experiment based on the EnOI-IAU are lower than that based on the IAU. The AMO has two activity centers, located in the subpolar and tropical North Atlantic. The skills of the experiment based on the EnOI are close to that based on the IAU in the tropical North Atlantic, while much lower than the latter in the extratropical region due to a false simulation of the warming trend in the region.